Citation
Comparative responses of five sympatric species of mice to overwintering colonies of monarch butterflies in Mexico

Material Information

Title:
Comparative responses of five sympatric species of mice to overwintering colonies of monarch butterflies in Mexico
Creator:
Glendinning, John Ingersoll, 1959-
Publication Date:
Language:
English
Physical Description:
vii, 209 leaves : ill. ; 28 cm.

Subjects

Subjects / Keywords:
Dissertations, Academic -- Zoology -- UF
Mice -- Reproduction ( lcsh )
Mice -- Food ( lcsh )
Monarch butterfly -- Wintering ( lcsh )
Zoology thesis Ph. D
City of Gainesville ( local )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1989.
Bibliography:
Includes bibliographical references (leaves 194-208)
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by John Ingersoll Glendinning.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
020383018 ( ALEPH )
21576332 ( OCLC )
AHC0796 ( NOTIS )
AA00004780_00001 ( sobekcm )

Downloads

This item has the following downloads:


Full Text
COMPARATIVE RESPONSES OF FIVE SYMPATRIC SPECIES OF MICE
TO OVERWINTERING COLONIES OF MONARCH BUTTERFLIES IN MEXICO
By
JOHN INGERSOLL GLENDINNING
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1989


ACKNOWLEDGEMENT S
I thank the members of my PhD committee, Donald Dewsbury, John
Eisenberg, Lou Guillette and Frank Slansky, for their guidance and
critical evaluation of my ideas, and my major advisor, Lincoln Brower,
for struggling with me through the painful process of learning how to
write, for his insight, expertise and support, and for showing me how
to camp in style and safety in Mexico. I also thank Tonya Van Hook,
Jim Anderson, Alfonso Alonso, Alfredo Arellano, and Bill Calvert for
their camaraderie and help while camping on Sierra Chincua; Alfonso
Alonso for help collecting the data in Chapter 7; Michael Carleton of
the the National Museum of Natural History for identifying the species
of mice; Monarca A.C., Bernardo Villa-R., and Jorge Sobern for
logistical support in Mexico; Hilda Flores and Jose Luis Villaseor,
Alfonso Alonso, Alfredo Arellano and Ron Kelley for help identifying
the understory vegetation on Sierra Chincua; the Direccin General de
Flora y Fauna Silvestre of SEDUE for a scientific collecting permit in
Mexico; Steve Malcolm and Tonya Van Hook for patiently teaching me the
lipid analysis and spectrophotometric techniques; Mark Yang and Carlos
Martinez del Rio for statistical advice; Daryl Harrison for help with
the illustrations; Steve Malcolm, Carlos Martinez del Rio, Doug Levey,
Linda Fink, and Myron Zalucki for editorial help; Fred Morrison for
help collecting and rearing monarchs on Asclepias syriaca: Mark
Stelljes for the monocrotaline N-oxides; Mary Allen, Alfredo Arellano,



William Calvert, Carlos Galindo, Victor Snchez, Mark Stelljes, and
Jacqueline Roy for the unpublished data; and the Dept, of Zoology,
Sigma Xi Society, and Explorer's Club for financial assistance. Last
but not least, I thank Diana Schulmann for her editorial help,
unfailing support and love throughout the greater and lesser moments
of my doctoral program.


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
COMPARATIVE RESPONSES OF FIVE SYMPATRIC SPECIES OF MICE TO
OVERWINTERING COLONIES OF MONARCH BUTTERFLIES IN MEXICO
By
John Ingersoll Glendinning
August 1989
Chairperson: Lincoln P. Brower
Major Department: Zoology
I compared five species of sympatric mice (Peromyscus melanotis.
E-. aaLSCUS hylocetes, Eeithrodontomys sumichrasti. Microtus mexicanus
aalyua and Neotomodon alstoni alstoni) in terms of their ability to
take advantage of overwintering colonies of monarch butterflies
(Danaus plexippus) in Mexico. To eat this superabundant food
resource, the mice must overcome the monarch's chemical defense
system, which consists of bitter-tasting and potentially toxic
cardiac glycosides (CGs) and pyrrolizidine alkaloids (PAs).
Peromyscus melanotis appeared to be the only species of mouse that
breached these defenses: large numbers of £_*. melanotis immigrated to
the colonies, fed on the monarchs, and initiated winter reproduction.
In contrast, the other species of mice appeared to avoid the
colonies. On grids outside the colonies, individuals of all five
species were common but reproductively inactive throughout most of
iv


the winter. To explain these demographic results, I hypothesized
that E^. melanotis 1) was the only species tolerant to the
microhabitat features of the overwintering sites, 2) aggressively
excluded the other species, and 3) was the only species that
tolerated the monarch's defensive compounds. Because the understory
vegetation patterns were not significantly different from those
outside colonies, I rejected the first hypothesis. The second
hypothesis was rejected because resident E^. melanotis were unable to
dominate the other species in captive agonistic encounters. The
third hypothesis was accepted because only E^_ melanotis was able to
1) thrive on a diet of pure monarchs and 2) learn how to reject the
CG-laden cuticle and feed selectively on the low-CG internal tissues.
To determine whether S^. melanotis was uniquely able to overcome the
taste and toxicity of the monarch's defensive compounds, several
further feeding experiments were initiated to compare E^. melanotis
with E^_ hylocetes and E^. sumichrasti. None of the species was
sensitive to the toxic effects of digitoxin (a CG), which suggests
that all could have ingested the monarch's CGs with impunity.
However, E_*. hylocetes and sumichrasti were much more sensitive
than E_*. melanotis to the taste of digitoxin, suggesting that the
divergent demographic responses of the different mouse species can be
explained by their differential responses to the bitter taste of the
CG-laden monarch butterflies. Monocrotaline (a PA) did not seem to
have an influence on the feeding behavior of either species.


TABLE OF CONTENTS
Page
ACKNOWLEDGEMENTS ii
ABSTRACT vi
CHAPTERS
1 GENERAL INTRODUCTION 1
Overwintering Biology of Monarch Butterflies in Mexico 3
Known and Potential Predators of Overwintering Monarchs 4
Overview of Dissertation 6
2 DEFENSIVE COMPOUNDS IN THE MONARCH BUTTERFLY 7
Cardiac Glycosides 8
Chemistry and Toxicology 8
Defensive Value 11
Pyrrolizidine Alkaloids 12
Chemistry and Toxicology 12
Defensive Value 16
3 DEMOGRAPHIC RESPONSES TO COLONIES OF MONARCH BUTTERFLIES 19
Methods 20
Results 34
Discussion 66
4 COMPARATIVE FEEDING RESPONSES TO OVERWINTERING MONARCHS 73
Methods 73
Results 82
Discussion 100
5 CONSUMPTION AND AVOIDANCE OF MONARCHS: THE INFLUENCE OF
MONARCH ABUNDANCE AND CARDIAC GLYCOSIDE CONCENTRATION 105
Methods 105
Results 112
Discussion 120


6 COMPARATIVE TASTE AND TOXIC RESPONSES TO CARDIAC GLYCOSIDES
AND PYRROLIZIDINE ALKALOIDS 127
Methods 127
Results 135
Discussion 153
7 BEHAVIORAL AND ECOLOGICAL INTERACTIONS OF FORAGING
P MELANOTIS WITH OVERWINTERING MONARCHS 163
Methods 164
Results 169
Discussion 175
8 GENERAL CONCLUSIONS 182
Impact of Mouse Predation on Overwintering Monarch
Butterflies 182
Why is Ej. melanotis the only Mouse Species That Feeds on the
Monarchs? 183
Are Popluations of melanotis in the Overwintering Areas
Specifically Adapted for Overcoming the Monarch's Defensive
Compounds 185
Benefits and Costs of Being Tolerant to Bitter Foods 186
APPENDICES
A PAIRED COMPARISONS OF THE PRECENT SIMILARITY OF UNDERSTORY
VEGETATION ON GRIDS LOCATED INSIDE AND OUTSIDE THE MONARCH
COLONIES 190
B PERCENTAGE IMPORTANCE VALUES FOR THE SPECIES OF UNDERSTORY
VEGETATION ON THE EIGHT TRAPPING GRIDS LOCATED INSIDE OR
OUTSIDE A MONARCH COLONY ON SIERRA CHINCUA 191
REFERENCES 194
BIOGRAPHICAL SKETCH
209


CHAPTER 1
GENERAL INTRODUCTION
Arthropods possess a diverse array of chemical defenses that deter,
sicken, or even kill vertebrate predators (Eisner 1970, Duffey 1976,
Blum 1981, Pasteis et al. 1983). However, the actual effectiveness of
these defenses in naturally occurring predator-prey complexes is
largely unknown because most studies have been conducted in the
laboratory with unnatural predators (Duffey 1976, Pasteis et al.
1983). The few studies with natural predators have simply described 1)
how a single predator species responds in the laboratory to the
defenses of several prey species (Cyr 1972, Langley 1981, Redford 1984,
Whitman et al. 1985, Whitman et al. 1986), or 2) the extent to which
sympatric species of birds in the field utilize one or several
chemically defended butterfly species (Fink et al. 1983, Collins and
Watson 1983, Sargent 1987). A notable study by Pearson (1985) examined
how effectively lizards, birds, and robber flies could breach the
multiple defenses (chemical, physical, and behavioral) of adult tiger
beetles. However, to my knowledge, no study has compared
experimentally an entire assemblage of sympatric predator species of
the same taxon in terms of their ability to overcome the chemical
defenses of a single, shared prey species.
1


2
Mammals are thought to have evolved a variety of attributes for
dealing with potentially poisonous foods (Eisner 1970, Freeland and
Janzen 1974, Brower 1984) They include the ability to 1) learn from
their mother which foods are safe (Galef 1977, Vaughan and Czaplewski
1985), 2) detoxify certain compounds (e.g., Freeland and Janzen 1974,
Carlson and Breeze 1984), 3) feed selectively on nutritious tissues and
reject unpalatable and/or potentially toxic parts of a food item
(Benson and Borell 1931, Hamilton 1962, Kenagy 1972, Cyr 1972, Vaughan
and Czaplewski 1985, Brower et al. 1988, Roy and Bergeron unpuhl.
data), 4) avoid foods by developing a conditioned feeding aversion
(CFA) to them (Rzska 1953, Garcia and Koelling 1966, Brower 1969,
Kalat and Rozin 1973, Robbins 1978, Jacobs and Labows 1979, Daly et al.
1982, Brower and Fink 1985), and 5) avoid all bitter-tasting foods
(Bate-Smith 1972). This latter attribute is derived from the
observation that many mammals are averse to a variety of bitter
compounds at extremely low concentrations (Patton and Ruch 1944,
Carpenter 1956, Chrzanowski 1965, Pressman and Doolittle 1966, Ganchrow
1977, Jacobs et al. 1978). Virtually all poisons of biological origin
are bitter-tasting to humans (Richter 1950, Bate-Smith 1972, Garcia and
Hankins 1975, Brower 1984). Thus a general aversion to bitter foods
may be an effective way to avoid ingesting such compounds.
The relative importance of these five attributes to foraging wild
animals is unknown. This is because, in most cases, each attribute has
been studied in isolation from the others with only one species of
predator. In this dissertation, I will examine the relative importance
of feeding selectively, CFAs, and avoiding bitter-tasting foods in
determining the responses of 5 species of wild mice to the vast


3
overwintering colonies of monarch butterflies (Panana plex-i ppns L.)
that form each year in Mexico.
Overwintering Biology of Monarch Butterflies in Mexico
Each fall, virtually the entire eastern North American population
of monarch butterflies migrates to several, remote, high-altitude
overwintering sites in central Mexico (Brower and Calvert 1985, Calvert
and Brower 1986) There, within the oyamel fir forests (Abies
religiosa H.B.K.), they form colonies of tens of millions of
\
individuals. The colonies range in size from 0.3 to 3.3 ha, attain
densities of about 10 million per ha, cover the trees and understory
vegetation, and stay in approximately the same location from late
December through late February (Brower and Calvert 1985, Calvert and
Brower 1986) Although the colonies form on the same mountainsides
each year, they do not always form in the same location; colony sites
are frequently 1 to 2 kilometers from one another among years (Calvert
and Brower 1986, Figure 2-la, b; Glendinning pers. observ.).
Individual butterflies have extensive lipid reserves (mean in January =
35% of total dry weight; Brower 1985). They are vulnerable to
terrestrially foraging predators because low temperatures constrain the
butterflies' predator avoidance behaviors such as flying back to
communal roosts and crawling up vegetation (Calvert and Cohen 1983,
Masters et al. 1988, Alonso et al. in press). As such, these colonies
constitute a potential, superabundant, lipid-rich food bonanza for
mice.
However, overwintering monarchs contain cardiac glycosides (CGs)
and pyrrolizidine alkaloids (PAs), both of which are thought to protect


4
monarchs against most vertebrate predators. I describe the chemistry,
toxicology, and purported defensive value of these compounds in the
next chapter.
Known and Potential Predators of Overwintering Monarchs
Of the 37 species of omnivorous and insectivorous birds present in
the Sierra Chincua overwintering area (Arellano et al. in press) r only
black-headed grosbeaks and black-backed orioles (Calvert et al. 1979,
Fink and Brower 1981, Fink et al., 1983; Brower and Calvert, 1985,
Arellano et al. in press) feed extensively on the monarchs. Whereas
grosbeaks are relatively insensitive to the toxic effects of CGs,
orioles avoid ingesting large amounts of CGs through taste rejection of
the more toxic monarchs (Fink and Brower 1981) Mixed flocks of these
two species were estimated to kill 4,550 to 34,300 butterflies per day
in a 2.25 ha colony, or about 9% of the colony (Brower and Calvert
1985) There is some evidence that prolonged consumption of monarchs
caused one or both of these species to accumulate toxic levels of CGs
and possibly PAs (Brower and Calvert 1985, Arellano et al. in press)
Several species of mice may also feed extensively upon the
monarchs. Individual black-eared mice Peromyscus melanotis J.A. Allen
and Chapman, Neotomodon alstoni alstoni Merriam, and Microtus mexicanus
salvus Saussure that were captured near overwintering colonies in
Mexico have been found to eat monarchs to varying degrees while in
captivity (Brower et al. 1985). (The former two species were
erroneously referred to as Peromyscus maniculatus labecula Elliot and
P. spicilegus J.A. Allen, respectively, in Brower et al., 1985; see
Brower et al. 1988 for proper identification). Stomach content


5
analyses and short-term trapping studies suggest that female P.
melanotis inside monarch colonies feed upon monarchs and are larger and
more reproductively active than female conspecifics outside colonies
(Brower et al. 1985). Thus the evidence suggests that melanotis
actually benefits from feeding upon the monarchs.
There are several other animals known or suspected to feed upon
overwintering monarchs. Both cattle and domestic turkeys occasionally
eat the butterflies (Urquhart 1976, A. Arellano unpub1. data'. Calvert
et al. (1979) found scats presumed to be from the hog-nosed skunk
Conepatus mesoleucus nelsoni that contained monarch parts. I also
observed an individual hog-nosed skunk inside a monarch colony during
the night on Sierra Chincua during February 1986, but I could not
determine whether it was feeding on monarchs. Brower et al. (1985)
caught one Sorex saussurei near a monarch colony on Sierra Chincua but
were unable to determine whether it had fed upon monarchs. Between
1985 to 1989, I trapped 34 additional saussurei (5 inside and 29
outside monarch colonies) but did not analyze their stomach contents.
During my field work in the Sierra Chincua overwintering area
(Figure 3-la and b), I discovered several additional species of small
mammals. They included four muroid rodents (Peromyscus aztecus
hylocetes, Reithrodontomys sumichrasti, Reithrodontomys cf. megalotis.
Nelson! neotomodon goldmani). a squirrel (Sciurus cf. aureogaster). a
shrew (Cryptotis goldmani alticola), a weasel (Mustela frenata), and a
bat (Lasiurus cinereus). However, all species except the former 2 were
extremely rare; < 2 individuals of each were trapped and/or sighted.


6
Overview of Dissertation
The initial goal of this project was to compare the demographic
responses of Peromyscus melanotis J. A. Allen and Chapmen, E_*. aztecus
hylocetes Merriam, Reithrodontomys sumichrasti Saussure, Microtus
mexicanus salvus Saussure, and Neotomodon alstoni alstoni Merriam to
the overwintering monarchs. In Chapter 3, I describe the discovery
that all species but E^. melanotis largely avoided the monarch colonies.
Large numbers of E^. melanotis immigrated into the colonies, fed
extensively on monarchs, and initiated high levels of reproduction. To
explain this finding, I examine three questions: 1) are there unique
plant species compositions in the sites where monarch colonies form
that render these areas suitable only to E^. melanotis: 2) can P.
melanotis aggressively exclude the other species of mice from the
colonies; or 3) is melanotis better able to breach the monarch's
chemical defense system? I rule out the first two hypotheses in
Chapter 3 and devote Chapters 4 through 6 to testing the third
hypothesis by means of an integrated series of feeding experiments with
monarchs and artificial diets. In Chapter 7, I examine the potential
impact of predation by E^. melanotis on the live and dead butterflies
that accumulate on the ground and in low vegetation, and how this
predation is influenced by the accessibility and degree of desiccation
of the monarchs.


CHAPTER 2
DEFENSIVE COMPOUNDS IN OVERWINTERING MONARCH BUTTERFLIES
Adult monarchs contain cardiac glycosides (CGs) that were
sequestered and stored from the species of milkweed plant (Asclepia.q
spp.) that they fed upon as larvae (Brower and Moffit 1974, Roeske et
al. 1976, Malcolm and Brower 1989). Available evidence indicates that
85 to 92% of overwintering monarchs in Mexico fed as larvae on the
common milkweed A. syriaca L. (Seiber et al. 1986, Malcolm and Brower
1989). Whereas freshly-eclosed A. syriaca-reared individuals initially
have relatively high concentrations of CG (mean = 278 |lg/0.1 g), they
lose on average 86% of these CGs during the long migration to Mexico
(mean = 40 |ig/0.1 g) (Malcolm et al. 1989, Malcolm and Brower 1989) .
Approximately 80% of the CGs in A. syriaca-reared monarchs are located
in the cuticle and wings (Brower et al. 1988) .
Overwintering monarchs in Mexico also sequester and store
pyrrolizidine alkaloids (PAs) from PA containing plant families, such
as the Asteraceae (Compositae), Boraginaceae, Apocynaceae, and Fabaceae
(Leguminosae) (Edgar et al. 1976, Brower 1984, Mattocks 1986, Kelley et
al. 1987, M. Stelljes unpubl. data). The alkaloid appears to be
acquired by imbibing nectar both en route to the overwintering sites
(Kelley et al. 1987) and while overwintering (Glendinning and Kelley
unpubl. data). Preliminary evidence indicates that PAs, like CGs, are
concentrated in the cuticular material (M. Stelljes unpubl. data)
7


8
Cardiac Glycosides
Chemistry and Toxicology
All CGs have as their basic structural component C-23 steroid
glycosides (Figure 2-1). The use of the adjective cardiac stems from
the ability of these compounds to increase drastically myocardial
contractile force in vertebrates. Most biologically active CGs have an
-OH group at position C14, and the steroid moiety (rings A-D) is
completely saturated. There is usually an unsaturated five-member
lactone ring at position C17 (ring E); biological activity is abolished
by saturating the lactone or opening its ring structure. The steroid
nucleus and the lactone ring comprise the aglycone or genin portion of
the molecule and confer pharmacological activity to the structure (Moe
and Farah 1975) There are usually 1 to 4 sugar residues, or
glycosides, bonded to the steroid nucleus by an ether linkage at
position C3.
The potency and duration of activity of CGs are determined
primarily by the polarity of the glycosides and various steroid
substituents (Moe and Farah 1975, Detweiler 1967) That is, polarity
strongly influences absorption, distribution, metabolism, excretion and
possibly gustation of CGs in vertebrates. As compared to polar CGs,
less polar ones are 1) absorbed more easily across the lipoid membrane
of the intestine (Herman et al. 1962, Lauterbach 1981), 2) bound more
readily by serum albumin and thus rendered unavailable for
pharmacological or toxicological action or metabolism and excretion
(Lukas and deMartino 1969, Moe and Farah 1975), 3) metabolized less
readily in the liver (Detweiler 1967), 4) excreted more slowly because


o
.0
Figure 2-1. General structure of cardiac glycosides.
R = a glycoside.


A. Aspecioside
B. Uscharidin
Figure 2-2. Some representative cardiac glycosides.


11
they get caught in the enterohepatic cycle (i.e., absorbed CGs and
their metabolites are excreted in the bile and then reabsorbed; Okita
1967, Detweiler 1967), and 5) taste rejected by birds at higher
concentrations (Brower and Fink 1985) .
The structures of 5 CGs relevant to this dissertation are presented
in Figure 2-2. Aspecioside is the dominant CG in A^. syr-i ar.a L.
(Malcolm et al. 1989), and because 85 to 92% of the monarchs at the
Mexican overwintering sites appear to have fed on this foodplant, it is
also the dominant CG in their bodies (Seiber et al. 1986, Malcolm and
Brower 1989). Calotropin, calactin and uscharidin are the dominant CGs
in monarchs reared on A^. curassavica L. and are much less polar than
aspecioside (Roeske et al. 1976); mice were offered monarchs reared on
both A*, curassavica and A^ syriaca in Chapter 5. Digitoxin is derived
from several Digitalis species (Scrophulariaceae) and is of
intermediate polarity; mice were offered diets treated with digitoxin
in Chapter 6.
Defensive Value
Cardiac glycosides stored in monarch butterflies may deter
predators because of their bitter taste and/or toxic effects. CGs are
intensely bitter-tasting to humans (see references in Brower and
Glazier 1975, Glendinning pers. observ.) and are avoided by blue jays
Cyanocitta cristata at extremely low concentrations in food (Brower and
Fink 1985). Thus taste alone may be sufficient to render monarchs
unpalatable to mice.
Oral dosages of a variety of different types of CGs, including
those in adult monarchs, induce emesis, cardiac failure, neurotoxicity,


12
ataxia, dyspnea, severe tremors, and convulsions in many vertebrate
species (Detweiler 1967, Marty 1983, Brower 1984). However,
sensitivity to oral dosages of CGs is species-specific: dogs, some
species of birds, and humans are highly sensitive, whereas the rodents
Mus musculus, Rattus norvegicus and Peromyscus maniculatus are not
(Detweiler 1967, Barnes and Eltherington 1973, Tanz and Urquilla 1982,
Marty 1983, Brower and Fink 1985) The insensitivity of these rodents
appears to be due primarily to poor absorption of CGs across the
gastrointestinal wall (Marty 1983). Thus, given that the mice studied
in this dissertation are closely related to the 3 listed above, it is
likely that they are similarly insensitive to the toxic effects of oral
dosages of CGs.
Pyrrolizidine Alkaloids
Chemistry and Toxicology
The ester derivative of all toxic PAs and its numbering is shown in
Figure 2-3a. Two five-membered rings share a common nitrogen and
carbon at positions 4 and 8, respectively. Ester linkages can form at
positions 1 and 7. Pyrrolizidine alkaloids are frequently grouped
based on the number and type of ester linkages they have: 1) monoesters
have a single ester (e.g., intermedine, Figure 2-3b), usually at
position 1; 2) open diesters have two esters that do not join one
another; and 3) macrocyclic diesters have 2 esters that join to form a
ring structure (e.g., senecionine, Figure 2-3c).
The nitrogen atom of all PAs readily undergoes oxidation, which
converts the PA from a free amine to an N-oxide. An example of this
oxidation-reduction reaction is presented in Figure 2-4 for the


A. Amino alcohol derivative
of all toxic PAs with numbering.
R = H or OH
B. Intermedlne, an open monoester,
with numbering on the CHj i ester-
linked acid group.
Lx)
C. Seneclonlne, a
macrocycllc dlester.
OR CHjOR
0. Reactive Pyrrole.
R organic acid
or proton.
Figure 2 3. Some representative pyrrolizidine alkaloids and their derivatives.


Free Amine
N-oxide
Figure 2-4. Reaction involved in the conversion of free amine monocrotaline
to its N-oxide form.


15
macrocyclic diester monocrotaline. The bond between the nitrogen and
oxygen atoms in an N-oxide contains an unequally shared electron.
Because this electron tends to reside closer to the oxygen atom, it
gives the oxygen atom a net negative charge (e.g., see Figure 2-4) and
makes the N-oxide more polar than the free amine. Nevertheless, the
free amine and N-oxide forms of the same PA produce the same type of
toxicity. Both forms are commonly found together in a variety of
plants (Mattocks 1986) and insects (Brown 1984), including
overwintering monarchs in Mexico (Kelley et al. 1987).
There are 3 structural attributes essential for toxicity (Mattocks
198 6) : 1) the ring nucleus must be double-bonded between Cl and C2 (see
Figure 2-3a); 2) there must be at least one esterified hydroxyl group
(e.g., between C9 and CIO, Figure 2-3b); and 3) at least one of the
ester-linked side chains must contain a branched carbon chain (e.g.,
between Cll and C14, Figure 2-3-b). However, PAs are not themselves
toxic. They are rendered toxic by a dehydrogenation reaction in the
liver microsomes that normally metabolize alkaloids like PAs into more
polar and excretable derivatives. This reaction converts PAs into less
polar and highly toxic metabolites referred to as reactive pyrroles
(e.g., see Figure 2-3d; Mattocks 1972). Accordingly, the most toxic
PAs are those most readily coverted to reactive pyrroles.
There are two important structural features of toxic PAs that
facilitate breakdown into pyrrolic metabolites (Mattocks 1981) First,
low polarity PAs are metabolized into reactive pyrroles at faster rates
because they gain better access to the surfaces of microsomes.
Generally, polarity varies as follows in PAs: monoesters > open
diesters > macrocyclic diesters. Second, the type of ester linkage


16
strongly influences the proportions of pyrroles and N-oxides formed
from the free amines during the dehydrogenation reaction. (Recall that
both pyrroles and N-oxides are formed by this same reaction.) The
conformation of monoesters and macrocyclic diesters favors production
of reactive pyrroles, whereas that of open diesters gives relatively
more N-oxides.
Available evidence indicates that monoesters (e.g., intermedine)
are more common than macrocyclic diesters (e.g., senecionine) in
overwintering monarchs in Mexico; no open diesters were reported
(Kelley et al. 1987). However, this study was based on butterflies
collected during December, which would have probably derived their PAs
from nectar sources along the migration routes (Kelley et al. 1987).
More work is needed to determine whether the monarchs modify the
quantity and/or types of PAs in their bodies over the course of an
overwintering season by nectaring on the locally abundant PA containing
plants (e.g., Senecio and Eupatorium spp.; see Appendix 2 for species
names and their relative abundances; Robbins 1982) .
Defensive Value
As with CGs, PAs may deter predators by means of their bitter taste
and/or toxic effects. PAs are bitter-tasting to humans (Boppr 1986,
Glendinning pers. observ.) and their presence in the arctiid moth
Utetheisa ornatrix. ithomiine butterflies, and artificial foods causes
rejection by a variety of invertebrate and vertebrate predators (Eisner
1980, Brown 1984, Boppr 1986). The relevance of these findings to
monarch's PA-defense is uncertain. The PA content in monarchs is
extremely low relative to that of the ithomiines tested (see Kelley et


17
al. 1987, Brown 1984), and thus may not be sufficiently high to deter
predators. Moreover, the findings from the studies with JL. omatrix
and PA-adulterated diets cannot be evaluated because they were from
unpublished studies that were referred to in secondary sources (Eisner
1980, Boppr 1986). The relationship of bitterness to polarity and
molecular conformation in PAs has not been studied as it has in CGs
(see above).
Oral dosages of PAs are known to cause a variety of toxic effects
primarily in the liver of mammals, apparently because that is where the
reactive pyrroles are formed (Mattocks 1986). Relatively high dosages
produce acute effects and usually kill mice within 3-7 days by causing
severe hemorrhagic necrosis in liver cells. Lower dosages produce more
chronic effects and may take several months to kill mice by inducing
formation of 1) giant liver cells (megalocytes), 2) veno-occlusive
disease of the liver, and/or 3) depressed rates of growth and food
intake (Schoental and Magee 1957, McLean et al. 1964, Hooper 1978,
Goeger et al. 1983, Cheeke and Pierson-Goeger 1983, Mattocks 1986).
Even though large interspecific differences exist among laboratory
animals (mice, rats, guinea pigs, rabbits, and hamsters) in terms of
sensitivity to particular PAs, no species appears to be insensitive to
all or even a large number of PAs (Mattocks 1986).
Chronic PA toxicity does not usually result in death for several
months, and thus it is unclear whether mammals feel sick immediately
after ingesting sublethal doses of PAs. It may be that the action of
PAs is so insidious, that a mammal would not feel sickened until
several weeks after ingesting them. If this is the case, then it would
be unlikely that mammals could develop CFAs after ingesting them


18
(Brower 1984).
mm


CHAPTER 3
DEMOGRAPHIC RESPONSES TO COLONIES OF MONARCH BUTTERFLIES
Numerous workers have documented strong numerical and reproductive
responses by small mammals to areas with heavy seed crops (Jameson
1953, 1955, Gashweiler 1979, Halvorson 1982, King 1983) and dense
concentrations of insects (Holling 1959, Hanski and Parviainen 1985,
Hahus and Smith ill press). When these food bonanzas occurred in
habitats containing several small species of mammals, however, certain
species invariably responded more strongly than others (Halvorson 1982,
King 1983, Holling 1959, Hanski and Parviainen 1985, Hahus and Smith in
press). The factors that enabled particular species to take greater
advantage of the food bonanzas and thereby increase their reproductive
output proportionately are largely unknown. Possible factors include
1) competitive exclusion (Halvorson 1982), 2) greater tolerance to
microhabitat features of the areas with the food resource, and/or 3)
less sensitivity to the behavioral, physical and/or chemical defenses
of the food resource. In this chapter, I compare the demographic and
feeding responses of several species of mice to overwintering colonies
of monarch butterflies (Danaus plexippus L.) in Mexico, and attempt to
explain why one species enjoys exclusive access to the butterflies.
Four questions regarding the response of nearby mouse populations
to the monarch colonies are explored: 1) what is the density and
biomass of monarchs inside colonies available to terrestrially foraging
19


20
mice; 2) what species of mice occur in the overwintering areas and what
is their relative abundance; 3) are winter demographic patterns of
populations of the different species of mice, such as immigration,
breeding activity, density, home range size and age-class distribution,
influenced by the the colonies; 4) how do the diets of mice inside and
outside the colonies compare? In the course of this investigation, I
discovered that £_*. melanotis was the only species that immigrated to
the colonies and fed extensively on the monarchs. To explain this, I
tested two hypotheses. First, were there unique understory plant
species compositions in the sites where monarch colonies formed that
rendered these areas suitable only to L melanotis? Second, did P.
melanotis aggressively exclude the other species of mice from the
monarch colonies?
Methods
The Study Areas
I trapped and conducted experiments from January to March, 1985 to
1988, on the slope of the Arroyo La Plancha of the Sierra Chincua
mountain massif in northeastern Michoacn, Mexico (Figure 3-la, Anon.
1976: topographic map). I also trapped on 3 February and 10 March 1986
on the slope of the Arroyo Los Conejos of the nearby Sierra El
Campanario, at 1900'N and 10000'W. These 2 sites are among the 5
principal overwintering areas of the eastern population of the monarch
butterfly in the region (Calvert and Brower 1986).
All work was done between 3000 and 3320 m. The vegetation is
montane, boreal, coniferous forest dominated by the oyamel fir (Abies
religiosa H.B.K.). The understory vegetation is described below. The


21
period from late December to early March is the coldest and driest time
of year (Calvert et al. in pnenn), with night-time forest temperatures
on Sierra Chincua usually ranging between 0 to 6C (Calvert and Brower
1986, Alonso et al. in press). and with total monthly precipitation
averaging less than 23.3 mm (Calvert et al. in press. Anon. 1982).
Amount of Monarch Material Available to Foraging Mice
To estimate the amount of butterfly material on the forest floor
inside a monarch colony during the night, I ran 3 transects each for 2
consecutive days between 15 and 20 January 1988 inside a 0.44 ha colony
(W. Calvert pers. comm.) on Sierra Chincua. The transects were 80 m
long, each with 10 one-meter2 sampling quadrats set at 8 m intervals.
Because the colony was only about 45 m wide, I used a stratified random
sampling scheme (Zar 1984) to choose lengthwise transect locations.
Each morning between 700 and 730 h, the quadrats were cleared of all
monarch remains. Then, between 1900 and 1930 h on the same day, I
tallied the number of butterflies in each quadrat that were either on
the ground or less than 3 cm above it. Monarchs were classified as
live, moribund, cuticle pierced (i.e., sampled but not eaten), abdomen
removed, abdomen deviscerated, or thorax contents eaten. I disregarded
the few monarchs (< 3% of total monarchs tallied) that did not fall
into one of these 6 categories.
For each daily tally, X determined the mean density of each type of
monarch per m2 quadrat per day. I also estimated the mean grams of
tissue per m2 quadrat per day represented by each type of monarch,
using the following steps. First, I knew the approximate amount of
tissue in the abdomen (123 mg; dry weight) and thorax (41 mg; d.w.) of


22
overwintering monarchs (Chapter 4: Table 4-2). Second, I assumed that
all deviscerated abdomens and damaged thoraces still had on average 50%
of the internal tissues remaining (Fink and Brower 1981, Glendinning
pers. observ.). Then, the number of grams of tissue/m^/day in each
damage category was calculated using the equations provided in Table 3-
1.
Mark-recapture Trapping on Sierra Chincua
I compared the numerical responses of the different species of mice
to the monarch colonies and determined whether those mice living inside
the monarch colonies showed higher levels of reproduction. To do this,
mark-recapture studies were run between 20 January and 16 March 1985-
1987, and between 15 and 20 January 1988, in a total of 8 grid
locations that were either inside or at least 120 m to either side of
the colonies' outer edge (henceforth, outside; Figure 3-la and b). All
of the colonies had formed compact aggregations several weeks before X
set grids inside of them. In the 1985-87 seasons, the inside grids
were within the colonies until the last week of February, at which time
the colonies moved at least several hundred meters downhill.
In the 1985-1987 seasons, each grid was trapped one night per week
for 8 consecutive weeks and consisted of 60 trap stations in a 6 x 10
array with 10 m spacing, resulting in a sampling area of 0.6 ha. Two
traps were set on the ground within 1 m of each trapping station during
all seasons, except in 1985, when only 1 trap was set per station.
During the 1988 season, two grids were trapped for 3 consecutive nights
and consisted of 30 trap stations in a 3 x 10 array with 10 m spacing,
resulting in a sampling area of 0.3 ha. As in previous seasons, two


23
Table 3-1. Equations used to calculate the grams of tissue/m2/day in
the different damage categories that were available to terrestrially
foraging mice on the forest floor inside monarch colonies. For example,
to estimate the amount of live monarch tissue available to mice, I
multiplied the mean number of monarchs per m^ that fell to the monarch
floor per day (mean no./m^/day) times the combined dry weights of the
tissue in an overwintering monarchs's abdomen (123 mg) and thorax (41
mg). Based on Fink and Brower (1981) and Glendinning (pars. observ.). I
assumed that deviscerated abdomens and thoraces still had 50% of the
internal tissues remaining.
Damage category
Equation
live:
moribund:
cuticle pierced:
abdomen removed:
abdomen deviscerated:
thorax deviscerated:
(mean no./m^/day)
(mean no./m^/day)
(mean no. /re?/day)
(mean no./m^/day)
(mean no./m^/day)
(mean no./m?/day)
(123 + 41 mg)
(123 + 41 mg)
(123 + 41 mg)
(41 mg)
[ (123 mg) (50%) + (41 mg)]
[(123 mg) + (41 mg)(50%)]


24
traps were set on the ground. To determine whether my terrestrial
trapping regime had missed an arboreal rodent fauna, a third trap was
also secured to the tree trunk nearest each station, assuming one
occurred within 2 m from the station, at a height of 2 m; a total of 43
arboreal traps were set.
Folding Sherman live traps (8 x 9 x 23 cm) were baited with rolled
oats and provided with a compressed cotton ball for nesting material.
Mice were toe-clipped for permanent identification, and species, sex,
age, weight (to the nearest 0.1 g with 100 g Pesla scales),
reproductive condition, and trap station were recorded. Peromyscus
melanotis age was recorded as juvenile (<15.5 g) or adult (>15.5 g).
Only adults of the other species of mice were captured. Males of all
species were recorded as reproductively active if their testes were
descended (i.e., scrotal) or inactive if their testes were within their
abdominal cavity (i.e., abdominal). Females of all species were
recorded as pregnant if they had swollen abdomens (and/or weighed > 28
g, if they were E^. melanotis), as lactating if they had large nipples,
or as nonbreeding if they had small nipples and no swollen abdomen.
Traps were set between 1700 and 1900 h and checked the following
morning between 700 and 1000 h. Soiled traps were cleaned thoroughly
with a paper towel and water to reduce the influence of residual odors
from the previous occupant on subsequent captures (Mazdzer et al. 1976,
Stoddard and Smith 1984).
Mice were considered resident if they were present on the same grid
for 2 or more continuous weeks. The minimum number alive (MNA)
enumeration technique (Hilborn et al. 1976) was used to estimate both
the total number of mice on each grid each week and the number of weeks


25
each mouse was present on a grid. To validate use of this enumeration
technique, I also determined the minimum trappability of the population
(Krebs and Boonstra 1984) .
Home range sizes were estimated by calculating the area within a
convex polygon drawn around the outermost trap locations where each
mouse was caught. Home ranges were determined only for those mice 1)
caught at least 5 times, 2) trapped at a station on the edge of the
grid < 1 time, and 3) whose home range size did not increase after the
fifth capture.
To determine the mean weights of the 5 common species of mice on
Sierra Chincua, I combined the results of all mark-recapture studies on
Sierra Chincua between 1985 and 1988. If an individual was captured
more than once, I used its mean weight. Weights of pregnant females
were excluded. However, because most breeding females were caught more
than 4 times, I was able to determine their mean weight before and/or
after their pregnancy. Intersexual comparisons within each species
were made with the unpaired (two-tailed) t-test. In this and all
subsequent statistical tests, I tested for normality with the
Kolmogorov-Smirnov test (Zar 1984) If a significant departure was
found (i.e., P < 0.05), I used a nonparametric test.
To determine the numerical and reproductive responses of the
different species of mice to the monarch colonies, I treated each grid
as an independent sample. Because the population density estimates
among the successive trapping periods in each grid were not independent
(i.e., many of the mice remained on the same grid for more than one
trapping period), I simply describe trends in these data (Hurlbert
1984). However, I used a two factor ANOVA (repeated over time) to


26
determine the effects of colony proximity and time on the appearance of
new juveniles in the grids.
For several analyses, the combined results from all grids inside
colonies (n = 5) were compared with those from all grids outside
colonies (n = 6). Even though each grid should be analyzed separately
(Hurlbert 1984) I felt justified combining them because all grids had
relatively similar understory vegetation (see below), species
compositions of mice, and trends in the MNA estimates and levels of
reproductive activity over time. Moreover, sample sizes were small in
each grid. Mice from inside and outside colonies were compared in
terms of 1) the percentage of adult females of each species that were
reproductive, 2) the number of weeks that adult mice of each species
were present on the grids, using the Mann-Whitney U test (with the
normal approximation), 3) the home range sizes of male and female P.
melanotis. using the Mann-Whitney U test, 4) the proportion of juvenile
and adult £_*. melanotis that remained on the grids for greater than one
week, and 5) the effects of colony proximity and sex on the body
weights of male and female E^. melanotis, using the two factor ANOVA.
For comparisons 2 and 4, I excluded those mice first trapped during the
last three weeks of the season (i.e., from 23 February to 9 March,
1985-87).
Snap-trapping of mice at Sierra Chincua and Sierra El Campanario
In addition to these mark-recapture studies, I compared the
demographic responses of mice to the monarch colonies on Sierra
Campanario and Sierra Chincua. To do this, I ran 4 parallel transects
on Sierra Campanario on 3 February and on 15 March 1986 in a 2.1 ha


monarch colony (W. Calvert pers. comm.) r as well as on Sierra Chincua
on 15 February 1986 in a 0.6 ha colony on grid 7 (Figure 3-lb). During
each night, two of the transects were set inside the colony and the
other two 300 m to either side of the colonies' edges. Each transect
was 80 m long with 10 trapping stations set at 8 m intervals. I also
snap-trapped on 21 and 22 March 1986 in grids 1 and 2 (Figure 3-lb) at
the same stations used for the above-described mark-recapture studies.
I set two Museum Special snap-traps per station and baited each
with rolled oats and peanut butter. Traps were set between 1700 and
1900 h and rechecked the next morning between 700 and 900 h. I
recorded species, sex, age, weight, and reproductive condition and used
the same indices of age and reproductive status as described above for
the mark-recapture study.
The results from Sierra Campanario were used to compare the species
composition and reproductive activity of mice inside and outside of
colonies. Because of small sample sizes, I combined the results of the
two sampling nights when making the reproductive comparisons. These
results were compared to those from the mark-recapture studies on
Sierra Chincua. The stomachs from mice captured on both massifs were
included in the dietary analyses described below.
Stomach Content Analyses
I compared the stomach contents of mice from inside and outside
monarch colonies in terms of 1) the occurrence of monarch material, 2)
dry weight, 3) lipid weight, 4) percent lipid, 5) total cardiac
glycoside (CG) content, and 6) CG concentration. All weight
measurements were made to the nearest 0.0001 g with a Mettler AK 160


28
balance. To get representative stomach samples from naturally foraging
mice, I used the mice caught in snap-traps, to assure that they had
died immediately upon capture. (Food boluses pass through the
digestive tracts of live, captive Peromyscus maniculatus in 2 to 7
hours (Marty 1983).)
Each animal's stomach was removed and preserved in 50 ml of 95%
ethanol, and then frozen (within 3 weeks) in a domestic freezer until
analyzed in October 1988. To quantify the occurrence of monarch
material, I first removed and discarded all bait; this bait was not
included in the lipid and CG analyses. Next, I placed the contents of
each stomach in a petri dish, while suspended in about 10 ml of 95%
ethanol, and observed 20 fields under lOOx magnification. If I found
no food items in a field, I chose another one. I determined the
presence or absence of monarch material (e.g.. small pieces of
exoskeleton, testes pigment, fat, and tracheal material) in each field,
and then computed the frequency of occurrence of monarch material in
all 20 fields for each stomach sample. Reference samples of monarch
material were made by feeding test mice a diet consisting solely of
monarchs and running their stomach contents through the procedure
described above. The monarch material and ethanol in which it was
stored (henceforth, stomach content sample) were together subjected to
the lipid and CG analyses described below.
Each stomach content sample was dried for 16 h at 60C in a forced-
draft oven, then cooled under vacuum with dry-rite, and weighed.
Neutral lipids were extracted with petroleum ether (see methods in
Walford 1980, Brower ei. ni. in prep.) and CGs from the lean residue
with ethanol (see methods in Brower et al. 1975) Gravimetric


29
techniques were used to determine total lipid content per stomach
sample, and spectrophotometric techniques to determine the
concentration (p.g/0.1 g dry weight) and weight of CGs in each stomach
(Brower et al. 1975, Brower et al. 1985).
It is unlikely that digestive enzymes or food items in the stomach
interfered with my lipid and CG determinations. First, the small
intestine is the major site of lipid metabolism and absorption in
rodents; some lipolysis but no absorption occurs in the stomach (Booth
et al. 1961, Harrison and Leat 1975) Second, CGs are metabolized
poorly if at all in the stomachs of melanotis and maniculatns as
well as in that of birds (Marty 1983, Brower et al. 1988, Brower et al.
1985) Third, virtually all ingested bait, which contained lipids and
also could have interfered with the spectroassay (Brower et al. 1985),
was physically removed from the stomach samples prior to the
quantitative analyses; the oats and peanut butter had formed into a
hard clump, which was easy to separate from the other food.
I used the one-way ANOVA and Scheff F-test to compare the five
species' stomach samples in terms of 1) the frequency of occurrence of
monarch material, 2) dry weight, 3) lipid weight, 4) percent lipid, 5)
gross CG content, and 6) CG concentration. The percent lipid data were
arc-sine transformed (Zar 1984) for the analysis. I also performed a
two-way ANOVA on the results from melanotis to examine the effects
of colony proximity and sex on the same 5 variables.
Comparison of the Understory Vegetation in the Trapping Grids
Numerous studies suggest that microhabitat features (i.e., species
composition and density of understory vegetation) strongly influence


30
the distribution in space and time of species of mice (e.g., Brown
1964, Hansen and Fleharty 1974, M'Closkey and Lajoie 1975, Huntly and
Inouye 1987) I thus hypothesized that the preponderance of P.
melanotis inside the colonies may have been influenced by unique plant
species compositions found in the colony sites. To test this, I
described and compared the understory vegetation in the eight trapping
grids, employing standard methods (Risser and Rice 1971, Whittaker
1970, Mueller-Dombois and Ellenberg 1974). Voucher specimens of all
plant species were identified by Hilda Flores and Jose Luis Villaseor
of the MEXU-herbarium at the Universidad Nacional Autonoma de Mxico.
I used a random stratified sampling scheme to select 12 out of the
total of 45 quadrats per grid; each grid was divided into 4 equal
blocks, and then within each block I randomly selected 3 quadrats.
Meter-squared areas in the center of each quadrat were sampled and the
frequency and percent coverage of each plant species was determined by
visual inspection. Then, I determined the importance percentage of
each species by summing its relative dominance (using the percent
coverage data) and relative frequency and dividing by 2. Finally, I
used the following equation to make pair-wise comparisons between the 8
trapping grids:
Percent similarity = 1 0.5 X IPa^-Pb^l
where p is the importance percentage divided by 100, a and b refer to
the two grids being compared, and i to each species.
To determine whether the sites where monarchs form colonies possess
unique microhabitat characteristics, I categorized each percent
similarity value into one of three groups, based on the proximity of
the two grids in the pair-wise comparison to a monarch colony: inside


31
inside, outside vs.. outside, and inside as., outside. I used the
Kruskal-Wallis test to compare the three groups and expected the inside
xs. inside PS values to be consistently higher.
Aggressive Relations Among the Different Mouse Species
I also hypothesized that the preponderance of £_^ melanot-i s inside
the monarch colonies could have been due to E^. melanotis aggressively
excluding the other species of mice. To test this, I studied the
aggressive relations of E_^ melanotis. E^. a^. hylocetes. R. sumichrasti.
and iL. salvus by staging 70 intrasexual, paired interactions between
a reproductively active E^. melanotis in its home range inside a monarch
colony (henceforth, resident) and a reproductively inactive,
conspecific or heterospecific from outside the colony (henceforth,
intruder). I thus had a total of eight pairing combinations: 1
conspecific and 3 heterospecific ones for each sex. Sexually active
residents and inactive intruders were used because this pattern
mimicked my trapping results most closely. I selected mice of
comparable body mass ( 2 g) for the conspecific pairings; this was not
possible for the heterospecific pairings, given the large size
differences (Table 3-4). I predicted that resident E^_ melanotis would
dominate intruders. Recently Wolff et al. (1983) showed that dominance
among 2 similarly sized Peromyscus species is site-specific and not
species-specific. That is, an individual is more likely to win an
aggressive encounter with a conspecific or heterospecific of the same
sex when in its own home range.
Paired encounters were staged between 0800 and 1100 h in a clear,
plexi-glass arena (32 x 32 x 48 cm), with an open bottom and top,


32
following the methodology of Wolff et al. (1983) The animals were
thus on natural substrate. Each trial was carried out at the trap
station where the resident was captured the previous night, and
consisted of a 30 s acclimation and a 5 m interaction period. During
the acclimation period, each mouse was placed in the arena on opposite
sides of a removable cardboard partition. The interaction period was
video-taped and began immediately after the partition was removed.
Because it was difficult to distinguish resident and intruder during
conspecific pairings, I painted a 0.5 cm diameter, red or gold spot of
nail polish on the pelage between each animals ears. The mice did not
appear to have been disturbed by the nail polish. The assignment of
gold or red for resident or intruder was determined randomly for each
mouse. This color labelling was not necessary for the heterospecific
pairings.
For each interaction period, I recorded the number of non-offensive
and offensive approaches, attacks and retreats exhibited by the
resident and intruder (Table 3-2). To determine the outcome of the
encounters, I modified slightly the criteria of Wolff et al. (1983):
1) Winner: The animal with the fewest number of retreats.
2) Draw: Both animals had a similar number of retreats (i.e., the
total number by each individual differed by < 2).
3) Ha aggression: Neither animal displayed any aggression towards
the other.
To determine whether resident £_*. melanotis won a greater number of
trials than expected by chance in each of the eight pairing
combinations, I used the binomial test. To compare the number of non
offensive approaches, offensive approaches, attacks and retreats


33
Table 3-2. Partial ethogram of behaviors displayed by malanotis. p. a.
hylocetes, E^. sumichrasti and M^. salvus during paired interactions .
Many of the behavioral descriptions are modified from Eisenberg's (1968)
more complete ethogram of Peromyscns.
Behavior
Description
Non-offensive approach Mouse approaches opponent slowly with body
Offensive approach
contours relaxed.
Mouse approaches opponent with its body
stretched out. Its body shows rigidity and
muscular tension as it moves. Frequently, the
tail is rigid and the ears and vibrissae are
extended forward.
Attack
a. Rush
Mouse suddenly accelerates approach towards
opponent, usually following an offensive
approach.
b. Attack leap
Mouse leaps into the air at opponent, striking
it with its limbs and/or body.
c. Lunae
Mouse strikes at a nearby opponent with
forepaws, but keeps feet planted on substrate.
Retreat
a. Escape leap
Attacked mouse flees from attacking mouse with
wild and erratic leaps.
b. Fliaht
Attacked mouse runs away from a rush, attack
c. Flight after fight
leap or lunge.
Mouse flees following a locked fight, which is
when the two animals lock their ventrums
together while rolling on the ground.


34
exhibited by the resident and intruder, I used the Wilcoxon matched-
pairs test (Zar 1984). For this analysis, I combined the results from
both sexes, thereby reducing the number of pairing combinations to
four, because no intersexual difference was apparent.
Results
Density of Monarchs on Forest Floor Inside Colonies
An average of 6.5 monarchs/m^ was added to the forest floor inside
a monarch colony each night (Table 3-3). Perhaps more significant is
that 83% of these monarchs had intact abdomens and thoraces (i.e., were
live, moribund and/or sampled) and thus still possessed the majority of
their tissues and body water. An average total of 0.93 gram (dry
weight) of lipid-rich tissue per m^, or 9,300 g per hectare, was
available to foraging mice. This is an underestimate of the total
amount of monarch material available to mice because I did not include
the dead monarchs that had accumulated on the forest floor prior to
sampling or the live ones roosting on low-lying understory vegetation
(Alonso et al. in press).
Irappability and. ..Reliability of Data
Minimum trappabilities for mice marked on Sierra Chincua between
1985 and 1986 were extremely high: for £^. melanotis. 90% inside and 88%
outside the monarch colonies; for sumichrasti. 93%; for a.
hylocetes. 89%; for 1L_ alstoni. 91%; and for nu. salvus. 82%.
Trappability estimates for the latter 4 species were from individuals
trapped on grids outside colonies. Thus, the MNA estimates were
considered to be accurate within 5% (Hilborn ££. nl., 197 6) My


35
Table 3-3. Estimated density of live, moribund and bird-damaged
monarchs (mean no./m^; S.E.), and quantity of monarch tissue (mean
no. grams/m2; dry weight) on the forest floor at the beginning of
the night inside a 0.44 ha monarch colony on Sierra Chincua.
Samples were made over six nights between 15 and 20 January 1988.
State of
monarch
Density
(no. per m2)
Tissue (g; d.w.)
per m2
live
2.3
(0.9)
0.38
moribund
1.1
(0.3)
0.18
cuticle pierced
1.8
(0.1)
0.30
abdomen removed
1.1
(0.2)
0.04
abdomen deviscerated
0.1
(0.1)
0.02
thorax contents eaten
0.1
(0.1)
0.02
Total
6.5
0.93


36
treatment of the trapping grids as independent samples was
substantiated because 1) I recorded only one instance of a mouse moving
between trapping grids (an adult male melanotis moved from grid 7
(inside colony) to 8 (outside colony) during 1985), and 2) I removed
(i.e., snap-trapped) nearly the entire population of mice from grid 2
at the end of the 1986 season and thus diminished the influence of
temporal pseudoreplication (Hurlbert 1984) on my results from the same
grid during the 1987 season. Finally, my use of terrestrial trapping
results was justified because no mice were caught in the arboreal traps
set during 1988.
Species Composition of Mice Inside and Outside Monarch Colonies
I marked a total of 634 adult and juvenile mice in my mark-
recapture studies on Sierra Chincua between 1985-88. Of this total,
71% were £_. melanotis, 11% were £_. hylocetesf 8% were R.
aumichrasti, 8% were n*. nu. salvus, 2% were a_^ alstoni. Clearly, e^.
melanotis was the most common species.
All species differed greatly in size, except for a^. hylocetes
and 1L_ alstoni (Table 3-4). Peromyscus melanotis and nu salvus were
the only species with apparent sexual dimorphisms, with the former
having larger females and the latter larger males.
The species composition of mice on grids inside and outside monarch
colonies differed markedly and in a similar manner on both Sierra
Chincua and Sierra Campanario (Table 3-5). At least 90% of the mice
captured on grids inside colonies were E_*. melanotis. whereas only 45 to
69% of those captured on grids outside colonies were E_^ melanotis.
These findings were consistent both among and within seasons.


Table 3-4. Mean body weights (S.E., in grams) of adult mice of 5 species caught inside and outside
monarch colonies. Mice were weighed during mark-recapture studies on Sierra Chincua between 20
January and 16 March, 1985-1988; an average weight was calculated for those mice caught more than
once. Weights of pregnant female £_*. melanotis were excluded; no pregnant females of the other species
were trapped. However, because most breeding, female £_*. melanotis were caught more than 4 times, I
was able to determine their mean weight before and/or after their pregnancy. Sample sizes are in
parentheses. I compared the weights of both sexes with the unpaired t-test (two-tailed).
Gender
R. a. hylocetes
N. alstoni
M. m
salvus
P. melanotis
R. sumichrasti
male
36.2 0.5
37.2 1.1
29.1
0.9
18.8 0.1
12.7 0.3
(40)
(4)
(28)
(220)
(21)
female
35.9 0.9
34.4 1.5
26.6
0.5
21.2 0.2
13.6 0.4
(30)
(8)
(24)
(227)
(32)
t value -0.26 -1.29 -2.29 11.66 1.64
0.793 0.226 0.026 0.0001 0.107
P value


Table 3-5. Total number of mice (adult and juvenile) of 5 species captured inside and outside monarch
colonies on Sierra Chincua and Sierra Campanario, and the percentage of these mice that were P.
melanotis. I show results from mark-recapture studies on Sierra Chincua between 20 January and 16
March, 1985-88, and from snap-trap studies on Sierra Campanario between 3 February and 10 March, 1986.
Total number
of individual mice
captured
Percent
Year
Grid
R. melanotis
£. a. hvlocetes
R. sumichrasti u
. l*. alstoni
M. m... salvos
R. melanotis
SIERRA CHINCUA
inside colony
1985
7
58
2
1
0
l
93.6
1986
2
99
5
4
0
2
90.0
1987
2
153
2
4
0
li
90.0
5
78
1
3
0
2
92.8
1988
1
39
0
1
0
0
97.5
Total
427
1 3
13
0
1 6
91 0
1985
gut-Side colony
62.5
69.1
b
8
1 D
47
9
6
7
0
1
0
4
1986
1
26
12
9
2
9
44.8
3
36
14
10
7
11
46.2
1987
4
36
15
4
2
8
55.4
1988
2
15
5
4
2
4
50.0
Total
175
5 8
4 0
1 4
36
54.1
Continued.


Table 3-5Continued.
1986
2 Feb.
18
0
SIERRA CAMPANARIO
inside colony
0
0
0
100.0
10 March
37
0
1
0
0
97.4
Total
55
0
1
0
0
98.2
1986
2 Feb.
8
1
aulside salany
2
0
1
66.0
10 March
15
3
2
0
5
60.0
Total
23
4
4
0
6
62.1


40
Moreover, they are greatly strengthened because my comparisons on
Sierra Chincua included grids that were inside a colony during some
years and outside a colony during others (see grid 2 during 1986-88,
and grid 1 during 1986 and 1988; Figure 3-lb).
Of the mice captured inside colonies, only melanotis individuals
established residency; they were trapped on average for 3.7 consecutive
weeks (Table 3-6). All individuals of the other species were trapped
only once inside the colonies, except for one 1L m*. salvus that was
caught for 2 consecutive weeks. Thus, even though immigrants of all
five species encountered the monarch colonies, only individual P.
melanotis remained and established residency. This contrasts with the
results from outside the colonies, where individuals of all species
remained on average for at least 2.6 consecutive weeks. In fact, the
mean duration of residence on grids outside versus those inside monarch
colonies was significantly greater for £_*. hylocetes r R. snmir.hrasti
and nL_ salvus. but not for melanotis (Table 3-6) .
Female £_*. melanotis trapped inside colonies exhibited the highest
levels of reproduction, both on Sierra Chincua and Sierra Campanario
(Table 3-7). Female £^_ melanotis trapped outside colonies also showed
relatively high reproductive activity compared to the other four
species. However, the majority of the females of all five species
outside the colonies did not initiate breeding until late February and
March. This pattern contrasts sharply with that of female melanotis
inside colonies on Sierra Chincua, which were breeding extensively by
20 January (Figure 3-3a).


Figure 3-1. (A) Location of mark-recapture study grids for 4 years from 1985-88 (black squares
inside hatched area) in relation to topography and drainage patterns on Sierra Chincua (adapted
from Anon., 1981). Contour interval = 100 m. (B) The hatched area has been expanded and
presented separately for all four trapping years. Trapping grids are drawn to scale and each grid
is given a unique number. Solid grid squares indicate which grids were trapped each year. Colony
locations are indicated by the stippled area. Even though a colony formed in grid 7 during 1986,
I did not conduct a mark-recapture study there during that year.


Continued.


Figure 3-1continued.


Table 3-6. Mean number of weeks (S.E.) that adult mice of the 5 species remained on the grids inside
(n = 4) and outside (n = 5) monarch colonies on Sierra Chincua. Only those mice first trapped between
20 January and 23 February, 1985 to 1987, were included in the analyses. Sample sizes are in
parentheses. Comparisons are made between mice from inside and outside colonies with the Mann-Whitney
U test (normal approximation).
Colony
proximity
EL_ melanotis
P. a. hylocetes
R. sumichrasti
i. alstoni
salvus
inside
3.7 0.2
(210)
1.0 0.0
(11)
1.0 0.0
(13)
1.1 0.1
(14)
outside
3.8 0.3
(77)
3.2 0.3
(45)
2.6 0.3
(31)
3.5 0.5
(11)
2.7 0.4
(29)
Z value*
0.020
3.307
3.670
3.190
P value
0.9838
0.0009
0.0002

0.0014
* corrected for ties.


Table 3-7. Percentage of adult females of 5 species that were pregnant and/or lactating. I combined
the data for all females trapped in mark-recapture studies on Sierra Chincua between 20 January and 16
March, 1985-88, and in snap-trap studies on Sierra Campanario on 3 February and 10 March, 1986.
Percentage reproductive (total number captured)
Proximity
to colony
£. melanatia
P. a. hvlocet.es
B- sumichrasti
L. alstoni
M. m
. salvus
SIERRA CHINCUA
inside
88.2
(271)
0.0 ( 4)
11.1 ( 9)
-.- (0)
0.0
( 6)
outside
54.5
( 55)
4.2 (24)
5.0 (20)
20.0 (5)
7.1
(14)
SIERRA CAMPANARIO
inside
82.6
( 23)
-.- ( 0)
0.0 ( 1)
-.- (0)
-.-
( 0)
outside
37.5
( 8)
0.0 ( 2)
0.0 ( 1)
-.- (0)
0.0
( 2)


46
Demographic Responses of P. melanotis to the Monarch Colonies
Initial population densities (i.e., those during early January) of
mice on the grids inside and outside the colonies were similar, except
on grid 2 in 1986 and 1987, where female density was already high
(Figure 3-2). However, subsequent population densities increased more
than two-fold on the grids inside colonies, while remaining roughly
constant on the grids outside colonies. Peak densities of botn sexes
reached 50 to 97 mice per ha on grids inside colonies, and only 13 to
32 mice per ha on grids outside colonies. A comparison of the
proportion of adult and juvenile mice that established residency inside
colonies suggests that adult immigration, rather than juvenile
recruitment, was primarily responsible for the increased density over
time (Table 3-8) In the grids outside of the colonies, juveniles and
adults occurred in similar proportions. Immigration of adults was
consistently female-biased to grids inside colonies and male-biased to
grids outside colonies (Table 3-9; Figure 3-2).
Whereas females inside colonies showed high levels of reproduction
throughout the winter, those outside colonies did not initiate high
levels of reproduction until late February (Figure 3-3a). There were
also many more new juveniles caught inside colonies (Figure 3-3b). A
two factor ANOVA revealed a significant effect of colony proximity (F =
19.21, df 1, P = 0.0032), time (F = 7.67, df 7, P = 0.0001), and
colony proximity x time (F = 4.86, df = 7, P = 0.0003) on the number of
juveniles captured. The significant interaction indicates that the
number of new juveniles increased with time on the grids inside but not
on those outside colonies. The large number of juveniles inside


Figure 3-2. Minimum number alive estimates of male and female P.
melanotis present on the 4 trapping grids inside and 5 outside monarch
colonies from 20 January to 16 March, 1985 to 1987. The numbers to the
right of each line indicate the specific trapping grid. Because grid 2
was trapped for two seasons, I distinguish the results from 1986 as 2,
and those from 1987 as 2*.


Minimum Number Alive
48
January February March
^OD^jrocn


49
Table 3-8 Proportion of juvenile and adult me la not is remaining
for greater than one week on 4 grids inside and 5 grids outside
monarch colonies on Sierra Chincua. Only those mice that were first
trapped between 20 January and 23 February, 1985-87, were included.
Total
number
Age-class Number of weeks on grid trapped
one
> two
inside colony
juvenile
0.72
0.28
47
adult
0.24
0.76
210
outside
colony
juvenile
0.26
0.74
27
0.35
0.65
77
adult


50
Table 3-9. Percentage of E_^ melanotis that established resideny on
grids inside and outside of monarch colonies that were female.
Year
Inside
colony
Outside colony
Grid
no.
Female
%
Grid
no.
Female
%
1985
7
58.6
6
33.3
8
41.2
1986
2
66.1
1
44.4
3
50.0
1987
2
60.1
4
46.2
5
58.3
Mean S.E.
60.8 1.8
43.0 2.8


Figure 3-3. (A) Mean (S.E.) percentage of female P_*. melanotis in
reproductive condition on 4 grids inside and 5 grids outside monarch
colonies from 20 January to 16 March, 1985 to 1987. Sample sizes are
written alongside each symbol. (B) Mean (S.E.) number of new juvenile
P. melanotis trapped inside and outside monarch colonies on the same
grids and over the same time period as in A.


Mean (S.E.) no. of new juveniles Mean (S.E.) percent reproductive
52
A.
B.


53
colonies strongly suggests that the females were able to successfully
wean their young.
Forty-one £_^ melanotis fit my criteria for use in the home range
size determinations (in m2; mean S.E.). Home ranges for males inside
(274.4 96.1, n = 7) and outside (425.9 89.8, n = 8) colonies did not
differ significantly (Z = 1.46, P = 0.144), nor did those for females
inside (158.9 30.9, n = 16) and outside (187.7 47.5, n = 10) colonies
(Z = 0.24, P = 0.811]. However, when the results from grids inside and
outside colonies were combined, males had significantly larger home
ranges than did females (355.2 66.4 and 170.0 25.9, respectively; Z =
2.64, P = 0.0084) .
Nonpregnant, resident females were significantly heavier than
resident males in grids inside and outside colonies (Table 3-10) The
colony proximity x sex interaction was not significant. Neither sex
exhibited evidence of weight loss or deteriorating health over the
winter.
Stomach Content Analyses
Detailed visual examinations indicated that the stomach samples of
E^. melanotis collected inside colonies consisted solely of monarch
material (Figure 3-4). I also encountered variable amounts of monarch
material in the stomach of mice from grids outside colonies, with P.
melanotis having significantly more than the other species, and 1L m.
salvus having significantly less. Live and/or dead monarchs do occur
on the forest floor outside colonies, albeit at a relatively low
density (Glendinning unpub1. observ.).


54
Table 3-10. Comparison of the body weights (mean S.E., in
grams) of resident £. melanotis trapped inside and outside
monarch colonies between 27 January and 23 February, 1985-87, on
Sierra Chincua. As mice were caught at least two times, an
average weight was calculated for each mouse. Pregnant female
weights were excluded. However, because most breeding females
were caught more than 4 times, I was able to determine their
mean weight before and/or after their pregnancy. Sample sizes
are in parentheses. The results of a 2-factor ANOVA are
presented.
Proximity
to colony
Males
Females
inside
18.5 0.3
(75)
21.2 0.3
(110)
outside
17.7 0.4
(48)
19.8 0.5
(31)
Source of
Variation
df F
-ratio
P
Colony
Proximity
1
8.89
0.0031
Sex
1
45.57
0.0001
Interaction
1
5.11
0.3937


Figure 3-4. Mean (S.E.) occurrence of monarch material in the stomach contents of 74 L.
melanotis collected inside and 38 £j_ melanotis. 17 sumichrasti. 15 L i. hyloceteSr 5 EL.
alstoni. and 10 L nu. salvus collected outside monarch colonies on Sierra Chincua and Sierra
Campanario. I present results of a one-way ANOVA. Different subscripts (a, b, c, d) indicate
significant differences among means (P < 0.05; Scheff F-test).


o
o
c
CD
O
O
O
c
cd
(D
1001
80-
60-
¡nside colony
outside colony
Kruskal-Wallis H = 140.7
P = 0.0001
VJl
ON
40-
20-
P. melanotis R. sumichrasti P. a. hylocetes
N. a. alstoni
M. m. salvus


57
Even though the dry weights of £^_ hylocetes, u_._ alstoni and
1L. KU salvus1 stomach contents were significantly higher than those of
£_^ melanotis (collected outside colonies) and sumichrastif all
species had statistically equal amounts of lipid (Table 3-11) .
However, melanotis had a significantly higher percentage of lipid,
CG content and CG concentration than did the other species. Together
with the results in Figure 3-4, these results suggest strongly that on
grids outside colonies, E^. melanotis consumed substantially more
monarch material than did the other four species.
Two way ANOVA's on the stomach samples from £_*. melanotis inside and
outside colonies demonstrated significant effects of colony proximity
and sex on dry weight and lipid weight, and of colony proximity on
percentage of lipid (Table 3-12). The same effects were significant
for CG concentration, but in the opposite direction (i.e., males both
inside and outside colonies had higher values). Cardiac glycoside
content was not influenced significantly by colony proximity or sex.
These results suggest that E^. melanotis inside colonies fed exclusively
on monarch material and thus accumulated large quantities of lipid,
amounting to approximately 43% of the total dry mass of their stomach
contents. However^ even though conspecifics outside colonies ingested
less monarch materials, the CG concentration in their stomach samples
was significantly greater. Females both inside and outside colonies
had greater dry weights and lipid weights; this intersexual difference
was greatest for mice from inside colonies.


Table 3-11. Comparison of stomach content samples from 5 mouse species for the variables dry weight,
lipid weight, percent lipid, and CG content and concentration (mean S.E.). All mice were collected
outside monarch colonies on Sierra Chincua and Sierra Campanario. Interspecific comparisons are
made with one-way ANOVA and Scheff F-tests. Subscripts, a and b, indicate significant differences
among species within each column (P < 0.05).
Dry
Lipid
Lipid
CG
weight
weight
as %
CG content
concentration
Species
N
(g)
(g)
dry weight
(|lg/0.1 g; d.w.)
P. melanotis
38
0.14 0.02
b
0.03 .004
a
25.6 1.5 a
105.7 10.4
a
100.1 14.4 a
E- sumichrasti
17
0.13 0.01
b
0.02 .002
a
14.8 1.2 b
1.7 0.8
b
1.3 0.6 b
P. .. hvlocetes
15
0.44 0.07
a
0.04 .007
a
9.8 1.2 b
2.5 1.1
b
1.3 0.7 b
N. alstoni
5
0.39 0.09
a
0.04 .007
a
4.1 1.8 b
o.o b
0.0 b
M. nu. salvus
10
0.34 0.05
a
0.02 .002
a
7.0 1.1 b
o.o b
0.0 b
F-ratio
17.78
3.13
23.94
22.21
14.20
df
4
4
4
4
4
P-value
0.0001
0.0191
0.0001
0.0001
0.0001


Table 3-12. Comparison of stomach content samples from male and female P. melanotis collected inside and
outside monarch colonies for the variables dry weight, lipid weight, percent lipid, and CG content and
concentration (mean S.E.). Mice were trapped on Sierra Chincua and Sierra Campanario. The effects of
sex and colony proximity on each variable were determined with a two-factor ANOVA. NS P > 0.05, P <
0.05, ** P < 0.005, P < 0.0005.
Proximity
to colony
Sex
N
Dry
weight
(g)
Lipid
weight
(g)
Lipid
as %
dry weight
CG content
<^g)
CG
concentration
((lg/0.1 g; d.w.)
inside
female
37
0.32 0.04
0.14 0.02
43.6 2.0
126.4 14.3
52.5 6.7
male
37
0.20 0.02
0.08 0.01
42.9 1.4
91.3 9.5
62.8 7.3
outside
female
14
0.18 0.03
0.05 0.007
26.9 2.4
96.1 14.8
74.4 17.9
male
24
0.11 0.02
0.03 0.005
24.9 2.0
111.2 14.1
114.8 20.3
Source of
Variation
df
F-ratios
Colony proximity
1
8.95 **
19.45 ***
70.31 ***
0.48 NS
-8.75 **
Sex
1
6.82 *
5.53 *
0.46 NS
0.13 NS
-4.14 *
Interaction
1
0.65 NS
1.56 NS
0.12 NS
2.98 NS
1.47 NS


60
Comparison of the Understory Vegetation in the Eight Grids
My results suggested that the understory vegetation on the grids
inside monarch colonies did not differ significantly from that on the
grids outside colonies. The pair-wise, percent similarity values from
inside 2S.. inside comparisons were not significantly higher than those
of outside ys.. outside and inside 2S. outside comparisons (mean S.E.
in respective order = 47.5 4.0, 40.4 6.0 and 48.4 3.6; Kruskal-
Wallis H value = 1.135, P = 0.567; Appendix A). The percentage
importance values of the 34 understory plant species encountered on the
8 grids are in Appendix B.
Aggressive Relations Among the Different Mouse Species
The results of the 70 pairings in which a win/loss decision was
made are presented in Table 3-13. The number of wins by resident P-
melanotis was significantly greater than expected by chance in
conspecific pairings, but not in any of the heterospecific pairings.
In fact, resident melanotis lost the vast majority of pairings with
E-*. hylocetes and H*. nu. salvus. whereas the results of their pairings
with fL_ sumichrasti were equivocal (Figure 3-5). In conspecific
pairings, the resident E_^ melanotis exhibited significantly more non
offensive and offensive approaches and attacks (rushes) than did the
intruders (Figure 3-6) Interactions between melanotis and a.
hylocetes were the least aggressive; the majority of the former's
retreats followed non-offensive approaches by the latter. Most of
HU salvus' attacks involved lunges following non-offensive approaches
by melanotis. Interactions between melanotis and sumichrasti
were the most balanced, with the former exhibiting more offensive


61
Table 3-13. Outcomes of intrasexual, paired trials between a
resident £. melanotis and a visiting conspecific or heterospecific
in which a win/loss decision was made. The winner of a trial had
the fewest retreats. The first animal of each pair is the resident
and the second animal is the visitor. The number of wins by the
resident and visitor were compared with the binomial test.
Species
Sex
No. of
Trials
No.of
Wins
Binomial
Prob.
£. melanotis
F
10
9
0.039
E. melanotis
F
1
E. melanotis
M
10
9
0.039
£. melanotis
M
1
£. melanotis
F
9
3
0.870
B. sumichrasti
F
5
£. melanotis
M
8
5
0.453
R. sumichrasti
M
2
£. melanotis
F
8
1
0.070
H*. nb. aalaus
F
7
£. melanotis
M
8
0
0.008
Hl. nu. salvus
M
8
£. melanotis
F
9
0
0.004
P. a. hylonetes
F
9
£ melanotis
M
8
0
0.016
£. a. hylocetes
M
7


Figure 3-5. Percentage of resident wins, intruder wins, draws, and no aggression outcomes in 70
intra- and interspecific pairs of melanotis (£_*. hlJ fL_ sumichrasti (E^ Z+), nu. salvus (M.
HL_ s.), and fL. hylocetes (P. a. h.) N = number of pairings for both sexes combined.


Percent
0 No Aggression
Draws
0 Intruder Wins
Resident Wins


Figure 3-6. Mean (S.E.) number of non-offensive and offensive approaches, attacks and retreats
exhibited by resident E^. melanotis and intruder conspecifics or heterospecifics (fL*. sumichrasti.
M. m. salvus. or E,. hylocetes) during the 5 minute trials. For each behavior, I compare the
number of times it was exhibited by the resident and intruder with the Wilcoxon matched-pairs
signed-rank test; an asterisk indicates significance at the 0.05 level. Species abbreviations are
as in Figure 5.


Mean No. of Occurrences Mean No. of Occurences
P. m. x P. a. h.
Non- Offensive
Offensive Approach
Approach
Attack Retreat
P. m. x M. m. s.
Non- Offensive Attack
Offensive Approach
Approach
Retreat


66
approaches and the latter more attack leaps, commonly when the P.
melanotis was oriented towards another object in the arena. These
results suggest that resident £_*. melanotis could not prevent
individuals of the other species from foraging and establishing
residency inside monarch colonies.
Discussion
Comparison of the Responses of the Five Mouse Species to the
Monarch Colonies
Even though melanotis was the most common species of mice on
Sierra Chincua and Sierra Campanario, my results suggest that
disproportionately large numbers of this species immigrated into the
colonies and established residency. Individuals of the other species
also immigrated into the colonies, but none of them established
residency, with the exception of one £L. iil. salvus. These results
contrast with those from grids outside colonies, where all five species
of mice commonly established residency. Thus it appears that all
species but melanotis left the monarch colonies soon after
encountering them.
The diets of melanotis inside colonies consisted almost entirely
of monarch material and averaged 43% lipid. Females had more than 2.5
times more lipid than conspecifics and heterospecifics outside
colonies. My data suggest that the energy and nutrients the mice
derived from the monarchs enabled them to initiate high levels of
winter reproduction, confirming the findings of Brower et al. (1985).
Similar responses to natural and experimentally created food abundance
in winter have been reported in other rodent populations (Linduska


67
1942, Watts 1970, Andrzejewski 1975, Gashweiler 1979, Taitt 1981,
Jenson 1982, Eriksson 1984, Briggs 1986) Moreover, diets with
comparable levels of fat (40 to 60%) were found to stimulate
reproduction and reduce infant mortality in laboratory rats (Scheer et
al. 1947, Innami et al. 1973).
Even though the £^_ melanotis and other species outside the colonies
bred during the winter, the vast majority did not begin until late
February and it is notable that a much larger percentage of the female
melanotis initiated reproduction than did females of the other
species. This may be due to supplemental nutrients derived from
monarch material, which augmented their winter diet.
Rodent populations do not normally breed in winter in temperate
and/or high altitude regions (Rintamaa et al. 1976, Millar 1984,
Bronson 1985, Kenagy and Barnes 1988), and previously studied
populations of the same species studied herein in high-altitude regions
of the Mexican Transvolcanic Range are no exception (Villa 1952, Canela
and Snchez 1984, Robertson 1975, V. Snchez pers. comm.). a variety
of factors interrelate to produce this pattern, including energy,
nutrients, photoperiod, humidity, temperature and social factors.
However, energetic constraints are thought to play a major role, owing
to the high costs of thermoregulation, searching for scarce food, and
reproduction (Schipp et al. 1963, Sadlier et al. 1973, Stebbins 1977,
Millar 1979, Porter and McClure 1984, Bronson and Perrigo 1987, Perrigo
1987). Apparently, gaining access to monarchs allows melanotis to
overcome these energetic constraints and breed during the winter.


68
Demographic Responses of P. melanotis to the Monarch Colonies
Peak MNA estimates of mouse densities inside colonies were much
higher (50 to 97/ha) than those outside (13 to 32/ha) and those
reported for other Peromyscus populations whose diets had been
supplemented artificially (13 to 59/ha; Fordham 1971, Hansen and Batzli
1978, Gilbert and Krebs 1981, Taitt 1981, Briggs 1986, Young and Stout
1986, Wolff 1986). The unusual densities inside colonies is
attributable almost exclusively to adult immigration. Such high
densities and rates of immigration must have created high levels of
intraspecific aggression, associated with the defense of core areas
against intruders (Watson and Moss 1970, Wolff et al. 1983, Wolff
1985). In this study, I found same-sexed, resident melanotis to be
extremely aggressive towards conspecific intruders (Figure 3-6, Table
3-13).
Even with this aggression, home ranges of 2_*. melanotis inside
colonies were not significantly smaller than those of conspecifics
outside colonies. This suggests that neither food abundance nor
population density are major factors determining home range size in
these mice. In other studies of £_*. leucopus and maniculatus
populations, workers have drawn both similar (Stickel 1960, Sheppe
1966, Hansen and Batzli 1978, Wolff 1985, Wolff 1986) and contrasting
(Bendell 1959, Taitt 1981) conclusions. Clearly, the factors governing
home range size in E^_ melanotis deserve further study, particularly
since the average home range sizes of both sexes are considerably
smaller than those reported for other Peromyscus species (Taitt 1981,
Wolff 1985, Wolff 1986, Vessey 1987).


69
In this study and that of Brower ££. al. (1985), immigration to
monarch colonies was female-biased, and the females inside colonies
were significantly larger than the males both inside and outside
colonies. Whereas several workers have reported similar female-biased
responses to high quality habitats in other Peromyscus populations
(Bowers and Smith 1979, Fordham 1971), many other workers have not
found such a response (review in Vessey 1987). The reason for the
preponderance of females inside the colonies is unclear. Because
females had smaller home ranges than did males, they may have been able
to pack themselves more tightly within colonies without extensive home
range overlap. Second, the females may limit male immigration as
suggested for several other Peromyscus species (Metzgar 1971, Bowers
and Smith 1979) and Tamias striatus (Wolfe 1966, Brenner et al. 1978).
Unfortunately, I did not examine the intersexual, agonistic relations
of £_*. melanotis. Third, it is possible that the natural sex-ratio on
Sierra Chincua was female-biased. However, this is unlikely given that
the mean percentage of females (+S.E.) in my 23 captive litters of P.
melanotis (from 15 female and 14 male adult parents) collected on
Sierra Chincua was 50.1 14.0.
The high densities of breeding females inside colonies also may
have contributed to the low levels of juvenile recruitment. Other
workers have reported negative correlations between juvenile densities
and those of breeding females in Peromyscus populations, and they
concluded that breeding females had aggressively excluded young mice
from their home ranges (Hansen and Batzli 1978, Galindo and Krebs
1987). Clearly, the role of aggression by breeding females towards


70
conspecific males and juveniles in mediating access to high quality
resources deserves further investigation.
Consumption of Monarchy
I was surprised by the relatively large amounts of monarch material
present in the stomachs of £*. melanotis collected outside colonies.
Brower et. lL. (1985) also found monarch material in the stomachs of P.
melanotis from outside colonies, but did not quantify the amount.
There are two ways in which these mice could have accessed monarch
material: 1) by making nightly forays into the nearby colony; and/or 2)
by scavenging within their own home ranges on dead or moribund
monarchs. Many butterflies die outside colonies because of bird
predation, exhaustion due to inadequate lipid reserves, and dehydration
(Walford 1980, Brower 1985). The second possibility is most likely
because my trappability estimates in grids outside colonies were so
high and I did not record any mice moving from grids outside to ones
inside colonies. Therefore, there must have been a substantial number
of dead and/or moribund monarchs on the forest floor each night in the
grids outside of the colonies. This implies that all species of mice,
except for £. melanotis. not only avoided the colonies, but also
avoided the dead and/or moribund monarchs in the grids outside of the
colonies.
I can offer two explanations for why the stomach contents of P.
melanotis outside monarch colonies contained significantly higher
concentrations of CGs than did those of conspecifics inside colonies.
First, because the density of monarchs is so much higher inside
colonies, mice foraging there could feed selectively on monarchs having


71
low levels of CGs with limited searching costs. Caged melanotis are
able to distinguish between monarchs with low and high levels of CGs
(Chapters 4 and 5). For those mice outside colonies, the caloric costs
of extensive searching for low CG monarchs would have been much higher
and thus may not have been energetically worthwhile. Second, when
offered hydrated monarchs (i.e., ones that are live, moribund, or
recently dead), captive EL*. melanotis most commonly consume the
abdominal material by first discarding the cuticle and then eating the
internal tissues (Chapter 3). Monarch cuticle is known to contain high
concentrations of CGs (Brower et al. 1988). In contrast, when offered
desiccated monarchs, captive £^_ melanotis eat both the cuticle and
contents, apparently because the cuticle became tightly bound to the
internal tissues during the desiccation process (Chapter 7, Brower et
al. 1988) Therefore, given that many desiccated monarchs occur
outside colonies (Glendinning pers. observ.), and that hydrated ones
are common inside colonies (Table 3-3), mice that were foraging outside
colonies may have ingested higher concentrations of cuticle, and hence
of CGs.
Why is P. melanotis the only species of mice that feeds on the monarch
butterflies?
I tested the hypotheses that the four other species of mice did not
eat monarchs because 1) the microhabitat characteristics (e.g..
understory vegetation) of overwintering areas may have only suited P.
melanotis. or 2) melanotis competitively excluded them. I rejected
the first hypothesis because the understory vegetation patterns of
grids inside colonies were not significantly different from those of
grids outside colonies. I also rejected the second hypothesis because


72
resident melanotis were unable to dominate nu. salvus and a .
hylocetesf and were able to dominate only about half of the R.
sumichrasti. This indicated that resident melanotis could not
prevent any of the species from feeding on the monarchs and
establishing residency inside the colonies.
The results reported herein and in Chapters 4 to 6 suggest that all
species of mice except E^. melanotis avoided the colonies because of an
aversion to the monarchs. I examine the feeding responses of P-
melanotis. £_*. sumichrasti. E^. hylocetes and nu. salvus to
overwintering monarchs and their defensive compounds in the next
chapter.


CHAPTER 4
COMPARATIVE FEEDING RESPONSES TO OVERWINTERING MONARCH BUTTERFLIES
To compare the feeding responses of £_*. melanotis. P. a. hylocetes.
E^. sumichrasti. and EL. HL*. salvus to overwintering monarchs, I conducted
3 experiments with caged individuals. I measured how many monarchs
each species ate and the degree to which they avoided cuticular
material. Monarchs store higher concentrations of CGs in their cuticle
compared to their body contents (Table 4-1; Brower et al. 1988) .
Second, I further explored CG avoidance in melanotis. which feeds
naturally on monarchs, by offering them a choice between male and
female monarchs. Male monarchs have on average 30% lower CG
concentrations than females (Brower and Calvert 1985). I compared how
many abdomens of each sex were eaten and the extent to which each
sexes' cuticle was avoided. Third, I examined how well all four
species could maintain weight on a diet consisting solely of monarchs.
Methods
Trapping and Maintenance of Mice
I trapped and conducted experiments from 20 January through 28
February 1986 on the slope of the Arroyo La Plancha of the Sierra
Chincua mountain massif (Figure 3-la). Mice were trapped in Sherman
live traps (7.6 x 8.9 x 22.9 cm) at least 800 m from the monarch colony
to reduce the possibility that they had previously encountered a
73


74
monarch colony. Because of a high variance in food consumption in
reproductive mice, particularly females (Sadlier et al. 1973, Stebbins
1977, Millar 1979), only reproductively inactive adults (non-lactating
or non-pregnant females and non-scrotal males) were used in the
experiments. Each mouse was used only once.
The mice were housed and tested individually in wire mesh cages
(about 30 cm high x 25 cm in diameter) set on a tarpaulin-covered table
in the shade. Dacron batting was added for nesting material and pieces
of cardboard were placed between the cages to isolate the mice
visually. For six consecutive nights prior to the feeding experiments,
each mouse was maintained on Purina laboratory chow no. 5001
(henceforth, mouse chow) and water. All of them fed and drank
regularly and either maintained or gained weight.
Monarch Collection
Monarchs were collected with butterfly nets from accessible
clusters on fir branches within the butterfly colony. For Experiments
1 and 3, monarchs were offered to mice without reference to their sex.
Branch clusters were about 58% females during January and February (n =
7 samples, for a total of 1095 butterflies; T. Van Hook unpubl. data).
For Experiment 2, equal numbers of each sex were offered to mice.
Experiment 1: Patterns of Feeding by Four Mouse Species
The feeding responses of the 4 species to monarchs were compared.
Each mouse was weighed at the beginning of the experiment. Then 4
males and 4 females of each species were individually caged and each
was offered 40 monarchs, mouse chow d libitum, and water for two


75
consecutive nights. In this and the next experiment, all mice received
live (and active) monarchs, except for nu. salvus. which received
inactive monarchs (i.e., ones whose thorax had been squeezed firmly)
because they would not approach live ones. At 0900 h all mouse chow
was removed from each mouse's cage. At 1900 h on the same day, the
monarchs and fresh mouse chow were given to each mouse. At 0900 h the
following morning, each mouse was removed from its cage to tally
patterns of feeding damage to dead monarchs, then replaced and deprived
of food until 1900 h that day, at which time the feeding trial was
repeated.
I determined from preliminary feeding trials that mice consumed
variable portions of thoraces, but always ate the cuticle together with
the contents. In contrast, when feeding upon abdomens, they
characteristically ate either 100% of the cuticle and contents (i.e.,
fed non-selectively) or 100% of the contents only and discarded the
cuticle (i.e., fed selectively). Based on these observations, I chose
. priori 6 categories of monarch damage (Table 4-1). I assume that
sampling a monarch represents active rejection of it based on taste.
Determination of the amount of monarch tissue, cuticle, and CGs
eaten by each mouse in Experiments 1 and 3 involved several steps.
First, I estimated the amount of tissue, cuticle and CGs in the body
parts of overwintering monarchs (see Table 4-2). For these
estimations, I modified the data from Brower £t. al. (1988) because they
were derived from freshly-eclosed, Asclepias syriaca-reared monarchs,
which have higher CG contents and substantially lower amounts of fat in
their abdomens than monarchs reared on the same food plant that have
migrated to the Mexican overwintering grounds (Brower 1985, Malcolm and


76
Table 4-1. Categories of monarch damage by mice.
1. No visible damage
2. Sampled: <25% of abdomen and/or thorax eaten
3. 25-50% (cuticle and contents) of thorax eaten
4. 51-100% (cuticle and contents) of thorax eaten
5. Abdomen eaten non-selectively (> 25% of cuticle and contents)
6. Abdomen eaten selectively (> 25% of contents; cuticle rejected)


77
Brower 1989) Second, I assumed that whenever a mouse ate 1) an
abdomen non-selectively, it ate 100% of the cuticle and contents, 2) an
abdomen selectively, it ate 100% of the contents, 3) 25 to 50% of a
thorax, it ate 37.5% of the cuticle and contents, and 4) 51 to 100% of
a thorax, it ate 75% of the cuticle and contents. These assumptions
agreed generally with my preliminary observations. Third, I calculated
the total quantity of tissue, cuticle and CGs eaten by a given mouse
over a given time period with the following equations, which
incorporate data from Table 4-2:
tissue eaten = [(no. abdomens eaten selectively and non-selectively)
(123 mg)] + [(no. thoraces eaten 25-50%) (37.5%) (41 mg)]
+ [(no. thoraces eaten 51-100%)(75%)(41 mg)];
cuticle eaten = [(no. abdomens eaten non-selectively)(11 mg)] +
[(no. thoraces eaten 25-50%) (37.5%) (22 mg)] +
[(no. thoraces eaten 51-100%) (75%) (22 mg) ] ;
CG eaten = (no. abdomens eaten selectively)(10 |lg)] +
[(no. abdomens eaten non-selectively) (10 + 24 ]lg) ] +
[(no. thoraces eaten 25-50%) (37.5%) (10 + 11 (ig) ] +
(no. thoraces eaten 51-100%) (75%) (10 + 11 [ig) ] .
Fourth, I added together the predation records from both nights in
Experiment 1, whereas for Experiment 3, I analyzed each night
separately so as to monitor changes in consumption of monarch tissue,
cuticle and CGs over the 6 nights.


Table 4-2. Mean dry weights and CG contents of the abdominal cuticle and tissue, thoracic cuticle
and tissue, and wings of freshly-eclosed monarchs reared on Asclepias syriaca (from Brower et al.
1988: Table 3). These data, along with the mean dry weights and CG contents (all body parts
combined) of monarchs overwintering in Mexico (Malcolm and Brower 1989), were used to estimate the
dry weights and CG contents of the different body parts in overwintering monarchs. Data from A.
syriaca-reared monarchs are appropriate for these estimations because the evidence suggests that 85
to 92% of the overwintering monarchs in Mexico have fed as larvae on this milkweed (Seiber et al.
1986; Malcolm et al. 1989). I assume that: 1) the dry weights of the cuticle, wings and thoracic
tissue were the same in both types of monarchs, and that only the abdominal tissue weights differed,
owing to increased amounts of fat associated with diapause (Brower 1985); and 2) the distribution of
CGs in the different body parts was the same in both types of monarchs, but that the CG contents were
proportionally lower
in
each body
part in
overwintering
monarchs.
Freshly-eclosed monarchs
Estimated
values
for overwintering monarchs
Dry
weight3
CG content3
Dry
weight
CG
content
CG concentration
Body Part
mg
percent
ug
percent
mg
percent
ng
percent
\lq/0 lg d. w.
Abdominal cuticle
ii
6%
91
26%
ii
5%
24
26%
218
Abdominal tisssue
70
37%
37
11%
123
51%
10
11%
8
Thorax cuticle
22
11%
41
12%
22
9%
11
12%
50
Thorax tissue
41
22%
37
11%
41
17%
10
11%
24
Wings
45
24%
145
41%
45
19%
39
41%
87
Continued.


Table 4-2Continued.
Contents of
abdomen + thorax
111
59%
74
21%
165
68%
20
21%
12
Wings + cuticle
78
41%
277
79%
78
32%
74
79%
95
Sum of all parts
189
100%
351
100%
242b
100%
94b
100%
39
a Values are rounded to the nearest whole value.
13 These are means derived from 563 overwintering monarch butterflies at Sierra Chincua
(Malcolm and Brower, 1989).


80
Because the 4 species of mice differed greatly in size (Table 4-
6), I could not compare directly the amounts of monarch tissue, cuticle
or CG eaten. Therefore, I standardized consumption by computing the
0 7
ratio of the weight of tissue, cuticle, or CG eaten to mouse weight .
A n
The power function, weight is derived from a regression of
consumption rate on weight for a wide range of mammals (Farlow 1976,
Peters 1983). Mass-specific consumption values were calculated
separately for each mouse, using their weight at the beginning of the
experiment.
Experiment 2: Predation by P. melanotis in Relation to Butterfly Sex
The previous experiment examined whether either of the species
discriminated between abdominal cuticle and contents. This experiment
addressed a more complex question: When given a choice between monarchs
of differing CG concentrations (in this case males versus females),
will £_*. melanotis eat more low CG ones and feed selectively on more of
the high CG ones? Male monarchs have on average 30% lower CG
concentrations than females (Brower and Calvert 1985). I could not
estimate separately for both sexes the amounts of tissue, cuticle and
CGs eaten because the data were not available; those in Table 4-2 are
from pooled samples of male and female monarchs.
Seven female and 8 male melanotis were each offered 25 female
and 25 male monarchs and mouse chow and water d libitum per night for
two consecutive nights. Five hours before each feeding trial, I
squeezed the thoraces of the female and male monarchs and then placed a
1 mm diameter, red or gold spot of nail polish on each butterfly
abdomen, thorax, head, and wings so as to enable tallying of the parts


81
by sex after the mice dismembered them. The assignment of gold or red
for males or females was determined randomly for each mouse. Because
sexual differences in wing length are negligible and wet weights vary
unpredictabilty (Brower and Calvert 1985), I considered size and weight
unlikely bases for killing an excess of one sex.
Experiment 5: Effects of Long-Term Consumption of Monarchs
In this experiment, I explored whether all 4 species could maintain
or gain weight on a pure diet of monarchs. Each mouse (6 Ej_ melanotis.
5 £u. sumichrasti. 4 L. 1. hylocetes. and 5 L. IL salvua; sex-ratios
were approximately equal) was pre-exposed to 40 monarchs and mouse chow
ad 1ibitum per night for 2 nights. Then for the next 6 nights, each
mouse received 55 monarchs per night and water ad libitum, without
mouse chow. Preliminary feeding trials indicated that all species ate
fewer than this number of monarchs per night. Each mouse's weight was
taken at the beginning and end of the experiment, 8 days later. All
mice were tested between 7 and 21 February.
Statistical Analyses
All statistical tests followed Zar (1984) In Experiment 1, paired
(2-tailed) t-tests were run separately on each species to compare the
number of abdomens its., thoraces eaten, and the number of abdomens eaten
selectively xs.. non-selectively. One-way ANOVA's and Scheff F-tests
were used to compare the species in terms of the mass-specific amounts
of monarch tissue, cuticle or CGs eaten, as well as the ratios of the
amounts of CGs to tissues eaten. I was justified in using the mass-
specific consumption values in the ANOVA's because the coefficient of


82
variation of the scaling variable (i.e., mouse weight) was consistently
less than that of the dependent variable (i.e., amount of tissue,
cuticle or CG consumed)(Anderson and Lydic 1977, Packard and Boardman
1988) The unpaired t-test was used to compare the percentage of
abdomens eaten selectively by £_*. melanotis and R_*. sumichrasti. Data
were transformed (arcsin Vx) for this and all subsequent statistical
comparisons of percentage data.
Paired t-test comparisons were used in Experiment 2 to compare the
total male and female monarch abdomens eaten (selectively and non-
selectively), as well as the percentage of those abdomens that were
eaten selectively.
Two-factor ANOVA's (repeated on time) were used in Experiment 3 to
determine the effects of species (E_^ melanotis. hylocetes. and £_*.
sumichrasti) and time on tissue, cuticle, and CG consumption. The same
test was used again to determine the effect of species and time on the
percentage of abdomens eaten selectively by melanotis and
sumichrasti. M. m. salvus was excluded from all statistical analyses
in this experiment.
Results
Experiment 1: Patterns of Feeding by all Four Mouse Species
Peromyscus melanotis. sumichrasti and hylocet,a.S each ate
more than 52% of the monarchs offered to them (Table 4-3) and
significantly more abdomens than thoraces (Figure 4-1). In contrast,
M. m. salvus ate less than 13% of the monarchs offered to them and
virtually identical numbers of abdomens and thoraces. Moreover, nu.


Table 4-3. Comparison of the feeding patterns of all 4 mouse species upon monarch butterflies. Eight
individuals per species were each offered 40 monarchs and mouse chow d libitum per night for two
consecutive nights. For each species, the mean number of monarchs eaten, sampled or left undamaged (
S.E) by each species are presented. Because mice ate both the abdomen and thorax of some monarchs, the
row totals are all greater than 80.
Mean
no. of monarchs S.E. in each feeding category
No visible
Thoraces eaten: Abdomens eaten:
No.
offered/
Species
damage
Sampled 25-50% 51-100% non-select, selectively
mouse
melanotis
14.3
3
15.4
4
5.1
2
0.0
0
18.8
1
31.3
2
80
L. sumichrasti
26.0
3
10.9
2
4.5
1
0.9
.4
19.3
2
23.3
1
80
£j. aztecus
20.5
1
17.8
2
8.5
1
4.8
1
39.4
4
2.8
1
80
1L. salvus
57.6
5
11.9
5
4.1
2
3.8
1
10.1
1
0.1
.l
80
oo
oo




mm
Figure 4-1. Mean number of monarch abdomens and thoraces (+S.E.) eaten by 8 individuals of each
of 4 mouse species. Each mouse received 40 monarchs and mouse chow libitum per night for two
consecutive nights. The significant within species comparisons between the number of abdomens and
thoraces eaten are indicated by asterisks.


Mean number eaten
55 -
50 -
45 -
40 -
35 -
30 -
25 -
20 -
15 -
10 -
5-
0
P.
melanotis
*
*
R.
sumichrasti
Abdomens
H Thoraces
**P< 0.0001 ;df =
paired t-test
7
P. a. M. m.
hylocetes salvus


86
salvus approached only immobilized monarchs. None of the mice consumed
the wings, head or legs.
The standardized consumption values are shown in Table 4-4.
Peromyscus melanotis and L. sumichrasti consumed significantly more
monarch tissue than did E_*. hylocetes and H*. Eb salvus. However,
a. hylocetes ate significantly more cuticle and both E^_ sumichrasti and
P. a. hylocetes ate more CGs than did the other 2 species. To help
interpret these patterns of consumption, I computed for each species
the ratio of the amounts of CGs to tissues eaten. The ratios for
melanotis and Ed. sumichrasti were significantly lower than those for
the other 2 species (Table 4-4).
The large interspecific differences in these ratios can be
explained by the way each mouse species consumed monarch abdomens.
Peromyscus a. hylocetes and KL. Eb salvus rarely ate abdomens
selectively, whereas E^. melanotis and E^. sumichrasti both ate
significantly more abdomens selectively than non-selectively (Figure 4-
2). Peromyscus melanotis fed selectively on a significantly greater
percentage of abdomens than did E^. sumichrasti (means = 62.2 and 55.1,
respectively; unpaired t-value = 2.89; df = 14; P < 0.013). Thus by
feeding selectively, E^. melanotis and E^. sumichrasti reduced
substantially the CG concentration of the monarch tissue they ingested.
By examining mouse-damaged monarchs and watching mice eat them, I
discovered interspecific differences in the way the four species
extracted abdominal contents. Peromyscus melanotis and E^ sumichrasti
commonly made longitudinal slits down the full length of the abdomen
and then sucked and/or licked out the abdominal contents. In contrast,


Table 4-4. Interspecific comparison of the standardized amounts of monarch tissue, cuticle, and CGs
eaten. The mean ratio of the amounts of CGs to tissue ingested ( S.E.) is also compared among
species. Eight mice from each species were each offered forty monarchs and mouse chow ad libitum
per night for two consecutive nights. Between species comparisons are made with one-way ANOVA.
Different subscripts (a, b and c) indicate significant differences among species within each column
(P < 0.05; Scheff F-test).
Mouse
species
Standardized amounts eaten (mean S.E.)
CGs/tissue
ratio*
Tissue
(g/kg0>7 mouse)
Cuticle
(g/kg0-7 mouse)
CGs
(mg/kg0-7 mouse)
E-r. melanotis
676.9 25.9 a
32.0 1.8 a
96.4 4.2
a
0.14 0.01
a
sumichrasti
703.4 22.6 a
40.9 2.6 a/ b
120.0 7.0
b
0.17 0.01
a
3^. hylocetes
378.4 13.2 b
46.1 1.8 b
126.4 4.3
b
0.33 0.01
b
£L_ nu. salvus
120.4 15.4 c
18.8 2.7 c
46.1 5.8
c
0.39 0.01
c
F-value
190.4
27.1
44.7
212.3
df
3
3
3
3
P
0.0001
0.0001
0.0001
0.0001
* Determined from individual ratios.


Figure 4-2. Mean number of monarch abdomens (S.E.) eaten selectively and non-selectively by four
mouse species (n = 8/species). Each mouse received 40 monarchs and mouse chow ad libitum per
night for two consecutive nights.


45 n
P.
melanotis
R.
sumichrasti
*
*
T
P. a.
hylocetes
|H Non-Selectively
Selectively
P < 0.0004; P < 0.05
df = 7; paired t-test
*
*
M. m.
salvus


90
a__ hylocetes and nu. salvus usually bit off the end of the abdomen
and pulled the contents out with their teeth.
Experiment 2: Predation by P. melanotis in Relation to Butterfly Sex
Peromyscus melanotis killed (feeding categories 2 -6; Table 4-1)
equal numbers of male and female monarchs, which suggests no initial
difference in the risk of male and female monarchs to mouse attack
(Table 4-5). However, adding together the number of abdomens eaten
selectively and non-selectively reveals that R^. melanotis ate a larger
number of male than female abdomens (means = 28.4 and 22.7,
respectively; paired t-value = 3.198, df = 14; P = 0.0064). Of those
abdomens eaten, R^. melanotis fed selectively on greater percentage of
female abdomens (female and male means = 68.1 and 50.0, respectively;
paired t-value = 10.63; df = 14; P = 0.0001) These results support
the hypothesis that both the quantity of tissue eaten and the tendency
to feed selectively are influenced by CG concentration.
Experiment 3: Effects of Long-Term Consumption of Monarchs
All R* melanotis gained weight over the six nights, whereas 60-100%
of the individuals of each of the other species lost weight (Table 4-
6). Microtus su. salvus were excluded from the statistical analyses
because they ate very few monarchs and all died within four days.
The mass-specific amounts of monarch tissue, cuticle and CGs eaten
nightly by R^ melanotis. hylocetes and sumichrasti are plotted
in Figure 4-3a-c. The two-factor ANOVA's performed separately on
tissue, cuticle and CGs all indicated significant effects of species
(for tissue, F = 25.5, df = 2, P < 0.0001; for cuticle, F = 6.41, df =


Table 4-5. Comparative feeding patterns by melanotis upon male and female monarch butterflies.
Fifteen mice were each offered 25 male and 25 female butterflies together with mouse chow d libitum
per night for two consecutive nights. For each monarch sex, the mean number eaten, sampled or left
undamaged ( S.E) by the 15 mice are presented. Because mice ate both the abdomen and thorax of some
monarchs, the row totals are all greater than 50.
Mean no. of monarchs S.E. in each feeding category
Thoraces
eaten:
Abdomens
; eaten:
No.
Monarch
sex
No visible
damage
Sampled
25-50% 51-100%
non-select.
selectively
offered/
mouse
male
8.5 1
12.8 1
1.8 .4
3.6 1
14.1 1
14.3 1
50
VO
female
8.7 2
17.8 2
1.4 .4
1.3 .3
7.3 1
15.4 1
50


Table 4-6. Initial weights (mean S.E.), weight changes and % mortality in P_*. melanotis. R.
sumichrasti. P. a. hylocetes and HU. salvus given 55 monarchs per mouse per night for 6 consecutive
nights, without alternative food.
Mouse
Species
N
Initial
weight (g)
Weight
change (g)
% weight
change
% of mice
that lost
weight
%
mortality
melanotis
6
19.6
(0.4)
+0.83 (0.2)
+4.2
0
0
Rj. sumichrasti
5
13.4
(0.3)
-0.38 (0.2)
-2.8
60
0
P. a. hvlocetes
4
36.8
(+0.6)
-1.70 (0.7)
-4.6
75
0
1L. EL. salvus
5
30.0
(0.6)
-4.30 (0.7)
-14.3
100
100*

All died within 4 days.


Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EW7DIXER6_QXIQ0S INGEST_TIME 2011-11-08T19:49:34Z PACKAGE AA00004780_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES



PAGE 1

&203$5$7,9( 5(63216(6 2) ),9( 6<03$75,& 63(&,(6 2) 0,&( 72 29(5:,17(5,1* &2/21,(6 2) 021$5&+ %877(5)/,(6 ,1 0(;,&2 %\ -2+1 ,1*(562// */(1',11,1* $ ',66(57$7,21 35(6(17(' 72 7+( *5$'8$7( 6&+22/ 2) 7+( 81,9(56,7< 2) )/25,'$ ,1 3$57,$/ )8/),//0(17 2) 7+( 5(48,5(0(176 )25 7+( '(*5(( 2) '2&725 2) 3+,/2623+< 81,9(56,7< 2) )/25,'$

PAGE 2

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
PAGE 3

:LOOLDP &DOYHUW &DUORV *DOLQGR 9LFWRU 6£QFKH] 0DUN 6WHOOMHV DQG -DFTXHOLQH 5R\ IRU WKH XQSXEOLVKHG GDWD DQG WKH 'HSW RI =RRORJ\ 6LJPD ;L 6RFLHW\ DQG ([SORUHUnV &OXE IRU ILQDQFLDO DVVLVWDQFH /DVW EXW QRW OHDVW WKDQN 'LDQD 6FKXOPDQQ IRU KHU HGLWRULDO KHOS XQIDLOLQJ VXSSRUW DQG ORYH WKURXJKRXW WKH JUHDWHU DQG OHVVHU PRPHQWV RI P\ GRFWRUDO SURJUDP

PAGE 4

$EVWUDFW RI 'LVVHUWDWLRQ 3UHVHQWHG WR WKH *UDGXDWH 6FKRRO RI WKH 8QLYHUVLW\ RI )ORULGD LQ 3DUWLDO )XOILOOPHQW RI WKH 5HTXLUHPHQWV IRU WKH 'HJUHH RI 'RFWRU RI 3KLORVRSK\ &203$5$7,9( 5(63216(6 2) ),9( 6<03$75,& 63(&,(6 2) 0,&( 72 29(5:,17(5,1* &2/21,(6 2) 021$5&+ %877(5)/,(6 ,1 0(;,&2 %\ -RKQ ,QJHUVROO *OHQGLQQLQJ $XJXVW &KDLUSHUVRQ /LQFROQ 3 %URZHU 0DMRU 'HSDUWPHQW =RRORJ\ FRPSDUHG ILYH VSHFLHV RI V\PSDWULF PLFH 3HURP\VFXV PHODQRWLV (} DD/6&86 K\ORFHWHV (HLWKURGRQWRP\V VXPLFKUDVWL 0LFURWXV PH[LFDQXV DDO\XD DQG 1HRWRPRGRQ DOVWRQL DOVWRQLf LQ WHUPV RI WKHLU DELOLW\ WR WDNH DGYDQWDJH RI RYHUZLQWHULQJ FRORQLHV RI PRQDUFK EXWWHUIOLHV 'DQDXV SOH[LSSXVf LQ 0H[LFR 7R HDW WKLV VXSHUDEXQGDQW IRRG UHVRXUFH WKH PLFH PXVW RYHUFRPH WKH PRQDUFKnV FKHPLFDO GHIHQVH V\VWHP ZKLFK FRQVLVWV RI ELWWHUWDVWLQJ DQG SRWHQWLDOO\ WR[LF FDUGLDF JO\FRVLGHV &*Vf DQG S\UUROL]LGLQH DONDORLGV 3$Vf 3HURP\VFXV PHODQRWLV DSSHDUHG WR EH WKH RQO\ VSHFLHV RI PRXVH WKDW EUHDFKHG WKHVH GHIHQVHV ODUJH QXPEHUV RI eBr PHODQRWLV LPPLJUDWHG WR WKH FRORQLHV IHG RQ WKH PRQDUFKV DQG LQLWLDWHG ZLQWHU UHSURGXFWLRQ ,Q FRQWUDVW WKH RWKHU VSHFLHV RI PLFH DSSHDUHG WR DYRLG WKH FRORQLHV 2Q JULGV RXWVLGH WKH FRORQLHV LQGLYLGXDOV RI DOO ILYH VSHFLHV ZHUH FRPPRQ EXW UHSURGXFWLYHO\ LQDFWLYH WKURXJKRXW PRVW RI LY

PAGE 5

WKH ZLQWHU 7R H[SODLQ WKHVH GHPRJUDSKLF UHVXOWV K\SRWKHVL]HG WKDW (A PHODQRWLV f ZDV WKH RQO\ VSHFLHV WROHUDQW WR WKH PLFURKDELWDW IHDWXUHV RI WKH RYHUZLQWHULQJ VLWHV f DJJUHVVLYHO\ H[FOXGHG WKH RWKHU VSHFLHV DQG f ZDV WKH RQO\ VSHFLHV WKDW WROHUDWHG WKH PRQDUFKnV GHIHQVLYH FRPSRXQGV %HFDXVH WKH XQGHUVWRU\ YHJHWDWLRQ SDWWHUQV ZHUH QRW VLJQLILFDQWO\ GLIIHUHQW IURP WKRVH RXWVLGH FRORQLHV UHMHFWHG WKH ILUVW K\SRWKHVLV 7KH VHFRQG K\SRWKHVLV ZDV UHMHFWHG EHFDXVH UHVLGHQW (A PHODQRWLV ZHUH XQDEOH WR GRPLQDWH WKH RWKHU VSHFLHV LQ FDSWLYH DJRQLVWLF HQFRXQWHUV 7KH WKLUG K\SRWKHVLV ZDV DFFHSWHG EHFDXVH RQO\ (AB PHODQRWLV ZDV DEOH WR f WKULYH RQ D GLHW RI SXUH PRQDUFKV DQG f OHDUQ KRZ WR UHMHFW WKH &*ODGHQ FXWLFOH DQG IHHG VHOHFWLYHO\ RQ WKH ORZ&* LQWHUQDO WLVVXHV 7R GHWHUPLQH ZKHWKHU 6A PHODQRWLV ZDV XQLTXHO\ DEOH WR RYHUFRPH WKH WDVWH DQG WR[LFLW\ RI WKH PRQDUFKnV GHIHQVLYH FRPSRXQGV VHYHUDO IXUWKHU IHHGLQJ H[SHULPHQWV ZHUH LQLWLDWHG WR FRPSDUH (A PHODQRWLV ZLWK (AB K\ORFHWHV DQG (A VXPLFKUDVWL 1RQH RI WKH VSHFLHV ZDV VHQVLWLYH WR WKH WR[LF HIIHFWV RI GLJLWR[LQ D &*f ZKLFK VXJJHVWV WKDW DOO FRXOG KDYH LQJHVWHG WKH PRQDUFKnV &*V ZLWK LPSXQLW\ +RZHYHU (Br K\ORFHWHV DQG VXPLFKUDVWL ZHUH PXFK PRUH VHQVLWLYH WKDQ (Br PHODQRWLV WR WKH WDVWH RI GLJLWR[LQ VXJJHVWLQJ WKDW WKH GLYHUJHQW GHPRJUDSKLF UHVSRQVHV RI WKH GLIIHUHQW PRXVH VSHFLHV FDQ EH H[SODLQHG E\ WKHLU GLIIHUHQWLDO UHVSRQVHV WR WKH ELWWHU WDVWH RI WKH &*ODGHQ PRQDUFK EXWWHUIOLHV 0RQRFURWDOLQH D 3$f GLG QRW VHHP WR KDYH DQ LQIOXHQFH RQ WKH IHHGLQJ EHKDYLRU RI HLWKHU VSHFLHV

PAGE 6

7$%/( 2) &217(176 3DJH $&.12:/('*(0(176 LL $%675$&7 YL &+$37(56 *(1(5$/ ,1752'8&7,21 2YHUZLQWHULQJ %LRORJ\ RI 0RQDUFK %XWWHUIOLHV LQ 0H[LFR .QRZQ DQG 3RWHQWLDO 3UHGDWRUV RI 2YHUZLQWHULQJ 0RQDUFKV 2YHUYLHZ RI 'LVVHUWDWLRQ '()(16,9( &203281'6 ,1 7+( 021$5&+ %877(5)/< &DUGLDF *O\FRVLGHV &KHPLVWU\ DQG 7R[LFRORJ\ 'HIHQVLYH 9DOXH 3\UUROL]LGLQH $ONDORLGV &KHPLVWU\ DQG 7R[LFRORJ\ 'HIHQVLYH 9DOXH '(02*5$3+,& 5(63216(6 72 &2/21,(6 2) 021$5&+ %877(5)/,(6 0HWKRGV 5HVXOWV 'LVFXVVLRQ &203$5$7,9( )((',1* 5(63216(6 72 29(5:,17(5,1* 021$5&+6 0HWKRGV 5HVXOWV 'LVFXVVLRQ &2168037,21 $1' $92,'$1&( 2) 021$5&+6 7+( ,1)/8(1&( 2) 021$5&+ $%81'$1&( $1' &$5',$& */<&26,'( &21&(175$7,21 0HWKRGV 5HVXOWV 'LVFXVVLRQ

PAGE 7

&203$5$7,9( 7$67( $1' 72;,& 5(63216(6 72 &$5',$& */<&26,'(6 $1' 3<552/,=,',1( $/.$/2,'6 0HWKRGV 5HVXOWV 'LVFXVVLRQ %(+$9,25$/ $1' (&2/2*,&$/ ,17(5$&7,216 2) )25$*,1* 3 0(/$127,6 :,7+ 29(5:,17(5,1* 021$5&+6 0HWKRGV 5HVXOWV 'LVFXVVLRQ *(1(5$/ &21&/86,216 ,PSDFW RI 0RXVH 3UHGDWLRQ RQ 2YHUZLQWHULQJ 0RQDUFK %XWWHUIOLHV :K\ LV (M PHODQRWLV WKH RQO\ 0RXVH 6SHFLHV 7KDW )HHGV RQ WKH 0RQDUFKV" $UH 3RSOXDWLRQV RI PHODQRWLV LQ WKH 2YHUZLQWHULQJ $UHDV 6SHFLILFDOO\ $GDSWHG IRU 2YHUFRPLQJ WKH 0RQDUFKnV 'HIHQVLYH &RPSRXQGV %HQHILWV DQG &RVWV RI %HLQJ 7ROHUDQW WR %LWWHU )RRGV $33(1',&(6 $ 3$,5(' &203$5,6216 2) 7+( 35(&(17 6,0,/$5,7< 2) 81'(56725< 9(*(7$7,21 21 *5,'6 /2&$7(' ,16,'( $1' 2876,'( 7+( 021$5&+ &2/21,(6 % 3(5&(17$*( ,03257$1&( 9$/8(6 )25 7+( 63(&,(6 2) 81'(56725< 9(*(7$7,21 21 7+( (,*+7 75$33,1* *5,'6 /2&$7(' ,16,'( 25 2876,'( $ 021$5&+ &2/21< 21 6,(55$ &+,1&8$ 5()(5(1&(6 %,2*5$3+,&$/ 6.(7&+

PAGE 8

&+$37(5 *(1(5$/ ,1752'8&7,21 $UWKURSRGV SRVVHVV D GLYHUVH DUUD\ RI FKHPLFDO GHIHQVHV WKDW GHWHU VLFNHQ RU HYHQ NLOO YHUWHEUDWH SUHGDWRUV (LVQHU 'XIIH\ %OXP 3DVWHLV HW DO f +RZHYHU WKH DFWXDO HIIHFWLYHQHVV RI WKHVH GHIHQVHV LQ QDWXUDOO\ RFFXUULQJ SUHGDWRUSUH\ FRPSOH[HV LV ODUJHO\ XQNQRZQ EHFDXVH PRVW VWXGLHV KDYH EHHQ FRQGXFWHG LQ WKH ODERUDWRU\ ZLWK XQQDWXUDO SUHGDWRUV 'XIIH\ 3DVWHLV HW DO f 7KH IHZ VWXGLHV ZLWK QDWXUDO SUHGDWRUV KDYH VLPSO\ GHVFULEHG f KRZ D VLQJOH SUHGDWRU VSHFLHV UHVSRQGV LQ WKH ODERUDWRU\ WR WKH GHIHQVHV RI VHYHUDO SUH\ VSHFLHV &\U /DQJOH\ 5HGIRUG :KLWPDQ HW DO :KLWPDQ HW DO f RU f WKH H[WHQW WR ZKLFK V\PSDWULF VSHFLHV RI ELUGV LQ WKH ILHOG XWLOL]H RQH RU VHYHUDO FKHPLFDOO\ GHIHQGHG EXWWHUIO\ VSHFLHV )LQN HW DO &ROOLQV DQG :DWVRQ 6DUJHQW f $ QRWDEOH VWXG\ E\ 3HDUVRQ f H[DPLQHG KRZ HIIHFWLYHO\ OL]DUGV ELUGV DQG UREEHU IOLHV FRXOG EUHDFK WKH PXOWLSOH GHIHQVHV FKHPLFDO SK\VLFDO DQG EHKDYLRUDOf RI DGXOW WLJHU EHHWOHV +RZHYHU WR P\ NQRZOHGJH QR VWXG\ KDV FRPSDUHG H[SHULPHQWDOO\ DQ HQWLUH DVVHPEODJH RI V\PSDWULF SUHGDWRU VSHFLHV RI WKH VDPH WD[RQ LQ WHUPV RI WKHLU DELOLW\ WR RYHUFRPH WKH FKHPLFDO GHIHQVHV RI D VLQJOH VKDUHG SUH\ VSHFLHV

PAGE 9

0DPPDOV DUH WKRXJKW WR KDYH HYROYHG D YDULHW\ RI DWWULEXWHV IRU GHDOLQJ ZLWK SRWHQWLDOO\ SRLVRQRXV IRRGV (LVQHU )UHHODQG DQG -DQ]HQ %URZHU f 7KH\ LQFOXGH WKH DELOLW\ WR f OHDUQ IURP WKHLU PRWKHU ZKLFK IRRGV DUH VDIH *DOHI 9DXJKDQ DQG &]DSOHZVNL f f GHWR[LI\ FHUWDLQ FRPSRXQGV HJ )UHHODQG DQG -DQ]HQ &DUOVRQ DQG %UHH]H f f IHHG VHOHFWLYHO\ RQ QXWULWLRXV WLVVXHV DQG UHMHFW XQSDODWDEOH DQGRU SRWHQWLDOO\ WR[LF SDUWV RI D IRRG LWHP %HQVRQ DQG %RUHOO +DPLOWRQ .HQDJ\ &\U 9DXJKDQ DQG &]DSOHZVNL %URZHU HW DO 5R\ DQG %HUJHURQ XQSXKO GDWDf f DYRLG IRRGV E\ GHYHORSLQJ D FRQGLWLRQHG IHHGLQJ DYHUVLRQ &)$f WR WKHP 5]VND *DUFLD DQG .RHOOLQJ %URZHU .DODW DQG 5R]LQ 5REELQV -DFREV DQG /DERZV 'DO\ HW DO %URZHU DQG )LQN f DQG f DYRLG DOO ELWWHUWDVWLQJ IRRGV %DWH6PLWK f 7KLV ODWWHU DWWULEXWH LV GHULYHG IURP WKH REVHUYDWLRQ WKDW PDQ\ PDPPDOV DUH DYHUVH WR D YDULHW\ RI ELWWHU FRPSRXQGV DW H[WUHPHO\ ORZ FRQFHQWUDWLRQV 3DWWRQ DQG 5XFK &DUSHQWHU &KU]DQRZVNL 3UHVVPDQ DQG 'RROLWWOH *DQFKURZ -DFREV HW DO f 9LUWXDOO\ DOO SRLVRQV RI ELRORJLFDO RULJLQ DUH ELWWHUWDVWLQJ WR KXPDQV 5LFKWHU %DWH6PLWK *DUFLD DQG +DQNLQV %URZHU f 7KXV D JHQHUDO DYHUVLRQ WR ELWWHU IRRGV PD\ EH DQ HIIHFWLYH ZD\ WR DYRLG LQJHVWLQJ VXFK FRPSRXQGV 7KH UHODWLYH LPSRUWDQFH RI WKHVH ILYH DWWULEXWHV WR IRUDJLQJ ZLOG DQLPDOV LV XQNQRZQ 7KLV LV EHFDXVH LQ PRVW FDVHV HDFK DWWULEXWH KDV EHHQ VWXGLHG LQ LVRODWLRQ IURP WKH RWKHUV ZLWK RQO\ RQH VSHFLHV RI SUHGDWRU ,Q WKLV GLVVHUWDWLRQ ZLOO H[DPLQH WKH UHODWLYH LPSRUWDQFH RI IHHGLQJ VHOHFWLYHO\ &)$V DQG DYRLGLQJ ELWWHUWDVWLQJ IRRGV LQ GHWHUPLQLQJ WKH UHVSRQVHV RI VSHFLHV RI ZLOG PLFH WR WKH YDVW

PAGE 10

RYHUZLQWHULQJ FRORQLHV RI PRQDUFK EXWWHUIOLHV 3DQDQD SOH[L SSQV /f WKDW IRUP HDFK \HDU LQ 0H[LFR 2YHUZLQWHULQJ %LRORJ\ RI 0RQDUFK %XWWHUIOLHV LQ 0H[LFR (DFK IDOO YLUWXDOO\ WKH HQWLUH HDVWHUQ 1RUWK $PHULFDQ SRSXODWLRQ RI PRQDUFK EXWWHUIOLHV PLJUDWHV WR VHYHUDO UHPRWH KLJKDOWLWXGH RYHUZLQWHULQJ VLWHV LQ FHQWUDO 0H[LFR %URZHU DQG &DOYHUW &DOYHUW DQG %URZHU f 7KHUH ZLWKLQ WKH R\DPHO ILU IRUHVWV $ELHV UHOLJLRVD +%.f WKH\ IRUP FRORQLHV RI WHQV RI PLOOLRQV RI ? LQGLYLGXDOV 7KH FRORQLHV UDQJH LQ VL]H IURP WR KD DWWDLQ GHQVLWLHV RI DERXW PLOOLRQ SHU KD FRYHU WKH WUHHV DQG XQGHUVWRU\ YHJHWDWLRQ DQG VWD\ LQ DSSUR[LPDWHO\ WKH VDPH ORFDWLRQ IURP ODWH 'HFHPEHU WKURXJK ODWH )HEUXDU\ %URZHU DQG &DOYHUW &DOYHUW DQG %URZHU f $OWKRXJK WKH FRORQLHV IRUP RQ WKH VDPH PRXQWDLQVLGHV HDFK \HDU WKH\ GR QRW DOZD\V IRUP LQ WKH VDPH ORFDWLRQ FRORQ\ VLWHV DUH IUHTXHQWO\ WR NLORPHWHUV IURP RQH DQRWKHU DPRQJ \HDUV &DOYHUW DQG %URZHU )LJXUH OD E *OHQGLQQLQJ SHUV REVHUYf ,QGLYLGXDO EXWWHUIOLHV KDYH H[WHQVLYH OLSLG UHVHUYHV PHDQ LQ -DQXDU\ b RI WRWDO GU\ ZHLJKW %URZHU f 7KH\ DUH YXOQHUDEOH WR WHUUHVWULDOO\ IRUDJLQJ SUHGDWRUV EHFDXVH ORZ WHPSHUDWXUHV FRQVWUDLQ WKH EXWWHUIOLHVn SUHGDWRU DYRLGDQFH EHKDYLRUV VXFK DV IO\LQJ EDFN WR FRPPXQDO URRVWV DQG FUDZOLQJ XS YHJHWDWLRQ &DOYHUW DQG &RKHQ 0DVWHUV HW DO $ORQVR HW DO LQ SUHVVf $V VXFK WKHVH FRORQLHV FRQVWLWXWH D SRWHQWLDO VXSHUDEXQGDQW OLSLGULFK IRRG ERQDQ]D IRU PLFH +RZHYHU RYHUZLQWHULQJ PRQDUFKV FRQWDLQ FDUGLDF JO\FRVLGHV &*Vf DQG S\UUROL]LGLQH DONDORLGV 3$Vf ERWK RI ZKLFK DUH WKRXJKW WR SURWHFW

PAGE 11

PRQDUFKV DJDLQVW PRVW YHUWHEUDWH SUHGDWRUV GHVFULEH WKH FKHPLVWU\ WR[LFRORJ\ DQG SXUSRUWHG GHIHQVLYH YDOXH RI WKHVH FRPSRXQGV LQ WKH QH[W FKDSWHU .QRZQ DQG 3RWHQWLDO 3UHGDWRUV RI 2YHUZLQWHULQJ 0RQDUFKV 2I WKH VSHFLHV RI RPQLYRURXV DQG LQVHFWLYRURXV ELUGV SUHVHQW LQ WKH 6LHUUD &KLQFXD RYHUZLQWHULQJ DUHD $UHOODQR HW DO LQ SUHVVf U RQO\ EODFNKHDGHG JURVEHDNV DQG EODFNEDFNHG RULROHV &DOYHUW HW DO )LQN DQG %URZHU )LQN HW DO %URZHU DQG &DOYHUW $UHOODQR HW DO LQ SUHVVf IHHG H[WHQVLYHO\ RQ WKH PRQDUFKV :KHUHDV JURVEHDNV DUH UHODWLYHO\ LQVHQVLWLYH WR WKH WR[LF HIIHFWV RI &*V RULROHV DYRLG LQJHVWLQJ ODUJH DPRXQWV RI &*V WKURXJK WDVWH UHMHFWLRQ RI WKH PRUH WR[LF PRQDUFKV )LQN DQG %URZHU f 0L[HG IORFNV RI WKHVH WZR VSHFLHV ZHUH HVWLPDWHG WR NLOO WR EXWWHUIOLHV SHU GD\ LQ D KD FRORQ\ RU DERXW b RI WKH FRORQ\ %URZHU DQG &DOYHUW f 7KHUH LV VRPH HYLGHQFH WKDW SURORQJHG FRQVXPSWLRQ RI PRQDUFKV FDXVHG RQH RU ERWK RI WKHVH VSHFLHV WR DFFXPXODWH WR[LF OHYHOV RI &*V DQG SRVVLEO\ 3$V %URZHU DQG &DOYHUW $UHOODQR HW DO LQ SUHVVf 6HYHUDO VSHFLHV RI PLFH PD\ DOVR IHHG H[WHQVLYHO\ XSRQ WKH PRQDUFKV ,QGLYLGXDO EODFNHDUHG PLFH 3HURP\VFXV PHODQRWLV -$ $OOHQ DQG &KDSPDQ 1HRWRPRGRQ DOVWRQL DOVWRQL 0HUULDP DQG 0LFURWXV PH[LFDQXV VDOYXV 6DXVVXUH WKDW ZHUH FDSWXUHG QHDU RYHUZLQWHULQJ FRORQLHV LQ 0H[LFR KDYH EHHQ IRXQG WR HDW PRQDUFKV WR YDU\LQJ GHJUHHV ZKLOH LQ FDSWLYLW\ %URZHU HW DO f 7KH IRUPHU WZR VSHFLHV ZHUH HUURQHRXVO\ UHIHUUHG WR DV 3HURP\VFXV PDQLFXODWXV ODEHFXOD (OOLRW DQG 3 VSLFLOHJXV -$ $OOHQ UHVSHFWLYHO\ LQ %URZHU HW DO VHH %URZHU HW DO IRU SURSHU LGHQWLILFDWLRQf 6WRPDFK FRQWHQW

PAGE 12

DQDO\VHV DQG VKRUWWHUP WUDSSLQJ VWXGLHV VXJJHVW WKDW IHPDOH 3 PHODQRWLV LQVLGH PRQDUFK FRORQLHV IHHG XSRQ PRQDUFKV DQG DUH ODUJHU DQG PRUH UHSURGXFWLYHO\ DFWLYH WKDQ IHPDOH FRQVSHFLILFV RXWVLGH FRORQLHV %URZHU HW DO f 7KXV WKH HYLGHQFH VXJJHVWV WKDW PHODQRWLV DFWXDOO\ EHQHILWV IURP IHHGLQJ XSRQ WKH PRQDUFKV 7KHUH DUH VHYHUDO RWKHU DQLPDOV NQRZQ RU VXVSHFWHG WR IHHG XSRQ RYHUZLQWHULQJ PRQDUFKV %RWK FDWWOH DQG GRPHVWLF WXUNH\V RFFDVLRQDOO\ HDW WKH EXWWHUIOLHV 8UTXKDUW $ $UHOODQR XQSXE GDWDn &DOYHUW HW DO f IRXQG VFDWV SUHVXPHG WR EH IURP WKH KRJQRVHG VNXQN &RQHSDWXV PHVROHXFXV QHOVRQL WKDW FRQWDLQHG PRQDUFK SDUWV DOVR REVHUYHG DQ LQGLYLGXDO KRJQRVHG VNXQN LQVLGH D PRQDUFK FRORQ\ GXULQJ WKH QLJKW RQ 6LHUUD &KLQFXD GXULQJ )HEUXDU\ EXW FRXOG QRW GHWHUPLQH ZKHWKHU LW ZDV IHHGLQJ RQ PRQDUFKV %URZHU HW DO f FDXJKW RQH 6RUH[ VDXVVXUHL QHDU D PRQDUFK FRORQ\ RQ 6LHUUD &KLQFXD EXW ZHUH XQDEOH WR GHWHUPLQH ZKHWKHU LW KDG IHG XSRQ PRQDUFKV %HWZHHQ WR WUDSSHG DGGLWLRQDO VDXVVXUHL LQVLGH DQG RXWVLGH PRQDUFK FRORQLHVf EXW GLG QRW DQDO\]H WKHLU VWRPDFK FRQWHQWV 'XULQJ P\ ILHOG ZRUN LQ WKH 6LHUUD &KLQFXD RYHUZLQWHULQJ DUHD )LJXUH OD DQG Ef GLVFRYHUHG VHYHUDO DGGLWLRQDO VSHFLHV RI VPDOO PDPPDOV 7KH\ LQFOXGHG IRXU PXURLG URGHQWV 3HURP\VFXV D]WHFXV K\ORFHWHV 5HLWKURGRQWRP\V VXPLFKUDVWL 5HLWKURGRQWRP\V FI PHJDORWLV 1HOVRQ QHRWRPRGRQ JROGPDQLf D VTXLUUHO 6FLXUXV FI DXUHRJDVWHUf D VKUHZ &U\SWRWLV JROGPDQL DOWLFRODf D ZHDVHO 0XVWHOD IUHQDWDf DQG D EDW /DVLXUXV FLQHUHXVf +RZHYHU DOO VSHFLHV H[FHSW WKH IRUPHU ZHUH H[WUHPHO\ UDUH LQGLYLGXDOV RI HDFK ZHUH WUDSSHG DQGRU VLJKWHG

PAGE 13

2YHUYLHZ RI 'LVVHUWDWLRQ 7KH LQLWLDO JRDO RI WKLV SURMHFW ZDV WR FRPSDUH WKH GHPRJUDSKLF UHVSRQVHV RI 3HURP\VFXV PHODQRWLV $ $OOHQ DQG &KDSPHQ (Br D]WHFXV K\ORFHWHV 0HUULDP 5HLWKURGRQWRP\V VXPLFKUDVWL 6DXVVXUH 0LFURWXV PH[LFDQXV VDOYXV 6DXVVXUH DQG 1HRWRPRGRQ DOVWRQL DOVWRQL 0HUULDP WR WKH RYHUZLQWHULQJ PRQDUFKV ,Q &KDSWHU GHVFULEH WKH GLVFRYHU\ WKDW DOO VSHFLHV EXW (A PHODQRWLV ODUJHO\ DYRLGHG WKH PRQDUFK FRORQLHV /DUJH QXPEHUV RI (A PHODQRWLV LPPLJUDWHG LQWR WKH FRORQLHV IHG H[WHQVLYHO\ RQ PRQDUFKV DQG LQLWLDWHG KLJK OHYHOV RI UHSURGXFWLRQ 7R H[SODLQ WKLV ILQGLQJ H[DPLQH WKUHH TXHVWLRQV f DUH WKHUH XQLTXH SODQW VSHFLHV FRPSRVLWLRQV LQ WKH VLWHV ZKHUH PRQDUFK FRORQLHV IRUP WKDW UHQGHU WKHVH DUHDV VXLWDEOH RQO\ WR (A PHODQRWLV f FDQ 3 PHODQRWLV DJJUHVVLYHO\ H[FOXGH WKH RWKHU VSHFLHV RI PLFH IURP WKH FRORQLHV RU f LV PHODQRWLV EHWWHU DEOH WR EUHDFK WKH PRQDUFKnV FKHPLFDO GHIHQVH V\VWHP" UXOH RXW WKH ILUVW WZR K\SRWKHVHV LQ &KDSWHU DQG GHYRWH &KDSWHUV WKURXJK WR WHVWLQJ WKH WKLUG K\SRWKHVLV E\ PHDQV RI DQ LQWHJUDWHG VHULHV RI IHHGLQJ H[SHULPHQWV ZLWK PRQDUFKV DQG DUWLILFLDO GLHWV ,Q &KDSWHU H[DPLQH WKH SRWHQWLDO LPSDFW RI SUHGDWLRQ E\ (A PHODQRWLV RQ WKH OLYH DQG GHDG EXWWHUIOLHV WKDW DFFXPXODWH RQ WKH JURXQG DQG LQ ORZ YHJHWDWLRQ DQG KRZ WKLV SUHGDWLRQ LV LQIOXHQFHG E\ WKH DFFHVVLELOLW\ DQG GHJUHH RI GHVLFFDWLRQ RI WKH PRQDUFKV

PAGE 14

&+$37(5 '()(16,9( &203281'6 ,1 29(5:,17(5,1* 021$5&+ %877(5)/,(6 $GXOW PRQDUFKV FRQWDLQ FDUGLDF JO\FRVLGHV &*Vf WKDW ZHUH VHTXHVWHUHG DQG VWRUHG IURP WKH VSHFLHV RI PLONZHHG SODQW $VFOHSLDT VSSf WKDW WKH\ IHG XSRQ DV ODUYDH %URZHU DQG 0RIILW 5RHVNH HW DO 0DOFROP DQG %URZHU f $YDLODEOH HYLGHQFH LQGLFDWHV WKDW WR b RI RYHUZLQWHULQJ PRQDUFKV LQ 0H[LFR IHG DV ODUYDH RQ WKH FRPPRQ PLONZHHG $ V\ULDFD / 6HLEHU HW DO 0DOFROP DQG %URZHU f :KHUHDV IUHVKO\HFORVHG $ V\ULDFDUHDUHG LQGLYLGXDOV LQLWLDOO\ KDYH UHODWLYHO\ KLJK FRQFHQWUDWLRQV RI &* PHDQ _OJ Jf WKH\ ORVH RQ DYHUDJH b RI WKHVH &*V GXULQJ WKH ORQJ PLJUDWLRQ WR 0H[LFR PHDQ _LJ Jf 0DOFROP HW DO 0DOFROP DQG %URZHU f $SSUR[LPDWHO\ b RI WKH &*V LQ $ V\ULDFDUHDUHG PRQDUFKV DUH ORFDWHG LQ WKH FXWLFOH DQG ZLQJV %URZHU HW DO f 2YHUZLQWHULQJ PRQDUFKV LQ 0H[LFR DOVR VHTXHVWHU DQG VWRUH S\UUROL]LGLQH DONDORLGV 3$Vf IURP 3$ FRQWDLQLQJ SODQW IDPLOLHV VXFK DV WKH $VWHUDFHDH &RPSRVLWDHf %RUDJLQDFHDH $SRF\QDFHDH DQG )DEDFHDH /HJXPLQRVDHf (GJDU HW DO %URZHU 0DWWRFNV .HOOH\ HW DO 0 6WHOOMHV XQSXEO GDWDf 7KH DONDORLG DSSHDUV WR EH DFTXLUHG E\ LPELELQJ QHFWDU ERWK HQ URXWH WR WKH RYHUZLQWHULQJ VLWHV .HOOH\ HW DO f DQG ZKLOH RYHUZLQWHULQJ *OHQGLQQLQJ DQG .HOOH\ XQSXEO GDWDf 3UHOLPLQDU\ HYLGHQFH LQGLFDWHV WKDW 3$V OLNH &*V DUH FRQFHQWUDWHG LQ WKH FXWLFXODU PDWHULDO 0 6WHOOMHV XQSXEO GDWDf

PAGE 15

&DUGLDF *O\FRVLGHV &KHPLVWU\ DQG 7R[LFRORJ\ $OO &*V KDYH DV WKHLU EDVLF VWUXFWXUDO FRPSRQHQW & VWHURLG JO\FRVLGHV )LJXUH f 7KH XVH RI WKH DGMHFWLYH FDUGLDF VWHPV IURP WKH DELOLW\ RI WKHVH FRPSRXQGV WR LQFUHDVH GUDVWLFDOO\ P\RFDUGLDO FRQWUDFWLOH IRUFH LQ YHUWHEUDWHV 0RVW ELRORJLFDOO\ DFWLYH &*V KDYH DQ 2+ JURXS DW SRVLWLRQ & DQG WKH VWHURLG PRLHW\ ULQJV $'f LV FRPSOHWHO\ VDWXUDWHG 7KHUH LV XVXDOO\ DQ XQVDWXUDWHG ILYHPHPEHU ODFWRQH ULQJ DW SRVLWLRQ & ULQJ (f ELRORJLFDO DFWLYLW\ LV DEROLVKHG E\ VDWXUDWLQJ WKH ODFWRQH RU RSHQLQJ LWV ULQJ VWUXFWXUH 7KH VWHURLG QXFOHXV DQG WKH ODFWRQH ULQJ FRPSULVH WKH DJO\FRQH RU JHQLQ SRUWLRQ RI WKH PROHFXOH DQG FRQIHU SKDUPDFRORJLFDO DFWLYLW\ WR WKH VWUXFWXUH 0RH DQG )DUDK f 7KHUH DUH XVXDOO\ WR VXJDU UHVLGXHV RU JO\FRVLGHV ERQGHG WR WKH VWHURLG QXFOHXV E\ DQ HWKHU OLQNDJH DW SRVLWLRQ & 7KH SRWHQF\ DQG GXUDWLRQ RI DFWLYLW\ RI &*V DUH GHWHUPLQHG SULPDULO\ E\ WKH SRODULW\ RI WKH JO\FRVLGHV DQG YDULRXV VWHURLG VXEVWLWXHQWV 0RH DQG )DUDK 'HWZHLOHU f 7KDW LV SRODULW\ VWURQJO\ LQIOXHQFHV DEVRUSWLRQ GLVWULEXWLRQ PHWDEROLVP H[FUHWLRQ DQG SRVVLEO\ JXVWDWLRQ RI &*V LQ YHUWHEUDWHV $V FRPSDUHG WR SRODU &*V OHVV SRODU RQHV DUH f DEVRUEHG PRUH HDVLO\ DFURVV WKH OLSRLG PHPEUDQH RI WKH LQWHVWLQH +HUPDQ HW DO /DXWHUEDFK f f ERXQG PRUH UHDGLO\ E\ VHUXP DOEXPLQ DQG WKXV UHQGHUHG XQDYDLODEOH IRU SKDUPDFRORJLFDO RU WR[LFRORJLFDO DFWLRQ RU PHWDEROLVP DQG H[FUHWLRQ /XNDV DQG GH0DUWLQR 0RH DQG )DUDK f f PHWDEROL]HG OHVV UHDGLO\ LQ WKH OLYHU 'HWZHLOHU f f H[FUHWHG PRUH VORZO\ EHFDXVH

PAGE 16

R )LJXUH *HQHUDO VWUXFWXUH RI FDUGLDF JO\FRVLGHV 5 D JO\FRVLGH

PAGE 17

$ $VSHFLRVLGH % 8VFKDULGLQ )LJXUH 6RPH UHSUHVHQWDWLYH FDUGLDF JO\FRVLGHV

PAGE 18

WKH\ JHW FDXJKW LQ WKH HQWHURKHSDWLF F\FOH LH DEVRUEHG &*V DQG WKHLU PHWDEROLWHV DUH H[FUHWHG LQ WKH ELOH DQG WKHQ UHDEVRUEHG 2NLWD 'HWZHLOHU f DQG f WDVWH UHMHFWHG E\ ELUGV DW KLJKHU FRQFHQWUDWLRQV %URZHU DQG )LQN f 7KH VWUXFWXUHV RI &*V UHOHYDQW WR WKLV GLVVHUWDWLRQ DUH SUHVHQWHG LQ )LJXUH $VSHFLRVLGH LV WKH GRPLQDQW &* LQ $A V\UL DUD / 0DOFROP HW DO f DQG EHFDXVH WR b RI WKH PRQDUFKV DW WKH 0H[LFDQ RYHUZLQWHULQJ VLWHV DSSHDU WR KDYH IHG RQ WKLV IRRGSODQW LW LV DOVR WKH GRPLQDQW &* LQ WKHLU ERGLHV 6HLEHU HW DO 0DOFROP DQG %URZHU f &DORWURSLQ FDODFWLQ DQG XVFKDULGLQ DUH WKH GRPLQDQW &*V LQ PRQDUFKV UHDUHG RQ $A FXUDVVDYLFD / DQG DUH PXFK OHVV SRODU WKDQ DVSHFLRVLGH 5RHVNH HW DO f PLFH ZHUH RIIHUHG PRQDUFKV UHDUHG RQ ERWK $r FXUDVVDYLFD DQG $A V\ULDFD LQ &KDSWHU 'LJLWR[LQ LV GHULYHG IURP VHYHUDO 'LJLWDOLV VSHFLHV 6FURSKXODULDFHDHf DQG LV RI LQWHUPHGLDWH SRODULW\ PLFH ZHUH RIIHUHG GLHWV WUHDWHG ZLWK GLJLWR[LQ LQ &KDSWHU 'HIHQVLYH 9DOXH &DUGLDF JO\FRVLGHV VWRUHG LQ PRQDUFK EXWWHUIOLHV PD\ GHWHU SUHGDWRUV EHFDXVH RI WKHLU ELWWHU WDVWH DQGRU WR[LF HIIHFWV &*V DUH LQWHQVHO\ ELWWHUWDVWLQJ WR KXPDQV VHH UHIHUHQFHV LQ %URZHU DQG *OD]LHU *OHQGLQQLQJ SHUV REVHUYf DQG DUH DYRLGHG E\ EOXH MD\V &\DQRFLWWD FULVWDWD DW H[WUHPHO\ ORZ FRQFHQWUDWLRQV LQ IRRG %URZHU DQG )LQN f 7KXV WDVWH DORQH PD\ EH VXIILFLHQW WR UHQGHU PRQDUFKV XQSDODWDEOH WR PLFH 2UDO GRVDJHV RI D YDULHW\ RI GLIIHUHQW W\SHV RI &*V LQFOXGLQJ WKRVH LQ DGXOW PRQDUFKV LQGXFH HPHVLV FDUGLDF IDLOXUH QHXURWR[LFLW\

PAGE 19

DWD[LD G\VSQHD VHYHUH WUHPRUV DQG FRQYXOVLRQV LQ PDQ\ YHUWHEUDWH VSHFLHV 'HWZHLOHU 0DUW\ %URZHU f +RZHYHU VHQVLWLYLW\ WR RUDO GRVDJHV RI &*V LV VSHFLHVVSHFLILF GRJV VRPH VSHFLHV RI ELUGV DQG KXPDQV DUH KLJKO\ VHQVLWLYH ZKHUHDV WKH URGHQWV 0XV PXVFXOXV 5DWWXV QRUYHJLFXV DQG 3HURP\VFXV PDQLFXODWXV DUH QRW 'HWZHLOHU %DUQHV DQG (OWKHULQJWRQ 7DQ] DQG 8UTXLOOD 0DUW\ %URZHU DQG )LQN f 7KH LQVHQVLWLYLW\ RI WKHVH URGHQWV DSSHDUV WR EH GXH SULPDULO\ WR SRRU DEVRUSWLRQ RI &*V DFURVV WKH JDVWURLQWHVWLQDO ZDOO 0DUW\ f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f PRQRHVWHUV KDYH D VLQJOH HVWHU HJ LQWHUPHGLQH )LJXUH Ef XVXDOO\ DW SRVLWLRQ f RSHQ GLHVWHUV KDYH WZR HVWHUV WKDW GR QRW MRLQ RQH DQRWKHU DQG f PDFURF\FOLF GLHVWHUV KDYH HVWHUV WKDW MRLQ WR IRUP D ULQJ VWUXFWXUH HJ VHQHFLRQLQH )LJXUH Ff 7KH QLWURJHQ DWRP RI DOO 3$V UHDGLO\ XQGHUJRHV R[LGDWLRQ ZKLFK FRQYHUWV WKH 3$ IURP D IUHH DPLQH WR DQ 1R[LGH $Q H[DPSOH RI WKLV R[LGDWLRQUHGXFWLRQ UHDFWLRQ LV SUHVHQWHG LQ )LJXUH IRU WKH

PAGE 20

$ $PLQR DOFRKRO GHULYDWLYH RI DOO WR[LF 3$V ZLWK QXPEHULQJ 5 + RU 2+ % ,QWHUPHGOQH DQ RSHQ PRQRHVWHU ZLWK QXPEHULQJ RQ WKH &+M L HVWHU OLQNHG DFLG JURXS /[f & 6HQHFORQOQH D PDFURF\FOOF GOHVWHU 25 &+M25 5HDFWLYH 3\UUROH 5 RUJDQLF DFLG RU SURWRQ )LJXUH 6RPH UHSUHVHQWDWLYH S\UUROL]LGLQH DONDORLGV DQG WKHLU GHULYDWLYHV

PAGE 21

)UHH $PLQH 1R[LGH )LJXUH 5HDFWLRQ LQYROYHG LQ WKH FRQYHUVLRQ RI IUHH DPLQH PRQRFURWDOLQH WR LWV 1R[LGH IRUP

PAGE 22

PDFURF\FOLF GLHVWHU PRQRFURWDOLQH 7KH ERQG EHWZHHQ WKH QLWURJHQ DQG R[\JHQ DWRPV LQ DQ 1R[LGH FRQWDLQV DQ XQHTXDOO\ VKDUHG HOHFWURQ %HFDXVH WKLV HOHFWURQ WHQGV WR UHVLGH FORVHU WR WKH R[\JHQ DWRP LW JLYHV WKH R[\JHQ DWRP D QHW QHJDWLYH FKDUJH HJ VHH )LJXUH f DQG PDNHV WKH 1R[LGH PRUH SRODU WKDQ WKH IUHH DPLQH 1HYHUWKHOHVV WKH IUHH DPLQH DQG 1R[LGH IRUPV RI WKH VDPH 3$ SURGXFH WKH VDPH W\SH RI WR[LFLW\ %RWK IRUPV DUH FRPPRQO\ IRXQG WRJHWKHU LQ D YDULHW\ RI SODQWV 0DWWRFNV f DQG LQVHFWV %URZQ f LQFOXGLQJ RYHUZLQWHULQJ PRQDUFKV LQ 0H[LFR .HOOH\ HW DO f 7KHUH DUH VWUXFWXUDO DWWULEXWHV HVVHQWLDO IRU WR[LFLW\ 0DWWRFNV f f WKH ULQJ QXFOHXV PXVW EH GRXEOHERQGHG EHWZHHQ &O DQG & VHH )LJXUH Df f WKHUH PXVW EH DW OHDVW RQH HVWHULILHG K\GUR[\O JURXS HJ EHWZHHQ & DQG &,2 )LJXUH Ef DQG f DW OHDVW RQH RI WKH HVWHUOLQNHG VLGH FKDLQV PXVW FRQWDLQ D EUDQFKHG FDUERQ FKDLQ HJ EHWZHHQ &OO DQG & )LJXUH Ef +RZHYHU 3$V DUH QRW WKHPVHOYHV WR[LF 7KH\ DUH UHQGHUHG WR[LF E\ D GHK\GURJHQDWLRQ UHDFWLRQ LQ WKH OLYHU PLFURVRPHV WKDW QRUPDOO\ PHWDEROL]H DONDORLGV OLNH 3$V LQWR PRUH SRODU DQG H[FUHWDEOH GHULYDWLYHV 7KLV UHDFWLRQ FRQYHUWV 3$V LQWR OHVV SRODU DQG KLJKO\ WR[LF PHWDEROLWHV UHIHUUHG WR DV UHDFWLYH S\UUROHV HJ VHH )LJXUH G 0DWWRFNV f $FFRUGLQJO\ WKH PRVW WR[LF 3$V DUH WKRVH PRVW UHDGLO\ FRYHUWHG WR UHDFWLYH S\UUROHV 7KHUH DUH WZR LPSRUWDQW VWUXFWXUDO IHDWXUHV RI WR[LF 3$V WKDW IDFLOLWDWH EUHDNGRZQ LQWR S\UUROLF PHWDEROLWHV 0DWWRFNV f )LUVW ORZ SRODULW\ 3$V DUH PHWDEROL]HG LQWR UHDFWLYH S\UUROHV DW IDVWHU UDWHV EHFDXVH WKH\ JDLQ EHWWHU DFFHVV WR WKH VXUIDFHV RI PLFURVRPHV *HQHUDOO\ SRODULW\ YDULHV DV IROORZV LQ 3$V PRQRHVWHUV RSHQ GLHVWHUV PDFURF\FOLF GLHVWHUV 6HFRQG WKH W\SH RI HVWHU OLQNDJH

PAGE 23

VWURQJO\ LQIOXHQFHV WKH SURSRUWLRQV RI S\UUROHV DQG 1R[LGHV IRUPHG IURP WKH IUHH DPLQHV GXULQJ WKH GHK\GURJHQDWLRQ UHDFWLRQ 5HFDOO WKDW ERWK S\UUROHV DQG 1R[LGHV DUH IRUPHG E\ WKLV VDPH UHDFWLRQf 7KH FRQIRUPDWLRQ RI PRQRHVWHUV DQG PDFURF\FOLF GLHVWHUV IDYRUV SURGXFWLRQ RI UHDFWLYH S\UUROHV ZKHUHDV WKDW RI RSHQ GLHVWHUV JLYHV UHODWLYHO\ PRUH 1R[LGHV $YDLODEOH HYLGHQFH LQGLFDWHV WKDW PRQRHVWHUV HJ LQWHUPHGLQHf DUH PRUH FRPPRQ WKDQ PDFURF\FOLF GLHVWHUV HJ VHQHFLRQLQHf LQ RYHUZLQWHULQJ PRQDUFKV LQ 0H[LFR QR RSHQ GLHVWHUV ZHUH UHSRUWHG .HOOH\ HW DO f +RZHYHU WKLV VWXG\ ZDV EDVHG RQ EXWWHUIOLHV FROOHFWHG GXULQJ 'HFHPEHU ZKLFK ZRXOG KDYH SUREDEO\ GHULYHG WKHLU 3$V IURP QHFWDU VRXUFHV DORQJ WKH PLJUDWLRQ URXWHV .HOOH\ HW DO f 0RUH ZRUN LV QHHGHG WR GHWHUPLQH ZKHWKHU WKH PRQDUFKV PRGLI\ WKH TXDQWLW\ DQGRU W\SHV RI 3$V LQ WKHLU ERGLHV RYHU WKH FRXUVH RI DQ RYHUZLQWHULQJ VHDVRQ E\ QHFWDULQJ RQ WKH ORFDOO\ DEXQGDQW 3$ FRQWDLQLQJ SODQWV HJ 6HQHFLR DQG (XSDWRULXP VSS VHH $SSHQGL[ IRU VSHFLHV QDPHV DQG WKHLU UHODWLYH DEXQGDQFHV 5REELQV f 'HIHQVLYH 9DOXH $V ZLWK &*V 3$V PD\ GHWHU SUHGDWRUV E\ PHDQV RI WKHLU ELWWHU WDVWH DQGRU WR[LF HIIHFWV 3$V DUH ELWWHUWDVWLQJ WR KXPDQV %RSSU *OHQGLQQLQJ SHUV REVHUYf DQG WKHLU SUHVHQFH LQ WKH DUFWLLG PRWK 8WHWKHLVD RUQDWUL[ LWKRPLLQH EXWWHUIOLHV DQG DUWLILFLDO IRRGV FDXVHV UHMHFWLRQ E\ D YDULHW\ RI LQYHUWHEUDWH DQG YHUWHEUDWH SUHGDWRUV (LVQHU %URZQ %RSSU f 7KH UHOHYDQFH RI WKHVH ILQGLQJV WR PRQDUFKnV 3$GHIHQVH LV XQFHUWDLQ 7KH 3$ FRQWHQW LQ PRQDUFKV LV H[WUHPHO\ ORZ UHODWLYH WR WKDW RI WKH LWKRPLLQHV WHVWHG VHH .HOOH\ HW

PAGE 24

DO %URZQ f DQG WKXV PD\ QRW EH VXIILFLHQWO\ KLJK WR GHWHU SUHGDWRUV 0RUHRYHU WKH ILQGLQJV IURP WKH VWXGLHV ZLWK -/ RPDWUL[ DQG 3$DGXOWHUDWHG GLHWV FDQQRW EH HYDOXDWHG EHFDXVH WKH\ ZHUH IURP XQSXEOLVKHG VWXGLHV WKDW ZHUH UHIHUUHG WR LQ VHFRQGDU\ VRXUFHV (LVQHU %RSSU f 7KH UHODWLRQVKLS RI ELWWHUQHVV WR SRODULW\ DQG PROHFXODU FRQIRUPDWLRQ LQ 3$V KDV QRW EHHQ VWXGLHG DV LW KDV LQ &*V VHH DERYHf 2UDO GRVDJHV RI 3$V DUH NQRZQ WR FDXVH D YDULHW\ RI WR[LF HIIHFWV SULPDULO\ LQ WKH OLYHU RI PDPPDOV DSSDUHQWO\ EHFDXVH WKDW LV ZKHUH WKH UHDFWLYH S\UUROHV DUH IRUPHG 0DWWRFNV f 5HODWLYHO\ KLJK GRVDJHV SURGXFH DFXWH HIIHFWV DQG XVXDOO\ NLOO PLFH ZLWKLQ GD\V E\ FDXVLQJ VHYHUH KHPRUUKDJLF QHFURVLV LQ OLYHU FHOOV /RZHU GRVDJHV SURGXFH PRUH FKURQLF HIIHFWV DQG PD\ WDNH VHYHUDO PRQWKV WR NLOO PLFH E\ LQGXFLQJ IRUPDWLRQ RI f JLDQW OLYHU FHOOV PHJDORF\WHVf f YHQRRFFOXVLYH GLVHDVH RI WKH OLYHU DQGRU f GHSUHVVHG UDWHV RI JURZWK DQG IRRG LQWDNH 6FKRHQWDO DQG 0DJHH 0F/HDQ HW DO +RRSHU *RHJHU HW DO &KHHNH DQG 3LHUVRQ*RHJHU 0DWWRFNV f (YHQ WKRXJK ODUJH LQWHUVSHFLILF GLIIHUHQFHV H[LVW DPRQJ ODERUDWRU\ DQLPDOV PLFH UDWV JXLQHD SLJV UDEELWV DQG KDPVWHUVf LQ WHUPV RI VHQVLWLYLW\ WR SDUWLFXODU 3$V QR VSHFLHV DSSHDUV WR EH LQVHQVLWLYH WR DOO RU HYHQ D ODUJH QXPEHU RI 3$V 0DWWRFNV f &KURQLF 3$ WR[LFLW\ GRHV QRW XVXDOO\ UHVXOW LQ GHDWK IRU VHYHUDO PRQWKV DQG WKXV LW LV XQFOHDU ZKHWKHU PDPPDOV IHHO VLFN LPPHGLDWHO\ DIWHU LQJHVWLQJ VXEOHWKDO GRVHV RI 3$V ,W PD\ EH WKDW WKH DFWLRQ RI 3$V LV VR LQVLGLRXV WKDW D PDPPDO ZRXOG QRW IHHO VLFNHQHG XQWLO VHYHUDO ZHHNV DIWHU LQJHVWLQJ WKHP ,I WKLV LV WKH FDVH WKHQ LW ZRXOG EH XQOLNHO\ WKDW PDPPDOV FRXOG GHYHORS &)$V DIWHU LQJHVWLQJ WKHP

PAGE 25

%URZHU f PP

PAGE 26

&+$37(5 '(02*5$3+,& 5(63216(6 72 &2/21,(6 2) 021$5&+ %877(5)/,(6 1XPHURXV ZRUNHUV KDYH GRFXPHQWHG VWURQJ QXPHULFDO DQG UHSURGXFWLYH UHVSRQVHV E\ VPDOO PDPPDOV WR DUHDV ZLWK KHDY\ VHHG FURSV -DPHVRQ *DVKZHLOHU +DOYRUVRQ .LQJ f DQG GHQVH FRQFHQWUDWLRQV RI LQVHFWV +ROOLQJ +DQVNL DQG 3DUYLDLQHQ +DKXV DQG 6PLWK LOO SUHVVf :KHQ WKHVH IRRG ERQDQ]DV RFFXUUHG LQ KDELWDWV FRQWDLQLQJ VHYHUDO VPDOO VSHFLHV RI PDPPDOV KRZHYHU FHUWDLQ VSHFLHV LQYDULDEO\ UHVSRQGHG PRUH VWURQJO\ WKDQ RWKHUV +DOYRUVRQ .LQJ +ROOLQJ +DQVNL DQG 3DUYLDLQHQ +DKXV DQG 6PLWK LQ SUHVVf 7KH IDFWRUV WKDW HQDEOHG SDUWLFXODU VSHFLHV WR WDNH JUHDWHU DGYDQWDJH RI WKH IRRG ERQDQ]DV DQG WKHUHE\ LQFUHDVH WKHLU UHSURGXFWLYH RXWSXW SURSRUWLRQDWHO\ DUH ODUJHO\ XQNQRZQ 3RVVLEOH IDFWRUV LQFOXGH f FRPSHWLWLYH H[FOXVLRQ +DOYRUVRQ f f JUHDWHU WROHUDQFH WR PLFURKDELWDW IHDWXUHV RI WKH DUHDV ZLWK WKH IRRG UHVRXUFH DQGRU f OHVV VHQVLWLYLW\ WR WKH EHKDYLRUDO SK\VLFDO DQGRU FKHPLFDO GHIHQVHV RI WKH IRRG UHVRXUFH ,Q WKLV FKDSWHU FRPSDUH WKH GHPRJUDSKLF DQG IHHGLQJ UHVSRQVHV RI VHYHUDO VSHFLHV RI PLFH WR RYHUZLQWHULQJ FRORQLHV RI PRQDUFK EXWWHUIOLHV 'DQDXV SOH[LSSXV /f LQ 0H[LFR DQG DWWHPSW WR H[SODLQ ZK\ RQH VSHFLHV HQMR\V H[FOXVLYH DFFHVV WR WKH EXWWHUIOLHV )RXU TXHVWLRQV UHJDUGLQJ WKH UHVSRQVH RI QHDUE\ PRXVH SRSXODWLRQV WR WKH PRQDUFK FRORQLHV DUH H[SORUHG f ZKDW LV WKH GHQVLW\ DQG ELRPDVV RI PRQDUFKV LQVLGH FRORQLHV DYDLODEOH WR WHUUHVWULDOO\ IRUDJLQJ

PAGE 27

PLFH f ZKDW VSHFLHV RI PLFH RFFXU LQ WKH RYHUZLQWHULQJ DUHDV DQG ZKDW LV WKHLU UHODWLYH DEXQGDQFH f DUH ZLQWHU GHPRJUDSKLF SDWWHUQV RI SRSXODWLRQV RI WKH GLIIHUHQW VSHFLHV RI PLFH VXFK DV LPPLJUDWLRQ EUHHGLQJ DFWLYLW\ GHQVLW\ KRPH UDQJH VL]H DQG DJHFODVV GLVWULEXWLRQ LQIOXHQFHG E\ WKH WKH FRORQLHV f KRZ GR WKH GLHWV RI PLFH LQVLGH DQG RXWVLGH WKH FRORQLHV FRPSDUH" ,Q WKH FRXUVH RI WKLV LQYHVWLJDWLRQ GLVFRYHUHG WKDW eBr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£Q 0H[LFR )LJXUH OD $QRQ WRSRJUDSKLF PDSf DOVR WUDSSHG RQ )HEUXDU\ DQG 0DUFK RQ WKH VORSH RI WKH $UUR\R /RV &RQHMRV RI WKH QHDUE\ 6LHUUD (O &DPSDQDULR DW rn1 DQG rn: 7KHVH VLWHV DUH DPRQJ WKH SULQFLSDO RYHUZLQWHULQJ DUHDV RI WKH HDVWHUQ SRSXODWLRQ RI WKH PRQDUFK EXWWHUIO\ LQ WKH UHJLRQ &DOYHUW DQG %URZHU f $OO ZRUN ZDV GRQH EHWZHHQ DQG P 7KH YHJHWDWLRQ LV PRQWDQH ERUHDO FRQLIHURXV IRUHVW GRPLQDWHG E\ WKH R\DPHO ILU $ELHV UHOLJLRVD +%.f 7KH XQGHUVWRU\ YHJHWDWLRQ LV GHVFULEHG EHORZ 7KH

PAGE 28

SHULRG IURP ODWH 'HFHPEHU WR HDUO\ 0DUFK LV WKH FROGHVW DQG GULHVW WLPH RI \HDU &DOYHUW HW DO LQ SQHQQf ZLWK QLJKWWLPH IRUHVW WHPSHUDWXUHV RQ 6LHUUD &KLQFXD XVXDOO\ UDQJLQJ EHWZHHQ WR r& &DOYHUW DQG %URZHU $ORQVR HW DO LQ SUHVVf DQG ZLWK WRWDO PRQWKO\ SUHFLSLWDWLRQ DYHUDJLQJ OHVV WKDQ PP &DOYHUW HW DO LQ SUHVV $QRQ f $PRXQW RI 0RQDUFK 0DWHULDO $YDLODEOH WR )RUDJLQJ 0LFH 7R HVWLPDWH WKH DPRXQW RI EXWWHUIO\ PDWHULDO RQ WKH IRUHVW IORRU LQVLGH D PRQDUFK FRORQ\ GXULQJ WKH QLJKW UDQ WUDQVHFWV HDFK IRU FRQVHFXWLYH GD\V EHWZHHQ DQG -DQXDU\ LQVLGH D KD FRORQ\ : &DOYHUW SHUV FRPPf RQ 6LHUUD &KLQFXD 7KH WUDQVHFWV ZHUH P ORQJ HDFK ZLWK RQHPHWHU VDPSOLQJ TXDGUDWV VHW DW P LQWHUYDOV %HFDXVH WKH FRORQ\ ZDV RQO\ DERXW P ZLGH XVHG D VWUDWLILHG UDQGRP VDPSOLQJ VFKHPH =DU f WR FKRRVH OHQJWKZLVH WUDQVHFW ORFDWLRQV (DFK PRUQLQJ EHWZHHQ DQG K WKH TXDGUDWV ZHUH FOHDUHG RI DOO PRQDUFK UHPDLQV 7KHQ EHWZHHQ DQG K RQ WKH VDPH GD\ WDOOLHG WKH QXPEHU RI EXWWHUIOLHV LQ HDFK TXDGUDW WKDW ZHUH HLWKHU RQ WKH JURXQG RU OHVV WKDQ FP DERYH LW 0RQDUFKV ZHUH FODVVLILHG DV OLYH PRULEXQG FXWLFOH SLHUFHG LH VDPSOHG EXW QRW HDWHQf DEGRPHQ UHPRYHG DEGRPHQ GHYLVFHUDWHG RU WKRUD[ FRQWHQWV HDWHQ GLVUHJDUGHG WKH IHZ PRQDUFKV b RI WRWDO PRQDUFKV WDOOLHGf WKDW GLG QRW IDOO LQWR RQH RI WKHVH FDWHJRULHV )RU HDFK GDLO\ WDOO\ ; GHWHUPLQHG WKH PHDQ GHQVLW\ RI HDFK W\SH RI PRQDUFK SHU P TXDGUDW SHU GD\ DOVR HVWLPDWHG WKH PHDQ JUDPV RI WLVVXH SHU P TXDGUDW SHU GD\ UHSUHVHQWHG E\ HDFK W\SH RI PRQDUFK XVLQJ WKH IROORZLQJ VWHSV )LUVW NQHZ WKH DSSUR[LPDWH DPRXQW RI WLVVXH LQ WKH DEGRPHQ PJ GU\ ZHLJKWf DQG WKRUD[ PJ GZf RI

PAGE 29

RYHUZLQWHULQJ PRQDUFKV &KDSWHU 7DEOH f 6HFRQG DVVXPHG WKDW DOO GHYLVFHUDWHG DEGRPHQV DQG GDPDJHG WKRUDFHV VWLOO KDG RQ DYHUDJH b RI WKH LQWHUQDO WLVVXHV UHPDLQLQJ )LQN DQG %URZHU *OHQGLQQLQJ SHUV REVHUYf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n RXWHU HGJH KHQFHIRUWK RXWVLGH )LJXUH OD DQG Ef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

PAGE 30

7DEOH (TXDWLRQV XVHG WR FDOFXODWH WKH JUDPV RI WLVVXHPGD\ LQ WKH GLIIHUHQW GDPDJH FDWHJRULHV WKDW ZHUH DYDLODEOH WR WHUUHVWULDOO\ IRUDJLQJ PLFH RQ WKH IRUHVW IORRU LQVLGH PRQDUFK FRORQLHV )RU H[DPSOH WR HVWLPDWH WKH DPRXQW RI OLYH PRQDUFK WLVVXH DYDLODEOH WR PLFH PXOWLSOLHG WKH PHDQ QXPEHU RI PRQDUFKV SHU PA WKDW IHOO WR WKH PRQDUFK IORRU SHU GD\ PHDQ QRPAGD\f WLPHV WKH FRPELQHG GU\ ZHLJKWV RI WKH WLVVXH LQ DQ RYHUZLQWHULQJ PRQDUFKVnV DEGRPHQ PJf DQG WKRUD[ PJf %DVHG RQ )LQN DQG %URZHU f DQG *OHQGLQQLQJ SDUV REVHUYf DVVXPHG WKDW GHYLVFHUDWHG DEGRPHQV DQG WKRUDFHV VWLOO KDG b RI WKH LQWHUQDO WLVVXHV UHPDLQLQJ 'DPDJH FDWHJRU\ (TXDWLRQ OLYH PRULEXQG FXWLFOH SLHUFHG DEGRPHQ UHPRYHG DEGRPHQ GHYLVFHUDWHG WKRUD[ GHYLVFHUDWHG PHDQ QRPAGD\f PHDQ QRPAGD\f PHDQ QR UH"GD\f PHDQ QRPAGD\f PHDQ QRPAGD\f PHDQ QRP"GD\f PJf PJf PJf PJf > PJf bf PJf@ > PJf PJfbf@

PAGE 31

WUDSV ZHUH VHW RQ WKH JURXQG 7R GHWHUPLQH ZKHWKHU P\ WHUUHVWULDO WUDSSLQJ UHJLPH KDG PLVVHG DQ DUERUHDO URGHQW IDXQD D WKLUG WUDS ZDV DOVR VHFXUHG WR WKH WUHH WUXQN QHDUHVW HDFK VWDWLRQ DVVXPLQJ RQH RFFXUUHG ZLWKLQ P IURP WKH VWDWLRQ DW D KHLJKW RI P D WRWDO RI DUERUHDO WUDSV ZHUH VHW )ROGLQJ 6KHUPDQ OLYH WUDSV [ [ FPf ZHUH EDLWHG ZLWK UROOHG RDWV DQG SURYLGHG ZLWK D FRPSUHVVHG FRWWRQ EDOO IRU QHVWLQJ PDWHULDO 0LFH ZHUH WRHFOLSSHG IRU SHUPDQHQW LGHQWLILFDWLRQ DQG VSHFLHV VH[ DJH ZHLJKW WR WKH QHDUHVW J ZLWK J 3HVOD VFDOHVf UHSURGXFWLYH FRQGLWLRQ DQG WUDS VWDWLRQ ZHUH UHFRUGHG 3HURP\VFXV PHODQRWLV DJH ZDV UHFRUGHG DV MXYHQLOH Jf RU DGXOW Jf 2QO\ DGXOWV RI WKH RWKHU VSHFLHV RI PLFH ZHUH FDSWXUHG 0DOHV RI DOO VSHFLHV ZHUH UHFRUGHG DV UHSURGXFWLYHO\ DFWLYH LI WKHLU WHVWHV ZHUH GHVFHQGHG LH VFURWDOf RU LQDFWLYH LI WKHLU WHVWHV ZHUH ZLWKLQ WKHLU DEGRPLQDO FDYLW\ LH DEGRPLQDOf )HPDOHV RI DOO VSHFLHV ZHUH UHFRUGHG DV SUHJQDQW LI WKH\ KDG VZROOHQ DEGRPHQV DQGRU ZHLJKHG J LI WKH\ ZHUH (A PHODQRWLVf DV ODFWDWLQJ LI WKH\ KDG ODUJH QLSSOHV RU DV QRQEUHHGLQJ LI WKH\ KDG VPDOO QLSSOHV DQG QR VZROOHQ DEGRPHQ 7UDSV ZHUH VHW EHWZHHQ DQG K DQG FKHFNHG WKH IROORZLQJ PRUQLQJ EHWZHHQ DQG K 6RLOHG WUDSV ZHUH FOHDQHG WKRURXJKO\ ZLWK D SDSHU WRZHO DQG ZDWHU WR UHGXFH WKH LQIOXHQFH RI UHVLGXDO RGRUV IURP WKH SUHYLRXV RFFXSDQW RQ VXEVHTXHQW FDSWXUHV 0D]G]HU HW DO 6WRGGDUG DQG 6PLWK f 0LFH ZHUH FRQVLGHUHG UHVLGHQW LI WKH\ ZHUH SUHVHQW RQ WKH VDPH JULG IRU RU PRUH FRQWLQXRXV ZHHNV 7KH PLQLPXP QXPEHU DOLYH 01$f HQXPHUDWLRQ WHFKQLTXH +LOERUQ HW DO f ZDV XVHG WR HVWLPDWH ERWK WKH WRWDO QXPEHU RI PLFH RQ HDFK JULG HDFK ZHHN DQG WKH QXPEHU RI ZHHNV

PAGE 32

HDFK PRXVH ZDV SUHVHQW RQ D JULG 7R YDOLGDWH XVH RI WKLV HQXPHUDWLRQ WHFKQLTXH DOVR GHWHUPLQHG WKH PLQLPXP WUDSSDELOLW\ RI WKH SRSXODWLRQ .UHEV DQG %RRQVWUD f +RPH UDQJH VL]HV ZHUH HVWLPDWHG E\ FDOFXODWLQJ WKH DUHD ZLWKLQ D FRQYH[ SRO\JRQ GUDZQ DURXQG WKH RXWHUPRVW WUDS ORFDWLRQV ZKHUH HDFK PRXVH ZDV FDXJKW +RPH UDQJHV ZHUH GHWHUPLQHG RQO\ IRU WKRVH PLFH f FDXJKW DW OHDVW WLPHV f WUDSSHG DW D VWDWLRQ RQ WKH HGJH RI WKH JULG WLPH DQG f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f WWHVW ,Q WKLV DQG DOO VXEVHTXHQW VWDWLVWLFDO WHVWV WHVWHG IRU QRUPDOLW\ ZLWK WKH .ROPRJRURY6PLUQRY WHVW =DU f ,I D VLJQLILFDQW GHSDUWXUH ZDV IRXQG LH 3 f XVHG D QRQSDUDPHWULF WHVW 7R GHWHUPLQH WKH QXPHULFDO DQG UHSURGXFWLYH UHVSRQVHV RI WKH GLIIHUHQW VSHFLHV RI PLFH WR WKH PRQDUFK FRORQLHV WUHDWHG HDFK JULG DV DQ LQGHSHQGHQW VDPSOH %HFDXVH WKH SRSXODWLRQ GHQVLW\ HVWLPDWHV DPRQJ WKH VXFFHVVLYH WUDSSLQJ SHULRGV LQ HDFK JULG ZHUH QRW LQGHSHQGHQW LH PDQ\ RI WKH PLFH UHPDLQHG RQ WKH VDPH JULG IRU PRUH WKDQ RQH WUDSSLQJ SHULRGf VLPSO\ GHVFULEH WUHQGV LQ WKHVH GDWD +XUOEHUW f +RZHYHU XVHG D WZR IDFWRU $129$ UHSHDWHG RYHU WLPHf WR

PAGE 33

GHWHUPLQH WKH HIIHFWV RI FRORQ\ SUR[LPLW\ DQG WLPH RQ WKH DSSHDUDQFH RI QHZ MXYHQLOHV LQ WKH JULGV )RU VHYHUDO DQDO\VHV WKH FRPELQHG UHVXOWV IURP DOO JULGV LQVLGH FRORQLHV Q f ZHUH FRPSDUHG ZLWK WKRVH IURP DOO JULGV RXWVLGH FRORQLHV Q f (YHQ WKRXJK HDFK JULG VKRXOG EH DQDO\]HG VHSDUDWHO\ +XUOEHUW f IHOW MXVWLILHG FRPELQLQJ WKHP EHFDXVH DOO JULGV KDG UHODWLYHO\ VLPLODU XQGHUVWRU\ YHJHWDWLRQ VHH EHORZf VSHFLHV FRPSRVLWLRQV RI PLFH DQG WUHQGV LQ WKH 01$ HVWLPDWHV DQG OHYHOV RI UHSURGXFWLYH DFWLYLW\ RYHU WLPH 0RUHRYHU VDPSOH VL]HV ZHUH VPDOO LQ HDFK JULG 0LFH IURP LQVLGH DQG RXWVLGH FRORQLHV ZHUH FRPSDUHG LQ WHUPV RI f WKH SHUFHQWDJH RI DGXOW IHPDOHV RI HDFK VSHFLHV WKDW ZHUH UHSURGXFWLYH f WKH QXPEHU RI ZHHNV WKDW DGXOW PLFH RI HDFK VSHFLHV ZHUH SUHVHQW RQ WKH JULGV XVLQJ WKH 0DQQ:KLWQH\ 8 WHVW ZLWK WKH QRUPDO DSSUR[LPDWLRQf f WKH KRPH UDQJH VL]HV RI PDOH DQG IHPDOH 3 PHODQRWLV XVLQJ WKH 0DQQ:KLWQH\ 8 WHVW f WKH SURSRUWLRQ RI MXYHQLOH DQG DGXOW eBr PHODQRWLV WKDW UHPDLQHG RQ WKH JULGV IRU JUHDWHU WKDQ RQH ZHHN DQG f WKH HIIHFWV RI FRORQ\ SUR[LPLW\ DQG VH[ RQ WKH ERG\ ZHLJKWV RI PDOH DQG IHPDOH (A PHODQRWLV XVLQJ WKH WZR IDFWRU $129$ )RU FRPSDULVRQV DQG H[FOXGHG WKRVH PLFH ILUVW WUDSSHG GXULQJ WKH ODVW WKUHH ZHHNV RI WKH VHDVRQ LH IURP )HEUXDU\ WR 0DUFK f 6QDSWUDSSLQJ RI PLFH DW 6LHUUD &KLQFXD DQG 6LHUUD (O &DPSDQDULR ,Q DGGLWLRQ WR WKHVH PDUNUHFDSWXUH VWXGLHV FRPSDUHG WKH GHPRJUDSKLF UHVSRQVHV RI PLFH WR WKH PRQDUFK FRORQLHV RQ 6LHUUD &DPSDQDULR DQG 6LHUUD &KLQFXD 7R GR WKLV UDQ SDUDOOHO WUDQVHFWV RQ 6LHUUD &DPSDQDULR RQ )HEUXDU\ DQG RQ 0DUFK LQ D KD

PAGE 34

PRQDUFK FRORQ\ : &DOYHUW SHUV FRPPf U DV ZHOO DV RQ 6LHUUD &KLQFXD RQ )HEUXDU\ LQ D KD FRORQ\ RQ JULG )LJXUH OEf 'XULQJ HDFK QLJKW WZR RI WKH WUDQVHFWV ZHUH VHW LQVLGH WKH FRORQ\ DQG WKH RWKHU WZR P WR HLWKHU VLGH RI WKH FRORQLHVn HGJHV (DFK WUDQVHFW ZDV P ORQJ ZLWK WUDSSLQJ VWDWLRQV VHW DW P LQWHUYDOV DOVR VQDSWUDSSHG RQ DQG 0DUFK LQ JULGV DQG )LJXUH OEf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f WKH RFFXUUHQFH RI PRQDUFK PDWHULDO f GU\ ZHLJKW f OLSLG ZHLJKW f SHUFHQW OLSLG f WRWDO FDUGLDF JO\FRVLGH &*f FRQWHQW DQG f &* FRQFHQWUDWLRQ $OO ZHLJKW PHDVXUHPHQWV ZHUH PDGH WR WKH QHDUHVW J ZLWK D 0HWWOHU $.

PAGE 35

EDODQFH 7R JHW UHSUHVHQWDWLYH VWRPDFK VDPSOHV IURP QDWXUDOO\ IRUDJLQJ PLFH XVHG WKH PLFH FDXJKW LQ VQDSWUDSV WR DVVXUH WKDW WKH\ KDG GLHG LPPHGLDWHO\ XSRQ FDSWXUH )RRG EROXVHV SDVV WKURXJK WKH GLJHVWLYH WUDFWV RI OLYH FDSWLYH 3HURP\VFXV PDQLFXODWXV LQ WR KRXUV 0DUW\ ff (DFK DQLPDOnV VWRPDFK ZDV UHPRYHG DQG SUHVHUYHG LQ PO RI b HWKDQRO DQG WKHQ IUR]HQ ZLWKLQ ZHHNVf LQ D GRPHVWLF IUHH]HU XQWLO DQDO\]HG LQ 2FWREHU 7R TXDQWLI\ WKH RFFXUUHQFH RI PRQDUFK PDWHULDO ILUVW UHPRYHG DQG GLVFDUGHG DOO EDLW WKLV EDLW ZDV QRW LQFOXGHG LQ WKH OLSLG DQG &* DQDO\VHV 1H[W SODFHG WKH FRQWHQWV RI HDFK VWRPDFK LQ D SHWUL GLVK ZKLOH VXVSHQGHG LQ DERXW PO RI b HWKDQRO DQG REVHUYHG ILHOGV XQGHU O22[ PDJQLILFDWLRQ ,I IRXQG QR IRRG LWHPV LQ D ILHOG FKRVH DQRWKHU RQH GHWHUPLQHG WKH SUHVHQFH RU DEVHQFH RI PRQDUFK PDWHULDO HJ VPDOO SLHFHV RI H[RVNHOHWRQ WHVWHV SLJPHQW IDW DQG WUDFKHDO PDWHULDOf LQ HDFK ILHOG DQG WKHQ FRPSXWHG WKH IUHTXHQF\ RI RFFXUUHQFH RI PRQDUFK PDWHULDO LQ DOO ILHOGV IRU HDFK VWRPDFK VDPSOH 5HIHUHQFH VDPSOHV RI PRQDUFK PDWHULDO ZHUH PDGH E\ IHHGLQJ WHVW PLFH D GLHW FRQVLVWLQJ VROHO\ RI PRQDUFKV DQG UXQQLQJ WKHLU VWRPDFK FRQWHQWV WKURXJK WKH SURFHGXUH GHVFULEHG DERYH 7KH PRQDUFK PDWHULDO DQG HWKDQRO LQ ZKLFK LW ZDV VWRUHG KHQFHIRUWK VWRPDFK FRQWHQW VDPSOHf ZHUH WRJHWKHU VXEMHFWHG WR WKH OLSLG DQG &* DQDO\VHV GHVFULEHG EHORZ (DFK VWRPDFK FRQWHQW VDPSOH ZDV GULHG IRU K DW r& LQ D IRUFHG GUDIW RYHQ WKHQ FRROHG XQGHU YDFXXP ZLWK GU\ULWH DQG ZHLJKHG 1HXWUDO OLSLGV ZHUH H[WUDFWHG ZLWK SHWUROHXP HWKHU VHH PHWKRGV LQ :DOIRUG %URZHU HL QL LQ SUHSf DQG &*V IURP WKH OHDQ UHVLGXH ZLWK HWKDQRO VHH PHWKRGV LQ %URZHU HW DO f *UDYLPHWULF

PAGE 36

WHFKQLTXHV ZHUH XVHG WR GHWHUPLQH WRWDO OLSLG FRQWHQW SHU VWRPDFK VDPSOH DQG VSHFWURSKRWRPHWULF WHFKQLTXHV WR GHWHUPLQH WKH FRQFHQWUDWLRQ SJ J GU\ ZHLJKWf DQG ZHLJKW RI &*V LQ HDFK VWRPDFK %URZHU HW DO %URZHU HW DO f ,W LV XQOLNHO\ WKDW GLJHVWLYH HQ]\PHV RU IRRG LWHPV LQ WKH VWRPDFK LQWHUIHUHG ZLWK P\ OLSLG DQG &* GHWHUPLQDWLRQV )LUVW WKH VPDOO LQWHVWLQH LV WKH PDMRU VLWH RI OLSLG PHWDEROLVP DQG DEVRUSWLRQ LQ URGHQWV VRPH OLSRO\VLV EXW QR DEVRUSWLRQ RFFXUV LQ WKH VWRPDFK %RRWK HW DO +DUULVRQ DQG /HDW f 6HFRQG &*V DUH PHWDEROL]HG SRRUO\ LI DW DOO LQ WKH VWRPDFKV RI PHODQRWLV DQG PDQLFXODWQV DV ZHOO DV LQ WKDW RI ELUGV 0DUW\ %URZHU HW DO %URZHU HW DO f 7KLUG YLUWXDOO\ DOO LQJHVWHG EDLW ZKLFK FRQWDLQHG OLSLGV DQG DOVR FRXOG KDYH LQWHUIHUHG ZLWK WKH VSHFWURDVVD\ %URZHU HW DO f ZDV SK\VLFDOO\ UHPRYHG IURP WKH VWRPDFK VDPSOHV SULRU WR WKH TXDQWLWDWLYH DQDO\VHV WKH RDWV DQG SHDQXW EXWWHU KDG IRUPHG LQWR D KDUG FOXPS ZKLFK ZDV HDV\ WR VHSDUDWH IURP WKH RWKHU IRRG XVHG WKH RQHZD\ $129$ DQG 6FKHII )WHVW WR FRPSDUH WKH ILYH VSHFLHVn VWRPDFK VDPSOHV LQ WHUPV RI f WKH IUHTXHQF\ RI RFFXUUHQFH RI PRQDUFK PDWHULDO f GU\ ZHLJKW f OLSLG ZHLJKW f SHUFHQW OLSLG f JURVV &* FRQWHQW DQG f &* FRQFHQWUDWLRQ 7KH SHUFHQW OLSLG GDWD ZHUH DUFVLQH WUDQVIRUPHG =DU f IRU WKH DQDO\VLV DOVR SHUIRUPHG D WZRZD\ $129$ RQ WKH UHVXOWV IURP PHODQRWLV WR H[DPLQH WKH HIIHFWV RI FRORQ\ SUR[LPLW\ DQG VH[ RQ WKH VDPH YDULDEOHV &RPSDULVRQ RI WKH 8QGHUVWRU\ 9HJHWDWLRQ LQ WKH 7UDSSLQJ *ULGV 1XPHURXV VWXGLHV VXJJHVW WKDW PLFURKDELWDW IHDWXUHV LH VSHFLHV FRPSRVLWLRQ DQG GHQVLW\ RI XQGHUVWRU\ YHJHWDWLRQf VWURQJO\ LQIOXHQFH

PAGE 37

WKH GLVWULEXWLRQ LQ VSDFH DQG WLPH RI VSHFLHV RI PLFH HJ %URZQ +DQVHQ DQG )OHKDUW\ 0n&ORVNH\ DQG /DMRLH +XQWO\ DQG ,QRX\H f WKXV K\SRWKHVL]HG WKDW WKH SUHSRQGHUDQFH RI 3 PHODQRWLV LQVLGH WKH FRORQLHV PD\ KDYH EHHQ LQIOXHQFHG E\ XQLTXH SODQW VSHFLHV FRPSRVLWLRQV IRXQG LQ WKH FRORQ\ VLWHV 7R WHVW WKLV GHVFULEHG DQG FRPSDUHG WKH XQGHUVWRU\ YHJHWDWLRQ LQ WKH HLJKW WUDSSLQJ JULGV HPSOR\LQJ VWDQGDUG PHWKRGV 5LVVHU DQG 5LFH :KLWWDNHU 0XHOOHU'RPERLV DQG (OOHQEHUJ f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f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

PAGE 38

LQVLGH RXWVLGH YV RXWVLGH DQG LQVLGH DV RXWVLGH XVHG WKH .UXVNDO:DOOLV WHVW WR FRPSDUH WKH WKUHH JURXSV DQG H[SHFWHG WKH LQVLGH [V LQVLGH 36 YDOXHV WR EH FRQVLVWHQWO\ KLJKHU $JJUHVVLYH 5HODWLRQV $PRQJ WKH 'LIIHUHQW 0RXVH 6SHFLHV DOVR K\SRWKHVL]HG WKDW WKH SUHSRQGHUDQFH RI eBA PHODQRWL V LQVLGH WKH PRQDUFK FRORQLHV FRXOG KDYH EHHQ GXH WR (A PHODQRWLV DJJUHVVLYHO\ H[FOXGLQJ WKH RWKHU VSHFLHV RI PLFH 7R WHVW WKLV VWXGLHG WKH DJJUHVVLYH UHODWLRQV RI (BA PHODQRWLV (A DA K\ORFHWHV 5 VXPLFKUDVWL DQG L/ VDOYXV E\ VWDJLQJ LQWUDVH[XDO SDLUHG LQWHUDFWLRQV EHWZHHQ D UHSURGXFWLYHO\ DFWLYH (A PHODQRWLV LQ LWV KRPH UDQJH LQVLGH D PRQDUFK FRORQ\ KHQFHIRUWK UHVLGHQWf DQG D UHSURGXFWLYHO\ LQDFWLYH FRQVSHFLILF RU KHWHURVSHFLILF IURP RXWVLGH WKH FRORQ\ KHQFHIRUWK LQWUXGHUf WKXV KDG D WRWDO RI HLJKW SDLULQJ FRPELQDWLRQV FRQVSHFLILF DQG KHWHURVSHFLILF RQHV IRU HDFK VH[ 6H[XDOO\ DFWLYH UHVLGHQWV DQG LQDFWLYH LQWUXGHUV ZHUH XVHG EHFDXVH WKLV SDWWHUQ PLPLFNHG P\ WUDSSLQJ UHVXOWV PRVW FORVHO\ VHOHFWHG PLFH RI FRPSDUDEOH ERG\ PDVV s Jf IRU WKH FRQVSHFLILF SDLULQJV WKLV ZDV QRW SRVVLEOH IRU WKH KHWHURVSHFLILF SDLULQJV JLYHQ WKH ODUJH VL]H GLIIHUHQFHV 7DEOH f SUHGLFWHG WKDW UHVLGHQW (AB PHODQRWLV ZRXOG GRPLQDWH LQWUXGHUV 5HFHQWO\ :ROII HW DO f VKRZHG WKDW GRPLQDQFH DPRQJ VLPLODUO\ VL]HG 3HURP\VFXV VSHFLHV LV VLWHVSHFLILF DQG QRW VSHFLHVVSHFLILF 7KDW LV DQ LQGLYLGXDO LV PRUH OLNHO\ WR ZLQ DQ DJJUHVVLYH HQFRXQWHU ZLWK D FRQVSHFLILF RU KHWHURVSHFLILF RI WKH VDPH VH[ ZKHQ LQ LWV RZQ KRPH UDQJH 3DLUHG HQFRXQWHUV ZHUH VWDJHG EHWZHHQ DQG K LQ D FOHDU SOH[LJODVV DUHQD [ [ FPf ZLWK DQ RSHQ ERWWRP DQG WRS

PAGE 39

IROORZLQJ WKH PHWKRGRORJ\ RI :ROII HW DO f 7KH DQLPDOV ZHUH WKXV RQ QDWXUDO VXEVWUDWH (DFK WULDO ZDV FDUULHG RXW DW WKH WUDS VWDWLRQ ZKHUH WKH UHVLGHQW ZDV FDSWXUHG WKH SUHYLRXV QLJKW DQG FRQVLVWHG RI D V DFFOLPDWLRQ DQG D P LQWHUDFWLRQ SHULRG 'XULQJ WKH DFFOLPDWLRQ SHULRG HDFK PRXVH ZDV SODFHG LQ WKH DUHQD RQ RSSRVLWH VLGHV RI D UHPRYDEOH FDUGERDUG SDUWLWLRQ 7KH LQWHUDFWLRQ SHULRG ZDV YLGHRWDSHG DQG EHJDQ LPPHGLDWHO\ DIWHU WKH SDUWLWLRQ ZDV UHPRYHG %HFDXVH LW ZDV GLIILFXOW WR GLVWLQJXLVK UHVLGHQW DQG LQWUXGHU GXULQJ FRQVSHFLILF SDLULQJV SDLQWHG D FP GLDPHWHU UHG RU JROG VSRW RI QDLO SROLVK RQ WKH SHODJH EHWZHHQ HDFK DQLPDOfV HDUV 7KH PLFH GLG QRW DSSHDU WR KDYH EHHQ GLVWXUEHG E\ WKH QDLO SROLVK 7KH DVVLJQPHQW RI JROG RU UHG IRU UHVLGHQW RU LQWUXGHU ZDV GHWHUPLQHG UDQGRPO\ IRU HDFK PRXVH 7KLV FRORU ODEHOOLQJ ZDV QRW QHFHVVDU\ IRU WKH KHWHURVSHFLILF SDLULQJV )RU HDFK LQWHUDFWLRQ SHULRG UHFRUGHG WKH QXPEHU RI QRQRIIHQVLYH DQG RIIHQVLYH DSSURDFKHV DWWDFNV DQG UHWUHDWV H[KLELWHG E\ WKH UHVLGHQW DQG LQWUXGHU 7DEOH f 7R GHWHUPLQH WKH RXWFRPH RI WKH HQFRXQWHUV PRGLILHG VOLJKWO\ WKH FULWHULD RI :ROII HW DO f f :LQQHU 7KH DQLPDO ZLWK WKH IHZHVW QXPEHU RI UHWUHDWV f 'UDZ %RWK DQLPDOV KDG D VLPLODU QXPEHU RI UHWUHDWV LH WKH WRWDO QXPEHU E\ HDFK LQGLYLGXDO GLIIHUHG E\ f f +D DJJUHVVLRQ 1HLWKHU DQLPDO GLVSOD\HG DQ\ DJJUHVVLRQ WRZDUGV WKH RWKHU 7R GHWHUPLQH ZKHWKHU UHVLGHQW eBr PHODQRWLV ZRQ D JUHDWHU QXPEHU RI WULDOV WKDQ H[SHFWHG E\ FKDQFH LQ HDFK RI WKH HLJKW SDLULQJ FRPELQDWLRQV XVHG WKH ELQRPLDO WHVW 7R FRPSDUH WKH QXPEHU RI QRQn RIIHQVLYH DSSURDFKHV RIIHQVLYH DSSURDFKHV DWWDFNV DQG UHWUHDWV

PAGE 40

7DEOH 3DUWLDO HWKRJUDP RI EHKDYLRUV GLVSOD\HG E\ PDODQRWLV S D K\ORFHWHV (A VXPLFKUDVWL DQG 0A VDOYXV GXULQJ SDLUHG LQWHUDFWLRQV 0DQ\ RI WKH EHKDYLRUDO GHVFULSWLRQV DUH PRGLILHG IURP (LVHQEHUJnV f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

PAGE 41

H[KLELWHG E\ WKH UHVLGHQW DQG LQWUXGHU XVHG WKH :LOFR[RQ PDWFKHG SDLUV WHVW =DU f )RU WKLV DQDO\VLV FRPELQHG WKH UHVXOWV IURP ERWK VH[HV WKHUHE\ UHGXFLQJ WKH QXPEHU RI SDLULQJ FRPELQDWLRQV WR IRXU EHFDXVH QR LQWHUVH[XDO GLIIHUHQFH ZDV DSSDUHQW 5HVXOWV 'HQVLW\ RI 0RQDUFKV RQ )RUHVW )ORRU ,QVLGH &RORQLHV $Q DYHUDJH RI PRQDUFKVPA ZDV DGGHG WR WKH IRUHVW IORRU LQVLGH D PRQDUFK FRORQ\ HDFK QLJKW 7DEOH f 3HUKDSV PRUH VLJQLILFDQW LV WKDW b RI WKHVH PRQDUFKV KDG LQWDFW DEGRPHQV DQG WKRUDFHV LH ZHUH OLYH PRULEXQG DQGRU VDPSOHGf DQG WKXV VWLOO SRVVHVVHG WKH PDMRULW\ RI WKHLU WLVVXHV DQG ERG\ ZDWHU $Q DYHUDJH WRWDO RI JUDP GU\ ZHLJKWf RI OLSLGULFK WLVVXH SHU PA RU J SHU KHFWDUH ZDV DYDLODEOH WR IRUDJLQJ PLFH 7KLV LV DQ XQGHUHVWLPDWH RI WKH WRWDO DPRXQW RI PRQDUFK PDWHULDO DYDLODEOH WR PLFH EHFDXVH GLG QRW LQFOXGH WKH GHDG PRQDUFKV WKDW KDG DFFXPXODWHG RQ WKH IRUHVW IORRU SULRU WR VDPSOLQJ RU WKH OLYH RQHV URRVWLQJ RQ ORZO\LQJ XQGHUVWRU\ YHJHWDWLRQ $ORQVR HW DO LQ SUHVVf ,UDSSDELOLW\ DQG 5HOLDELOLW\ RI 'DWD 0LQLPXP WUDSSDELOLWLHV IRU PLFH PDUNHG RQ 6LHUUD &KLQFXD EHWZHHQ DQG ZHUH H[WUHPHO\ KLJK IRU eA PHODQRWLV b LQVLGH DQG b RXWVLGH WKH PRQDUFK FRORQLHV IRU VXPLFKUDVWL b IRU D K\ORFHWHV b IRU /B DOVWRQL b DQG IRU QX VDOYXV b 7UDSSDELOLW\ HVWLPDWHV IRU WKH ODWWHU VSHFLHV ZHUH IURP LQGLYLGXDOV WUDSSHG RQ JULGV RXWVLGH FRORQLHV 7KXV WKH 01$ HVWLPDWHV ZHUH FRQVLGHUHG WR EH DFFXUDWH ZLWKLQ b +LOERUQ ee QO f 0\

PAGE 42

7DEOH (VWLPDWHG GHQVLW\ RI OLYH PRULEXQG DQG ELUGGDPDJHG PRQDUFKV PHDQ QRPA s 6(f DQG TXDQWLW\ RI PRQDUFK WLVVXH PHDQ QR JUDPVP GU\ ZHLJKWf RQ WKH IRUHVW IORRU DW WKH EHJLQQLQJ RI WKH QLJKW LQVLGH D KD PRQDUFK FRORQ\ RQ 6LHUUD &KLQFXD 6DPSOHV ZHUH PDGH RYHU VL[ QLJKWV EHWZHHQ DQG -DQXDU\ 6WDWH RI PRQDUFK 'HQVLW\ QR SHU Pf 7LVVXH J GZf SHU P OLYH sf PRULEXQG sf FXWLFOH SLHUFHG sf DEGRPHQ UHPRYHG sf DEGRPHQ GHYLVFHUDWHG sf WKRUD[ FRQWHQWV HDWHQ sf 7RWDO

PAGE 43

WUHDWPHQW RI WKH WUDSSLQJ JULGV DV LQGHSHQGHQW VDPSOHV ZDV VXEVWDQWLDWHG EHFDXVH f UHFRUGHG RQO\ RQH LQVWDQFH RI D PRXVH PRYLQJ EHWZHHQ WUDSSLQJ JULGV DQ DGXOW PDOH PHODQRWLV PRYHG IURP JULG LQVLGH FRORQ\f WR RXWVLGH FRORQ\f GXULQJ f DQG f UHPRYHG LH VQDSWUDSSHGf QHDUO\ WKH HQWLUH SRSXODWLRQ RI PLFH IURP JULG DW WKH HQG RI WKH VHDVRQ DQG WKXV GLPLQLVKHG WKH LQIOXHQFH RI WHPSRUDO SVHXGRUHSOLFDWLRQ +XUOEHUW f RQ P\ UHVXOWV IURP WKH VDPH JULG GXULQJ WKH VHDVRQ )LQDOO\ P\ XVH RI WHUUHVWULDO WUDSSLQJ UHVXOWV ZDV MXVWLILHG EHFDXVH QR PLFH ZHUH FDXJKW LQ WKH DUERUHDO WUDSV VHW GXULQJ 6SHFLHV &RPSRVLWLRQ RI 0LFH ,QVLGH DQG 2XWVLGH 0RQDUFK &RORQLHV PDUNHG D WRWDO RI DGXOW DQG MXYHQLOH PLFH LQ P\ PDUN UHFDSWXUH VWXGLHV RQ 6LHUUD &KLQFXD EHWZHHQ 2I WKLV WRWDO b ZHUH eB‘ PHODQRWLV b ZHUH eB} K\ORFHWHVI b ZHUH 5 DXPLFKUDVWL b ZHUH Qr QX VDOYXV b ZHUH DBA DOVWRQL &OHDUO\ HA PHODQRWLV ZDV WKH PRVW FRPPRQ VSHFLHV $OO VSHFLHV GLIIHUHG JUHDWO\ LQ VL]H H[FHSW IRU DA K\ORFHWHV DQG /B DOVWRQL 7DEOH f 3HURP\VFXV PHODQRWLV DQG QX VDOYXV ZHUH WKH RQO\ VSHFLHV ZLWK DSSDUHQW VH[XDO GLPRUSKLVPV ZLWK WKH IRUPHU KDYLQJ ODUJHU IHPDOHV DQG WKH ODWWHU ODUJHU PDOHV 7KH VSHFLHV FRPSRVLWLRQ RI PLFH RQ JULGV LQVLGH DQG RXWVLGH PRQDUFK FRORQLHV GLIIHUHG PDUNHGO\ DQG LQ D VLPLODU PDQQHU RQ ERWK 6LHUUD &KLQFXD DQG 6LHUUD &DPSDQDULR 7DEOH f $W OHDVW b RI WKH PLFH FDSWXUHG RQ JULGV LQVLGH FRORQLHV ZHUH (Br PHODQRWLV ZKHUHDV RQO\ WR b RI WKRVH FDSWXUHG RQ JULGV RXWVLGH FRORQLHV ZHUH (BA PHODQRWLV 7KHVH ILQGLQJV ZHUH FRQVLVWHQW ERWK DPRQJ DQG ZLWKLQ VHDVRQV

PAGE 44

7DEOH 0HDQ ERG\ ZHLJKWV s6( LQ JUDPVf RI DGXOW PLFH RI VSHFLHV FDXJKW LQVLGH DQG RXWVLGH PRQDUFK FRORQLHV 0LFH ZHUH ZHLJKHG GXULQJ PDUNUHFDSWXUH VWXGLHV RQ 6LHUUD &KLQFXD EHWZHHQ -DQXDU\ DQG 0DUFK DQ DYHUDJH ZHLJKW ZDV FDOFXODWHG IRU WKRVH PLFH FDXJKW PRUH WKDQ RQFH :HLJKWV RI SUHJQDQW IHPDOH eBr PHODQRWLV ZHUH H[FOXGHG QR SUHJQDQW IHPDOHV RI WKH RWKHU VSHFLHV ZHUH WUDSSHG +RZHYHU EHFDXVH PRVW EUHHGLQJ IHPDOH eBr PHODQRWLV ZHUH FDXJKW PRUH WKDQ WLPHV ZDV DEOH WR GHWHUPLQH WKHLU PHDQ ZHLJKW EHIRUH DQGRU DIWHU WKHLU SUHJQDQF\ 6DPSOH VL]HV DUH LQ SDUHQWKHVHV FRPSDUHG WKH ZHLJKWV RI ERWK VH[HV ZLWK WKH XQSDLUHG WWHVW WZRWDLOHGf *HQGHU 5 D K\ORFHWHV 1 DOVWRQL 0 P VDOYXV 3 PHODQRWLV 5 VXPLFKUDVWL PDOH s s s s s f f f f f IHPDOH s s s s s f f f f f W YDOXH 3 YDOXH

PAGE 45

7DEOH 7RWDO QXPEHU RI PLFH DGXOW DQG MXYHQLOHf RI VSHFLHV FDSWXUHG LQVLGH DQG RXWVLGH PRQDUFK FRORQLHV RQ 6LHUUD &KLQFXD DQG 6LHUUD &DPSDQDULR DQG WKH SHUFHQWDJH RI WKHVH PLFH WKDW ZHUH 3 PHODQRWLV VKRZ UHVXOWV IURP PDUNUHFDSWXUH VWXGLHV RQ 6LHUUD &KLQFXD EHWZHHQ -DQXDU\ DQG 0DUFK DQG IURP VQDSWUDS VWXGLHV RQ 6LHUUD &DPSDQDULR EHWZHHQ )HEUXDU\ DQG 0DUFK 7RWDO QXPEHU RI LQGLYLGXDO PLFH FDSWXUHG 3HUFHQW
PAGE 46

7DEOH f§&RQWLQXHG )HE 6,(55$ &$03$1$5,2 LQVLGH FRORQ\ 0DUFK 7RWDO )HE DXOVLGH VDODQ\ 0DUFK 7RWDO

PAGE 47

0RUHRYHU WKH\ DUH JUHDWO\ VWUHQJWKHQHG EHFDXVH P\ FRPSDULVRQV RQ 6LHUUD &KLQFXD LQFOXGHG JULGV WKDW ZHUH LQVLGH D FRORQ\ GXULQJ VRPH \HDUV DQG RXWVLGH D FRORQ\ GXULQJ RWKHUV VHH JULG GXULQJ DQG JULG GXULQJ DQG )LJXUH OEf 2I WKH PLFH FDSWXUHG LQVLGH FRORQLHV RQO\ PHODQRWLV LQGLYLGXDOV HVWDEOLVKHG UHVLGHQF\ WKH\ ZHUH WUDSSHG RQ DYHUDJH IRU FRQVHFXWLYH ZHHNV 7DEOH f $OO LQGLYLGXDOV RI WKH RWKHU VSHFLHV ZHUH WUDSSHG RQO\ RQFH LQVLGH WKH FRORQLHV H[FHSW IRU RQH / Pr VDOYXV WKDW ZDV FDXJKW IRU FRQVHFXWLYH ZHHNV 7KXV HYHQ WKRXJK LPPLJUDQWV RI DOO ILYH VSHFLHV HQFRXQWHUHG WKH PRQDUFK FRORQLHV RQO\ LQGLYLGXDO 3 PHODQRWLV UHPDLQHG DQG HVWDEOLVKHG UHVLGHQF\ 7KLV FRQWUDVWV ZLWK WKH UHVXOWV IURP RXWVLGH WKH FRORQLHV ZKHUH LQGLYLGXDOV RI DOO VSHFLHV UHPDLQHG RQ DYHUDJH IRU DW OHDVW FRQVHFXWLYH ZHHNV ,Q IDFW WKH PHDQ GXUDWLRQ RI UHVLGHQFH RQ JULGV RXWVLGH YHUVXV WKRVH LQVLGH PRQDUFK FRORQLHV ZDV VLJQLILFDQWO\ JUHDWHU IRU eBr K\ORFHWHV U 5 VQPLUKUDVWL DQG Q/B VDOYXV EXW QRW IRU PHODQRWLV 7DEOH f )HPDOH eBr PHODQRWLV WUDSSHG LQVLGH FRORQLHV H[KLELWHG WKH KLJKHVW OHYHOV RI UHSURGXFWLRQ ERWK RQ 6LHUUD &KLQFXD DQG 6LHUUD &DPSDQDULR 7DEOH f )HPDOH eAB PHODQRWLV WUDSSHG RXWVLGH FRORQLHV DOVR VKRZHG UHODWLYHO\ KLJK UHSURGXFWLYH DFWLYLW\ FRPSDUHG WR WKH RWKHU IRXU VSHFLHV +RZHYHU WKH PDMRULW\ RI WKH IHPDOHV RI DOO ILYH VSHFLHV RXWVLGH WKH FRORQLHV GLG QRW LQLWLDWH EUHHGLQJ XQWLO ODWH )HEUXDU\ DQG 0DUFK 7KLV SDWWHUQ FRQWUDVWV VKDUSO\ ZLWK WKDW RI IHPDOH PHODQRWLV LQVLGH FRORQLHV RQ 6LHUUD &KLQFXD ZKLFK ZHUH EUHHGLQJ H[WHQVLYHO\ E\ -DQXDU\ )LJXUH Df

PAGE 48

)LJXUH $f /RFDWLRQ RI PDUNUHFDSWXUH VWXG\ JULGV IRU \HDUV IURP EODFN VTXDUHV LQVLGH KDWFKHG DUHDf LQ UHODWLRQ WR WRSRJUDSK\ DQG GUDLQDJH SDWWHUQV RQ 6LHUUD &KLQFXD DGDSWHG IURP $QRQ f &RQWRXU LQWHUYDO P %f 7KH KDWFKHG DUHD KDV EHHQ H[SDQGHG DQG SUHVHQWHG VHSDUDWHO\ IRU DOO IRXU WUDSSLQJ \HDUV 7UDSSLQJ JULGV DUH GUDZQ WR VFDOH DQG HDFK JULG LV JLYHQ D XQLTXH QXPEHU 6ROLG JULG VTXDUHV LQGLFDWH ZKLFK JULGV ZHUH WUDSSHG HDFK \HDU &RORQ\ ORFDWLRQV DUH LQGLFDWHG E\ WKH VWLSSOHG DUHD (YHQ WKRXJK D FRORQ\ IRUPHG LQ JULG GXULQJ GLG QRW FRQGXFW D PDUNUHFDSWXUH VWXG\ WKHUH GXULQJ WKDW \HDU

PAGE 49

&RQWLQXHG

PAGE 50

)LJXUH f§FRQWLQXHG

PAGE 51

7DEOH 0HDQ QXPEHU RI ZHHNV s6(f WKDW DGXOW PLFH RI WKH VSHFLHV UHPDLQHG RQ WKH JULGV LQVLGH Q f DQG RXWVLGH Q f PRQDUFK FRORQLHV RQ 6LHUUD &KLQFXD 2QO\ WKRVH PLFH ILUVW WUDSSHG EHWZHHQ -DQXDU\ DQG )HEUXDU\ WR ZHUH LQFOXGHG LQ WKH DQDO\VHV 6DPSOH VL]HV DUH LQ SDUHQWKHVHV &RPSDULVRQV DUH PDGH EHWZHHQ PLFH IURP LQVLGH DQG RXWVLGH FRORQLHV ZLWK WKH 0DQQ:KLWQH\ 8 WHVW QRUPDO DSSUR[LPDWLRQf &RORQ\ SUR[LPLW\ (/B PHODQRWLV 3 D K\ORFHWHV 5 VXPLFKUDVWL L DOVWRQL VDOYXV LQVLGH s f s f s f s f RXWVLGH s f s f s f s f s f = YDOXHr 3 YDOXH f f r FRUUHFWHG IRU WLHV

PAGE 52

7DEOH 3HUFHQWDJH RI DGXOW IHPDOHV RI VSHFLHV WKDW ZHUH SUHJQDQW DQGRU ODFWDWLQJ FRPELQHG WKH GDWD IRU DOO IHPDOHV WUDSSHG LQ PDUNUHFDSWXUH VWXGLHV RQ 6LHUUD &KLQFXD EHWZHHQ -DQXDU\ DQG 0DUFK DQG LQ VQDSWUDS VWXGLHV RQ 6LHUUD &DPSDQDULR RQ )HEUXDU\ DQG 0DUFK 3HUFHQWDJH UHSURGXFWLYH WRWDO QXPEHU FDSWXUHGf 3UR[LPLW\ WR FRORQ\ e PHODQDWLD 3 D KYORFHWHV % VXPLFKUDVWL / DOVWRQL 0 P VDOYXV 6,(55$ &+,1&8$ LQVLGH f f f f f RXWVLGH f f f f f 6,(55$ &$03$1$5,2 LQVLGH f f f f f RXWVLGH f f f f f

PAGE 53

'HPRJUDSKLF 5HVSRQVHV RI 3 PHODQRWLV WR WKH 0RQDUFK &RORQLHV ,QLWLDO SRSXODWLRQ GHQVLWLHV LH WKRVH GXULQJ HDUO\ -DQXDU\f RI PLFH RQ WKH JULGV LQVLGH DQG RXWVLGH WKH FRORQLHV ZHUH VLPLODU H[FHSW RQ JULG LQ DQG ZKHUH IHPDOH GHQVLW\ ZDV DOUHDG\ KLJK )LJXUH f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f ,Q WKH JULGV RXWVLGH RI WKH FRORQLHV MXYHQLOHV DQG DGXOWV RFFXUUHG LQ VLPLODU SURSRUWLRQV ,PPLJUDWLRQ RI DGXOWV ZDV FRQVLVWHQWO\ IHPDOHELDVHG WR JULGV LQVLGH FRORQLHV DQG PDOHELDVHG WR JULGV RXWVLGH FRORQLHV 7DEOH )LJXUH f :KHUHDV IHPDOHV LQVLGH FRORQLHV VKRZHG KLJK OHYHOV RI UHSURGXFWLRQ WKURXJKRXW WKH ZLQWHU WKRVH RXWVLGH FRORQLHV GLG QRW LQLWLDWH KLJK OHYHOV RI UHSURGXFWLRQ XQWLO ODWH )HEUXDU\ )LJXUH Df 7KHUH ZHUH DOVR PDQ\ PRUH QHZ MXYHQLOHV FDXJKW LQVLGH FRORQLHV )LJXUH Ef $ WZR IDFWRU $129$ UHYHDOHG D VLJQLILFDQW HIIHFW RI FRORQ\ SUR[LPLW\ ) GI 3 f WLPH ) GI 3 f DQG FRORQ\ SUR[LPLW\ [ WLPH ) GI 3 f RQ WKH QXPEHU RI MXYHQLOHV FDSWXUHG 7KH VLJQLILFDQW LQWHUDFWLRQ LQGLFDWHV WKDW WKH QXPEHU RI QHZ MXYHQLOHV LQFUHDVHG ZLWK WLPH RQ WKH JULGV LQVLGH EXW QRW RQ WKRVH RXWVLGH FRORQLHV 7KH ODUJH QXPEHU RI MXYHQLOHV LQVLGH

PAGE 54

)LJXUH 0LQLPXP QXPEHU DOLYH HVWLPDWHV RI PDOH DQG IHPDOH 3 PHODQRWLV SUHVHQW RQ WKH WUDSSLQJ JULGV LQVLGH DQG RXWVLGH PRQDUFK FRORQLHV IURP -DQXDU\ WR 0DUFK WR 7KH QXPEHUV WR WKH ULJKW RI HDFK OLQH LQGLFDWH WKH VSHFLILF WUDSSLQJ JULG %HFDXVH JULG ZDV WUDSSHG IRU WZR VHDVRQV GLVWLQJXLVK WKH UHVXOWV IURP DV DQG WKRVH IURP DV r

PAGE 55

0LQLPXP 1XPEHU $OLYH -DQXDU\ )HEUXDU\ 0DUFK A2'AMURFQ

PAGE 56

7DEOH 3URSRUWLRQ RI MXYHQLOH DQG DGXOW PH OD QRW LV UHPDLQLQJ IRU JUHDWHU WKDQ RQH ZHHN RQ JULGV LQVLGH DQG JULGV RXWVLGH PRQDUFK FRORQLHV RQ 6LHUUD &KLQFXD 2QO\ WKRVH PLFH WKDW ZHUH ILUVW WUDSSHG EHWZHHQ -DQXDU\ DQG )HEUXDU\ ZHUH LQFOXGHG 7RWDO QXPEHU $JHFODVV 1XPEHU RI ZHHNV RQ JULG WUDSSHG RQH WZR LQVLGH FRORQ\ MXYHQLOH DGXOW RXWVLGH FRORQ\ MXYHQLOH DGXOW

PAGE 57

7DEOH 3HUFHQWDJH RI (BA PHODQRWLV WKDW HVWDEOLVKHG UHVLGHQ\ RQ JULGV LQVLGH DQG RXWVLGH RI PRQDUFK FRORQLHV WKDW ZHUH IHPDOH
PAGE 58

)LJXUH $f 0HDQ s6(f SHUFHQWDJH RI IHPDOH 3Br PHODQRWLV LQ UHSURGXFWLYH FRQGLWLRQ RQ JULGV LQVLGH DQG JULGV RXWVLGH PRQDUFK FRORQLHV IURP -DQXDU\ WR 0DUFK WR 6DPSOH VL]HV DUH ZULWWHQ DORQJVLGH HDFK V\PERO %f 0HDQ s6(f QXPEHU RI QHZ MXYHQLOH 3 PHODQRWLV WUDSSHG LQVLGH DQG RXWVLGH PRQDUFK FRORQLHV RQ WKH VDPH JULGV DQG RYHU WKH VDPH WLPH SHULRG DV LQ $

PAGE 59

0HDQ s6(f QR RI QHZ MXYHQLOHV 0HDQ s6(f SHUFHQW UHSURGXFWLYH $ %

PAGE 60

FRORQLHV VWURQJO\ VXJJHVWV WKDW WKH IHPDOHV ZHUH DEOH WR VXFFHVVIXOO\ ZHDQ WKHLU \RXQJ )RUW\RQH eBA PHODQRWLV ILW P\ FULWHULD IRU XVH LQ WKH KRPH UDQJH VL]H GHWHUPLQDWLRQV LQ P PHDQ s6(f +RPH UDQJHV IRU PDOHV LQVLGH s Q f DQG RXWVLGH s Q f FRORQLHV GLG QRW GLIIHU VLJQLILFDQWO\ = 3 f QRU GLG WKRVH IRU IHPDOHV LQVLGH s Q f DQG RXWVLGH s Q f FRORQLHV = 3 @ +RZHYHU ZKHQ WKH UHVXOWV IURP JULGV LQVLGH DQG RXWVLGH FRORQLHV ZHUH FRPELQHG PDOHV KDG VLJQLILFDQWO\ ODUJHU KRPH UDQJHV WKDQ GLG IHPDOHV s DQG s UHVSHFWLYHO\ = 3 f 1RQSUHJQDQW UHVLGHQW IHPDOHV ZHUH VLJQLILFDQWO\ KHDYLHU WKDQ UHVLGHQW PDOHV LQ JULGV LQVLGH DQG RXWVLGH FRORQLHV 7DEOH f 7KH FRORQ\ SUR[LPLW\ [ VH[ LQWHUDFWLRQ ZDV QRW VLJQLILFDQW 1HLWKHU VH[ H[KLELWHG HYLGHQFH RI ZHLJKW ORVV RU GHWHULRUDWLQJ KHDOWK RYHU WKH ZLQWHU 6WRPDFK &RQWHQW $QDO\VHV 'HWDLOHG YLVXDO H[DPLQDWLRQV LQGLFDWHG WKDW WKH VWRPDFK VDPSOHV RI (A PHODQRWLV FROOHFWHG LQVLGH FRORQLHV FRQVLVWHG VROHO\ RI PRQDUFK PDWHULDO )LJXUH f DOVR HQFRXQWHUHG YDULDEOH DPRXQWV RI PRQDUFK PDWHULDO LQ WKH VWRPDFK RI PLFH IURP JULGV RXWVLGH FRORQLHV ZLWK 3 PHODQRWLV KDYLQJ VLJQLILFDQWO\ PRUH WKDQ WKH RWKHU VSHFLHV DQG / P VDOYXV KDYLQJ VLJQLILFDQWO\ OHVV /LYH DQGRU GHDG PRQDUFKV GR RFFXU RQ WKH IRUHVW IORRU RXWVLGH FRORQLHV DOEHLW DW D UHODWLYHO\ ORZ GHQVLW\ *OHQGLQQLQJ XQSXE REVHUYf

PAGE 61

7DEOH &RPSDULVRQ RI WKH ERG\ ZHLJKWV PHDQ s6( LQ JUDPVf RI UHVLGHQW e PHODQRWLV WUDSSHG LQVLGH DQG RXWVLGH PRQDUFK FRORQLHV EHWZHHQ -DQXDU\ DQG )HEUXDU\ RQ 6LHUUD &KLQFXD $V PLFH ZHUH FDXJKW DW OHDVW WZR WLPHV DQ DYHUDJH ZHLJKW ZDV FDOFXODWHG IRU HDFK PRXVH 3UHJQDQW IHPDOH ZHLJKWV ZHUH H[FOXGHG +RZHYHU EHFDXVH PRVW EUHHGLQJ IHPDOHV ZHUH FDXJKW PRUH WKDQ WLPHV ZDV DEOH WR GHWHUPLQH WKHLU PHDQ ZHLJKW EHIRUH DQGRU DIWHU WKHLU SUHJQDQF\ 6DPSOH VL]HV DUH LQ SDUHQWKHVHV 7KH UHVXOWV RI D IDFWRU $129$ DUH SUHVHQWHG 3UR[LPLW\ WR FRORQ\ 0DOHV )HPDOHV LQVLGH s f s f RXWVLGH s f s f 6RXUFH RI 9DULDWLRQ GI ) UDWLR 3 &RORQ\ 3UR[LPLW\ 6H[ ,QWHUDFWLRQ

PAGE 62

)LJXUH 0HDQ s6(f RFFXUUHQFH RI PRQDUFK PDWHULDO LQ WKH VWRPDFK FRQWHQWV RI / PHODQRWLV FROOHFWHG LQVLGH DQG eMB PHODQRWLV VXPLFKUDVWL / £L K\ORFHWH6U (/ DOVWRQL DQG / QX VDOYXV FROOHFWHG RXWVLGH PRQDUFK FRORQLHV RQ 6LHUUD &KLQFXD DQG 6LHUUD &DPSDQDULR SUHVHQW UHVXOWV RI D RQHZD\ $129$ 'LIIHUHQW VXEVFULSWV D E F Gf LQGLFDWH VLJQLILFDQW GLIIHUHQFHV DPRQJ PHDQV 3 6FKHII )WHVWf

PAGE 63

R R F &' 2 2 2 F FG cQVLGH FRORQ\ RXWVLGH FRORQ\ .UXVNDO:DOOLV + 3 9-O 21 3 PHODQRWLV 5 VXPLFKUDVWL 3 D K\ORFHWHV 1 D DOVWRQL 0 P VDOYXV

PAGE 64

(YHQ WKRXJK WKH GU\ ZHLJKWV RI eAB K\ORFHWHV XBB DOVWRQL DQG / .8 VDOYXV VWRPDFK FRQWHQWV ZHUH VLJQLILFDQWO\ KLJKHU WKDQ WKRVH RI eBA PHODQRWLV FROOHFWHG RXWVLGH FRORQLHVf DQG VXPLFKUDVWLI DOO VSHFLHV KDG VWDWLVWLFDOO\ HTXDO DPRXQWV RI OLSLG 7DEOH f +RZHYHU PHODQRWLV KDG D VLJQLILFDQWO\ KLJKHU SHUFHQWDJH RI OLSLG &* FRQWHQW DQG &* FRQFHQWUDWLRQ WKDQ GLG WKH RWKHU VSHFLHV 7RJHWKHU ZLWK WKH UHVXOWV LQ )LJXUH WKHVH UHVXOWV VXJJHVW VWURQJO\ WKDW RQ JULGV RXWVLGH FRORQLHV (A PHODQRWLV FRQVXPHG VXEVWDQWLDOO\ PRUH PRQDUFK PDWHULDO WKDQ GLG WKH RWKHU IRXU VSHFLHV 7ZR ZD\ $129$nV RQ WKH VWRPDFK VDPSOHV IURP eBr PHODQRWLV LQVLGH DQG RXWVLGH FRORQLHV GHPRQVWUDWHG VLJQLILFDQW HIIHFWV RI FRORQ\ SUR[LPLW\ DQG VH[ RQ GU\ ZHLJKW DQG OLSLG ZHLJKW DQG RI FRORQ\ SUR[LPLW\ RQ SHUFHQWDJH RI OLSLG 7DEOH f 7KH VDPH HIIHFWV ZHUH VLJQLILFDQW IRU &* FRQFHQWUDWLRQ EXW LQ WKH RSSRVLWH GLUHFWLRQ LH PDOHV ERWK LQVLGH DQG RXWVLGH FRORQLHV KDG KLJKHU YDOXHVf &DUGLDF JO\FRVLGH FRQWHQW ZDV QRW LQIOXHQFHG VLJQLILFDQWO\ E\ FRORQ\ SUR[LPLW\ RU VH[ 7KHVH UHVXOWV VXJJHVW WKDW (A PHODQRWLV LQVLGH FRORQLHV IHG H[FOXVLYHO\ RQ PRQDUFK PDWHULDO DQG WKXV DFFXPXODWHG ODUJH TXDQWLWLHV RI OLSLG DPRXQWLQJ WR DSSUR[LPDWHO\ b RI WKH WRWDO GU\ PDVV RI WKHLU VWRPDFK FRQWHQWV +RZHYHUA HYHQ WKRXJK FRQVSHFLILFV RXWVLGH FRORQLHV LQJHVWHG OHVV PRQDUFK PDWHULDOV WKH &* FRQFHQWUDWLRQ LQ WKHLU VWRPDFK VDPSOHV ZDV VLJQLILFDQWO\ JUHDWHU )HPDOHV ERWK LQVLGH DQG RXWVLGH FRORQLHV KDG JUHDWHU GU\ ZHLJKWV DQG OLSLG ZHLJKWV WKLV LQWHUVH[XDO GLIIHUHQFH ZDV JUHDWHVW IRU PLFH IURP LQVLGH FRORQLHV

PAGE 65

7DEOH &RPSDULVRQ RI VWRPDFK FRQWHQW VDPSOHV IURP PRXVH VSHFLHV IRU WKH YDULDEOHV GU\ ZHLJKW OLSLG ZHLJKW SHUFHQW OLSLG DQG &* FRQWHQW DQG FRQFHQWUDWLRQ PHDQ s 6(f $OO PLFH ZHUH FROOHFWHG RXWVLGH PRQDUFK FRORQLHV RQ 6LHUUD &KLQFXD DQG 6LHUUD &DPSDQDULR ,QWHUVSHFLILF FRPSDULVRQV DUH PDGH ZLWK RQHZD\ $129$ DQG 6FKHII )WHVWV 6XEVFULSWV D DQG E LQGLFDWH VLJQLILFDQW GLIIHUHQFHV DPRQJ VSHFLHV ZLWKLQ HDFK FROXPQ 3 f 'U\ /LSLG /LSLG &* ZHLJKW ZHLJKW DV b &* FRQWHQW FRQFHQWUDWLRQ 6SHFLHV 1 Jf Jf GU\ ZHLJKW QJf _OJ J GZf 3 PHODQRWLV s E s D s D s D s D ( VXPLFKUDVWL s E s D s E s E s E 3 £ KYORFHWHV s D s D s E s E s E 1 DOVWRQL s D s D s E RR E E 0 QX VDOYXV s D s D s E RR E E )UDWLR GI 3YDOXH

PAGE 66

7DEOH &RPSDULVRQ RI VWRPDFK FRQWHQW VDPSOHV IURP PDOH DQG IHPDOH 3 PHODQRWLV FROOHFWHG LQVLGH DQG RXWVLGH PRQDUFK FRORQLHV IRU WKH YDULDEOHV GU\ ZHLJKW OLSLG ZHLJKW SHUFHQW OLSLG DQG &* FRQWHQW DQG FRQFHQWUDWLRQ PHDQ s 6(f 0LFH ZHUH WUDSSHG RQ 6LHUUD &KLQFXD DQG 6LHUUD &DPSDQDULR 7KH HIIHFWV RI VH[ DQG FRORQ\ SUR[LPLW\ RQ HDFK YDULDEOH ZHUH GHWHUPLQHG ZLWK D WZRIDFWRU $129$ 16 3 r 3 rr 3 3 3UR[LPLW\ WR FRORQ\ 6H[ 1 'U\ ZHLJKW Jf /LSLG ZHLJKW Jf /LSLG DV b GU\ ZHLJKW &* FRQWHQW AJf &* FRQFHQWUDWLRQ OJ J GZf LQVLGH IHPDOH s s s s s PDOH s s s s s RXWVLGH IHPDOH s s s s s PDOH s s s s s 6RXUFH RI 9DULDWLRQ GI )UDWLRV &RORQ\ SUR[LPLW\ rr rrr rrr 16 rr 6H[ r r 16 16 r ,QWHUDFWLRQ 16 16 16 16 16

PAGE 67

&RPSDULVRQ RI WKH 8QGHUVWRU\ 9HJHWDWLRQ LQ WKH (LJKW *ULGV 0\ UHVXOWV VXJJHVWHG WKDW WKH XQGHUVWRU\ YHJHWDWLRQ RQ WKH JULGV LQVLGH PRQDUFK FRORQLHV GLG QRW GLIIHU VLJQLILFDQWO\ IURP WKDW RQ WKH JULGV RXWVLGH FRORQLHV 7KH SDLUZLVH SHUFHQW VLPLODULW\ YDOXHV IURP LQVLGH 6 LQVLGH FRPSDULVRQV ZHUH QRW VLJQLILFDQWO\ KLJKHU WKDQ WKRVH RI RXWVLGH \V RXWVLGH DQG LQVLGH 6 RXWVLGH FRPSDULVRQV PHDQ s6( LQ UHVSHFWLYH RUGHU s s DQG s .UXVNDO :DOOLV + YDOXH 3 $SSHQGL[ $f 7KH SHUFHQWDJH LPSRUWDQFH YDOXHV RI WKH XQGHUVWRU\ SODQW VSHFLHV HQFRXQWHUHG RQ WKH JULGV DUH LQ $SSHQGL[ % $JJUHVVLYH 5HODWLRQV $PRQJ WKH 'LIIHUHQW 0RXVH 6SHFLHV 7KH UHVXOWV RI WKH SDLULQJV LQ ZKLFK D ZLQORVV GHFLVLRQ ZDV PDGH DUH SUHVHQWHG LQ 7DEOH 7KH QXPEHU RI ZLQV E\ UHVLGHQW 3 PHODQRWLV ZDV VLJQLILFDQWO\ JUHDWHU WKDQ H[SHFWHG E\ FKDQFH LQ FRQVSHFLILF SDLULQJV EXW QRW LQ DQ\ RI WKH KHWHURVSHFLILF SDLULQJV ,Q IDFW UHVLGHQW PHODQRWLV ORVW WKH YDVW PDMRULW\ RI SDLULQJV ZLWK (r K\ORFHWHV DQG +r QX VDOYXV ZKHUHDV WKH UHVXOWV RI WKHLU SDLULQJV ZLWK I/B VXPLFKUDVWL ZHUH HTXLYRFDO )LJXUH f ,Q FRQVSHFLILF SDLULQJV WKH UHVLGHQW (BA PHODQRWLV H[KLELWHG VLJQLILFDQWO\ PRUH QRQn RIIHQVLYH DQG RIIHQVLYH DSSURDFKHV DQG DWWDFNV UXVKHVf WKDQ GLG WKH LQWUXGHUV )LJXUH f ,QWHUDFWLRQV EHWZHHQ PHODQRWLV DQG D K\ORFHWHV ZHUH WKH OHDVW DJJUHVVLYH WKH PDMRULW\ RI WKH IRUPHUnV UHWUHDWV IROORZHG QRQRIIHQVLYH DSSURDFKHV E\ WKH ODWWHU 0RVW RI +8 VDOYXVn DWWDFNV LQYROYHG OXQJHV IROORZLQJ QRQRIIHQVLYH DSSURDFKHV E\ PHODQRWLV ,QWHUDFWLRQV EHWZHHQ PHODQRWLV DQG VXPLFKUDVWL ZHUH WKH PRVW EDODQFHG ZLWK WKH IRUPHU H[KLELWLQJ PRUH RIIHQVLYH

PAGE 68

7DEOH 2XWFRPHV RI LQWUDVH[XDO SDLUHG WULDOV EHWZHHQ D UHVLGHQW e PHODQRWLV DQG D YLVLWLQJ FRQVSHFLILF RU KHWHURVSHFLILF LQ ZKLFK D ZLQORVV GHFLVLRQ ZDV PDGH 7KH ZLQQHU RI D WULDO KDG WKH IHZHVW UHWUHDWV 7KH ILUVW DQLPDO RI HDFK SDLU LV WKH UHVLGHQW DQG WKH VHFRQG DQLPDO LV WKH YLVLWRU 7KH QXPEHU RI ZLQV E\ WKH UHVLGHQW DQG YLVLWRU ZHUH FRPSDUHG ZLWK WKH ELQRPLDO WHVW 6SHFLHV 6H[ 1R RI 7ULDOV 1RRI :LQV %LQRPLDO 3URE e PHODQRWLV ) ( PHODQRWLV ) ( PHODQRWLV 0 e PHODQRWLV 0 e PHODQRWLV ) % VXPLFKUDVWL ) e PHODQRWLV 0 5 VXPLFKUDVWL 0 e PHODQRWLV ) +r QE DDODXV ) e PHODQRWLV 0 +O QX VDOYXV 0 e PHODQRWLV ) 3 D K\ORQHWHV ) ef PHODQRWLV 0 e D K\ORFHWHV 0

PAGE 69

)LJXUH 3HUFHQWDJH RI UHVLGHQW ZLQV LQWUXGHU ZLQV GUDZV DQG QR DJJUHVVLRQ RXWFRPHV LQ LQWUD DQG LQWHUVSHFLILF SDLUV RI PHODQRWLV eBr KOI/B VXPLFKUDVWL (A =f QX VDOYXV 0 +/B Vf DQG I/ K\ORFHWHV 3 D Kf 1 QXPEHU RI SDLULQJV IRU ERWK VH[HV FRPELQHG

PAGE 70

3HUFHQW 1R $JJUHVVLRQ ’ 'UDZV ,QWUXGHU :LQV ‘ 5HVLGHQW :LQV

PAGE 71

)LJXUH 0HDQ s6(f QXPEHU RI QRQRIIHQVLYH DQG RIIHQVLYH DSSURDFKHV DWWDFNV DQG UHWUHDWV H[KLELWHG E\ UHVLGHQW (A PHODQRWLV DQG LQWUXGHU FRQVSHFLILFV RU KHWHURVSHFLILFV I/r VXPLFKUDVWL 0 P VDOYXV RU ( K\ORFHWHVf GXULQJ WKH PLQXWH WULDOV )RU HDFK EHKDYLRU FRPSDUH WKH QXPEHU RI WLPHV LW ZDV H[KLELWHG E\ WKH UHVLGHQW DQG LQWUXGHU ZLWK WKH :LOFR[RQ PDWFKHGSDLUV VLJQHGUDQN WHVW DQ DVWHULVN LQGLFDWHV VLJQLILFDQFH DW WKH OHYHO 6SHFLHV DEEUHYLDWLRQV DUH DV LQ )LJXUH

PAGE 72

0HDQ 1R RI 2FFXUUHQFHV 0HDQ 1R RI 2FFXUHQFHV 3 P [ 3 D K 1RQ 2IIHQVLYH 2IIHQVLYH $SSURDFK $SSURDFK $WWDFN 5HWUHDW 3 P [ 0 P V 1RQ 2IIHQVLYH $WWDFN 2IIHQVLYH $SSURDFK $SSURDFK 5HWUHDW

PAGE 73

DSSURDFKHV DQG WKH ODWWHU PRUH DWWDFN OHDSV FRPPRQO\ ZKHQ WKH 3 PHODQRWLV ZDV RULHQWHG WRZDUGV DQRWKHU REMHFW LQ WKH DUHQD 7KHVH UHVXOWV VXJJHVW WKDW UHVLGHQW eBr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e/ LLO VDOYXV 7KHVH UHVXOWV FRQWUDVW ZLWK WKRVH IURP JULGV RXWVLGH FRORQLHV ZKHUH DOO ILYH VSHFLHV RI PLFH FRPPRQO\ HVWDEOLVKHG UHVLGHQF\ 7KXV LW DSSHDUV WKDW DOO VSHFLHV EXW PHODQRWLV OHIW WKH PRQDUFK FRORQLHV VRRQ DIWHU HQFRXQWHULQJ WKHP 7KH GLHWV RI PHODQRWLV LQVLGH FRORQLHV FRQVLVWHG DOPRVW HQWLUHO\ RI PRQDUFK PDWHULDO DQG DYHUDJHG b OLSLG )HPDOHV KDG PRUH WKDQ WLPHV PRUH OLSLG WKDQ FRQVSHFLILFV DQG KHWHURVSHFLILFV RXWVLGH FRORQLHV 0\ GDWD VXJJHVW WKDW WKH HQHUJ\ DQG QXWULHQWV WKH PLFH GHULYHG IURP WKH PRQDUFKV HQDEOHG WKHP WR LQLWLDWH KLJK OHYHOV RI ZLQWHU UHSURGXFWLRQ FRQILUPLQJ WKH ILQGLQJV RI %URZHU HW DO f 6LPLODU UHVSRQVHV WR QDWXUDO DQG H[SHULPHQWDOO\ FUHDWHG IRRG DEXQGDQFH LQ ZLQWHU KDYH EHHQ UHSRUWHG LQ RWKHU URGHQW SRSXODWLRQV /LQGXVND

PAGE 74

:DWWV $QGU]HMHZVNL *DVKZHLOHU 7DLWW -HQVRQ (ULNVVRQ %ULJJV f 0RUHRYHU GLHWV ZLWK FRPSDUDEOH OHYHOV RI IDW WR bf ZHUH IRXQG WR VWLPXODWH UHSURGXFWLRQ DQG UHGXFH LQIDQW PRUWDOLW\ LQ ODERUDWRU\ UDWV 6FKHHU HW DO ,QQDPL HW DO f (YHQ WKRXJK WKH e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f DQG SUHYLRXVO\ VWXGLHG SRSXODWLRQV RI WKH VDPH VSHFLHV VWXGLHG KHUHLQ LQ KLJKDOWLWXGH UHJLRQV RI WKH 0H[LFDQ 7UDQVYROFDQLF 5DQJH DUH QR H[FHSWLRQ 9LOOD &DQHOD DQG 6£QFKH] 5REHUWVRQ 9 6£QFKH] SHUV FRPPf D YDULHW\ RI IDFWRUV LQWHUUHODWH WR SURGXFH WKLV SDWWHUQ LQFOXGLQJ HQHUJ\ QXWULHQWV SKRWRSHULRG KXPLGLW\ WHPSHUDWXUH DQG VRFLDO IDFWRUV +RZHYHU HQHUJHWLF FRQVWUDLQWV DUH WKRXJKW WR SOD\ D PDMRU UROH RZLQJ WR WKH KLJK FRVWV RI WKHUPRUHJXODWLRQ VHDUFKLQJ IRU VFDUFH IRRG DQG UHSURGXFWLRQ 6FKLSS HW DO 6DGOLHU HW DO 6WHEELQV 0LOODU 3RUWHU DQG 0F&OXUH %URQVRQ DQG 3HUULJR 3HUULJR f $SSDUHQWO\ JDLQLQJ DFFHVV WR PRQDUFKV DOORZV PHODQRWLV WR RYHUFRPH WKHVH HQHUJHWLF FRQVWUDLQWV DQG EUHHG GXULQJ WKH ZLQWHU

PAGE 75

'HPRJUDSKLF 5HVSRQVHV RI 3 PHODQRWLV WR WKH 0RQDUFK &RORQLHV 3HDN 01$ HVWLPDWHV RI PRXVH GHQVLWLHV LQVLGH FRORQLHV ZHUH PXFK KLJKHU WR KDf WKDQ WKRVH RXWVLGH WR KDf DQG WKRVH UHSRUWHG IRU RWKHU 3HURP\VFXV SRSXODWLRQV ZKRVH GLHWV KDG EHHQ VXSSOHPHQWHG DUWLILFLDOO\ WR KD )RUGKDP +DQVHQ DQG %DW]OL *LOEHUW DQG .UHEV 7DLWW %ULJJV
PAGE 76

,Q WKLV VWXG\ DQG WKDW RI %URZHU ee DO f LPPLJUDWLRQ WR PRQDUFK FRORQLHV ZDV IHPDOHELDVHG DQG WKH IHPDOHV LQVLGH FRORQLHV ZHUH VLJQLILFDQWO\ ODUJHU WKDQ WKH PDOHV ERWK LQVLGH DQG RXWVLGH FRORQLHV :KHUHDV VHYHUDO ZRUNHUV KDYH UHSRUWHG VLPLODU IHPDOHELDVHG UHVSRQVHV WR KLJK TXDOLW\ KDELWDWV LQ RWKHU 3HURP\VFXV SRSXODWLRQV %RZHUV DQG 6PLWK )RUGKDP f PDQ\ RWKHU ZRUNHUV KDYH QRW IRXQG VXFK D UHVSRQVH UHYLHZ LQ 9HVVH\ f 7KH UHDVRQ IRU WKH SUHSRQGHUDQFH RI IHPDOHV LQVLGH WKH FRORQLHV LV XQFOHDU %HFDXVH IHPDOHV KDG VPDOOHU KRPH UDQJHV WKDQ GLG PDOHV WKH\ PD\ KDYH EHHQ DEOH WR SDFN WKHPVHOYHV PRUH WLJKWO\ ZLWKLQ FRORQLHV ZLWKRXW H[WHQVLYH KRPH UDQJH RYHUODS 6HFRQG WKH IHPDOHV PD\ OLPLW PDOH LPPLJUDWLRQ DV VXJJHVWHG IRU VHYHUDO RWKHU 3HURP\VFXV VSHFLHV 0HW]JDU %RZHUV DQG 6PLWK f DQG 7DPLDV VWULDWXV :ROIH %UHQQHU HW DO f 8QIRUWXQDWHO\ GLG QRW H[DPLQH WKH LQWHUVH[XDO DJRQLVWLF UHODWLRQV RI eBr PHODQRWLV 7KLUG LW LV SRVVLEOH WKDW WKH QDWXUDO VH[UDWLR RQ 6LHUUD &KLQFXD ZDV IHPDOHELDVHG +RZHYHU WKLV LV XQOLNHO\ JLYHQ WKDW WKH PHDQ SHUFHQWDJH RI IHPDOHV 6(f LQ P\ FDSWLYH OLWWHUV RI 3 PHODQRWLV IURP IHPDOH DQG PDOH DGXOW SDUHQWVf FROOHFWHG RQ 6LHUUD &KLQFXD ZDV s 7KH KLJK GHQVLWLHV RI EUHHGLQJ IHPDOHV LQVLGH FRORQLHV DOVR PD\ KDYH FRQWULEXWHG WR WKH ORZ OHYHOV RI MXYHQLOH UHFUXLWPHQW 2WKHU ZRUNHUV KDYH UHSRUWHG QHJDWLYH FRUUHODWLRQV EHWZHHQ MXYHQLOH GHQVLWLHV DQG WKRVH RI EUHHGLQJ IHPDOHV LQ 3HURP\VFXV SRSXODWLRQV DQG WKH\ FRQFOXGHG WKDW EUHHGLQJ IHPDOHV KDG DJJUHVVLYHO\ H[FOXGHG \RXQJ PLFH IURP WKHLU KRPH UDQJHV +DQVHQ DQG %DW]OL *DOLQGR DQG .UHEV f &OHDUO\ WKH UROH RI DJJUHVVLRQ E\ EUHHGLQJ IHPDOHV WRZDUGV

PAGE 77

FRQVSHFLILF PDOHV DQG MXYHQLOHV LQ PHGLDWLQJ DFFHVV WR KLJK TXDOLW\ UHVRXUFHV GHVHUYHV IXUWKHU LQYHVWLJDWLRQ &RQVXPSWLRQ RI 0RQDUFK\ ZDV VXUSULVHG E\ WKH UHODWLYHO\ ODUJH DPRXQWV RI PRQDUFK PDWHULDO SUHVHQW LQ WKH VWRPDFKV RI er PHODQRWLV FROOHFWHG RXWVLGH FRORQLHV %URZHU HW O/ f DOVR IRXQG PRQDUFK PDWHULDO LQ WKH VWRPDFKV RI 3 PHODQRWLV IURP RXWVLGH FRORQLHV EXW GLG QRW TXDQWLI\ WKH DPRXQW 7KHUH DUH WZR ZD\V LQ ZKLFK WKHVH PLFH FRXOG KDYH DFFHVVHG PRQDUFK PDWHULDO f E\ PDNLQJ QLJKWO\ IRUD\V LQWR WKH QHDUE\ FRORQ\ DQGRU f E\ VFDYHQJLQJ ZLWKLQ WKHLU RZQ KRPH UDQJHV RQ GHDG RU PRULEXQG PRQDUFKV 0DQ\ EXWWHUIOLHV GLH RXWVLGH FRORQLHV EHFDXVH RI ELUG SUHGDWLRQ H[KDXVWLRQ GXH WR LQDGHTXDWH OLSLG UHVHUYHV DQG GHK\GUDWLRQ :DOIRUG %URZHU f 7KH VHFRQG SRVVLELOLW\ LV PRVW OLNHO\ EHFDXVH P\ WUDSSDELOLW\ HVWLPDWHV LQ JULGV RXWVLGH FRORQLHV ZHUH VR KLJK DQG GLG QRW UHFRUG DQ\ PLFH PRYLQJ IURP JULGV RXWVLGH WR RQHV LQVLGH FRORQLHV 7KHUHIRUH WKHUH PXVW KDYH EHHQ D VXEVWDQWLDO QXPEHU RI GHDG DQGRU PRULEXQG PRQDUFKV RQ WKH IRUHVW IORRU HDFK QLJKW LQ WKH JULGV RXWVLGH RI WKH FRORQLHV 7KLV LPSOLHV WKDW DOO VSHFLHV RI PLFH H[FHSW IRU e PHODQRWLV QRW RQO\ DYRLGHG WKH FRORQLHV EXW DOVR DYRLGHG WKH GHDG DQGRU PRULEXQG PRQDUFKV LQ WKH JULGV RXWVLGH RI WKH FRORQLHV FDQ RIIHU WZR H[SODQDWLRQV IRU ZK\ WKH VWRPDFK FRQWHQWV RI 3 PHODQRWLV RXWVLGH PRQDUFK FRORQLHV FRQWDLQHG VLJQLILFDQWO\ KLJKHU FRQFHQWUDWLRQV RI &*V WKDQ GLG WKRVH RI FRQVSHFLILFV LQVLGH FRORQLHV )LUVW EHFDXVH WKH GHQVLW\ RI PRQDUFKV LV VR PXFK KLJKHU LQVLGH FRORQLHV PLFH IRUDJLQJ WKHUH FRXOG IHHG VHOHFWLYHO\ RQ PRQDUFKV KDYLQJ

PAGE 78

ORZ OHYHOV RI &*V ZLWK OLPLWHG VHDUFKLQJ FRVWV &DJHG PHODQRWLV DUH DEOH WR GLVWLQJXLVK EHWZHHQ PRQDUFKV ZLWK ORZ DQG KLJK OHYHOV RI &*V &KDSWHUV DQG f )RU WKRVH PLFH RXWVLGH FRORQLHV WKH FDORULF FRVWV RI H[WHQVLYH VHDUFKLQJ IRU ORZ &* PRQDUFKV ZRXOG KDYH EHHQ PXFK KLJKHU DQG WKXV PD\ QRW KDYH EHHQ HQHUJHWLFDOO\ ZRUWKZKLOH 6HFRQG ZKHQ RIIHUHG K\GUDWHG PRQDUFKV LH RQHV WKDW DUH OLYH PRULEXQG RU UHFHQWO\ GHDGf FDSWLYH (/r PHODQRWLV PRVW FRPPRQO\ FRQVXPH WKH DEGRPLQDO PDWHULDO E\ ILUVW GLVFDUGLQJ WKH FXWLFOH DQG WKHQ HDWLQJ WKH LQWHUQDO WLVVXHV &KDSWHU f 0RQDUFK FXWLFOH LV NQRZQ WR FRQWDLQ KLJK FRQFHQWUDWLRQV RI &*V %URZHU HW DO f ,Q FRQWUDVW ZKHQ RIIHUHG GHVLFFDWHG PRQDUFKV FDSWLYH eAB PHODQRWLV HDW ERWK WKH FXWLFOH DQG FRQWHQWV DSSDUHQWO\ EHFDXVH WKH FXWLFOH EHFDPH WLJKWO\ ERXQG WR WKH LQWHUQDO WLVVXHV GXULQJ WKH GHVLFFDWLRQ SURFHVV &KDSWHU %URZHU HW DO f 7KHUHIRUH JLYHQ WKDW PDQ\ GHVLFFDWHG PRQDUFKV RFFXU RXWVLGH FRORQLHV *OHQGLQQLQJ SHUV REVHUYf DQG WKDW K\GUDWHG RQHV DUH FRPPRQ LQVLGH FRORQLHV 7DEOH f PLFH WKDW ZHUH IRUDJLQJ RXWVLGH FRORQLHV PD\ KDYH LQJHVWHG KLJKHU FRQFHQWUDWLRQV RI FXWLFOH DQG KHQFH RI &*V :K\ LV 3 PHODQRWLV WKH RQO\ VSHFLHV RI PLFH WKDW IHHGV RQ WKH PRQDUFK EXWWHUIOLHV" WHVWHG WKH K\SRWKHVHV WKDW WKH IRXU RWKHU VSHFLHV RI PLFH GLG QRW HDW PRQDUFKV EHFDXVH f WKH PLFURKDELWDW FKDUDFWHULVWLFV HJ XQGHUVWRU\ YHJHWDWLRQf RI RYHUZLQWHULQJ DUHDV PD\ KDYH RQO\ VXLWHG 3 PHODQRWLV RU f PHODQRWLV FRPSHWLWLYHO\ H[FOXGHG WKHP UHMHFWHG WKH ILUVW K\SRWKHVLV EHFDXVH WKH XQGHUVWRU\ YHJHWDWLRQ SDWWHUQV RI JULGV LQVLGH FRORQLHV ZHUH QRW VLJQLILFDQWO\ GLIIHUHQW IURP WKRVH RI JULGV RXWVLGH FRORQLHV DOVR UHMHFWHG WKH VHFRQG K\SRWKHVLV EHFDXVH

PAGE 79

UHVLGHQW PHODQRWLV ZHUH XQDEOH WR GRPLQDWH QX VDOYXV DQG D K\ORFHWHVI DQG ZHUH DEOH WR GRPLQDWH RQO\ DERXW KDOI RI WKH 5 VXPLFKUDVWL 7KLV LQGLFDWHG WKDW UHVLGHQW PHODQRWLV FRXOG QRW SUHYHQW DQ\ RI WKH VSHFLHV IURP IHHGLQJ RQ WKH PRQDUFKV DQG HVWDEOLVKLQJ UHVLGHQF\ LQVLGH WKH FRORQLHV 7KH UHVXOWV UHSRUWHG KHUHLQ DQG LQ &KDSWHUV WR VXJJHVW WKDW DOO VSHFLHV RI PLFH H[FHSW (A PHODQRWLV DYRLGHG WKH FRORQLHV EHFDXVH RI DQ DYHUVLRQ WR WKH PRQDUFKV H[DPLQH WKH IHHGLQJ UHVSRQVHV RI 3 PHODQRWLV eBr VXPLFKUDVWL (A K\ORFHWHV DQG QX VDOYXV WR RYHUZLQWHULQJ PRQDUFKV DQG WKHLU GHIHQVLYH FRPSRXQGV LQ WKH QH[W FKDSWHU

PAGE 80

&+$37(5 &203$5$7,9( )((',1* 5(63216(6 72 29(5:,17(5,1* 021$5&+ %877(5)/,(6 7R FRPSDUH WKH IHHGLQJ UHVSRQVHV RI eBr PHODQRWLV 3 D K\ORFHWHV (A VXPLFKUDVWL DQG (/ +/r VDOYXV WR RYHUZLQWHULQJ PRQDUFKV FRQGXFWHG H[SHULPHQWV ZLWK FDJHG LQGLYLGXDOV PHDVXUHG KRZ PDQ\ PRQDUFKV HDFK VSHFLHV DWH DQG WKH GHJUHH WR ZKLFK WKH\ DYRLGHG FXWLFXODU PDWHULDO 0RQDUFKV VWRUH KLJKHU FRQFHQWUDWLRQV RI &*V LQ WKHLU FXWLFOH FRPSDUHG WR WKHLU ERG\ FRQWHQWV 7DEOH %URZHU HW DO f 6HFRQG IXUWKHU H[SORUHG &* DYRLGDQFH LQ PHODQRWLV ZKLFK IHHGV QDWXUDOO\ RQ PRQDUFKV E\ RIIHULQJ WKHP D FKRLFH EHWZHHQ PDOH DQG IHPDOH PRQDUFKV 0DOH PRQDUFKV KDYH RQ DYHUDJH b ORZHU &* FRQFHQWUDWLRQV WKDQ IHPDOHV %URZHU DQG &DOYHUW f FRPSDUHG KRZ PDQ\ DEGRPHQV RI HDFK VH[ ZHUH HDWHQ DQG WKH H[WHQW WR ZKLFK HDFK VH[HVn FXWLFOH ZDV DYRLGHG 7KLUG H[DPLQHG KRZ ZHOO DOO IRXU VSHFLHV FRXOG PDLQWDLQ ZHLJKW RQ D GLHW FRQVLVWLQJ VROHO\ RI PRQDUFKV 0HWKRGV 7UDSSLQJ DQG 0DLQWHQDQFH RI 0LFH WUDSSHG DQG FRQGXFWHG H[SHULPHQWV IURP -DQXDU\ WKURXJK )HEUXDU\ RQ WKH VORSH RI WKH $UUR\R /D 3ODQFKD RI WKH 6LHUUD &KLQFXD PRXQWDLQ PDVVLI )LJXUH ODf 0LFH ZHUH WUDSSHG LQ 6KHUPDQ OLYH WUDSV [ [ FPf DW OHDVW P IURP WKH PRQDUFK FRORQ\ WR UHGXFH WKH SRVVLELOLW\ WKDW WKH\ KDG SUHYLRXVO\ HQFRXQWHUHG D

PAGE 81

PRQDUFK FRORQ\ %HFDXVH RI D KLJK YDULDQFH LQ IRRG FRQVXPSWLRQ LQ UHSURGXFWLYH PLFH SDUWLFXODUO\ IHPDOHV 6DGOLHU HW DO 6WHEELQV 0LOODU f RQO\ UHSURGXFWLYHO\ LQDFWLYH DGXOWV QRQODFWDWLQJ RU QRQSUHJQDQW IHPDOHV DQG QRQVFURWDO PDOHVf ZHUH XVHG LQ WKH H[SHULPHQWV (DFK PRXVH ZDV XVHG RQO\ RQFH 7KH PLFH ZHUH KRXVHG DQG WHVWHG LQGLYLGXDOO\ LQ ZLUH PHVK FDJHV DERXW FP KLJK [ FP LQ GLDPHWHUf VHW RQ D WDUSDXOLQFRYHUHG WDEOH LQ WKH VKDGH 'DFURQ EDWWLQJ ZDV DGGHG IRU QHVWLQJ PDWHULDO DQG SLHFHV RI FDUGERDUG ZHUH SODFHG EHWZHHQ WKH FDJHV WR LVRODWH WKH PLFH YLVXDOO\ )RU VL[ FRQVHFXWLYH QLJKWV SULRU WR WKH IHHGLQJ H[SHULPHQWV HDFK PRXVH ZDV PDLQWDLQHG RQ 3XULQD ODERUDWRU\ FKRZ QR KHQFHIRUWK PRXVH FKRZf DQG ZDWHU $OO RI WKHP IHG DQG GUDQN UHJXODUO\ DQG HLWKHU PDLQWDLQHG RU JDLQHG ZHLJKW 0RQDUFK &ROOHFWLRQ 0RQDUFKV ZHUH FROOHFWHG ZLWK EXWWHUIO\ QHWV IURP DFFHVVLEOH FOXVWHUV RQ ILU EUDQFKHV ZLWKLQ WKH EXWWHUIO\ FRORQ\ )RU ([SHULPHQWV DQG PRQDUFKV ZHUH RIIHUHG WR PLFH ZLWKRXW UHIHUHQFH WR WKHLU VH[ %UDQFK FOXVWHUV ZHUH DERXW b IHPDOHV GXULQJ -DQXDU\ DQG )HEUXDU\ Q VDPSOHV IRU D WRWDO RI EXWWHUIOLHV 7 9DQ +RRN XQSXEO GDWDf )RU ([SHULPHQW HTXDO QXPEHUV RI HDFK VH[ ZHUH RIIHUHG WR PLFH ([SHULPHQW 3DWWHUQV RI )HHGLQJ E\ )RXU 0RXVH 6SHFLHV 7KH IHHGLQJ UHVSRQVHV RI WKH VSHFLHV WR PRQDUFKV ZHUH FRPSDUHG (DFK PRXVH ZDV ZHLJKHG DW WKH EHJLQQLQJ RI WKH H[SHULPHQW 7KHQ PDOHV DQG IHPDOHV RI HDFK VSHFLHV ZHUH LQGLYLGXDOO\ FDJHG DQG HDFK ZDV RIIHUHG PRQDUFKV PRXVH FKRZ G OLELWXP DQG ZDWHU IRU WZR

PAGE 82

FRQVHFXWLYH QLJKWV ,Q WKLV DQG WKH QH[W H[SHULPHQW DOO PLFH UHFHLYHG OLYH DQG DFWLYHf PRQDUFKV H[FHSW IRU QX VDOYXV ZKLFK UHFHLYHG LQDFWLYH PRQDUFKV LH RQHV ZKRVH WKRUD[ KDG EHHQ VTXHH]HG ILUPO\f EHFDXVH WKH\ ZRXOG QRW DSSURDFK OLYH RQHV $W K DOO PRXVH FKRZ ZDV UHPRYHG IURP HDFK PRXVHn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b RI WKH FXWLFOH DQG FRQWHQWV LH IHG QRQVHOHFWLYHO\f RU b RI WKH FRQWHQWV RQO\ DQG GLVFDUGHG WKH FXWLFOH LH IHG VHOHFWLYHO\f %DVHG RQ WKHVH REVHUYDWLRQV FKRVH £ SULRUL FDWHJRULHV RI PRQDUFK GDPDJH 7DEOH f DVVXPH WKDW VDPSOLQJ D PRQDUFK UHSUHVHQWV DFWLYH UHMHFWLRQ RI LW EDVHG RQ WDVWH 'HWHUPLQDWLRQ RI WKH DPRXQW RI PRQDUFK WLVVXH FXWLFOH DQG &*V HDWHQ E\ HDFK PRXVH LQ ([SHULPHQWV DQG LQYROYHG VHYHUDO VWHSV )LUVW HVWLPDWHG WKH DPRXQW RI WLVVXH FXWLFOH DQG &*V LQ WKH ERG\ SDUWV RI RYHUZLQWHULQJ PRQDUFKV VHH 7DEOH f )RU WKHVH HVWLPDWLRQV PRGLILHG WKH GDWD IURP %URZHU eW DO f EHFDXVH WKH\ ZHUH GHULYHG IURP IUHVKO\HFORVHG $VFOHSLDV V\ULDFDUHDUHG PRQDUFKV ZKLFK KDYH KLJKHU &* FRQWHQWV DQG VXEVWDQWLDOO\ ORZHU DPRXQWV RI IDW LQ WKHLU DEGRPHQV WKDQ PRQDUFKV UHDUHG RQ WKH VDPH IRRG SODQW WKDW KDYH PLJUDWHG WR WKH 0H[LFDQ RYHUZLQWHULQJ JURXQGV %URZHU 0DOFROP DQG

PAGE 83

7DEOH &DWHJRULHV RI PRQDUFK GDPDJH E\ PLFH 1R YLVLEOH GDPDJH 6DPSOHG b RI DEGRPHQ DQGRU WKRUD[ HDWHQ b FXWLFOH DQG FRQWHQWVf RI WKRUD[ HDWHQ b FXWLFOH DQG FRQWHQWVf RI WKRUD[ HDWHQ $EGRPHQ HDWHQ QRQVHOHFWLYHO\ b RI FXWLFOH DQG FRQWHQWVf $EGRPHQ HDWHQ VHOHFWLYHO\ b RI FRQWHQWV FXWLFOH UHMHFWHGf

PAGE 84

%URZHU f 6HFRQG DVVXPHG WKDW ZKHQHYHU D PRXVH DWH f DQ DEGRPHQ QRQVHOHFWLYHO\ LW DWH b RI WKH FXWLFOH DQG FRQWHQWV f DQ DEGRPHQ VHOHFWLYHO\ LW DWH b RI WKH FRQWHQWV f WR b RI D WKRUD[ LW DWH b RI WKH FXWLFOH DQG FRQWHQWV DQG f WR b RI D WKRUD[ LW DWH b RI WKH FXWLFOH DQG FRQWHQWV 7KHVH DVVXPSWLRQV DJUHHG JHQHUDOO\ ZLWK P\ SUHOLPLQDU\ REVHUYDWLRQV 7KLUG FDOFXODWHG WKH WRWDO TXDQWLW\ RI WLVVXH FXWLFOH DQG &*V HDWHQ E\ D JLYHQ PRXVH RYHU D JLYHQ WLPH SHULRG ZLWK WKH IROORZLQJ HTXDWLRQV ZKLFK LQFRUSRUDWH GDWD IURP 7DEOH WLVVXH HDWHQ >QR DEGRPHQV HDWHQ VHOHFWLYHO\ DQG QRQVHOHFWLYHO\f PJf@ >QR WKRUDFHV HDWHQ bf bf PJf@ >QR WKRUDFHV HDWHQ bfbf PJf@ FXWLFOH HDWHQ >QR DEGRPHQV HDWHQ QRQVHOHFWLYHO\f PJf@ >QR WKRUDFHV HDWHQ bf bf PJf@ >QR WKRUDFHV HDWHQ bf bf PJf @ &* HDWHQ QR DEGRPHQV HDWHQ VHOHFWLYHO\f _OJf@ >QR DEGRPHQV HDWHQ QRQVHOHFWLYHO\f @OJf @ >QR WKRUDFHV HDWHQ bf bf LJf @ QR WKRUDFHV HDWHQ bf bf >LJf @ )RXUWK DGGHG WRJHWKHU WKH SUHGDWLRQ UHFRUGV IURP ERWK QLJKWV LQ ([SHULPHQW ZKHUHDV IRU ([SHULPHQW DQDO\]HG HDFK QLJKW VHSDUDWHO\ VR DV WR PRQLWRU FKDQJHV LQ FRQVXPSWLRQ RI PRQDUFK WLVVXH FXWLFOH DQG &*V RYHU WKH QLJKWV

PAGE 85

7DEOH 0HDQ GU\ ZHLJKWV DQG &* FRQWHQWV RI WKH DEGRPLQDO FXWLFOH DQG WLVVXH WKRUDFLF FXWLFOH DQG WLVVXH DQG ZLQJV RI IUHVKO\HFORVHG PRQDUFKV UHDUHG RQ $VFOHSLDV V\ULDFD IURP %URZHU HW DO 7DEOH f 7KHVH GDWD DORQJ ZLWK WKH PHDQ GU\ ZHLJKWV DQG &* FRQWHQWV DOO ERG\ SDUWV FRPELQHGf RI PRQDUFKV RYHUZLQWHULQJ LQ 0H[LFR 0DOFROP DQG %URZHU f ZHUH XVHG WR HVWLPDWH WKH GU\ ZHLJKWV DQG &* FRQWHQWV RI WKH GLIIHUHQW ERG\ SDUWV LQ RYHUZLQWHULQJ PRQDUFKV 'DWD IURP $ V\ULDFDUHDUHG PRQDUFKV DUH DSSURSULDWH IRU WKHVH HVWLPDWLRQV EHFDXVH WKH HYLGHQFH VXJJHVWV WKDW WR b RI WKH RYHUZLQWHULQJ PRQDUFKV LQ 0H[LFR KDYH IHG DV ODUYDH RQ WKLV PLONZHHG 6HLEHU HW DO 0DOFROP HW DO f DVVXPH WKDW f WKH GU\ ZHLJKWV RI WKH FXWLFOH ZLQJV DQG WKRUDFLF WLVVXH ZHUH WKH VDPH LQ ERWK W\SHV RI PRQDUFKV DQG WKDW RQO\ WKH DEGRPLQDO WLVVXH ZHLJKWV GLIIHUHG RZLQJ WR LQFUHDVHG DPRXQWV RI IDW DVVRFLDWHG ZLWK GLDSDXVH %URZHU f DQG f WKH GLVWULEXWLRQ RI &*V LQ WKH GLIIHUHQW ERG\ SDUWV ZDV WKH VDPH LQ ERWK W\SHV RI PRQDUFKV EXW WKDW WKH &* FRQWHQWV ZHUH SURSRUWLRQDOO\ ORZHU LQ HDFK ERG\ SDUW LQ RYHUZLQWHULQJ PRQDUFKV )UHVKO\HFORVHG PRQDUFKV (VWLPDWHG YDOXHV IRU RYHUZLQWHULQJ PRQDUFKV 'U\ ZHLJKW &* FRQWHQW 'U\ ZHLJKW &* FRQWHQW &* FRQFHQWUDWLRQ %RG\ 3DUW PJ SHUFHQW XJ SHUFHQW PJ SHUFHQW QJ SHUFHQW ?OT OJ G Z $EGRPLQDO FXWLFOH LL b b LL b b $EGRPLQDO WLVVVXH b b b b 7KRUD[ FXWLFOH b b b b 7KRUD[ WLVVXH b b b b :LQJV b b b b &RQWLQXHG

PAGE 86

7DEOH f§&RQWLQXHG &RQWHQWV RI DEGRPHQ WKRUD[ b b b b :LQJV FXWLFOH b b b b 6XP RI DOO SDUWV b b E b E b D 9DOXHV DUH URXQGHG WR WKH QHDUHVW ZKROH YDOXH 7KHVH DUH PHDQV GHULYHG IURP RYHUZLQWHULQJ PRQDUFK EXWWHUIOLHV DW 6LHUUD &KLQFXD 0DOFROP DQG %URZHU f

PAGE 87

%HFDXVH WKH VSHFLHV RI PLFH GLIIHUHG JUHDWO\ LQ VL]H 7DEOH f FRXOG QRW FRPSDUH GLUHFWO\ WKH DPRXQWV RI PRQDUFK WLVVXH FXWLFOH RU &* HDWHQ 7KHUHIRUH VWDQGDUGL]HG FRQVXPSWLRQ E\ FRPSXWLQJ WKH UDWLR RI WKH ZHLJKW RI WLVVXH FXWLFOH RU &* HDWHQ WR PRXVH ZHLJKW f $ Q 7KH SRZHU IXQFWLRQ ZHLJKW LV GHULYHG IURP D UHJUHVVLRQ RI FRQVXPSWLRQ UDWH RQ ZHLJKW IRU D ZLGH UDQJH RI PDPPDOV )DUORZ 3HWHUV f 0DVVVSHFLILF FRQVXPSWLRQ YDOXHV ZHUH FDOFXODWHG VHSDUDWHO\ IRU HDFK PRXVH XVLQJ WKHLU ZHLJKW DW WKH EHJLQQLQJ RI WKH H[SHULPHQW ([SHULPHQW 3UHGDWLRQ E\ 3 PHODQRWLV LQ 5HODWLRQ WR %XWWHUIO\ 6H[ 7KH SUHYLRXV H[SHULPHQW H[DPLQHG ZKHWKHU HLWKHU RI WKH VSHFLHV GLVFULPLQDWHG EHWZHHQ DEGRPLQDO FXWLFOH DQG FRQWHQWV 7KLV H[SHULPHQW DGGUHVVHG D PRUH FRPSOH[ TXHVWLRQ :KHQ JLYHQ D FKRLFH EHWZHHQ PRQDUFKV RI GLIIHULQJ &* FRQFHQWUDWLRQV LQ WKLV FDVH PDOHV YHUVXV IHPDOHVf ZLOO eBr PHODQRWLV HDW PRUH ORZ &* RQHV DQG IHHG VHOHFWLYHO\ RQ PRUH RI WKH KLJK &* RQHV" 0DOH PRQDUFKV KDYH RQ DYHUDJH b ORZHU &* FRQFHQWUDWLRQV WKDQ IHPDOHV %URZHU DQG &DOYHUW f FRXOG QRW HVWLPDWH VHSDUDWHO\ IRU ERWK VH[HV WKH DPRXQWV RI WLVVXH FXWLFOH DQG &*V HDWHQ EHFDXVH WKH GDWD ZHUH QRW DYDLODEOH WKRVH LQ 7DEOH DUH IURP SRROHG VDPSOHV RI PDOH DQG IHPDOH PRQDUFKV 6HYHQ IHPDOH DQG PDOH PHODQRWLV ZHUH HDFK RIIHUHG IHPDOH DQG PDOH PRQDUFKV DQG PRXVH FKRZ DQG ZDWHU G OLELWXP SHU QLJKW IRU WZR FRQVHFXWLYH QLJKWV )LYH KRXUV EHIRUH HDFK IHHGLQJ WULDO VTXHH]HG WKH WKRUDFHV RI WKH IHPDOH DQG PDOH PRQDUFKV DQG WKHQ SODFHG D PP GLDPHWHU UHG RU JROG VSRW RI QDLO SROLVK RQ HDFK EXWWHUIO\ DEGRPHQ WKRUD[ KHDG DQG ZLQJV VR DV WR HQDEOH WDOO\LQJ RI WKH SDUWV

PAGE 88

E\ VH[ DIWHU WKH PLFH GLVPHPEHUHG WKHP 7KH DVVLJQPHQW RI JROG RU UHG IRU PDOHV RU IHPDOHV ZDV GHWHUPLQHG UDQGRPO\ IRU HDFK PRXVH %HFDXVH VH[XDO GLIIHUHQFHV LQ ZLQJ OHQJWK DUH QHJOLJLEOH DQG ZHW ZHLJKWV YDU\ XQSUHGLFWDELOW\ %URZHU DQG &DOYHUW f FRQVLGHUHG VL]H DQG ZHLJKW XQOLNHO\ EDVHV IRU NLOOLQJ DQ H[FHVV RI RQH VH[ ([SHULPHQW (IIHFWV RI /RQJ7HUP &RQVXPSWLRQ RI 0RQDUFKV ,Q WKLV H[SHULPHQW H[SORUHG ZKHWKHU DOO VSHFLHV FRXOG PDLQWDLQ RU JDLQ ZHLJKW RQ D SXUH GLHW RI PRQDUFKV (DFK PRXVH (MB PHODQRWLV eX VXPLFKUDVWL / K\ORFHWHV DQG / ,/ VDOYXD VH[UDWLRV ZHUH DSSUR[LPDWHO\ HTXDOf ZDV SUHH[SRVHG WR PRQDUFKV DQG PRXVH FKRZ DG LELWXP SHU QLJKW IRU QLJKWV 7KHQ IRU WKH QH[W QLJKWV HDFK PRXVH UHFHLYHG PRQDUFKV SHU QLJKW DQG ZDWHU DG OLELWXP ZLWKRXW PRXVH FKRZ 3UHOLPLQDU\ IHHGLQJ WULDOV LQGLFDWHG WKDW DOO VSHFLHV DWH IHZHU WKDQ WKLV QXPEHU RI PRQDUFKV SHU QLJKW (DFK PRXVHnV ZHLJKW ZDV WDNHQ DW WKH EHJLQQLQJ DQG HQG RI WKH H[SHULPHQW GD\V ODWHU $OO PLFH ZHUH WHVWHG EHWZHHQ DQG )HEUXDU\ 6WDWLVWLFDO $QDO\VHV $OO VWDWLVWLFDO WHVWV IROORZHG =DU f ,Q ([SHULPHQW SDLUHG WDLOHGf WWHVWV ZHUH UXQ VHSDUDWHO\ RQ HDFK VSHFLHV WR FRPSDUH WKH QXPEHU RI DEGRPHQV LWV WKRUDFHV HDWHQ DQG WKH QXPEHU RI DEGRPHQV HDWHQ VHOHFWLYHO\ [V QRQVHOHFWLYHO\ 2QHZD\ $129$nV DQG 6FKHII )WHVWV ZHUH XVHG WR FRPSDUH WKH VSHFLHV LQ WHUPV RI WKH PDVVVSHFLILF DPRXQWV RI PRQDUFK WLVVXH FXWLFOH RU &*V HDWHQ DV ZHOO DV WKH UDWLRV RI WKH DPRXQWV RI &*V WR WLVVXHV HDWHQ ZDV MXVWLILHG LQ XVLQJ WKH PDVV VSHFLILF FRQVXPSWLRQ YDOXHV LQ WKH $129$nV EHFDXVH WKH FRHIILFLHQW RI

PAGE 89

YDULDWLRQ RI WKH VFDOLQJ YDULDEOH LH PRXVH ZHLJKWf ZDV FRQVLVWHQWO\ OHVV WKDQ WKDW RI WKH GHSHQGHQW YDULDEOH LH DPRXQW RI WLVVXH FXWLFOH RU &* FRQVXPHGf$QGHUVRQ DQG /\GLF 3DFNDUG DQG %RDUGPDQ f 7KH XQSDLUHG WWHVW ZDV XVHG WR FRPSDUH WKH SHUFHQWDJH RI DEGRPHQV HDWHQ VHOHFWLYHO\ E\ eBr PHODQRWLV DQG 5Br VXPLFKUDVWL 'DWD ZHUH WUDQVIRUPHG DUFVLQ 9[f IRU WKLV DQG DOO VXEVHTXHQW VWDWLVWLFDO FRPSDULVRQV RI SHUFHQWDJH GDWD 3DLUHG WWHVW FRPSDULVRQV ZHUH XVHG LQ ([SHULPHQW WR FRPSDUH WKH WRWDO PDOH DQG IHPDOH PRQDUFK DEGRPHQV HDWHQ VHOHFWLYHO\ DQG QRQ VHOHFWLYHO\f DV ZHOO DV WKH SHUFHQWDJH RI WKRVH DEGRPHQV WKDW ZHUH HDWHQ VHOHFWLYHO\ 7ZRIDFWRU $129$nV UHSHDWHG RQ WLPHf ZHUH XVHG LQ ([SHULPHQW WR GHWHUPLQH WKH HIIHFWV RI VSHFLHV (BA PHODQRWLV K\ORFHWHV DQG eBr VXPLFKUDVWLf DQG WLPH RQ WLVVXH FXWLFOH DQG &* FRQVXPSWLRQ 7KH VDPH WHVW ZDV XVHG DJDLQ WR GHWHUPLQH WKH HIIHFW RI VSHFLHV DQG WLPH RQ WKH SHUFHQWDJH RI DEGRPHQV HDWHQ VHOHFWLYHO\ E\ PHODQRWLV DQG VXPLFKUDVWL 0 P VDOYXV ZDV H[FOXGHG IURP DOO VWDWLVWLFDO DQDO\VHV LQ WKLV H[SHULPHQW 5HVXOWV ([SHULPHQW 3DWWHUQV RI )HHGLQJ E\ DOO )RXU 0RXVH 6SHFLHV 3HURP\VFXV PHODQRWLV VXPLFKUDVWL DQG K\ORFHWD6 HDFK DWH PRUH WKDQ b RI WKH PRQDUFKV RIIHUHG WR WKHP 7DEOH f DQG VLJQLILFDQWO\ PRUH DEGRPHQV WKDQ WKRUDFHV )LJXUH f ,Q FRQWUDVW 0 P VDOYXV DWH OHVV WKDQ b RI WKH PRQDUFKV RIIHUHG WR WKHP DQG YLUWXDOO\ LGHQWLFDO QXPEHUV RI DEGRPHQV DQG WKRUDFHV 0RUHRYHU QX

PAGE 90

7DEOH &RPSDULVRQ RI WKH IHHGLQJ SDWWHUQV RI DOO PRXVH VSHFLHV XSRQ PRQDUFK EXWWHUIOLHV (LJKW LQGLYLGXDOV SHU VSHFLHV ZHUH HDFK RIIHUHG PRQDUFKV DQG PRXVH FKRZ G OLELWXP SHU QLJKW IRU WZR FRQVHFXWLYH QLJKWV )RU HDFK VSHFLHV WKH PHDQ QXPEHU RI PRQDUFKV HDWHQ VDPSOHG RU OHIW XQGDPDJHG s 6(f E\ HDFK VSHFLHV DUH SUHVHQWHG %HFDXVH PLFH DWH ERWK WKH DEGRPHQ DQG WKRUD[ RI VRPH PRQDUFKV WKH URZ WRWDOV DUH DOO JUHDWHU WKDQ 0HDQ QR RI PRQDUFKV s 6( LQ HDFK IHHGLQJ FDWHJRU\ 1R YLVLEOH 7KRUDFHV HDWHQ $EGRPHQV HDWHQ 1R RIIHUHG 6SHFLHV GDPDJH 6DPSOHG b b QRQVHOHFW VHOHFWLYHO\ PRXVH PHODQRWLV s s s s s s / VXPLFKUDVWL s s s s s s eM D]WHFXV s s s s s s / VDOYXV s s s s s sO RR RR

PAGE 91

‘ ‘ PP )LJXUH 0HDQ QXPEHU RI PRQDUFK DEGRPHQV DQG WKRUDFHV 6(f HDWHQ E\ LQGLYLGXDOV RI HDFK RI PRXVH VSHFLHV (DFK PRXVH UHFHLYHG PRQDUFKV DQG PRXVH FKRZ OLELWXP SHU QLJKW IRU WZR FRQVHFXWLYH QLJKWV 7KH VLJQLILFDQW ZLWKLQ VSHFLHV FRPSDULVRQV EHWZHHQ WKH QXPEHU RI DEGRPHQV DQG WKRUDFHV HDWHQ DUH LQGLFDWHG E\ DVWHULVNV

PAGE 92

0HDQ QXPEHU HDWHQ 3 PHODQRWLV r r 5 VXPLFKUDVWL Â’ $EGRPHQV + 7KRUDFHV rr3 GI SDLUHG WWHVW 3 D 0 P K\ORFHWHV VDOYXV

PAGE 93

VDOYXV DSSURDFKHG RQO\ LPPRELOL]HG PRQDUFKV 1RQH RI WKH PLFH FRQVXPHG WKH ZLQJV KHDG RU OHJV 7KH VWDQGDUGL]HG FRQVXPSWLRQ YDOXHV DUH VKRZQ LQ 7DEOH 3HURP\VFXV PHODQRWLV DQG / VXPLFKUDVWL FRQVXPHG VLJQLILFDQWO\ PRUH PRQDUFK WLVVXH WKDQ GLG (Br K\ORFHWHV DQG +r (E VDOYXV +RZHYHU D K\ORFHWHV DWH VLJQLILFDQWO\ PRUH FXWLFOH DQG ERWK (AB VXPLFKUDVWL DQG 3 D K\ORFHWHV DWH PRUH &*V WKDQ GLG WKH RWKHU VSHFLHV 7R KHOS LQWHUSUHW WKHVH SDWWHUQV RI FRQVXPSWLRQ FRPSXWHG IRU HDFK VSHFLHV WKH UDWLR RI WKH DPRXQWV RI &*V WR WLVVXHV HDWHQ 7KH UDWLRV IRU PHODQRWLV DQG (G VXPLFKUDVWL ZHUH VLJQLILFDQWO\ ORZHU WKDQ WKRVH IRU WKH RWKHU VSHFLHV 7DEOH f 7KH ODUJH LQWHUVSHFLILF GLIIHUHQFHV LQ WKHVH UDWLRV FDQ EH H[SODLQHG E\ WKH ZD\ HDFK PRXVH VSHFLHV FRQVXPHG PRQDUFK DEGRPHQV 3HURP\VFXV D K\ORFHWHV DQG ./ (E VDOYXV UDUHO\ DWH DEGRPHQV VHOHFWLYHO\ ZKHUHDV (A PHODQRWLV DQG (A VXPLFKUDVWL ERWK DWH VLJQLILFDQWO\ PRUH DEGRPHQV VHOHFWLYHO\ WKDQ QRQVHOHFWLYHO\ )LJXUH f 3HURP\VFXV PHODQRWLV IHG VHOHFWLYHO\ RQ D VLJQLILFDQWO\ JUHDWHU SHUFHQWDJH RI DEGRPHQV WKDQ GLG (A VXPLFKUDVWL PHDQV DQG UHVSHFWLYHO\ XQSDLUHG WYDOXH GI 3 f 7KXV E\ IHHGLQJ VHOHFWLYHO\ (A PHODQRWLV DQG (A VXPLFKUDVWL UHGXFHG VXEVWDQWLDOO\ WKH &* FRQFHQWUDWLRQ RI WKH PRQDUFK WLVVXH WKH\ LQJHVWHG %\ H[DPLQLQJ PRXVHGDPDJHG PRQDUFKV DQG ZDWFKLQJ PLFH HDW WKHP GLVFRYHUHG LQWHUVSHFLILF GLIIHUHQFHV LQ WKH ZD\ WKH IRXU VSHFLHV H[WUDFWHG DEGRPLQDO FRQWHQWV 3HURP\VFXV PHODQRWLV DQG (A VXPLFKUDVWL FRPPRQO\ PDGH ORQJLWXGLQDO VOLWV GRZQ WKH IXOO OHQJWK RI WKH DEGRPHQ DQG WKHQ VXFNHG DQGRU OLFNHG RXW WKH DEGRPLQDO FRQWHQWV ,Q FRQWUDVW

PAGE 94

7DEOH ,QWHUVSHFLILF FRPSDULVRQ RI WKH VWDQGDUGL]HG DPRXQWV RI PRQDUFK WLVVXH FXWLFOH DQG &*V HDWHQ 7KH PHDQ UDWLR RI WKH DPRXQWV RI &*V WR WLVVXH LQJHVWHG s 6(f LV DOVR FRPSDUHG DPRQJ VSHFLHV (LJKW PLFH IURP HDFK VSHFLHV ZHUH HDFK RIIHUHG IRUW\ PRQDUFKV DQG PRXVH FKRZ DG OLELWXP SHU QLJKW IRU WZR FRQVHFXWLYH QLJKWV %HWZHHQ VSHFLHV FRPSDULVRQV DUH PDGH ZLWK RQHZD\ $129$ 'LIIHUHQW VXEVFULSWV D E DQG Ff LQGLFDWH VLJQLILFDQW GLIIHUHQFHV DPRQJ VSHFLHV ZLWKLQ HDFK FROXPQ 3 6FKHII )WHVWf 0RXVH VSHFLHV 6WDQGDUGL]HG DPRXQWV HDWHQ PHDQ s 6(f &*VWLVVXH UDWLRr 7LVVXH JNJ! PRXVHf &XWLFOH JNJ PRXVHf &*V PJNJ PRXVHf (U PHODQRWLV s D s D s D s D VXPLFKUDVWL s D s D E s E s D A K\ORFHWHV s E s E s E s E e/B QX VDOYXV s F s F s F s F )YDOXH GI 3 r 'HWHUPLQHG IURP LQGLYLGXDO UDWLRV

PAGE 95

)LJXUH 0HDQ QXPEHU RI PRQDUFK DEGRPHQV s6(f HDWHQ VHOHFWLYHO\ DQG QRQVHOHFWLYHO\ E\ IRXU PRXVH VSHFLHV Q VSHFLHVf (DFK PRXVH UHFHLYHG PRQDUFKV DQG PRXVH FKRZ DG OLELWXP SHU QLJKW IRU WZR FRQVHFXWLYH QLJKWV

PAGE 96

Q 3 PHODQRWLV 5 VXPLFKUDVWL r r 7 3 D K\ORFHWHV _+ 1RQ6HOHFWLYHO\ Â’ 6HOHFWLYHO\ 3 r 3 GI SDLUHG WWHVW r r 0 P VDOYXV

PAGE 97

DBB K\ORFHWHV DQG QX VDOYXV XVXDOO\ ELW RII WKH HQG RI WKH DEGRPHQ DQG SXOOHG WKH FRQWHQWV RXW ZLWK WKHLU WHHWK ([SHULPHQW 3UHGDWLRQ E\ 3 PHODQRWLV LQ 5HODWLRQ WR %XWWHUIO\ 6H[ 3HURP\VFXV PHODQRWLV NLOOHG IHHGLQJ FDWHJRULHV 7DEOH f HTXDO QXPEHUV RI PDOH DQG IHPDOH PRQDUFKV ZKLFK VXJJHVWV QR LQLWLDO GLIIHUHQFH LQ WKH ULVN RI PDOH DQG IHPDOH PRQDUFKV WR PRXVH DWWDFN 7DEOH f +RZHYHU DGGLQJ WRJHWKHU WKH QXPEHU RI DEGRPHQV HDWHQ VHOHFWLYHO\ DQG QRQVHOHFWLYHO\ UHYHDOV WKDW 5A PHODQRWLV DWH D ODUJHU QXPEHU RI PDOH WKDQ IHPDOH DEGRPHQV PHDQV DQG UHVSHFWLYHO\ SDLUHG WYDOXH GI 3 f 2I WKRVH DEGRPHQV HDWHQ 5A PHODQRWLV IHG VHOHFWLYHO\ RQ JUHDWHU SHUFHQWDJH RI IHPDOH DEGRPHQV IHPDOH DQG PDOH PHDQV DQG UHVSHFWLYHO\ SDLUHG WYDOXH GI 3 f 7KHVH UHVXOWV VXSSRUW WKH K\SRWKHVLV WKDW ERWK WKH TXDQWLW\ RI WLVVXH HDWHQ DQG WKH WHQGHQF\ WR IHHG VHOHFWLYHO\ DUH LQIOXHQFHG E\ &* FRQFHQWUDWLRQ ([SHULPHQW (IIHFWV RI /RQJ7HUP &RQVXPSWLRQ RI 0RQDUFKV $OO 5r PHODQRWLV JDLQHG ZHLJKW RYHU WKH VL[ QLJKWV ZKHUHDV b RI WKH LQGLYLGXDOV RI HDFK RI WKH RWKHU VSHFLHV ORVW ZHLJKW 7DEOH f 0LFURWXV VX VDOYXV ZHUH H[FOXGHG IURP WKH VWDWLVWLFDO DQDO\VHV EHFDXVH WKH\ DWH YHU\ IHZ PRQDUFKV DQG DOO GLHG ZLWKLQ IRXU GD\V 7KH PDVVVSHFLILF DPRXQWV RI PRQDUFK WLVVXH FXWLFOH DQG &*V HDWHQ QLJKWO\ E\ 5A PHODQRWLV K\ORFHWHV DQG VXPLFKUDVWL DUH SORWWHG LQ )LJXUH DF 7KH WZRIDFWRU $129$nV SHUIRUPHG VHSDUDWHO\ RQ WLVVXH FXWLFOH DQG &*V DOO LQGLFDWHG VLJQLILFDQW HIIHFWV RI VSHFLHV IRU WLVVXH ) GI 3 IRU FXWLFOH ) GI

PAGE 98

7DEOH &RPSDUDWLYH IHHGLQJ SDWWHUQV E\ PHODQRWLV XSRQ PDOH DQG IHPDOH PRQDUFK EXWWHUIOLHV )LIWHHQ PLFH ZHUH HDFK RIIHUHG PDOH DQG IHPDOH EXWWHUIOLHV WRJHWKHU ZLWK PRXVH FKRZ G OLELWXP SHU QLJKW IRU WZR FRQVHFXWLYH QLJKWV )RU HDFK PRQDUFK VH[ WKH PHDQ QXPEHU HDWHQ VDPSOHG RU OHIW XQGDPDJHG s 6(f E\ WKH PLFH DUH SUHVHQWHG %HFDXVH PLFH DWH ERWK WKH DEGRPHQ DQG WKRUD[ RI VRPH PRQDUFKV WKH URZ WRWDOV DUH DOO JUHDWHU WKDQ 0HDQ QR RI PRQDUFKV s 6( LQ HDFK IHHGLQJ FDWHJRU\ 7KRUDFHV HDWHQ $EGRPHQV HDWHQ 1R 0RQDUFK VH[ 1R YLVLEOH GDPDJH 6DPSOHG b b QRQVHOHFW VHOHFWLYHO\ RIIHUHG PRXVH PDOH s s s s s s 92 IHPDOH s s s s s s

PAGE 99

7DEOH ,QLWLDO ZHLJKWV PHDQ s6(f ZHLJKW FKDQJHV DQG b PRUWDOLW\ LQ 3Br PHODQRWLV 5 VXPLFKUDVWL 3 D K\ORFHWHV DQG +8 VDOYXV JLYHQ PRQDUFKV SHU PRXVH SHU QLJKW IRU FRQVHFXWLYH QLJKWV ZLWKRXW DOWHUQDWLYH IRRG 0RXVH 6SHFLHV 1 ,QLWLDO ZHLJKW Jf :HLJKW FKDQJH Jf b ZHLJKW FKDQJH b RI PLFH WKDW ORVW ZHLJKW b PRUWDOLW\ PHODQRWLV sf sf 5M VXPLFKUDVWL sf sf 3 D KYORFHWHV f sf / (/ VDOYXV sf sf r $OO GLHG ZLWKLQ GD\V

PAGE 100

, )LJXUH 6WDQGDUGL]HG DPRXQWV RI PRQDUFK WLVVXH FXWLFOH DQG &*V HDWHQ E\ / PHODQRWLV -/ VXPLFKUDVWL DQG A K\ORFHWHV (DFK PRXVH ZDV JLYHQ PRQDUFKV SHU QLJKW IRU VL[ FRQVHFXWLYH QLJKWV DIWHU D WZR QLJKW SUHH[SRVXUH WR PRQDUFKV DQG PRXVH FKRZ

PAGE 101

0HDQ s6(f DPRXQW HDWHQ R 3 PHRQRWLV f 5 VXPFKUDVWL

PAGE 102

3 DQG IRU &*V ) GI 3 f DQG RI WLPH IRU &*V ) GI 3 f 7KHUH ZHUH QRW VLJQLILFDQW HIIHFWV RI WLPH IRU WLVVXH ) GI 3 f QRU FXWLFOH ) GI 3 f 7KH LQWHUDFWLRQ RI VSHFLHV DQG WLPH ZDV VLJQLILFDQW IRU WLVVXH ) GI 3 f FXWLFOH ) GI 3 f DQG &*V ) GI 3 f 7KHVH UHVXOWV GHPRQVWUDWH WKDW f PHODQRWLV DQG VXPLFKUDVWL FRQVLVWHQWO\ DWH PRUH WLVVXH WKDQ GLG A K\ORFHWHV f (BB K\ORFHWHV FRQVLVWHQWO\ DWH PRUH FXWLFOH DQG &*V WKDQ GLG WKH RWKHU WZR VSHFLHV RI PLFH DQG f QLJKWO\ FRQVXPSWLRQ RI FXWLFOH DQG &*V E\ VXPLFKUDVWL DQG D K\ORFHWHV UHPDLQHG URXJKO\ FRQVWDQW RYHU WKH VL[ QLJKWV ZKHUHDV WKDW RI PHODQRWLV WHQGHG WR GLPLQLVK $V LQ ([SHULPHQW (BA PHODQRWLV DQG VXPLFKUDVWL RIWHQ IHG XSRQ DEGRPHQV VHOHFWLYHO\ )LJXUH f 7KH WZRIDFWRU $129$ LQGLFDWHG D VLJQLILFDQW HIIHFW RI WLPH ) GI 3 f EXW QRW RI VSHFLHV ) GI 3 f RU RI WKH VSHFLHV [ WLPH LQWHUDFWLRQ ) GI 3 f RQ WKH QLJKWO\ UDWH RI VHOHFWLYH IHHGLQJ E\ WKHVH WZR VSHFLHV %HFDXVH (AB K\ORFHWHV UDUHO\ IHG VHOHFWLYHO\ RQ PRQDUFK DEGRPHQV PHDQ b s RI WRWDO DEGRPHQV HDWHQf LW ZDV H[FOXGHG IURP )LJXUH DQG WKH DQDO\VLV )LJXUH VXJJHVWV WKDW RQO\ (BA PHODQRWLV LQFUHDVHG LWV QLJKWO\ UDWH RI VHOHFWLYH IHHGLQJ ZLWK WLPH 7KLV DSSDUHQW LQWHUVSHFLILF GLIIHUHQFH PD\ QRW KDYH EHHQ GHWHFWHG E\ WKH $129$ EHFDXVH RI WKH ODUJH DPRXQW RI LQWUDVSHFLILF YDULDWLRQ DQG VPDOO VDPSOH VL]HV 7KH UDWLR RI &*V WR WLVVXH LQJHVWHG UHPDLQHG QHDUO\ FRQVWDQW RYHU WKH VL[ QLJKWV IRU DOO WKUHH VSHFLHV )LJXUH f 7KH WZR IDFWRU $129$ LQGLFDWHG D VLJQLILFDQW HIIHFW RI VSHFLHV ) GI 3

PAGE 103

)LJXUH 0HDQ SHUFHQWDJH RI DEGRPHQV s 6(f HDWHQ VHOHFWLYHO\ E\ eBr PHODQRWLV DQG 5 VXPLFKUDVWL (DFK PRXVH ZDV JLYHQ PRQDUFKV SHU QLJKW IRU VL[ FRQVHFXWLYH QLJKWV DIWHU D WZR QLJKW SUHH[SRVXUH WR PRQDUFKV DQG PRXVH FKRZ

PAGE 104

0HDQ s6(fb RI DEGRPHQV HDWHQ VHOHFWLYHO\ R 5 PHRQRWLV 92

PAGE 105

)LJXUH 5DWLRV PHDQ s 6(f RI PRQDUFK &*V PJf WR WLVVXHV Jf HDWHQ E\ VL[ PHODQRWLV ILYH 5 VXPLFKUDVWL DQG IRXU / eBm K\ORFHWHV RYHU VL[ FRQVHFXWLYH QLJKWV (DFK PRXVH UHFHLYHG PRQDUFKV SHU QLJKW DQG QR PRXVH FKRZf DIWHU D WZR QLJKW SUHH[SRVXUH WR PRQDUFKV DQG PRXVH FKRZ

PAGE 106

0HDQ UDWLR RI &*V WR WLVVXH HDWHQ PJJf S S S S S S R 2 f§ UR RM t ML Dr 5 VXPLFKUDVWL

PAGE 107

f EXW QRW RI WLPH ) GI 3 f RU RI WKH VSHFLHV [ WLPH LQWHUDFWLRQ ) GI 3 f 7KXV E\ IHHGLQJ VHOHFWLYHO\ RQ DEGRPHQV (Br PHODQRWLV DQG eBr VXPLFKUDVWL UHGXFHG VXEVWDQWLDOO\ WKH UDWLRV RI &*V WR WLVVXH LQ WKHLU GLHW RYHU WKH VL[ QLJKWV 'LVFXVVLRQ 3DWWHUQV RI )HHGLQJ E\ 0LFH 7KH WZR VSHFLHV WKDW IHG VHOHFWLYHO\ RQ PRQDUFK DEGRPHQV 3 PHODQRWLV DQG %A VXPLFKUDVWL DOVR DWH WKH PRVW PRQDUFK WLVVXH RQ D ZHLJKWVSHFLILF EDVLV %\ GLVFDUGLQJ WKH DEGRPLQDO FXWLFOH WKH\ VXEVWDQWLDOO\ UHGXFHG LQJHVWLRQ RI &*V VHH 7DEOH f 3HURP\VFXV D K\ORFHWHV DQG +r HX VDOYXV U RQ WKH RWKHU KDQG UDUHO\ GLVFDUGHG FXWLFXODU PDWHULDO DQG WKXV DWH VLJQLILFDQWO\ PRUH FXWLFOH DQG &*V WKDQ GLG PHODQRWLV DQG (A VXPLFKUDVWL 7KHUH DUH VHYHUDO SRVVLEOH UHDVRQV ZK\ EO VDOYXV DQG eBrB D K\ORFHWHV DWH VR OLWWOH PRQDUFK PDWHULDO DV FRPSDUHG WR WKH RWKHU WZR VSHFLHV (YHQ WKRXJK 0LFURWXV VSHFLHV DUH NQRZQ WR HDW LQVHFWV WR YDU\LQJ GHJUHHV %DW]OL +DQVNL DQG 3DUYLDLQHQ f WKH\ WHQG WR EH VWURQJO\ KHUELYRURXV DQG WKXV PD\ EH OHVV PRWLYDWHG WR DWWDFN DQG HDW LQVHFWV SDUWLFXODUO\ PRYLQJ RQHV WKDQ DUH PRUH RPQLYRURXV VSHFLHV &\U f ,Q IDFW '/r VDOYXV ZDV WKH RQO\ VSHFLHV WKDW ZRXOG QRW DWWDFN DFWLYH PRQDUFKV 1HYHUWKHOHVV WKH IDFW WKDW ,/ QX VDOYXV DOVR UHIXVHG WR HDW LQDFWLYH PRQDUFKV VXJJHVWV WKDW WKH\ ZHUH DOVR DYHUVH WR WKH PRQDUFKnV GHIHQVLYH FRPSRXQGV 3HURP\VFXV D K\ORFHWHV RQ WKH RWKHU KDQG DSSHDUV WR KDYH D JUHDWHU SUHGLOHFWLRQ IRU HDWLQJ LQVHFWV JLYHQ WKH VWURQJO\ RPQLYRURXV IHHGLQJ KDELWV RI LWV JHQXV /DQGU\

PAGE 108

:KLWDNHU :ROII HW DO f DQG WKH ODUJH QXPEHUV RI PRQDUFKV WKDW LQGLYLGXDOV DWH )LJXUH DQG 7DEOH f ,WV DYHUVLRQ WR WKH PRQDUFKV LV PRUH OLNHO\ UHODWHG WR LWV LQWROHUDQFH WR &*V &KDSWHU f (YHQ WKRXJK eAB VXPLFKUDVWL LV DOVR VWURQJO\ GHWHUUHG E\ WKHVH FRPSRXQGV LW OLPLWHG LQJHVWLRQ RI WKHP WKURXJK LWV VHOHFWLYH IHHGLQJ EHKDYLRU ,W LV XQFOHDU ZK\ RQO\ PHODQRWLV DQG (A VXPLFKUDVWL FRPPRQO\ VKRZHG WKH VHOHFWLYH IHHGLQJ EHKDYLRU 7KH\ PD\ KDYH PRGLILHG QRUPDO IUXLW RU LQVHFWHDWLQJ EHKDYLRUV DV KDV EHHQ VXJJHVWHG IRU EODFN EDFNHG RULROHV ZKLFK DOVR IHHG VHOHFWLYHO\ RQ PRQDUFK DEGRPHQV %URZHU DQG )LQN f +RZHYHU ZK\ ZRXOG (A K\ORFHWHV DQG +r VDOYXV QRW KDYH PRGLILHG IHHGLQJ EHKDYLRUV RI WKHLU RZQ" 0RUHRYHU WKH VHOHFWLYH IHHGLQJ EHKDYLRU LV QRW XQLTXH WR (AB PHODQRWLV DQG 5 VXPLFKUDVWL 0HDGRZ YROHV 0LFURWXV SHQQV\OYDQLFXV UHPRYH WKH SKHQROLF ULFK EDUN RI 1RUZD\ SLQH 3LQXV UHVLQRVD DQG :KLWH VSUXFH 3LFHD JODXFD VDSOLQJV EHIRUH HDWLQJ WKH SKHQROLFSRRU FDPELXP 5R\ DQG -0 %HUJHURQ XQSXE GDWDf 6HFRQG UHG WUHH PLFH 3KHQDFRP\V ORQJLFDXGQV SHHO RII DQG GLVFDUG WKH H[WHUQDOO\ ORFDWHG UHVLQ GXFWV RI 'RXJODV ILU 3VHXGRWVXJD WD[LIROLD QHHGOHV EHIRUH HDWLQJ WKHP %HQVRQ DQG %RUHOO +DPLOWRQ f 7KLUG FKLVHOWRRWKHG NDQJDURR UDWV 'LSRGRP\V PLFURSV VKDYH RII WKH K\SHUVDOLQH SHULSKHUDO WLVVXH RI WKH OHDYHV RI VDOWEXVK $WULSOH[ FRQIHUWLIROLD VR WKDW WKH LQWHUQDO WLVVXH FDQ EH LQJHVWHG .HQDJ\ f $QG IRXUWK ZKHQ JUDVVKRSSHU PLFH 2Q\FKRP\V WRUULGXV DUH RIIHUHG KRXVH FULFNHWV $FKHWD GRPHVWLFXV ZKRVH KHDGV KDYH EHHQ WUHDWHG ZLWK D KLJKO\ ELWWHU VROXWLRQ TXLQLQHK\GURFKORULGH DQG HWKDQROf WKH\ ILUVW FKHZ RII DQG GLVFDUG WKH KHDGV EHIRUH HDWLQJ WKH ERGLHV &\U f )XUWKHU VWXG\ RI LQVHFWLYRU\ DQG IRRG KDQGOLQJ

PAGE 109

EHKDYLRU LQ WKH IRXU VSHFLHV RI PLFH DVVRFLDWHG ZLWK WKH PRQDUFK FRORQLHV ZLOO EH YDOXDEOH LQ H[SODLQLQJ WKH GLIIHUHQW ZD\V WKH\ IHG XSRQ PRQDUFKV ,Q DGGLWLRQ WR &*DYRLGDQFH WKUHH RWKHU UHDVRQV PD\ H[SODLQ ZK\ PLFH IHG VHOHFWLYHO\ RQ PRQDUFKV 7KH\ PD\ KDYH EHHQ DYRLGLQJ 3$V ZKLFK DOVR DSSHDU WR EH FRQFHQWUDWHG LQ WKH FXWLFOH 0 6WHOOMHV XQSXEOGDWDf $OWKRXJK WKLV UHPDLQV D GLVWLQFW SRVVLELOLW\ UHVXOWV SUHVHQWHG LQ WKH QH[W &KDSWHU GHPRQVWUDWH FOHDUO\ WKDW ERWK 3 PHODQRWLV DQG VXPLFKUDVWL IHHG VHOHFWLYHO\ RQ PRQDUFKV WKDW ODFN 3$V 6HFRQG EHFDXVH VPDOO PDPPDOV Jf DSSHDU LQFDSDEOH RI GLJHVWLQJ FXWLFOH WR DQ\ VLJQLILFDQW H[WHQW *ULIILWK DQG *DWHV 0 $OOHQ XQSXEO GDWD EXW VHH -HXQLDX[ f WKH\ FRXOG LQFUHDVH WKHLU QHW HQHUJ\ \LHOG RI IRUDJLQJ E\ DYRLGLQJ FXWLFOH DQG VHYHUDO IRUDJLQJ PRGHOV SUHGLFW DQLPDOV VKRXOG DYRLG LQGLJHVWLEOH IRRGV HJ .UHEV %HORYVN\ f &XWLFOH UHSUHVHQWV b RI DQ RYHUZLQWHULQJ PRQDUFKnV DEGRPLQDO ELRPDVV DQG b RI LWV WKRUDFLF ELRPDVV 7DEOH f 7KLUG PLFH PD\ KDYH IRXQG WKH FXWLFOH WRR WRXJK DQG OHDWKHU\ +RZHYHU WKH UHVXOWV IURP ([SHULPHQW IDLO WR VXSSRUW WKHVH ODWWHU WZR H[SODQDWLRQV 3HURP\VFXV PHODQRWLV DWH PRUH PDOH DEGRPHQV DQG IHG VHOHFWLYHO\ RQ D JUHDWHU SHUFHQWDJH RI IHPDOH RQHV 6HH &KDSWHU IRU D PRUH GHWDLOHG H[DPLQDWLRQ RI ZK\ WKHVH PLFH HDW PRQDUFKV VHOHFWLYHO\ 7KHVH UHVXOWV DUH LQ FRQWUDVW WR %URZHU HW DO f ZKR UDUHO\ REVHUYHG VHOHFWLYH IHHGLQJ E\ (A PHODQRWLV HUURQHRXVO\ UHIHUUHG WR DV (A PDQLFXODWXV LQ WKHLU SDSHU %URZHU HW DO f 7KLV GLVFUHSDQF\ LV H[SORUHG LQ &KDSWHU

PAGE 110

$EVHQFH RI &RQGLWLRQHG )HHGLQJ $YHUVLRQV WR 0RQDUFK &*V 3HURP\VFXV PHODQRWLV VXPLFKUDVWL DQG eBr K\ORFHWHV DOO FRQVXPHG H[WUHPHO\ KLJK QLJKWO\ GRVDJHV RI PRQDUFK &*V LQ ([SHULPHQWV DQG PHDQV DQG PJNJ PRXVH UHVSHFWLYHO\f ZLWKRXW GHYHORSLQJ DQ\ DSSDUHQW VLJQV RI &* WR[LFLW\ HJ DWD[LD WUHPEOLQJ UROOLQJ FRQYXOVLRQV K\SHUH[FLWDELOLW\ 0DUW\ f 7KH /'J IRU DQ RUDO GRVDJH RI GLJLWR[LQ LQ ODERUDWRU\ PLFH 0XV GRPHVWLFXV U UDWV 5DWWXV QRUYHJLFXV DQG FDWV )HOLV GRPHVWLFD LV DQG UHVSHFWLYHO\ %DUQHV DQG (OWKHULQJWRQ f 7KH PLFH LQ WKLV VWXG\ PD\ KDYH EHHQ DEOH WR WROHUDWH VXFK KLJK GRVDJHV EHFDXVH WKH\ DEVRUEHG RQO\ D VPDOO SHUFHQWDJH RI WKH PRQDUFK &*V WKH\ LQJHVWHG &DUGLDF JO\FRVLGH DEVRUSWLRQ PD\ EH OLPLWHG IRU RWKHU UHDVRQV )LUVW WKH JDVWURLQWHVWLQDO WUDFWV RI WZR VSHFLHV RI PLFH FORVHO\ UHODWHG WR WKH VSHFLHV VWXGLHG KHUHLQ DUH KLJKO\ LPSHUPHDEOH WR PRQDUFK &*V 3HURP\VFXV PDQLFXODWXV DQG 0XV GRPHVWLFXV 0DUW\ f 6HFRQG RYHUZLQWHULQJ PRQDUFKV LQ 0H[LFR KDYH D KLJK SURSRUWLRQ RI KLJK SRODULW\ &*V 6HLEHU HW DO f ZKLFK DUH DEVRUEHG E\ 0A GRPHVWLFXV OHVV UHDGLO\ WKDQ DUH ORZ SRODULW\ RQHV /DXWHUEDFK f 7KLUG VRPH RI WKH LQJHVWHG PRQDUFK &*V PD\ KDYH UHPDLQHG ERXQG WR WKH FXWLFOH GXULQJ GLJHVWLRQ DQG ZHUH WKXV XQDYDLODEOH IRU DEVRUSWLRQ VHH &OHPHQW %URZHU HW DO f )XUWKHU VWXG\ RI WKH WR[LFRORJLFDO VHQVLWLYLW\ RI WKHVH WKUHH VSHFLHV WR &*V DV ZHOO DV WKH ELRDYDLODELOLW\ RI &*V ERXQG LQ FXWLFOH ZLOO KHOS GHWHUPLQH ZKHWKHU HLWKHU VSHFLHV FRXOG GHYHORS &* WR[LFLW\ DQG KHQFH FRQGLWLRQHG IHHGLQJ DYHUVLRQV IROORZLQJ SURORQJHG FRQVXPSWLRQ RI PRQDUFKV

PAGE 111

(IIHFWV RI /RQJ7HUP &RQVXPSWLRQ RI 0RQDUFKV $OO VSHFLHV RI PLFH H[FHSW IRU .r QX VDOYXV DWH QHDUO\ FRQVWDQW DPRXQWV RI PRQDUFK WLVVXH RYHU WKH VL[ QLJKWV :KDW GLIIHUHG DPRQJ WKHVH VSHFLHV ZDV WKH VWDQGDUGL]HG DPRXQW RI WLVVXH HDWHQ SHU QLJKW 3HURP\VFXV K\ORFHWHV DWH FRQVLVWHQWO\ OHVV WKDQ GLG PHODQRWLV DQG 5r VXPLFKUDVWL 7KHVH UHVXOWV VXJJHVW WKDW WKUHHIRXUWKV RI WKH -/ K\ORFHWHV ORVW ZHLJKW EHFDXVH WKH\ DWH LQVXIILFLHQW DPRXQWV RI IRRG +RZHYHU LW VHHPV SDUDGR[LFDO WKDW WKUHHILIWKV RI WKH VXPLFKUDVWL DOVR ORVW ZHLJKW FRQVLGHULQJ WKH\ DWH VLPLODU VWDQGDUGL]HG DPRXQWV RI WLVVXH DV GLG WKH (M PHODQRWLV 2QH SRVVLEOH H[SODQDWLRQ LV WKDW EHFDXVH 5 VXPLFKUDVWL LV KLJKO\ DYHUVH WR WKH WDVWH RI &*V &KDSWHU f WKH PRQDUFK GLHW PD\ KDYH FDXVHG VWUHVV VHQVX 6HO\H f D FRQGLWLRQ NQRZQ WR LQFUHDVH KHDUW UDWH EORRG SUHVVXUH R[\JHQ FRQVXPSWLRQ 3RUJHV f DQG SRVVLEO\ VHDUFK DFWLYLW\ DOO RI WKHVH IDFWRUV HOHYDWH HQHUJHWLF GHPDQGV 6WUHVV DOVR OHDGV WR ZHLJKW ORVV LQ D YDULHW\ RI PDPPDOV DSSDUHQWO\ RZLQJ WR UHGXFHG IRRG LQWDNH DQGRU D GHFUHDVH LQ GLJHVWLYH HIILFLHQF\ .ODVLQJ f 8QGHU VXFK VWUHVV 5 VXPLFKUDVWL ZRXOG KDYH ORVW ZHLJKW HYHQ ZLWK WKH VDPH PDVVVSHFLILF HQHUJ\ LQWDNH DV PHODQRWLV

PAGE 112

&+$37(5 &2168037,21 $1' $92,'$1&( 2) 021$5&+6 7+( ,1)/8(1&( 2) 021$5&+ $%81'$1&( $1' &$5',$& */<&26,'( &21&(175$7,21 ,Q WKH SUHYLRXV &KDSWHU IRXQG WKDW ZKHQ RIIHUHG 0H[LFDQ RYHUZLQWHULQJ PRQDUFKV KHQFHIRUWK 0H[LFDQ PRQDUFKVf SHU QLJKW ZLWKRXW DQ DOWHUQDWH IRRG LQGLYLGXDO eB} PHODQRWLV DWH RQ DYHUDJH PRQDUFKV IHG SUHIHUHQWLDOO\ RQ ORZ &* PRQDUFKV LH WKH\ DWH PRUH PDOHV WKDQ IHPDOHVf DQG IUHTXHQWO\ DWH DEGRPHQV VHOHFWLYHO\ LH UHMHFWHG WKH &*ODGHQ DEGRPLQDO FXWLFOH DQG DWH WKH ORZ&* LQWHUQDO WLVVXHVf +RZHYHU %URZHU HW DO f UHSRUWHG WKDW ZKHQ RIIHUHG 0H[LFDQ PRQDUFKV RU PRQDUFKV ZLWK YDU\LQJ FRQFHQWUDWLRQV DQG NLQGV RI &*V LQGLYLGXDO PHODQRWLV DWH RQ DYHUDJH PRQDUFKV GLG QRW IHHG SUHIHUHQWLDOO\ RQ ORZ &* PRQDUFKV DQG UDUHO\ IHG VHOHFWLYHO\ RQ DEGRPHQV 7R UHVROYH WKHVH GLVFUHSDQW ILQGLQJV H[DPLQHG KRZ PRQDUFK FRQVXPSWLRQ SDWWHUQV E\ PHODQRWLV A K\ORFHWHV DQG VXPLFKUDVWL DUH LQIOXHQFHG E\ f YDU\LQJ WKH QXPEHU RI PRQDUFKV RIIHUHG GXULQJ RQH QLJKW DQG f YDU\LQJ &* FRQFHQWUDWLRQ DQG SRODULW\ ZKHQ PRQDUFKV DUH VXSHUDEXQGDQW 0HWKRGV ([SHULPHQW (IIHFW RI 0RQDUFK $EXQGDQFH RQ )RUDJLQJ 3DWWHUQV RI 0LFH )HHGLQJ WULDOV ZHUH FRQGXFWHG IURP WKURXJK )HEUXDU\ RQ WKH VORSH RI WKH $UUR\R /D 3ODQFKD RI WKH 6LHUUD &KLQFXD PRXQWDLQ

PAGE 113

PDVVLI LQ QRUWKHDVWHUQ 0LFKRDF£Q 0H[LFR DW DSSUR[LPDWHO\ r1 ODWLWXGH 7KLV VLWH LV RQH RI WKH SULQFLSDO RYHUZLQWHULQJ DUHDV RI WKH HDVWHUQ SRSXODWLRQ RI WKH PRQDUFK EXWWHUIO\ &DOYHUW DQG %URZHU f $W DQ DOWLWXGH RI DSSUR[LPDWHO\ P PLFH ZHUH WUDSSHG LQ 6KHUPDQ OLYH WUDSV [ [ FPf DW OHDVW P IURP WKH PRQDUFK FRORQ\ WR UHGXFH WKH SRVVLELOLW\ WKDW WKH\ KDG SULRU H[SHULHQFH HDWLQJ PRQDUFKV %HFDXVH RI D KLJK YDULDQFH LQ IRRG FRQVXPSWLRQ LQ UHSURGXFWLYH PLFH SDUWLFXODUO\ IHPDOHV 6DGOLHU HW DO 6WHEELQV 0LOODU f RQO\ UHSURGXFWLYHO\ LQDFWLYH DGXOWV QRQODFWDWLQJ RU QRQSUHJQDQW IHPDOHV DQG QRQVFURWDO PDOHVf ZHUH XVHG LQ WKH H[SHULPHQWV $ WRWDO RI WR PLFH RI HDFK VSHFLHV ZDV KRXVHG DQG WHVWHG LQGLYLGXDOO\ LQ ZLUH PHVK FDJHV DERXW FP KLJK [ FP LQ GLDPHWHUf VHW RQ D WDUSDXOLQFRYHUHG WDEOH LQ WKH VKDGH 'DFURQ EDWWLQJ ZDV DGGHG IRU QHVWLQJ PDWHULDO DQG SLHFHV RI FDUGERDUG ZHUH SODFHG EHWZHHQ WKH FDJHV WR LVRODWH WKH PLFH YLVXDOO\ )RU WKUHH FRQVHFXWLYH QLJKWV SULRU WR WKH IHHGLQJ H[SHULPHQWV HDFK PRXVH ZDV PDLQWDLQHG RQ 3XULQD ODERUDWRU\ FKRZ QR KHQFHIRUWK PRXVH FKRZf DQG ZDWHU $OO RI WKHP IHG DQG GUDQN UHJXODUO\ DQG HLWKHU PDLQWDLQHG RU JDLQHG ZHLJKW 0RQDUFKV ZHUH FROOHFWHG ZLWK EXWWHUIO\ QHWV IURP DFFHVVLEOH FOXVWHUV RQ R\DPHO ILU EUDQFKHV $ELHV UHOLJLRVD +%.f ZLWKLQ WKH EXWWHUIO\ FRORQ\ DQG ZHUH RIIHUHG LQ WKHLU QDWXUDO VH[UDWLRV WR PLFH GLG QRW GLVWLQJXLVK PRQDUFK PRUWDOLW\ ZLWK UHVSHFW WR PRQDUFK VH[ (DFK H[SHULPHQWDO UXQ ODVWHG IRU QLJKWV 'XULQJ WKH ILUVW QLJKW DW K OLYH PRQDUFKV DQG PRXVH FKRZ DG OLELWXP ZHUH SODFHG LQ HDFK PRXVHnV FDJH $OO PLFH FRQVXPHG QHDUO\ DOO RI WKHVH PRQDUFKV QRQVHOHFWLYHO\ 7KLV SUHH[SRVXUH SHULRG LQVXUHG WKDW DOO PLFH KDG

PAGE 114

VRPH SULRU H[SRVXUH WR PRQDUFKV $W K WKH IROORZLQJ PRUQLQJ DOO UHPDLQLQJ PRQDUFK PDWHULDO ZDV UHPRYHG 7KHQ DW K RQ WKH VDPH GD\ D VSHFLILHG QXPEHU VHH EHORZf RI PRQDUFKV DQG IUHVK PRXVH FKRZ DG OLELWXP ZHUH JLYHQ WR HDFK PRXVH $W K WKH IROORZLQJ PRUQLQJ HDFK PRXVH ZDV UHPRYHG IURP LWV FDJH WR WDOO\ SDWWHUQV RI IHHGLQJ GDPDJH WR GHDG PRQDUFKV 7KH VDPH PRXVH ZDV QHYHU WHVWHG PRUH WKDQ RQFH LQ WKLV DQG WKH QH[W H[SHULPHQW $V UHSRUWHG LQ &KDSWHU DOO WKUHH VSHFLHV RI PLFH IHHG SULPDULO\ RQ PRQDUFK DEGRPLQDO PDWHULDO WKRUDFHV DQG ZLQJV DUH UDUHO\ HDWHQ LQFOXGHG RQO\ WKH WZR PRVW FRPPRQO\ H[KLELWHG SDWWHUQV RI DEGRPLQDO FRQVXPSWLRQ f QRQVHOHFWLYH IHHGLQJ LH D PRXVH HDWV YLUWXDOO\ DOO RI WKH FXWLFOH DQG FRQWHQWVf DQG f VHOHFWLYH IHHGLQJ LH D PRXVH HDWV YLUWXDOO\ DOO RI WKH FRQWHQWV DQG GLVFDUGV WKH FXWLFOHf $OWKRXJK PLFH VDPSOHG LH DWH bf PRQDUFK DEGRPHQV DQG FRQVXPHG WKRUDFLF PDWHULDO WR YDU\LQJ GHJUHHV LQ b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

PAGE 115

SHU QLJKW DQG H[SHULPHQWDOO\ YDULHG &* FRQFHQWUDWLRQ $OO VSHFLHV ZHUH FRPSDUHG LQ WHUPV RI KRZ &* FRQFHQWUDWLRQ LQIOXHQFHG f WKH QXPEHU RI PRQDUFK DEGRPHQV HDWHQ DQG f VHOHFWLYH IHHGLQJ RQ PRQDUFK DEGRPHQV ,Q 0DUFK FROOHFWHG PHODQRWLV / A K\ORFHWHV DQG VXPLFKUDVWL ILYH RI HDFK VH[ IRU HDFK VSHFLHVf LQ WKH VDPH DUHD GHVFULEHG DERYH DQG WUDQVSRUWHG WKHP WR WKH 8QLYHUVLW\ RI )ORULGD LQ *DLQHVYLOOH )ORULGD 7KH\ ZHUH KRXVHG LQ SODVWLF FDJHV [ [ FPf DQG PDLQWDLQHG RQ PRXVH FKRZ DQG ZDWHU DG OLELWXP LQ D WHPSHUDWXUHKXPLGLW\ FRQWUROOHG URRP RQ D K OLJKW K GDUN F\FOH XVHG )@B RIIVSULQJ RI WKHVH PLFH PHODQRWLV VXPLFKUDVWL K\ORFHWHVf IRU WHVWLQJ ZKHQ WKH\ ZHUH WR GD\V ROG $V D UHVXOW DOO PLFH KDG VLPLODU IHHGLQJ KLVWRULHV 7KUHH GLIIHUHQW FDWHJRULHV RI PRQDUFKV ZHUH XVHG LQ WKH IHHGLQJ H[SHULPHQWV RQHV FROOHFWHG DW WKH 0H[LFDQ RYHUZLQWHULQJ VLWHV RQHV UHDUHG RQ $VFOHSLDV V\ULDFD / KHQFHIRUWK $A Lr PRQDUFKVf DQG RQHV UHDUHG RQ $r FXUDVVDYLFD / KHQFHIRUWK $A 6/r PRQDUFKVf 7KH 0H[LFDQ PRQDUFKV ZHUH FROOHFWHG DV LQ ([SHULPHQW DQG WKHQ VHDOHG LQVLGH WZR SODVWLF EDJV WR SUHYHQW GHVLFFDWLRQ WUDQVSRUWHG RQ LFH WR WKH 8QLYHUVLW\ RI )ORULGD DQG IUR]HQ 7KH $A A PRQDUFKV ZHUH FROOHFWHG LQ WKH ZLOG DV IRXUWK DQG ILIWK LQVWDU ODUYDH LQ ZHVWHUQ 0DVVDFKXVHWWV +DPSVKLUH DQG )UDQNOLQ &RXQWLHVf LQ $XJXVW DQG 6HSWHPEHU DQG IHG $ V\ULDFD OHDYHV IURP WKH VDPH ILHOGV DQG VKLSSHG DV FKU\VDOLGV WR WKH 8QLYHUVLW\ RI )ORULGD ZKHUH WKH\ KDWFKHG 7KH $r PRQDUFKV ZHUH WKH ILUVW JHQHUDWLRQ RI DGXOWV FDSWXUHG QHDUE\ WKH 8QLYHUVLW\ RI )ORULGD GXULQJ 6HSWHPEHU DQG 7KH\ ZHUH UDLVHG IURP HJJ WR PDWXULW\ RQ $A FXUDVVDYLFD OHDYHV IURP SRWWHG SODQWV

PAGE 116

RULJLQDWLQJ IURP VRXWKHUQ )ORULGD 'DGH &RXQW\f 7KH $ $ DQG $ / PRQDUFKV ZHUH IUR]HQ LQVLGH WZR SODVWLF EDJV K DIWHU HFORVLQ (YHQ WKRXJK WR b RI DOO 0H[LFDQ PRQDUFKV DSSHDU WR KDYH IHG DV ODUYDH RQ $ V\ULDFD SODQWV 6HLEHU HW DO 0DOFROP DQG %URZHU f WKH\ SRVVHVV VXEVWDQWLDOO\ ORZHU &* FRQFHQWUDWLRQV WKDQ GR PRQDUFKV WKDW ZHUH HLWKHU ODERUDWRU\UHDUHG RQ $ V\ULDFD NLOOHG DERXW K DIWHU HFORVLQ RU ZLOGUHDUHG RQ $ V\ULDFD DQG FROOHFWHG LQ 0DVVDFKXVHWWV 7DEOH f 7KLV LV EHFDXVH PRQDUFKV ORVH &*V GXULQJ WKH ORQJ PLJUDWLRQ WR WKH RYHUZLQWHULQJ VLWHV 0DOFROP DQG %URZHU f $V D UHVXOW WKH 0H[LFDQ DQG $ $ PRQDUFKV VKRXOG GLIIHU LQ &* FRQFHQWUDWLRQ EXW QRW SRODULW\ 7DEOH f ,Q FRQWUDVW ODERUDWRU\ UHDUHG $ $ PRQDUFKV QRW RQO\ KDYH D KLJKHU &* FRQFHQWUDWLRQ WKDQ ERWK $ 6M DQG 0H[LFDQ PRQDUFKV EXW DOVR GLIIHUHQW NLQGV RI &*V DV LQGLFDWHG E\ SRODULW\ GLIIHUHQFHV 7DEOH 6HLEHU HW DO f (YHQ WKRXJK WKH &*V LQ $ $ PRQDUFKV DOVR KDYH D JUHDWHU HPHWLF SRWHQF\ LQ EOXH MD\V 5RHVNH HW DO 7DEOH 0DOFROP DQG %URZHU 7DEOH f WKHVH ILQGLQJV GR QRW VHHP UHOHYDQW KHUH EHFDXVH PLFH GR QRW YRPLW /DQGDXHU HW DO f (DFK H[SHULPHQWDO UXQ ODVWHG IRU WZR QLJKWV ,PPHGLDWHO\ EHIRUH D WULDO D PRXVH ZDV WUDQVIHUUHG WR D FOHDQ FDJH $W K HLWKHU 0H[LFDQ PRQDUFKV $ $ PRQDUFKV RU $ $ DQG $ $ ZHUH SODFHG LQ WKH FDJH $OO PLFH DW OHDVW SDUWLDOO\ FRQVXPHG DOO RI WKHVH PRQDUFKV 0RXVH FKRZ DQG ZDWHU ZHUH DYDLODEOH DG OLELWXP $W K WKH IROORZLQJ PRUQLQJ DOO UHPDLQLQJ PRQDUFK PDWHULDO DQG PRXVH FKRZ ZDV UHPRYHG 7KHQ DW K RQ WKH VDPH GD\ HLWKHU 0H[LFDQ PRQDUFKV $ $ PRQDUFKV RU D FRPELQDWLRQ RI $ $ DQG $ $ PRQDUFKV ZHUH SODFHG LQ WKH PRXVHnV FDJH 7KH WKLUG WUHDWPHQW LQFOXGHG

PAGE 117

7DEOH 0HDQ &* FRQFHQWUDWLRQV _IJ &* J SRZGHUHG EXWWHUIO\f DQG HVWLPDWHV RI UHODWLYH SRODULW\ 5G YDOXHVf RI PRQDUFK EXWWHUIOLHV SUHVXPHG RU NQRZQ WR KDYH IHG RQ HLWKHU $r V\ULDFD RU $AB FXUDVVDYLFD 5G YDOXHV DUH GHULYHG IURP WKLQOD\HU FKURPDWRJUDSK\ DQG DUH VWDQGDUGL]HG WR GLJLWR[LQ KLJKHU YDOXHV LQGLFDWH ORZHU SRODULW\ 7KH 5G YDOXH IRU WKH GRPLQDQW &* LQ $A V\ULDFDUHDUHG PRQDUFKV DVSHFLRVLGH 6HLEHU HW DO 0DOFROP HW DO f DQG WKH UDQJH RI 5G YDOXHV IRU WKH FRPPRQ &*V LQ $A FXUDVVDYLFDUHDUHG PRQDUFKV XVFKDULGLQ FDORWURSLQ DQG FDODFWLQ 5RHVNH HW DO f DUH SUHVHQWHG %HFDXVH b RI 0H[LFDQ RYHUZLQWHULQJ PRQDUFKV IHG DV ODUYDH RQ $r V\ULDFD SODQWV 6HLEHU HW DO 0DOFROP DQG %URZHU f DVVXPH WKDW DVSHFLRVLGH LV WKH GRPLQDQW &* LQ 0H[LFDQ PRQDUFKV 0RQDUFK 6DPSOH 0HDQ &* FRQFHQWUDWLRQ 5G YDOXH RI GRPLQDQW &* Vf 2YHUZLQWHULQJ 6LHUUD &KLQFXD 0H[LFR D E :LOGUHDUHG LQ 0$ RQ $A V\ULDFD E E /DERUDWRU\UHDUHG LQ 0$ RQ / V\ULDFD F R WU /DERUDWRU\UHDUHG LQ IO RQ $r FXUDVVDYLFD G OH D 0DOFROP DQG %URZHU E 0DOFROP HW DO F %URZHU DQG *OD]LHU XQSXE GDWD G *UDQG PHDQ RI UHSRUWHG PHDQ FRQFHQWUDWLRQV f Q 5RHVNH HW DO f PDOHV %URZHU HW DO f Q IHPDOHV %URZHU HW DO DQG f Q 6 0DOFROP XQSXEO GDWD H 5RHVNH HW DO )LJXUH f

PAGE 118

,OO ERWK $A 6O DQG $r ILM PRQDUFKV EHFDXVH RI D OLPLWHG QXPEHU RI DYDLODEOH $r PRQDUFKV 0LFH DOZD\V UHFHLYHG WKH VDPH W\SH RI PRQDUFKV GXULQJ ERWK QLJKWV $W K WKH IROORZLQJ PRUQLQJ WKH PRXVH ZDV UHPRYHG IURP LWV FDJH WR WDOO\ SDWWHUQV RI IHHGLQJ GDPDJH $V LQ ([SHULPHQW RQO\ DEGRPLQDO FRQVXPSWLRQ SDWWHUQV ZHUH UHFRUGHG 7KH VDPH QXPEHU RI PLFH RI HDFK VSHFLHV UHFHLYHG WKH 0H[LFDQ PRQDUFKV Q f WKH $r PRQDUFKV Q f DQG WKH $A DQG $A PRQDUFKV Q f 7R GLVWLQJXLVK WKH UHPDLQV RI $ /r DQG $A PRQDUFKV DIWHU PLFH KDG GLVPHPEHUHG WKHP SODFHG D PP GLDPHWHU UHG RU JROG VSRW RI QDLO SROLVK RQ HDFK EXWWHUIO\nV DEGRPHQ WKRUD[ DQG ZLQJV 7KH DVVLJQPHQW RI JROG RU UHG IRU $A RU $AB / PRQDUFKV ZDV GHWHUPLQHG UDQGRPO\ IRU HDFK WULDO 7KH QDLO SROLVK ZDV DSSOLHG K EHIRUH HDFK WULDO DOVR PDUNHG WKH $r DQG 0H[LFDQ PRQDUFKV LQ D VLPLODU PDQQHU H[FHSW WKDW RQO\ RQH FRORU ZDV XVHG WKLV ZDV GHWHUPLQHG UDQGRPO\ IRU HDFK PRXVH 6WDWLVWLFDO $QDO\VLV $OO VWDWLVWLFDO WHVWV DQG WUDQVIRUPDWLRQV IROORZHG =DU f ,Q ([SHULPHQW WKH RQHZD\ $129$ DQG 6FKHII )WHVW ZHUH XVHG WR FRPSDUH f WKH QXPEHU RI DEGRPHQV HDWHQ VHOHFWLYHO\ DQG QRQVHOHFWLYHO\f E\ HDFK VSHFLHV ZKHQ PRQDUFK ZHUH RIIHUHG DQG f WKH SHUFHQWDJH RI DEGRPHQV HDWHQ VHOHFWLYHO\ E\ HDFK VSHFLHV DW WKH KLJKHVW GHQVLW\ 7KHVH DQG DOO VXEVHTXHQW GDWD ZHUH WHVWHG IRU QRUPDOLW\ ZLWK WKH .ROPRJRURY6PLUQRY WHVW WKH SHUFHQWDJH GDWD ZHUH DUFVLQH WUDQVIRUPHG SULRU WR WKH QRUPDOLW\ WHVW $ QRQSDUDPHWULF WHVW ZDV XVHG LI WKH GDWD ZHUH QRW QRUPDOO\ GLVWULEXWHG 3 f

PAGE 119

,Q ([SHULPHQW D RQHZD\ $129$ DQG 6FKHII )WHVW ZHUH SHUIRUPHG VHSDUDWHO\ IRU HDFK VSHFLHV WR GHWHUPLQH WKH HIIHFWV RI PRQDUFK W\SH LH 0H[LFDQ $r VBrB RU / L / LW PRQDUFKVf RQ WKH QXPEHU RI DEGRPHQV HDWHQ 7KH .UXVNDO:DOOLV RQHZD\ $129$ DQG 1RQSDUDPHWULF 7XNH\W\SH PXOWLSOH FRPSDULVRQ ZHUH SHUIRUPHG VHSDUDWHO\ IRU HDFK VSHFLHV WR DVFHUWDLQ WKH HIIHFW RI PRQDUFK W\SH RQ WKH SHUFHQWDJH RI DEGRPHQV HDWHQ VHOHFWLYHO\ :LOFR[RQ VLJQHGUDQN WHVWV ZHUH XVHG WR PDNH LQWUDVSHFLILF SDLUHG FRPSDULVRQV RI f WKH QXPEHU RI $ fV YHUVXV $A Qr DEGRPHQV HDWHQ VHOHFWLYHO\ DQG QRQVHOHFWLYHO\f IRU DOO VSHFLHV DQG f WKH SHUFHQWDJH RI $AB YHUVXV $} DEGRPHQV HDWHQ VHOHFWLYHO\ IRU PHODQRWLV 5HVXOWV ([SHULPHQW (IIHFW RI 0RQDUFK $EXQGDQFH RQ )RUDJLQJ 3DWWHUQV RI 0LFH 0RQDUFK DEXQGDQFH VWURQJO\ LQIOXHQFHG WKH QXPEHU RI DEGRPHQV HDWHQ E\ HDFK VSHFLHV )LJXUH f :KHQ PRQDUFKV ZHUH RIIHUHG DOO VSHFLHV DWH QHDUO\ DOO RI WKH DEGRPHQV +RZHYHU ZKHQ JUHDWHU QXPEHUV RI PRQDUFKV ZHUH RIIHUHG QLJKWO\ FRQVXPSWLRQ E\ PHODQRWLV DQG VXPLFKUDVWL FRQWLQXHG WR LQFUHDVH ZKHUHDV WKDW RI Br K\ORFHWHV GLG QRW 1LJKWO\ FRQVXPSWLRQ OHYHOOHG RII IRU DOO VSHFLHV ZKHQ RU PRUH PRQDUFKV ZHUH RIIHUHG $Q LQWHUVSHFLILF FRPSDULVRQ RI WKH QXPEHU RI DEGRPHQV HDWHQ ZKHQ PRQDUFKV ZHUH RIIHUHG ZDV KLJKO\ VLJQLILFDQW ) GI 3 f PHODQRWLV DWH VLJQLILFDQWO\ PRUH DEGRPHQV WKDQ GLG eBr K\ORFHWHV 6FKHII )WHVW 3 f 1RQH RI WKH RWKHU SDLUZLVH FRPSDULVRQV ZDV VLJQLILFDQW 3 f 0RQDUFK DEXQGDQFH DOVR LQIOXHQFHG WKH SHUFHQWDJH RI DEGRPHQV HDWHQ VHOHFWLYHO\ E\ PHODQRWLV DQG eAB VXPLFKUDVWL EXW QRW eBr L K\ORFHWHV

PAGE 120

5HODWLRQVKLS RI WKH QXPEHU RI DEGRPHQV HDWHQ PHDQ s6(f WR WKH QXPEHU RI PRQDUFKV PHODQRWLV ,/B VXPLFKUDVWL DQG (AB K\ORFHWHV ‘ 7KH OLQH RI HTXDOLW\ \ [f LV FRPSDULVRQ $OWHUQDWH IRRG PRXVH FKRZf ZDV SUHVHQW G OLELWXP )LJXUH RIIHUHG LQ eA SUHVHQWHG IRU

PAGE 121

0HDQ QR RI DEGRPHQV HDWHQ s6(f 1R RI PRQDUFKV RIIHUHG

PAGE 122

)LJXUH f 7KH ODWWHU VSHFLHV UDUHO\ IHG VHOHFWLYHO\ RQ DQ\ DEGRPHQV 7KH UHODWLRQVKLS RI WKH UDWH RI VHOHFWLYH IHHGLQJ DJDLQVW PRQDUFK DEXQGDQFH ZDV 6VKDSHG IRU ERWK PHODQRWLV DQG 5AB VLLPLFKUDVWL 7KH XSSHU SODWHDX RI WKH FXUYHV IRU WKHVH WZR VSHFLHV GLIIHUHG VOLJKWO\ EXW QRW VLJQLILFDQWO\ $Q LQWHUVSHFLILF FRPSDULVRQ RI WKH SHUFHQWDJH RI DEGRPHQV HDWHQ VHOHFWLYHO\ ZKHQ PRQDUFKV ZHUH RIIHUHG ZDV KLJKO\ VLJQLILFDQW ) GI 3 f PHODQRWLV DQG VXPLFKUDVWL DWH VLJQLILFDQWO\ KLJKHU SHUFHQWDJHV RI DEGRPHQV VHOHFWLYHO\ WKDQ K\ORFHWHV 6FKHII )WHVW 3 f 7KH SHUFHQWDJH RI DEGRPHQV HDWHQ E\ (BA PHODQRWLV DQG %A VXPLFKUDVWL GLG QRW GLIIHU VLJQLILFDQWO\ 3 f )[SHULPHQW (IIHFW RI &* &RQFHQWUDWLRQ RQ )RUDJLQJ 3DWWHUQV RI 0LFH 7KH WRWDO QXPEHU RI DEGRPHQV HDWHQ E\ PHODQRWLV ZDV QRW LQIOXHQFHG E\ PRQDUFK W\SH 7DEOH f ,Q FRQWUDVW ,/} VXPLFKUDVWL DWH VLJQLILFDQWO\ PRUH DEGRPHQV IURP 0H[LFDQ PRQDUFKV WKDQ WKH\ GLG IURP $r A DQG WKH FRPELQDWLRQ $r DQG $AB O PRQDUFKV 3HURP\VFXV K\ORFHWHV DWH VLJQLILFDQWO\ PRUH DEGRPHQV IURP 0H[LFDQ DQG $A 6A PRQDUFKV WKDQ WKH\ GLG IURP WKH FRPELQDWLRQ RI $A Lr DQG $A PRQDUFKV ,QWHUVSHFLILF FRPSDULVRQV VKRZ WKDW (BA PHODQRWLV DWH PRUH DEGRPHQV IURP DOO WUHDWPHQWV WKDQ GLG WKH RWKHU VSHFLHV 7KHVH UHVXOWV VXJJHVW WKDW K\ORFHWHV DQG 5r VXPLFKUDVWL ZHUH GHWHUUHG E\ WKH DEGRPHQV ZLWK KLJKHU OHYHOV RI &*V ZKLOH PHODQRWLV ZDV QRW 0RQDUFK W\SH VLJQLILFDQWO\ LQIOXHQFHG WKH SHUFHQWDJH RI DEGRPHQV HDWHQ VHOHFWLYHO\ E\ ERWK eBA PHODQRWLV DQG -6/r VXPLFKUDVWL EXW LQ RSSRVLWH GLUHFWLRQV 7DEOH f 7KH IRUPHU VSHFLHV DWH D VLJQLILFDQWO\ KLJKHU SHUFHQWDJH RI WKH KLJK &* DEGRPHQV LH $r 6Br

PAGE 123

)LJXUH 5HODWLRQVKLS RI WKH SHUFHQWDJH RI DEGRPHQV HDWHQ VHOHFWLYHO\ LH FXWLFOH ZDV UHMHFWHG DQG WKH ORZ &* LQWHUQDO FRQWHQWV HDWHQ PHDQ I6(f WR WKH PRQDUFKV RIIHUHG LQ eBr PHODQRWLV VXPLFKUDVWL DQG K\ORFHWHV ‘ $OWHUQDWH ZDV SUHVHQW G OLELWXP WKH &*ODGHQ QXPEHU RI IRRG PRXVH FKRZf

PAGE 124

0HDQ b RI DEGRPHQV HDWHQ VHOHFWLYHO\ s6(f 1XPEHU RI PRQDUFKV RIIHUHG

PAGE 125

7DEOH 0HDQ QXPEHU RI PRQDUFK DEGRPHQV HDWHQ ERWK VHOHFWLYHO\ DQG QRQVHOHFWLYHO\ s6(f GXULQJ QLJKW E\ HDFK VSHFLHV ,QGLYLGXDO PLFH ZHUH RIIHUHG HLWKHU 0H[LFDQ PRQDUFKV $ V PRQDUFKV RU D FRPELQDWLRQ RI $r DQG $A PRQDUFKV 1R DOWHUQDWH IRRG ZDV SUHVHQW 7KH VDPH QXPEHU RI PLFH RI HDFK VSHFLHV UHFHLYHG WKH 0H[LFDQ PRQDUFKV Q f WKH $A PRQDUFKV Q f DQG WKH $} IX DQG $r c/r PRQDUFKV Q f ,QWUDVSHFLILF FRPSDULVRQV DUH PDGH ZLWKLQ HDFK FROXPQ ZLWK WKH RQHZD\ $129$ 'LIIHUHQW VXEVFULSWV D Ef LQGLFDWH VLJQLILFDQW GLIIHUHQFHV DPRQJ PHDQ ZLWKLQ HDFK FROXPQ 6FKHII )WHVW 3 f 7\SH RI PRQDUFK (B PHODQRWLV VXPLFKUDVWL (A f§ K\ORFHWHV 0H[LFDQ sD sD s $ ‘ 6 ‘ sOOD sE s $ V $r eB! sD sE s )UDWLR GI 3

PAGE 126

7DEOH 0HDQ SHUFHQWDJH RI HDFK W\SH RI PRQDUFK DEGRPHQ HDWHQ VHOHFWLYHO\ s6(f GXULQJ QLJKW E\ HDFK VSHFLHV ,QGLYLGXDO PLFH UHFHLYHG HLWKHU 0H[LFDQ PRQDUFKV $r PRQDUFKV RU D FRPELQDWLRQ RI $A $r DQG $r PRQDUFKV 1R DOWHUQDWH IRRG ZDV SUHVHQW 7KH VDPH QXPEHU RI PLFH RI HDFK VSHFLHV UHFHLYHG WKH 0H[LFDQ PRQDUFKV Q f WKH $r $r PRQDUFKV Q f DQG WKH $r $ DQG $r &Z PRQDUFKV Q f ,QWUDVSHFLILF FRPSDULVRQV DUH PDGH ZLWKLQ HDFK FROXPQ ZLWK WKH .UXVNDO:DOOLV RQHZD\ $129$ 'LIIHUHQW VXEVFULSWV D Ef ZLWKLQ HDFK FROXPQ LQGLFDWH VLJQLILFDQW GLIIHUHQFHV DPRQJ FRPSDULVRQf PHDQV 3 1RQSDUDPHWULF 7XNH\W\SH PXOWLSOH 7\SH RI PRQDUFK P PHODQRWLV 5 VXPLUKUDVWL 3 D KYORQIDIILV 0DtLVDD sD sE sD $r DA sD sE sD $ V $M R ‘ sE sD RRD +YDOXHr 1R RI JURXSV 3 r FRUUHFWHG IRU WLHV

PAGE 127

$A RQHVf ZKHUHDV WKH ODWWHU VSHFLHV DWH D VLJQLILFDQWO\ ORZHU SHUFHQWDJH RI WKH KLJK &* DEGRPHQV VHOHFWLYHO\ 7KH K\ORFHWHV UDUHO\ IHG VHOHFWLYHO\ RQ HLWKHU W\SH RI DEGRPHQ )RU PRQDUFK WUHDWPHQW LQGLYLGXDO PLFH UHFHLYHG HTXDO QXPEHUV RI / A YHUVXV $A PRQDUFKV $QDO\VLV RI WKH SUHGDWLRQ UHFRUGV IRU WKLV WUHDWPHQW DORQH GHPRQVWUDWHV WKDW DOO WKUHH VSHFLHV IHG SUHIHUHQWLDOO\ RQ / A PRQDUFKV 7DEOH f +RZHYHU GHVSLWH WKLV DSSDUHQW LQWHUVSHFLILF VLPLODULW\ LW LV LPSRUWDQW WR QRWH WKDW 3 PHODQRWLV DWH VXEVWDQWLDOO\ PRUH RI ERWK WKH $r A DQG WKH $A F DEGRPHQV WKDQ GLG WKH RWKHU VSHFLHV 2I WKH $ VA DQG $A DEGRPHQV HDWHQ PHODQRWLV DWH D VLJQLILFDQWO\ JUHDWHU SHUFHQWDJH RI WKH $r RQHV VHOHFWLYHO\ WKDQ WKH\ GLG RI WKH $r RQHV PHDQ DQG UHVSHFWLYHO\ =YDOXH ZLWK QRUPDO DSSUR[LPDWLRQ Q 3 f 7KXV eBr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eBr K\ORFHWHV GLG QRW LQFUHDVH ZKHUHDV WKDW RI VXPLFKUDVWL DQG (A PHODQRWLV LQFUHDVHG

PAGE 128

7DEOH 0HDQ QXPEHU RI $r DA DQG $X eBr DEGRPHQV HDWHQ ERWK VHOHFWLYHO\ DQG QRQVHOHFWLYHO\ s6(f GXULQJ QLJKW E\ HDFK VSHFLHV (DFK PRXVH UHFHLYHG VLPXOWDQHRXVO\ $ Lr DQG $r er PRQDUFKV 1R DOWHUQDWH IRRG ZDV SUHVHQW ,QWUDVSHFLILF SDLUHG FRPSDULVRQV DUH PDGH ZLWKLQ HDFK FROXPQ ZLWK WKH ZLOFR[RQ VLJQHG UDQN WHVW ZLWK QRUPDO DSSUR[LPDWLRQf 7\SH RI PRQDUFK eBf! PHODQRWLV e/ VXPLFKUDVWL 3 D KYORILVWVV $A s s s $r s s s =YDOXH Q S

PAGE 129

XQWLO LW SODWHDXHG DW DERXW DQG DEGRPHQV UHVSHFWLYHO\ 0RUHRYHU WKH SHUFHQWDJH RI PRQDUFKV HDWHQ VHOHFWLYHO\ LQFUHDVHG GUDPDWLFDOO\ IRU eBr PHODQRWLV DQG 5r VXPLFKUDVWL EXW QRW IRU (A D K\ORFHWHVf 7KH ORZHU LQWDNH RI DEGRPHQV E\ / L K\ORFHWHV FDQQRW EH H[SODLQHG E\ ORZHU HQHUJ\ GHPDQGV $V FRPSDUHG WR WKH RWKHU VSHFLHV LW LV A WLPHV KHDYLHU DQG HDWV RQ DYHUDJH WLPHV PRUH PRXVH FKRZ &KDSWHU f 7KXV (A K\ORFHWHV PXVW KDYH EHHQ PRUH DYHUVH WR WKH 0H[LFDQ PRQDUFKV WKDQ ZHUH WKH RWKHU VSHFLHV FDQ RIIHU SRVVLEOH H[SODQDWLRQV IRU ZK\ WKH VHOHFWLYH IHHGLQJ EHKDYLRU RI (A PHODQRWLV DQG 5r VXPLFKUDVWL ZDV LQIOXHQFHG VR VWURQJO\ E\ WKH QXPEHU RI PRQDUFKV RIIHUHG $V PRQDUFKV DEXQGDQFH LQFUHDVHV WKH PLFH PD\ VLPSO\ EHFRPH OHVV WROHUDQW RI ELWWHUWDVWLQJ ERG\ SDUWV LH FXWLFOHf 6HFRQG WKH PLFH PD\ KDYH WR HDW D ODUJH QXPEHU RI PRQDUFKV HJ f EHIRUH WKH\ OHDUQ KRZ WR IHHG RQ DEGRPHQV LQ WKLV PDQQHU ERWK TXLFNO\ DQG HDVLO\ 7KHUH LV VRPH HPSLULFDO VXSSRUW IRU WKH ODWWHU K\SRWKHVLV :KHQ PDLQWDLQHG RQ D GLHW RI PRQDUFKV SHU QLJKW RYHU QLJKWV WKH SHUFHQWDJH RI DEGRPHQV HDWHQ VHOHFWLYHO\ E\ 3 PHODQRWLV LQFUHDVHG VWHDGLO\ IURP FD b RQ QLJKW WR DERXW b RQ QLJKW &KDSWHU f &XULRXVO\ LQ WKH VDPH H[SHULPHQW WKH SHUFHQWDJH RI DEGRPHQV HDWHQ E\ 5r VXPLFKUDVWL GLG QRW LQFUHDVH RYHU WKH QLJKWV 'HWDLOHG REVHUYDWLRQDO DQDO\VLV RI ERWK VSHFLHV IHHGLQJ RQ YDU\LQJ QXPEHUV RI PRQDUFKV DUH QHHGHG WR DGGUHVV WKHVH K\SRWKHVHV ,QIOXHQFH RI &* &RQFHQWUDWLRQ DQG 3RODULW\ RQ $EGRPLQDO &RQVXPSWLRQ E\ 0LFH 5HLWKURGRQWRP\V VXPLFKUDVWL DQG (A L K\ORFHWHV DWH VOLJKWO\ PRUH 0H[LFDQ WKDQ PRQDUFKV ZKHUHDV (A PHODQRWLV DWH YLUWXDOO\

PAGE 130

LGHQWLFDO QXPEHUV RI ERWK W\SHV RI PRQDUFKV 7KHVH ILQGLQJV VXJJHVW WKDW er VXPLFKUDVWL DQG K\ORFHWHV EXW QRW PHODQRWLV ZHUH GHWHUUHG E\ WKH KLJKHU FRQFHQWUDWLRQV RI WKH &*V LQ / PRQDUFKV $OO VSHFLHV DYRLGHG WKH X PRQDUFKV PRVW VWURQJO\ +RZHYHU EHFDXVH WKH &*V LQ $BL PRQDUFKV DUH OHVV SRODU DQG PRUH FRQFHQWUDWHG WKDQ WKRVH LQ PRQDUFKV LW LV GLIILFXOW WR GHWHUPLQH ZKHWKHU LW ZDV RQH RU ERWK RI WKHVH IDFWRUV WKDW PDGH WKH $A PRQDUFKV VR XQSDODWDEOH %URZHU DQG )LQN f IRXQG WKDW EOXH MD\V &\DQRFLWWD FULVWDWDf ZHUH OHVV GHWHUUHG JXVWDWRULDOO\ E\ ORZ WKDQ KLJK SRODULW\ &*V 0RUHRYHU ZKHQ WDVWHG WKH $A Dr DQG $A PRQDUFKV WKH\ VHHPHG TXDOLWDWLYHO\ EXW QRW TXDQWLWDWLYHO\ VLPLODU WKDW LV ERWK W\SHV RI PRQDUFKV KDG D VLPLODU ELWWHU WDVWH EXW WKH ELWWHUQHVV RI $A &f§ PRQDUFKV ZDV PRUH LQWHQVH 7KHVH WZR ILQGLQJV VXJJHVW WKDW LW ZDV WKH KLJKHU FRQFHQWUDWLRQ RI &*V LQ WKH $r PRQDUFKV WKDW PDGH WKHP OHVV SDODWDEOH WR WKH PLFH 3DWWHUQV RI 6HOHFWLYH )HHGLQJ :KHQ (BA PHODQRWLV ZDV RIIHUHG ERWK $A DQG $r PRQDUFKV LW IHG VHOHFWLYHO\ RQ D VLJQLILFDQWO\ JUHDWHU SHUFHQWDJH RI WKH $A PRQDUFKV 7KLV ILQGLQJ VXSSRUWV WKH K\SRWKHVLV &KDSWHU f WKDW 3 PHODQRWLV VHOHFWLYHO\ DYRLGV WKH FXWLFOH RI DEGRPHQV ZLWK KLJK FRQFHQWUDWLRQV RI &*V WKHUHE\ UHQGHULQJ VXFK DEGRPHQV PRUH SDODWDEOH 7ZR DOWHUQDWH H[SODQDWLRQV IRU WKH VHOHFWLYH IHHGLQJ EHKDYLRU ZKLFK SUHGLFW WKDW WKH VHOHFWLYH IHHGLQJ EHKDYLRU VKRXOG RFFXU LQGHSHQGHQWO\ RI &* FRQFHQWUDWLRQ DQG SDODWDELOLW\ DUH QRW VXSSRUWHG E\ WKLV ILQGLQJ )LUVW EHFDXVH VPDOO PDPPDOV GR QRW DSSHDU WR GLJHVW FXWLFOH WR DQ\ VLJQLILFDQW H[WHQW *ULIILWK DQG *DWHV 0 $OOHQ XQSXEO

PAGE 131

GDWD EXW VHH -HXQLDX[ f WKH\ VKRXOG DYRLG LQJHVWLQJ LW ZKHQ SRVVLEOH DQG WKHUHE\ LQFUHDVH WKHLU QHW HQHUJ\ \LHOG RI IRUDJLQJ VHH .UHEV %HORYVN\ f 6HFRQG PLFH DYRLG FXWLFOH EHFDXVH RI LWV WRXJK DQG OHDWKHU\ TXDOLW\ 8QOLNH =r PHODQRWLV (/ VXPLFKUDVWL IHG VHOHFWLYHO\ RQ D JUHDWHU SHUFHQWDJH RI DEGRPHQV IURP ORZ&* PRQDUFKV LH 0H[LFDQ DQG VAB RQHVf VXJJHVW WKDW WKLV FDQ EH H[SODLQHG E\ WKHLU UHODWLYHO\ KLJK VHQVLWLYLW\ WR WKH WDVWH RI &*V &KDSWHU f 7KH KLJK &* FRQFHQWUDWLRQV LQ WKH $ V DQG $r PRQDUFKV PD\ KDYH EHHQ VR DYHUVLYH WKDW WKH\ GLVUXSWHG ,/ VXPLFKUDVWLV QRUPDO IRUDJLQJ EHKDYLRU WKXV FDXVLQJ LW WR HDW IHZHU DEGRPHQV DQG WR IHHG VHOHFWLYHO\ OHVV RIWHQ ,W LV OLNHO\ WKDW WKH LQWHUQDO WLVVXHV RI PDQ\ RI WKHVH PRQDUFKV KDG LQWROHUDEOH OHYHOV RI &*V 7DEOH DQG )LJXUH f ,Q WKH QH[W &KDSWHU PDLQWDLQ DOO WKUHH VSHFLHV RQ D PRXVH FKRZ GLHW WUHDWHG ZLWK D UHODWLYHO\ KLJK FRQFHQWUDWLRQ _LJ J FKRZf RI D &* GLJLWR[LQf :KHUHDV ERWK (/ PHODQRWLV DQG (A K\ORFHWHV HYHQWXDOO\ KDELWXDWHG WR WKH GLHWnV WDVWH DQG IHG DQG JDLQHG ZHLJKW QRUPDOO\ 5 VXPLFKUDVWL GLG QRW KDELWXDWH DQG HYHQWXDOO\ UHMHFWHG WKH GLHW DOWRJHWKHU ,W LV XQFOHDU ZK\ K\ORFHWHV UDUHO\ IHG VHOHFWLYHO\ RQ DEGRPHQV 7KH\ PD\ SRVVHVV PXFK OHVV IOH[LEOH IRRG KDQGOLQJ FDSDELOLWLHV WKDQ GR WKH RWKHU VSHFLHV ZKLFK OHDUQHG WKH VHOHFWLYH IHHGLQJ EHKDYLRU GXULQJ WKHLU ILUVW QLJKW RI H[SRVXUH WR PRQDUFKV VHH &\U f 3HURP\VFXV K\ORFHWHV GLG QRW DFTXLUH WKLV EHKDYLRU DIWHU IHHGLQJ RQ D GLHW FRQVLVWLQJ VROH\ RI 0H[LFDQ PRQDUFKV IRU FRQVHFXWLYH QLJKWV &KDSWHU f ,I K\ORFHWHV KDG OHDUQHG WR

PAGE 132

UHMHFW WKH &*ODGHQ FXWLFOH LW LV OLNHO\ WKDW LW ZRXOG KDYH LQJHVWHG VXEVWDQWLDOO\ JUHDWHU QXPEHUV RI PRQDUFK DEGRPHQV 3RWHQWLDO ,QIOXHQFH RI 3$V RQ 0RQDUFK &RQVXPSWLRQ 6HYHUDO DXWKRUV KDYH VXJJHVWHG WKDW 3$V PD\ SOD\ DQ LPSRUWDQW GHIHQVLYH UROH IRU PRQDUFKV (GJDU HW DO 5RWKVFKLOG DQG 0DUVK (LVQHU .HOOH\ HW DO EXW VHH %URZHU f +RZHYHU QR VWXG\ WR P\ NQRZOHGJH KDV VKRZQ WKLV WR EH WKH FDVH ,Q IDFW WKH UHVXOWV SUHVHQWHG KHUHLQ DQG LQ &KDSWHU VXJJHVW WKDW 3$V PD\ SOD\ D PLQRU GHIHQVLYH UROH ,Q WKLV FKDSWHU %A VXPLFKUDVWL DQG (A D K\ORFHWHV ZHUH PRUH VWURQJO\ GHWHUUHG E\ WKH 5r A PRQDUFKV ZKLFK ODFNHG 3$V EHFDXVH WKH\ KDG QR DFFHVV WR 3$ VRXUFHV DV ODUYDH DQG DGXOWV WKDQ WKH\ ZHUH E\ WKH ZLOGFDXJKW 0H[LFDQ PRQDUFKV ZKLFK DUH NQRZQ WR SRVVHVV 3$V .HOOH\ HW DO f 1HYHUWKHOHVV LW LV FRQFHLYDEOH WKDW (BA PHODQRWLV DQG VXPLFKUDVWL IHG VHOHFWLYHO\ RQ D VOLJKWO\ EXW QRW VLJQLILFDQWO\ KLJKHU SHUFHQWDJH RI 0H[LFDQ WKDQ PRQDUFKV EHFDXVH WKH DGGLWLRQDO SUHVHQFH RI WKH 3$V PD\ KDYH UHQGHUHG WKH FXWLFOH OHVV SDODWDEOH :K\ GR 5 VXPLFKUDVWL DQG 3 D K\ORFHWHV $YRLG (DWLQJ 2YHUZLQWHULQJ 0RQDUFKV LQ 0V[LUQ" ,Q WKLV FKDSWHU QHLWKHU VXPLFKUDVWL QRU (A A K\ORFHWHV DSSHDUHG WR EH VWURQJO\ DYHUVH WR 0H[LFDQ PRQDUFKV +RZHYHU VXJJHVW WKDW IUHHUDQJLQJ 5r VXPLFKUDVWL DQG eBA K\ORFHWHV DUH PXFK OHVV OLNHO\ WKDQ FDSWLYH RQHV WR HDW 0H[LFDQ PRQDUFKV DQG WKDW IUHHUDQJLQJ LQGLYLGXDOV RI ERWK VSHFLHV DUH VXIILFLHQWO\ GHWHUUHG E\ WKH WDVWH RI WKH &*V WR DYRLG WDNLQJ DGYDQWDJH RI FRORQLHV 7KLV K\SRWKHVLV LV VXSSRUWHG E\ WZR OLQHV RI HYLGHQFH :KHQ 5A VXPLFKUDVWL eBA

PAGE 133

K\ORFHWHV DQG PHODQRWLV ZHUH IHG D GLHW FRQVLVWLQJ VROH\ RI 0H[LFDQ PRQDUFKV IRU QLJKWV PHODQRWLV ZDV WKH RQO\ VSHFLHV DEOH WR PDLQWDLQ ZHLJKW &KDSWHU f 6HFRQG ZKHQ RIIHUHG SDLUHG FKRLFHV EHWZHHQ FRQWURO GLHWV DQG GLHWV WUHDWHG ZLWK D VHULHV RI GLIIHUHQW FRQFHQWUDWLRQV RI GLJLWR[LQ D W\SH RI &*f ERWK VXPLFKUDVWL DQG eBr K\ORFHWHV DYRLGHG WKH GLJLWR[LQ GLHWV DW FRQFHQWUDWLRQV ZHOO EHORZ WKH PHDQ IRU 0H[LFDQ PRQDUFKV &KDSWHU f ,Q FRQWUDVW 5A PHODQRWLV GLG QRW UHMHFW WKH GLJLWR[LQ GLHWV XQWLO D FRQFHQWUDWLRQ DERYH WKH PHDQ IRU RYHUZLQWHULQJ PRQDUFKV

PAGE 134

&+$37(5 &203$5$7,9( 5(63216(6 72 7+( 7$67( $1' 72;,& ())(&76 2) &$5',$& */<&26,'(6 $1' 3<552/,=,',1( $/.$/2,'6 7KH UHVXOWV LQ &KDSWHUV DQG VXJJHVW VWURQJO\ WKDW eB PHODQRILV LV PXFK PRUH WROHUDQW WR WKH WDVWH RI WKH &*V DQGRU 3$V LQ RYHUZLQWHULQJ PRQDUFK EXWWHUIOLHV LQ 0H[LFR ,Q WKLV &KDSWHU IXUWKHU H[DPLQH WKLV K\SRWKHVLV E\ FRPSDULQJ WKH UHVSRQVHV RI 3 PHODQRWLV Dr KYORFHWHV DQG ,/ VXPLFKUDVWL WR UHSUHVHQWDWLYH &*V DQG 3$V LQ DUWLILFLDO GLHWV 7KH H[SHULPHQWV ZHUH GHVLJQHG WR GHWHUPLQH WKH UHODWLYH LPSRUWDQFH RI SRVWLQJHVWLRQDO YHUVXV WDVWH FXHV LQ GHWHUPLQLQJ HDFK VSHFLHVn UHVSHFWLYH UHVSRQVHV WR WKH FRPSRXQGV 0HWKRGV DQG 0DWHULDOV ([SHULPHQW 'HWHUPLQDWLRQ RI 7DVWH 6HQVLWLYLW\ ,Q WKLV H[SHULPHQW LQGLYLGXDO PLFH PHODQRWLV 3 D K\ORFHWHV DQG VXPLFKUDVWLf ZHUH VXEMHFWHG WR D VHULHV RI SDLUHGFKRLFHV EHWZHHQ D FRQWURO DQG D WDVWDQW GLHW WUHDWHG ZLWK VXFFHVVLYHO\ LQFUHDVLQJ RU GHFUHDVLQJ FRQFHQWUDWLRQV RI D UHSUHVHQWDWLYH &* GLJLWR[LQf RU RQH RI WZR UHSUHVHQWDWLYH 3$V IUHHEDVH DQG 1R[LGH PRQRFURWDOLQHf %RWK IRUPV RI PRQRFURWDOLQH ZHUH XVHG EHFDXVH WKH 3$V LQ RYHUZLQWHULQJ PRQDUFKV LQ 0H[LFR DOVR RFFXU LQ WKHVH WZR DOWHUQDWH IRUPV .HOOH\ HW DO f XVHG GLJLWR[LQ DQG PRQRFURWDOLQH DV UHSUHVHQWDWLYH &*V DQG 3$V UHVSHFWLYHO\ EHFDXVH f WKH SDUWLFXODU &*V

PAGE 135

DQG 3$V LQ RYHUZLQWHULQJ PRQDUFKV DUH QRW FRPPHUFLDOO\ DYDLODEOH DQG f RYHUZLQWHULQJ PRQDUFKV KDYH D ELWWHU WDVWH VLPLODU WR GLJLWR[LQ DQG PRQRFURWDOLQH *OHQGLQQLQJ SHUV REVHUYf 7R TXDQWLI\ HDFK VSHFLHVn WDVWH VHQVLWLYLW\ D PRGLILHG IRUP RI WKH VR FDOOHG PHWKRG RI OLPLWV ZDV XVHG WR GHWHUPLQH HDFK VSHFLHVn UHVSHFWLYH DYRLGDQFH WKUHVKROG IRU HDFK FRPSRXQG VHH GLIIHUHQFH WKUHVKROG *HVFKHLGHU f ,Q 0DUFK WUDSSHG (A PHODQRWLV / L K\ORFHWHV DQG % VXPLFKUDVWL ILYH RI HDFK VH[ IRU HDFK VSHFLHVf P IURP WKH RYHUZLQWHULQJ FRORQ\ RI PRQDUFK EXWWHUIOLHV RQ 6LHUUD &KLQFXD LQ 0LFKRDF£Q 0[LFR &DOYHUW DQG %URZHU f DQG WUDQVSRUWHG WKHP WR WKH 8QLYHUVLW\ RI )ORULGD LQ *DLQHVYLOOH )ORULGD 7KH\ ZHUH KRXVHG LQ SODVWLF FDJHV [ [ FPf ZLWK GDFURQ EDWWLQJ DQG PDLQWDLQHG RQ 3XULQD 5RGHQW /DE &KRZ KHQFHIRUWK PRXVH FKRZf DQG ZDWHU DG OLELWXP LQ D WHPSHUDWXUHKXPLGLW\ FRQWUROOHG URRP RQ D K OLJKW K GDUN F\FOH XVHG WKH )AB RIIVSULQJ RI WKHVH PLFH PHODQRWLV D K\ORFHWHV DQG VXPLFKUDVWLf IRU WHVWLQJ LQ *DLQHVYLOOH )ORULGD ZKHQ WKH\ ZHUH WR GD\V ROG DOVR OLYHWUDSSHG QLQH DGXOW A PHODQRWLV NP WR WKH HDVW RI WKH VLWH RI WKH PRQDUFK FRORQ\ RQ DQG -DQXDU\ VHH )LJXUH f 7KLV DUHD ZDV RQ WKH HDVWHUQ VORSH RI 6LHUUD &KLQFXD ZKHUH PRQDUFK FRORQLHV GR QRW IRUP &DOYHUW DQG %URZHU f DQG DVVXPH WKDW SRSXODWLRQV RI PLFH LQ WKLV DUHD KDYH QRW HQFRXQWHUHG RYHUZLQWHULQJ PRQDUFKV 7KHVH PLFH ZHUH KRXVHG DQG PDLQWDLQHG LQ D PDQQHU VLPLODU WR WKDW GHVFULEHG DERYH H[FHSW WKDW WKH\ ZHUH NHSW LQ D WHQW LQ WKH VKDGH QH[W WR RXU FDPSVLWH RQ 6LHUUD &KLQFXD 7KH\ H[SHULHQFHG QDWXUDO GDLO\ IOXFWXDWLRQV LQ WHPSHUDWXUH DERXW WR r

PAGE 136

&f DQG OLJKW DERXW K OLJKW K GDUNf $IWHU WHVWLQJ DOO RI WKHVH PLFH ZHUH UHOHDVHG )RU WHVWLQJ LQ *DLQHVYLOOH UHKRXVHG WKH PLFH LQGLYLGXDOO\ LQ VWDLQOHVV VWHHO ZLUHERWWRPHG FDJHV [ [ FPf ZLWK GDFURQ EDWWLQJ 7KHVH FDJHV KDG WZR JODVV IRRG FXSV FP WDOO FP GLDPHWHU ZLWK D FP GLDPHWHU KROH LQ WKH SODVWLF OLGf VHFXUHG WR WKH IURQW VLGH RI WKH FDJH HDFK HLJKW FP IURP WKH UHVSHFWLYH FRUQHUV 6SLOODJH ZDV LQWHUFHSWHG ZLWK WZR SHWUL GLVKHV FP GLDPHWHUf SODFHG XQGHUQHDWK HDFK IRRG FXS RXWVLGH WKH FDJH 2QH ZDWHU ERWWOH ZDV DWWDFKHG WR WKH FDJH GLUHFWO\ LQ EHWZHHQ WKH IRRG FXSV 7KH WDVWDQW DQG FRQWURO GLHWV ZHUH SUHVHQWHG WR WKH PLFH LQ WKH IRRG FXSV )RU WHVWLQJ RQ 6LHUUD &KLQFXD UHKRXVHG WKH (A PHODQRWLV LQGLYLGXDOO\ LQ SODVWLF FDJHV LGHQWLFDO WR WKRVH GHVFULEHG DERYH H[FHSW WKDW WZR IRRG FXSV ZHUH VHFXUHG WR WKH IURQW ULJKW DQG OHIWKDQG FRUQHUV 6SLOODJH ZDV FROOHFWHG E\ UHVWLQJ HDFK IRRG FXS LQ D SHWUL GLVK FP GLDPHWHUf $ ZDWHU ERWWOH ZDV SODFHG GLUHFWO\ LQ EHWZHHQ WKH IRRG FXSV 7R SUHSDUH WKH WDVWDQW GLHWV WUHDWHG 3XULQD 5RGHQW /DE 0HDO QR KHQFHIRUWK PRXVH FKRZ PHDOf ZLWK HLWKHU GLJLWR[LQ IUHH DPLQH PRQRFURWDOLQH RU 1R[LGH PRQRFURWDOLQH 7KH GLJLWR[LQ DQG IUHH DPLQH PRQRFURWDOLQH ZHUH SXUFKDVHG IURP 6LJPD &KHPLFDO &R 6W /RXLV 02 ZKLOH WKH 1R[LGH PRQRFURWDOLQH ZDV GHULYHG IURP WKH IUHH DPLQH XVLQJ WKH WHFKQLTXH RXWOLQHG LQ 6FKZDUW] DQG %OXPEHUJV f DQG &UDLJ DQG 3XUXVKRWKDPDQ f ([DFW FRQFHQWUDWLRQV RI WKHVH FRPSRXQGV ZHUH HVWDEOLVKHG LQ WKH WDVWDQW GLHWV E\ f GLVVROYLQJ D JLYHQ TXDQWLW\ RI RQH RI WKH FRPSRXQGV LQ PO RI b HWKDQRO f PL[LQJ WKDW VROXWLRQ WKRURXJKO\ ZLWK J RI ODE FKRZ PHDO DQG f GU\LQJ WKH GLHW IRU

PAGE 137

KRXUV DW r& LQ D IRUFHGGUDIW RYHQ 7KH FRQWURO GLHW ZDV SUHSDUHG VLPLODUO\ H[FHSW WKDW QR WDVWDQW ZDV DGGHG $OO GLHWV ZHUH VWRUHG LQ D UHIULJHUDWRU 0LFH ZHUH VXEMHFWHG WR D VHULHV RI FRQVHFXWLYH K WZRFKRLFH SUHIHUHQFH WHVWV EHWZHHQ WKH WDVWDQW DQG FRQWURO GLHWV 7KH FRQFHQWUDWLRQ _LJ RI FRPSRXQG J PRXVH FKRZf RI GLJLWR[LQ IUHH DPLQH PRQRFURWDOLQH RU 1R[LGH PRQRFURWDOLQH LQ WKH WDVWDQW GLHW ZDV V\VWHPDWLFDOO\ LQFUHDVHG RU GHFUHDVHG HDFK K LQ QDWXUDO ORJ LQFUHPHQWV )RU H[DPSOH PHODQRWLV UHFHLYHG DVFHQGLQJ FRQFHQWUDWLRQV RI GLJLWR[LQ H H H H Hp SJ GLJLWR[LQ J PRXVH FKRZf 7KH UDQJH RI FRQFHQWUDWLRQV HDFK PRXVH UHFHLYHG ZDV VSHFLHV DQG FRPSRXQGVSHFLILF VHH 7DEOH f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f 7KH IRRG FXSV ZHUH WKHQ UHILOOHG DQG WKHLU SRVLWLRQV LQ WKH WHVWLQJ FDJH UHYHUVHG IRU WKH QH[W K SHULRG $IWHU WKLV DQG DOO VXEVHTXHQW K SHULRGV WKH VDPH SURFHGXUH ZDV UHSHDWHG H[FHSW WKDW f WKH WDVWDQW GLHW ZDV SODFHG LQ RQH FXS DQG WKH FRQWURO GLHW LQ WKH RWKHU f ZH XVHG FOHDQ IRRG FXSV DQG f IUHVK FRQWURO GLHW ZDV XVHG $OO

PAGE 138

7DEOH 7KH UDQJH RI FRQFHQWUDWLRQV _OJ FRPSRXQG J PRXVH FKRZf RI GLJLWR[LQ IUHH DPLQH PRQRFURWDOLQH DQG 1R[LGH PRQRFURWDOLQH XVHG LQ WKH WDVWDQW GLHWV (DFK PRXVH UHFHLYHG DQ DVFHQGLQJ $f RU GHVFHQGLQJ 'f VHULHV RI FRQFHQWUDWLRQV RI RQH RI WKH FRPSRXQGV LQ QDWXUDO ORJ VWHSV )RU H[DPSOH eAB PHODQRWLV UHFHLYHG WKH IROORZLQJ DVFHQGLQJ VHULHV RI GLJLWR[LQ FRQFHQWUDWLRQV H H H H DQG H ILJ GLJLWR[LQ J PRXVH FKRZ PHDO 7KH UDQJHV RI FRQFHQWUDWLRQV XVHG IRU HDFK FRPSRXQG DQG VSHFLHV ZHUH GHWHUPLQHG GXULQJ SUHOLPLQDU\ IHHGLQJ WULDOV $OO PLFH ZHUH FROOHFWHG P IURP D PRQDUFK FRORQ\ RQ 6LHUUD &KLQFXD H[FHSW IRU WKH (Br PHODQRW L V GHVLJQDWHG DV IDUf ZKLFK ZHUH FROOHFWHG NP IURP WKH VDPH PRQDUFK FRORQ\ 7KH QXPEHU RI PLFH RI HDFK VSHFLHV H[SRVHG WR HDFK FRPSRXQG DQG VHULHV RI FRQFHQWUDWLRQV LV LQ SDUHQWKHVHV 0RXVH 7\SH RI 'LJLWR[LQ )UHH DPLQH 1R[LGH VSHFLHV VHULHV PRQRFURWDOLQH PRQRFURWDOLQH PHODQRWLV $ f f PHODQRWLV IDUf $ f (M K\ORFHWHV $ f f $ f f f f f f f§ f§ f f f f f f f f 0 8 0 W / VXPLFKUDVWL

PAGE 139

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n PHDQ SUHIHUHQFH UDWLR IRU HDFK RI WKH FRQFHQWUDWLRQV RI GLJLWR[LQ DQG ERWK W\SHV RI PRQRFURWDOLQH 7KH DYRLGDQFH WKUHVKROG IRU ERWK DVFHQGLQJ DQG GHVFHQGLQJ VHULHV ZHUH VHW D SULRUL DV WKH ORZHVW FRQFHQWUDWLRQ DW ZKLFK WKH PHDQ SUHIHUHQFH UDWLR ZDV VLJQLILFDQWO\ EHORZ >3 RQHVDPSOH RQHWDLOHGf WWHVW@ $OO SUHIHUHQFH UDWLRV ZHUH WUDQVIRUPHG DUFVLQH f9[f EHIRUH VWDWLVWLFDO DQDO\VHV 7KHQ WR HVWLPDWH HDFK VSHFLHV DFWXDO DYRLGDQFH WKUHVKROG IRU HDFK FRPSRXQG WRRN WKH DYHUDJH RI WKH DVFHQGLQJ DQG GHVFHQGLQJ WKUHVKROGV DOVR ZDQWHG WR DVVHVV ZKHWKHU WKH WRWDO DPRXQW RI PRXVH FKRZ HDWHQ FRQWURO DQG WDVWDQW GLHWV FRPELQHGf E\ PLFH UHPDLQHG FRQVWDQW DV YDULHG f WKH FRQFHQWUDWLRQV RI WKH FRPSRXQGV LQ WKH WDVWDQW GLHWV DQG f WKH RUGHU RI WKHLU SUHVHQWDWLRQ 7R GR WKLV UDQ D VHSDUDWH IDFWRU UHSHDWHG PHDVXUHV $129$ IRU HDFK VSHFLHV RQ WKH WRWDO DPRXQW RI GLHW HDWHQ HDFK KRXU SHULRG E\ FRPSRXQG W\SH RUGHU RI SUHVHQWDWLRQ DQG FRQFHQWUDWLRQ WKH UHSHDWHG PHDVXUHf +RZHYHU EHFDXVH HDFK VSHFLHV ZDV H[SRVHG WR D GLIIHUHQW UDQJH RI FRQFHQWUDWLRQV RI HDFK FRPSRXQG 7DEOH f RQO\ WKH IRXU ORZHVW FRQFHQWUDWLRQV ZHUH LQFOXGHG 7KH UDQJH RI FRQFHQWUDWLRQV XVHG LQ WKH DQDO\VLV IRU f r

PAGE 140

PHODQRWLV ZDV WR SJ J GLHW IRU DOO FRPSRXQGV f 3 D K\ORFHWHV ZDV WR IRU GLJLWR[LQ DQG WR IRU ERWK IRUPV RI PRQRFURWDOLQH DQG f (r VXPLFKUDVWL ZDV WR IRU GLJLWR[LQ DQG WR IRU ERWK IRUPV RI PRQRFURWDOLQH 7KH GDWD IURP WKH (r PHODQRWLV FROOHFWHG NP IURP WKH PRQDUFK FRORQ\ ZHUH QRW LQFOXGHG LQ WKH DQDO\VLV ,Q WKLV DQG DOO VXEVHTXHQW $129$V WKH GDWD ZHUH VKRZQ WR EH QRUPDOO\ GLVWULEXWHG ZLWK WKH .ROPRJRURY6PLUQRY WHVW 3 f ([SHULPHQW 'HWHUPLQDWLRQ RI 7R[LF 6HQVLWLYLW\ ,Q WKLV H[SHULPHQW GHWHUPLQH IRU DOO WKUHH VSHFLHV WKH SK\VLRORJLFDO FRQVHTXHQFHV RI FKURQLFDOO\ HDWLQJ GLHWV FRQWDLQLQJ GLJLWR[LQ DQG PRQRFURWDOLQH DW FRQFHQWUDWLRQV VLPLODU WR WKRVH XVHG LQ ([SHULPHQW %HFDXVH RUDOO\ LQJHVWHG 1R[LGH PRQRFURWDOLQH LV FRQYHUWHG WR WKH IUHH DPLQH IRUP LQ WKH VPDOO LQWHVWLQH RI PLFH 0DWWRFNV f RQO\ WHVWHG WKH IUHH DPLQH IRUP RI PRQRFURWDOLQH 0LFH ZHUH KRXVHG DQG PDLQWDLQHG LQ *DLQHVYLOOH )ORULGD XQGHU WKH VDPH FRQGLWLRQV DV LQ ([SHULPHQW 7KLUW\HLJKW GD\ ROG )@B RIIVSULQJ RI ZLOGFDXJKW PLFH PHODQRWLV ( K\ORFHWHV DQG 5r VXPLFKUDVWLf ZHUH XVHG $OO WKUHH VSHFLHV ZHUH WHVWHG ZLWK GLJLWR[LQ ZKHUHDV RQO\ WKH IRUPHU WZR ZHUH WHVWHG ZLWK IUHH DPLQH PRQRFURWDOLQH )RU WHVWLQJ PLFH ZHUH UHKRXVHG LQ WKH VDPH VWDLQOHVV VWHHO FDJHV GHVFULEHG DERYH 2QH JODVV IRRG FXS DQG ZDWHU ERWWOH ZHUH VHFXUHG WR WKH IURQW RI HDFK FDJH DQG D SHWUL GLVK FP GLDPHWHUf ZDV SODFHG XQGHU HDFK IRRG FXS WR LQWHUFHSW VSLOODJH 7KH WDVWDQW GLHW ZDV SUHVHQWHG LQ D VLQJOH IRRG FXS

PAGE 141

, SUHSDUHG WKH FRQWURO DQG WDVWDQW GLHWV LQ WKH VDPH PDQQHU GHVFULEHG DERYH +RZHYHU XVHG RQO\ WZR FRQFHQWUDWLRQV RI IUHH DPLQH PRQRFURWDOLQH DQG GLJLWR[LQ DQG _LJ RI HLWKHU FRPSRXQG SHU J PRXVH FKRZ PHDO 7KHVH FRQFHQWUDWLRQV DUH VLPLODU WR WKRVH IRXQG LQ RYHUZLQWHULQJ PRQDUFKV VHH )LJXUHV D DQG Ef (DFK PRXVH ZDV JLYHQ DSSUR[LPDWHO\ J RI WKH FRQWURO GLHW GXULQJ WKH ILUVW WZR QLJKWV RI WHVWLQJ SUHOLPLQDU\ IHHGLQJ WULDOV GHPRQVWUDWHG WKDW LQGLYLGXDOV RI DOO VSHFLHV XVXDOO\ DWH OHVV WKDQ KDOI WKLV DPRXQW RI PRXVH FKRZ PHDO GXULQJ D JLYHQ QLJKW 7KHQ IRU HDFK RI WKH QH[W QLJKWV HDFK PRXVH ZDV RIIHUHG DQ HTXLYDOHQW DPRXQW RI HLWKHU WKH FRQWURO RU RQH RI WKH IRXU WDVWDQW GLHWV PRQLWRUHG HDFK PRXVHnV ZHLJKW WR WKH QHDUHVW J ZLWK D J 3HVOD VFDOH HYHU\ RWKHU GD\ EHWZHHQ K DQG IRRG FRQVXPSWLRQ WR WKH QHDUHVW J ZLWK D 0HWWOHU 3 EDODQFH HYHU\ GD\ GXULQJ WKH VDPH WLPH SHULRG /LWWHUPDWHV ZHUH GLVWULEXWHG URXJKO\ HYHQO\ DPRQJ WKH GLIIHUHQW GLHWDU\ WUHDWPHQWV 7ZRZD\ UHSHDWHG PHDVXUH $129$V ZHUH XVHG WR WHVW IRU WKH HIIHFWV RI DJH LQ GD\Vf DQG GLHW W\SH RQ f WKH ZHLJKW FKDQJH RI WKH PLFH DQG f WKH DPRXQW RI GLHW HDWHQ SHU GD\ 7KHVH WHVWV ZHUH SHUIRUPHG VHSDUDWHO\ IRU HDFK VSHFLHV DQG FRPSRXQG ,QWHUVSHFLILF FRPSDULVRQV RI WKH QLJKWO\ GRVDJHV PJ FRPSRXQGNJ PRXVHf RI GLJLWR[LQ DQG IUHH DPLQH PRQRFURWDOLQH ZHUH PDGH ZLWK D RQHn ZD\ $129$ DQG 6FKHII )WHVWV 7R FRPSXWH WKHVH QLJKWO\ GRVDJH YDOXHV GLYLGHG WKH DPRXQW RI FRPSRXQG LQJHVWHG HDFK QLJKW E\ D JLYHQ PRXVH ZKLFK FRXOG HVWLPDWH DFFXUDWHO\ EDVHG RQ WKH DPRXQW RI IRRG HDWHQf E\ LWV ZHLJKW HDFK QLJKW 6LQFH ZH FROOHFWHG ZHLJKWV HYHU\ RWKHU GD\

PAGE 142

, DVVXPHG WKDW HDFK PRXVHnV ZHLJKW UHPDLQHG FRQVWDQW RYHU WKH K SHULRGV EHWZHHQ ZHLJKLQJV 5HVXOWV ([SHULPHQW 'HWHUPLQDWLRQ RI 7DVWH 6HQVLWLYLW\ 7KH DYRLGDQFH WKUHVKROGV GHULYHG IURP DVFHQGLQJ DQG GHVFHQGLQJ VHULHV RI DOO WKUHH FRPSRXQGV ZHUH HLWKHU WKH VDPH RU ZLWKLQ RQH QDWXUDO ORJ VWHS RI RQH DQRWKHU LQ DOO FDVHV H[FHSW IRU WKRVH IURP (BA £B} K\ORFHWHV ZLWK 1R[LGH PRQRFURWDOLQH ZKLFK GLIIHUHG E\ QDWXUDO ORJ VWHSV 7DEOH f %HFDXVH WKH UHVSRQVHV WR WKH DVFHQGLQJ DQG GHVFHQGLQJ VHULHV ZHUH VR VLPLODU f GHWHUPLQHG WKH PHDQ DYRLGDQFH WKUHVKROG IRU HDFK FRPSRXQG DQG VSHFLHV E\ DYHUDJLQJ WKH PHDQV IURP ERWK VHULHV DQG f SRROHG WKH UHVXOWV IURP ERWK VHULHV DQG WKHQ SORWWHG HDFK VSHFLHVn PHDQ SUHIHUHQFH UDWLR DV D IXQFWLRQ RI WKH FRQFHQWUDWLRQ RI GLJLWR[LQ )LJXUH ODf 1R[LGH PRQRFURWDOLQH OEf DQG IUHH DPLQH PRQRFURWDOLQH OFf LQ WKH WDVWDQW GLHWV 7KH PHDQ DYRLGDQFH WKUHVKROGV DUH LQGLFDWHG ZLWK DUURZV RQ WKH [D[HV WKHVH SORWV $OO WKUHH VSHFLHV RI PLFH UHVSRQGHG TXLWH GLIIHUHQWO\ WR WKH GLJLWR[LQ GLHWV 3HURP\VFXV PHODQRWLVn PHDQ DYRLGDQFH WKUHVKROG ZDV DERXW WLPHV KLJKHU WKDQ VOA K\ORFHWHV n DQG K\ORFHWHV n ZDV DERXW WLPHV KLJKHU WKDQ VXPLFKUDVWLnV 7KH PHODQRWLV FROOHFWHG QHDUE\ Pf DQG IDU DZD\ NPf IURP 6LHUUD &KLQFXD H[KLELWHG LGHQWLFDO DVFHQGLQJ DYRLGDQFH WKUHVKROGV IRU GLJLWR[LQ 7DEOH f ,Q FRQWUDVW WR WKH UHVXOWV ZLWK GLJLWR[LQ DOO WKUHH VSHFLHV UHVSRQGHG UHODWLYHO\ VLPLODUO\ WR WKH 1R[LGH DQG IUHH DPLQH PRQRFURWDOLQH GLHWV 3HURP\VFXV PHODQRWLVn PHDQ DYRLGDQFH WKUHVKROG

PAGE 143

7DEOH $YRLGDQFH WKUHVKROG FRQFHQWUDWLRQV RI GLJLWR[LQ IUHH DPLQH PRQRFURWDOLQH DQG 1R[LGH PRQRFURWDOLQH IRU 5A PHODQRWLV $r K\ORFHWHV DQG 5A VXPLFKUDVWL IURP DVFHQGLQJ DQG GHVFHQGLQJ VHULHV RI SUHVHQWDWLRQV 7KH PHDQ DYRLGDQFH WKUHVKROG LV WKH DYHUDJH RI WKH DVFHQGLQJ DQG GHVFHQGLQJ WKUHVKROGV DOVR SUHVHQW WKH PHDQ DYRLGDQFH WKUHVKROG DV D PRODU FRQFHQWUDWLRQ $OO PLFH ZHUH FROOHFWHG P IURP D PRQDUFK FRORQ\ RQ 6LHUUD &KLQFXD H[FHSW IRU WKH 5B PHODQRW LV GHVLJQDWHG DV IDUf ZKLFK ZHUH FROOHFWHG NP IURP WKH VDPH PRQDUFK FRORQ\ $YRLGDQFH WKUHVKROG FRQFHQWUDWLRQV 0HDQ fOJ FRPSRXQG SHU 2OJ PRXVH FKRZf DYRLGDQFH WKUHVKROG &RPSRXQG 6SHFLHV $VFHQGLQJ VHULHV 'HVFHQGLQJ VHULHV 0HDQ DYRLGDQFH WKUHVKROG LQ PROHV FRPSG SHU J PRXVH FKRZ 'LJLWR[LQ (A PHODQRWLV ; (A PHODQRWLV IDUf (B A 5\ODFHWHV ; 5A 68PL&KUD6WL ; )UHH EDVH (B PHODQRWLV ; PRQRFURWDOLQH (MB L K\ORFHWHV ; f 5r 8PL&KUD6WL ; 1R[LGH (B PHODQRWLV ; PRQRFURWDOLQH (B DA K\ORFHWHV ; 5A VXPLFKUDVWL ;

PAGE 144

)LJXUH 0HDQ SUHIHUHQFH UDWLRV IRU GLIIHUHQW FRQFHQWUDWLRQV RI GLJLWR[LQ 1R[LGH PRQRFURWODLQH DQG IUHH DPLQH PRQRFURWDOLQH -J J GLHWf LQ PHLDQRWLV e (/f er KYORFHWHV (A -HDQG VXPLFKUDVWL (DFK PHDQ ZDV FDOFXODWHG IURP SUHIHUHQFH UDWLRV IURP ERWK DVFHQGLQJ DQG GHVFHQGLQJ VHULHV RI SUHVHQWDWLRQV 7KH VSHFLHVVSHFLILF PHDQ DYRLGDQFH WKUHVKROGV IRU HDFK FRPSRXQG DUH LQGLFDWHG ZLWK DUURZV RQ WKH UHVSHFWLYH [D[HV DQG DOVR LQ 7DEOH $ PHDQ SUHIHUHQFH UDWLR RI LQGLFDWHG E\ WKH GRWWHG OLQHf UHSUHVHQWV QR SUHIHUHQFH IRU WKH FRQWURO RU WDVWDQW GLHW ZKLOH D UDWLR RI UHSUHVHQWV WRWDO UHMHFWLRQ RI WKH WDVWDQW GLHW

PAGE 145

0HDQ SUHIHUHQFH UDWLR s6(f R 3 PHORQRWLV &RQFHQWUDWLRQ LJ2OJ GLHWf

PAGE 146

IRU WKH 1R[LGH ZDV DERXW WLPHV KLJKHU WKDQ WKDW RI WKH RWKHU VSHFLHV DQG LWV PHDQ DYRLGDQFH WKUHVKROG IRU WKH IUHH DPLQH ZDV HTXDO WR 5A VXPLFKUDVWLn V DQG RQO\ DERXW WLPHV KLJKHU WKDQ K\ORFHWHV 7KH UHVXOWV RI ZD\ $129$V SHUIRUPHG VHSDUDWHO\ RQ HDFK VSHFLHV LQGLFDWH WKDW WKH DPRXQW RI IRRG HDWHQ RYHU HDFK K SHULRG ZDV QRW LQIOXHQFHG VLJQLILFDQWO\ E\ WKH W\SH RI FRPSRXQG RUGHU RI SUHVHQWDWLRQ QRU FRQFHQWUDWLRQ RI WKH FRPSRXQG 7DEOH f 7KXV ZLWKLQ HDFK VSHFLHV DOO PLFH DWH D UHODWLYHO\ FRQVWDQW DQG VLPLODU DPRXQWV RI PRXVH FKRZ PHDO WKURXJKRXW WKH WZRFKRLFH IHHGLQJ H[SHULPHQWV PHDQ QR RI JUDPV LQJHVWHG s6( SHU K SHULRG E\ (Br PHODQRWLV s K\ORFHWHV s DQG 5AB VXPLFKUDVWL sf :KDW GLIIHUHG EHWZHHQ WKH K SHULRGV ZDV WKH UHODWLYH DPRXQWV RI FRQWURO 6 WDVWDQW GLHW HDWHQ ([SHULPHQW 'HWHUPLQDWLRQ RI 7R[LF 6HQVLWLYLW\ $OO PLFH IHG UHODWLYHO\ QRUPDOO\ RQ WKH FRQWURO GLHWV DQG WKH GLHWV WUHDWHG ZLWK GLJLWR[LQ DQG IUHH DPLQH PRQRFURWDOLQH H[FHSW IRU RI WKH (f§ K\ORFHWHV DQG DOO RI WKH VXPLFKUDVWL JLYHQ WKH KLJK GLJLWR[LQ GLHW 7DEOH f 7KHVH PLFH DWH H[WUHPHO\ VPDOO DPRXQWV RI WKH GLHW HDFK QLJKW DERXW Jf DQG ORVW ZHLJKW TXLFNO\ 7KH\ ZHUH UHPRYHG IURP WKH H[SHULPHQW ZLWKLQ GD\V VR WKH\ ZRXOG QRW GLH DQG ZHUH QRW LQFOXGHG LQ VXEVHTXHQW GDWD DQDO\VHV 7KH HIIHFWV RI WKH GLJLWR[LQ GLHWV RQ JURZWK DQG FRQVXPSWLRQ UDWH RQ DOO WKUHH VSHFLHV RI PLFH DUH LOOXVWUDWHG LQ )LJXUHV DQG DQG DQDO\]HG VWDWLVWLFDOO\ LQ 7DEOH $GGLWLRQ RI GLJLWR[LQ WR WKH GLHWV GLG QRW LQIOXHQFH (AB PHODQRWLVn DQG (A K\ORFHWHV JURZWK UDWH

PAGE 147

7DEOH 7KUHHZD\ UHSHDWHG PHDVXUH $129$ RI PRXVH FKRZ HDWHQ SHU K SHULRG E\ FRPSRXQG GLJLWR[LQ 1R[LGH PRQRFURWDOLQH DQG IUHH DPLQH PRQRFURWDOLQHf RUGHU RI SUHVHQWDWLRQ DVFHQGLQJ RU GHVFHQGLQJf DQG FRQFHQWUDWLRQ RI WKH FRPSRXQG LQ WKH WDVWDQW GLHW WKH UHSHDWHG PHDVXUHf 7KLV DQDO\VLV ZDV SHUIRUPHG VHSDUDWHO\ IRU er PHODQRWLV er D K\ORFHWHV DQG ,/ VXPLFKUDVWL 6RXUFH RI YDULDWLRQ GI ) 3 eB! PHODQRWLV &RPSRXQG $f 3UHVHQWDWLRQ RUGHU %f $ [ % &RQFHQWUDWLRQ &f $ [ & % [ & $ [ % [ & (M 6X K\ORFHWHV &RPSRXQG $f 3UHVHQWDWLRQ RUGHU %f $ [ % &RQFHQWUDWLRQ &f $ [ & % [ & $ [ % [ & 5 VXPLFKUDVWL &RPSRXQG $f 3UHVHQWDWLRQ RUGHU %f $ [ % &RQFHQWUDWLRQ &f $ [ & % [ & $ [ % [ &

PAGE 148

7DEOH 7KH QXPEHU RI MXYHQLOH PLFH WHVWHG XQGHU HDFK GLHWDU\ WUHDWPHQW DQG WKH SHUFHQWDJH RI WKHVH PLFH WKDW IHG QRUPDOO\ DQG JDLQHG ZHLJKW RQ WKH GLHWV &RPSRXQG 0RXVH 6SHFLHV &RQFHQWUDWLRQ _LJ FPSG J GLHWf Q L2 b Q b Q b 'LJLWR[LQ e/ PHODQRWLV (X 6/r K\ORFHWHV r 5r VXPLFKUDVWL r )UHH DPLQH (M PHODQRWLV PRQRFURWDOLQH (M K\ORFHWHV r 7KUHH RI WKH (O Dr K\ORFHWH DQG DOO 5r VXPLFKUDVWL DWH H[FHHGLQJO\ VPDOO DPRXQWV RI WKH GLHW DQG WKXV ORVW ZHLJKW SUHFLSLWRXVO\ 7KH\ ZHUH UHPRYHG IURP WKH H[SHULPHQW ZLWKLQ GD\V

PAGE 149

)LJXUH :HLJKW FKDQJHV RYHU WLPH LQ PHODQRWLV Q WUHDWPHQWf K\ORFHWHV Q WUHDWPHQWf DQG VXPLFKUDVWL Q WUHDWHPQWf WKDW ZHUH PDLQWDLQHG RQ RQH RI GLHWDU\ WUHDWPHQWV 7KH WUHDWPHQWV GLIIHUHG RQO\ ZLWK UHVSHFW WR WKH FRQFHQWUDWLRQ RI GLJLWR[LQ $OO PLFH EHJDQ WKH H[SHULPHQW DW GD\V RI DJH

PAGE 150

R &RQWURO GLHW

PAGE 151

)LJXUH $PRXQW RI GLHW HDWHQ RYHU WLPH E\ eBA PHODQRWLV Q WUHDWPHQWf 5r K\ORFHWHV Q WUHDWPHQWf DQG 5r VXPLFKUDVWL Q WUHDWHPQWf WKDW ZHUH PDLQWDLQHG RQ RQH RI GLHWDU\ WUHDWPHQWV 7KH WUHDWPHQWV GLIIHUHG RQO\ ZLWK UHVSHFW WR WKH FRQFHQWUDWLRQ RI GLJLWR[LQ

PAGE 152

R &RQWURO GLHW

PAGE 153

7DEOH 5HVXOWV RI WZRIDFWRU UHSHDWHG PHDVXUH $129$V RQ WKH HIIHFWV RI GLHW W\SH DQG DJH RQ WKH GHSHQGHQW PHDVXUHV f ZHLJKW RI PRXVHHYHU\ RWKHU QLJKW DQG f DPRXQW RI GLHW HDWHQPRXVHQLJKW 7KH GLHW IDFWRU LQFOXGHG WKUHH FRQFHQWUDWLRQV RI GLJLWR[LQ RU 0J J GLHWf IRU 3HURP\VFXV PHODQRWLV DQG A KYORFHWHV DQG RQO\ WZR DQG OJ J GLHWf IRU 5 VXPLFKUDVWL 6HH 7DEOH IRU WKH QXPEHU RI PLFH XVHG LQ HDFK $129$ DQG )LJXUHV DQG IRU JUDSKLFDO SUHVHQWDWLRQ RI WKH GDWD 'HSHQGHQW PHDVXUH 6SHFLHV 6RXUFH RI YDULDWLRQ ) GI 3 :HLJKW RI (X PHODQRWLV 'LHW W\SH PRXVH $JH ,QWHUDFWLRQ 3 D KYORFHWHV 'LHW W\SH $JH ,QWHUDFWLRQ %XX VXPLFKUDVWL 'LHW W\SH $JH ,QWHUDFWLRQ $PRXQW RI (r PHODQRWLV 'LHW W\SH GLHW HDWHQ $JH ,QWHUDFWLRQ 3 D KYORFHWHV 'LHW W\SH $JH ,QWHUDFWLRQ %XO VXPLFKUDVWL 'LHW W\SH $JH ,QWHUDFWLRQ

PAGE 154

7KH HIIHFW RI DJH ZDV VLJQLILFDQW IRU ERWK VSHFLHV DQG WKH LQWHUDFWLRQ RI GLHW W\SH DQG DJH ZDV QRQVLJQLILFDQW IRU (A PHODQRWLV DQG VLJQLILFDQW IRU (BA K\ORFHWHV ,Q FRQWUDVW DGGLWLRQ RI GLJLWR[LQ JUHDWO\ GLPLQLVKHG VXPLFKUDVWLnV JURZWK UDWH 7KHUH ZDV D VLJQLILFDQW HIIHFW RI GLHW W\SH DQG DJH RQ ZHLJKW FKDQJH WKH LQWHUDFWLRQ RI WKHVH IDFWRUV ZDV DOVR VLJQLILFDQW (YHQ WKRXJK DOO WKUHH VSHFLHV DSSHDUHG LQLWLDOO\ GHWHUUHG IURP HDWLQJ WKH GLJLWR[LQ WUHDWHG GLHWV VHH GD\V LQ )LJXUH f (BA DA K\ORFHWHV DQG (A PHODQRWLV HYHQWXDOO\ EHJDQ HDWLQJ HTXDO RU HYHQ JUHDWHU DPRXQW RI WKH FRQWURO DQG GLJLWR[LQWUHDWHG GLHWV ,Q FRQWUDVW VXPLFKUDVWL DSSHDUHG WR EH GHWHUUHG IURP HDWLQJ WKH GLJLWR[LQWUHDWHG GLHWV WKURXJKRXW WKH HQWLUH H[SHULPHQW ,W ODUJHO\ UHMHFWHG WKH GLHW ZLWK WKH KLJK FRQFHQWUDWLRQ RI GLJLWR[LQ DQG DWH VLJQLILFDQWO\ OHVV RI WKH RQH ZLWK WKH ORZ FRQFHQWUDWLRQ RI GLJLWR[LQ WKDQ LW GLG WKH FRQWURO GLHW 7KLV UHVXOW PD\ H[SODLQ VXPLFKUDVWLnV GHSUHVVHG JURZWK UDWH RQ WKH GLHW ZLWK ORZ FRQFHQWUDWLRQV RI GLJLWR[LQ 7KH HIIHFWV RI WKH IUHH DPLQH PRQRFURWDOLQH GLHWV RQ JURZWK DQG FRQVXPSWLRQ UDWH LQ (A PHODQRWLV DQG (A K\ORFHWHV DUH LOOXVWUDWHG LQ )LJXUHV DQG DQG DQDO\]HG VWDWLVWLFDOO\ LQ 7DEOH $GGLWLRQ RI IUHH DPLQH PRQRFURWDOLQH KDG QR LPSDFW RQ JURZWK UDWH LQ 3 PHODQRWLV EXW KDG D GRVHGHSHQGHQW QHJDWLYH LPSDFW LQ 3 D K\ORFHWHV 1HYHUWKHOHVV GLHW W\SH KDG QR LQIOXHQFH WKH DPRXQW RI IRRG HDWHQ E\ HLWKHU VSHFLHV (YHQ WKRXJK eBr DA K\ORFHWHV DWH PRUH RI WKH GLJLWR[LQ DQG PRQRFURWDOLQHWUHDWHG GLHWV RQ D JLYHQ QLJKW WKDQ GLG WKH WZR VPDOOHU VSHFLHV VHH )LJXUHV DQG f DQG WKXV DOVR LQJHVWHG JUHDWHU TXDQWLWLHV RI HDFK FRPSRXQG WKHLU PHDQ QLJKWO\ GRVDJH RI WKHVH

PAGE 155

)LJXUH :HLJKW FKDQJHV RYHU WLPH LQ PHODQRWLV Q WUHDWPHQWf DQG K\ORFHWHV Q WUHDWPHQWf WKDW ZHUH PDLQWDLQHG RQ RQH RI GLHWDU\ WUHDWPHQWV 7KH WUHDWPHQWV GLIIHUHG RQO\ ZLWK UHVSHFW WR WKH FRQFHQWUDWLRQ RI IUHH DPLQH PRQRFURWDOLQH $OO PLFH EHJDQ WKH H[SHULPHQW DW GD\V RI DJH

PAGE 156

r &RQWURO GLHW f O$ILT PRQRFURWROLQH2OJ GLHW

PAGE 157

)LJXUH $PRXQW RI GLHW HDWHQ RYHU WLPH E\ PHODQRWLV Q WUHDWPHQWf DQG K\ORFHWHV Q WUHDWPHQWf WKDW ZHUH PDLQWDLQHG RQ RQH RI GLHWDU\ WUHDWPHQWV 7KH WUHDWPHQWV GLIIHUHG RQO\ ZLWK UHVSHFW WR WKH FRQFHQWUDWLRQ RI IUHH DPLQH PRQRFURWDOLQH

PAGE 158

0HDQ DPRXQW RI GLHW HDWHQ Jf R &RQWURO GLHW f PRQRFURWDOLQH2, J GLHW

PAGE 159

7DEOH 5HVXOWV RI WZRIDFWRU UHSHDWHG PHDVXUH $129$V RQ WKH HIIHFWV RI GLHW DQG DJH RQ WKH GHSHQGHQW PHDVXUHV f ZHLJKW RI PRXVHHYHU\ RWKHU QLJKW DQG f WKH DPRXQW RI GLHW HDWHQPRXVHQLJKW 7KH GLHW IDFWRU LQFOXGHG FRQFHQWUDWLRQV RI IUHH DPLQH PRQRFURWDOLQH DQG _LJ J GLHWf 6HH 7DEOH IRU WKH QXPEHU RI PLFH XVHG LQ HDFK $129$ DQG )LJXUHV DQG IRU D JUDSKLFDO SUHVHQWDWLRQ RI WKH GDWD 'HSHQGHQW PHDVXUH 6SHFLHV 6RXUFH RI YDULDWLRQ ) GI 3 :HLJKW RI 6O PHODQRWLV 'LHW W\SH PRXVH $JH ,QWHUDFWLRQ 3 D KYORHHWHV 'LHW W\SH $JH ,QWHUDFWLRQ $PRXQW RI O PHODQRWLV 'LHW W\SH GLHW HDWHQ $JH ,QWHUDFWLRQ 3 D KYOQUHWHV 'LHW W\SH $JH ,QWHUDFWLRQ

PAGE 160

FRPSRXQGV ZDV VLJQLILFDQWO\ OHVV 7DEOH f 5HL WKPULQQWQP\V VXPLFKUDVWLnV QLJKWO\ GRVDJH RI GLJLWR[LQ LQ WKH QJ GLHW ZDV VLJQLILFDQWO\ KLJKHU WKDQ WKDW RI WKH RWKHU WZR VSHFLHV 3 6FKHII )WHVWf 'LVFXVVLRQ 'LIIHUHQFHV LQ 7DVWH 6HQVLWLYLWLHV $PRQJ WKH 7KUHH 6SHFLHV 7KH PHDQ DYRLGDQFH WKUHVKROGV IRU GLJLWR[LQ GLIIHUHG DPRQJ WKH VSHFLHV E\ VHYHUDO RUGHUV RI PDJQLWXGH ZLWK eBr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f 7KDW VXFK FORVHO\ UHODWHG VSHFLHV RI PLFH UHVSRQGHG VR GLIIHUHQWO\ WR WKH WDVWH RI GLJLWR[LQ LV QRW VXUSULVLQJ 6WXGLHV ZLWK ODERUDWRU\ PLFH KDYH UHYHDOHG LQWHUVWUDLQ GLIIHUHQFHV LQ WDVWH VHQVLWLYLW\ WR VHYHUDO ELWWHU FRPSRXQGV WKDW DUH RI FRPSDUDEOH PDJQLWXGH WR WKRVH UHSRUWHG LQ WKLV VWXG\ 7KH ELWWHU FRPSRXQGV LQFOXGH VXFURVH RFWD DFHWDWH :DUUHQ DQG /HZLV /XVK +DUGHU HW DO f TXLQLQH VXOIDWH /XVK f VWU\FKQLQH /XVK f DQG WKH DFHWDWHV RI UDIILQRVH JDODFWRVH DQG ODFWRVH /XVK f

PAGE 161

7DEOH 1LJKWO\ GRVDJH PJNJ PRXVHf DQG LQWDNH PJf RI GLJLWR[LQ DQG IUHH DPLQH PRQRFURWDOLQH PHDQ s6(f E\ PLFH RYHU QLJKWV LQ GLHWV ZLWK RQH RI WZR FRQFHQWUDWLRQV RI HLWKHU FRPSRXQG ,QWHUVSHFLILF FRPSDULVRQV DUH PDGH ZLWKLQ HDFK FROXPQ VHSDUDWHO\ IRU HDFK FRPSRXQG DQG FRQFHQWUDWLRQ ZLWK WKH RQHZD\ $129$ 'LIIHUHQW VXEVFULSWV D E Ff LQGLFDWH VLJQLILFDQW GLIIHUHQFHV DPRQJ PHDQV ZLWKLQ HDFK FROXPQ IRU WKH VSHFLHV FRPSDULVRQV 3 6FKHII )WHVWf r 3 r r 3 &RQFHQWUDWLRQ LQ GLHW 0RXVH &RPSRXQG _OJJf VSHFLHV Q 1LJKWO\ GRVDJH PJNJf $PRXQW LQJHVWHG QLJKW PJf 'LJLWR[LQ 3 D KYOQQHWSV s& sD PHODQRWLV sE sE / VXPLFKUDVWL sD sOF )UDWLR GI rr rr K\ODVHWHD s s eB! PHODQRWLV s s )UDWLR r r GI 0RQRFURWDOLQH WRODFVWDD s s PHODQRWLV s s )UDWLR GI r r 3 D KYORQHWHV s s PHODQRWLV s s )UDWLR GI r r

PAGE 162

,Q FRQWUDVW WR GLJLWR[LQ WKH PHDQ DYRLGDQFH WKUHVKROGV IRU PRQRFURWDOLQH ERWK IUHH DPLQH DQG 1R[LGH IRUPVf GLG QRW GLIIHU JUHDWO\ DPRQJ VSHFLHV 7KLV LQGLFDWHV WKDW f WKH GLYHUJHQW UHVSRQVHV WKHVH VSHFLHV RI PLFH VKRZ WR RYHUZLQWHULQJ PRQDUFK EXWWHUIOLHV LV SUREDEO\ QRW LQIOXHQFHG VWURQJO\ E\ WKHLU UHVSHFWLYH WDVWH VHQVLWLYLWLHV WR 3$V DQG f WKHUH LV QR VLJQLILFDQW GLIIHUHQFH LQ WKH JXVWDWRU\ SURSHUWLHV RI IUHH DPLQH DQG 1R[LGH IRUPV RI 3$V HYHQ WKRXJK WKH ODWHU IRUP LV VOLJKWO\ PRUH SRODU RZLQJ WR LW EHLQJ D VDOW ,W LV QRWDEOH WKDW WKH DYRLGDQFH WKUHVKROGV LQ PRODU XQLWVf IRU GLJLWR[LQ DQG ERWK IRUPV RI PRQRFURWDOLQH ZHUH VLPLODU LQ ( PHODQRWLV EXW ZLGHO\ GLYHUJHQW LQ / K\ORFHWHV DQG 5A VXPLFKUDVWL 7DEOH f 7KXV GLJLWR[LQ DSSHDUV WR EH PRUH HIIHFWLYH WKDQ PRQRFURWDOLQH DV DQ DYHUVLYH VWLPXOXV DJDLQVW 5A Lr K\ORFHWHV DQG 5r VXPLFKUDVWL ,W ZRXOG EH LQWHUHVWLQJ WR GHWHUPLQH ZKHWKHU WKLV DSSOLHV WR RWKHU NLQGV RI &*V DQG 3$V (FRORJLFDO 5HOHYDQFH RI WKH $YRLGDQFH 7KUHVKROGV 7R SODFH WKH DYRLGDQFH WKUHVKROGV RI WKH WKUHH VSHFLHV RI PLFH LQ DQ HFRORJLFDO FRQWH[W RYHUODLG HDFK VSHFLHVn DYRLGDQFH WKUHVKROG IRU GLJLWR[LQ DQG PRQRFURWDOLQH RQ WKH QDWXUDO IUHTXHQF\ GLVWULEXWLRQV RI &*V DQG 3$V UHVSHFWLYHO\ DV IRXQG LQ 0H[LFDQ RYHUZLQWHULQJ PRQDUFKV )LJXUHV D DQG Ef %HFDXVH WKH DYRLGDQFH WKUHVKROGV IRU WKH IUHH DPLQH DQG 1R[LGH IRUPV RI PRQRFURWDOLQH ZHUH VR VLPLODU DYHUDJHG WKHP IRU HDFK VSHFLHV DQG SUHVHQW WKHVH DYHUDJHV RQ )LJXUH Ef )LJXUH D LQGLFDWHV WKDW IRU 5BA DA K\ORFHWHV DQG 5 VXPLFKUDVWL PRUH WKDQ b RI WKH PRQDUFKV KDYH VXSUDWKUHVKROG FRQFHQWUDWLRQV RI &*V DQG IRU 5AB PHODQRWLV RQO\ b GR ,Q FRQWUDVW

PAGE 163

)LJXUH )UHTXHQF\ GLVWULEXWLRQ RI WKH ZKROH ERG\ FRQFHQWUDWLRQ RI $f &*V DQG %f 3$V LQ RYHUZLQWHULQJ PRQDUFKV RQ 6LHUUD &KLQFXD RQH RI WKH SULPDU\ VLWHV LQ 0H[LFR 7KH &* GDWD ZHUH GHULYHG IURP 0DOFROP DQG %URZHU f DQG WKH 3$ GDWD FDOFXODWHG IURP UHVXOWV LQ .HOOH\ HW DO f DQG %URZHU DQG &DOYHUW f 7KH PHDQ DYRLGDQFH WKUHVKROGV IRU GLJLWR[LQ DQG PRQRFURWDOLQH LQ PHODQRWLV 3 D K\ORFHWH DQG (A VXPLFKUDVWL DUH LQGLFDWHG RQ WKH UHVSHFWLYH &* DQG 3$ GLVWULEXWLRQV 5DWKHU WKDQ SUHVHQWLQJ VHSDUDWH PHDQ DYRLGDQFH WKUHVKROGV IRU IUHH DPLQH DQG 1R[LGH PRQRFURWDOLQH SUHVHQW DYHUDJHV RI WKHVH WKUHVKROGV IRU HDFK VSHFLHV

PAGE 164

b RI VDPSOH b RI VDPSOH ? &0 FR &2 R &0 &2 R &0 &2 R &0 ,7 &2 2 &0 &2 FR 2f R &0 &2 WR &2 *2 R! R &0 &0 &0 &0 &0 2 &0 &2 &2 G &0 LI FR &0 &r &2 Ir &2 WR &0 &0 ‘7 &2 G &0 R! R R &2 P FR R! R &0 &0 &0 &0 :KROH ERG\ FRQFHQWUDWLRQ SJ FRPSRXQG J GU\ ZHLJKW RI PRQDUFKf

PAGE 165

)LJXUH E LQGLFDWHV WKDW IRU DOO WKUHH VSHFLHV OHVV WKDQ b RI WKH PRQDUFKV KDYH VXSUDWKUHVKROG FRQFHQWUDWLRQV RI 3$V ,Q IDFW IRU 3 PHODQRWLV DOO PRQDUFKV KDYH VXEWKUHVKROG FRQFHQWUDWLRQV RI 3$V ([WUDSRODWLRQ RI WKHVH UHVXOWV WR IUHHUDQJLQJ PLFH VXJJHVWV WKDW WKH HQFRXQWHU UDWH RI SDODWDEOH PRQDUFKV LH RQHV ZLWK VXEWKUHVKROG FRQFHQWUDWLRQV RI 3$V DQG &*Vf RQ WKH IRUHVW IORRU LQVLGH D FRORQ\ E\ r DA K\ORFHWHV DQG VXPLFKUDVWL ZRXOG EH OHVV WKDQ RXW RI ZKHUHDV WKDW E\ PHODQRWLV ZRXOG EH RXW RI 7KH ORZ HQFRXQWHU UDWH IRU eBA K\ORFHWHV DQG 5r VXPLFKUDVWL PD\ H[SODLQ ZK\ WKH\ GR QRW IHHG QDWXUDOO\ RQ RYHUZLQWHULQJ PRQDUFKV ,Q PDNLQJ WKHVH H[WUDSRODWLRQV DP DVVXPLQJ WKDW WKH WDVWHV RI WKH &*V DQG 3$V LQ RYHUZLQWHULQJ PRQDUFKV DUH VLPLODU WR WKRVH RI GLJLWR[LQ DQG PRQRFURWDOLQH UHVSHFWLYHO\ LQ PRXVH FKRZ PHDO +RZHYHU WKHUH DUH VHYHUDO IDFWRUV WKDW PD\ FDXVH WKH WDVWHV WR GLIIHU TXDQWLWDWLYHO\ DQGRU TXDOLWDWLYHO\ )LUVW EHFDXVH DVSHFLRVLGH WKH GRPLQDQW &* LQ RYHUZLQWHULQJ PRQDUFKVf LV PRUH SRODU WKDQ GLJLWR[LQ 7DEOH f LW PD\ EH DYRLGHG DW D ORZHU FRQFHQWUDWLRQ %URZHU DQG )LQN f 6HFRQG &*V ERXQG LQ PRQDUFK FXWLFOH PD\ EH WDVWHG PRUH ZHDNO\ WKDQ RQHV PL[HG LQWR PRXVH FKRZ &OHPHQW f 7KLUG WKHUH PD\ EH V\QHUJLVWLF LQWHUDFWLRQV EHWZHHQ WKH &*V DQG 3$V LQ WKH VDPH PRQDUFK 6XFK WDVWH \QHUJLVPV KDYH EHHQ IRXQG WR RFFXU ZLWK FHUWDLQ VZHHWHQHUV LQ UDWV 6PLWK HW DO f 7KHVH SRVVLELOLWLHV UDLVH LQWHUHVWLQJ TXHVWLRQV IRU IXWXUH UHVHDUFK 7R[LF 6HQVLWLYLW\ RI 7KUHH 6SHFLHV WR 'LJLWR[LQ DQG 0RQRFURWDOLQH 'LJLWR[LQ )RU WKH (A PHODQRWLV DQG (A K\ORFHWHV WKDW KDELWXDWHG WR WKH ORZ DQG KLJK GLJLWR[LQ GLHWV WKHUH ZDV QR HIIHFW RI GLHWDU\

PAGE 166

WUHDWPHQW RQ ZHLJKW FKDQJH RU IRRG LQWDNH 7KLV UHVXOW EHFRPHV HYHQ PRUH VLJQLILFDQW ZKHQ RQH FRPSDUHV WKH QLJKWO\ GRVDJHV RI GLJLWR[LQ ERWK VSHFLHV H[SHULHQFHG XQGHU WKH KLJK &* WUHDWPHQW ZLWK WKH DFXWH /'4 YDOXHV IRU RUDO GRVDJHV RI GLJLWR[LQ LQ RWKHU VSHFLHV RI PDPPDOV 7KH PHDQ QLJKWO\ GRVDJHV IRU PHODQRWLV DQG K\OQQVWVA ZDV JUHDWHU WKDQ PJNJ PRXVH 7DEOH f ZKHUHDV WKH DFXWH /' IRU ODERUDWRU\ PLFH UDWV DQG FDWV LV DQG PJNJ UHVSHFWLYHO\ %DUQHV DQG (OWKHULQJWRQ 0DUW\ f 7KXV LW DSSHDUV WKDW eBA PHODQRWLV DQG K\ORFHWHV OLNH (A PDQLFXODWXV DQG ODERUDWRU\ PLFH 0DUW\ f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f 7KH VWURQJ

PAGE 167

DYHUVLRQ ,/r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f GRHV QRW LQIOXHQFH WKH KHSDWRWR[LF DFWLRQ RI D JLYHQ GRVDJH RI D 3$ 0DWWRFNV f LW LV SRVVLEOH WR FRPSDUH UHVXOWV IURP VWXGLHV LQ ZKLFK GLIIHUHQW W\SHV RI DGPLQLVWUDWLRQ SURFHGXUHV ZHUH XVHG $FXWH /'RnV IRU ODERUDWRU\ UDWV DQG PLFH WKDW UHFHLYHG LQWUDSHULWRQHDO LQMHFWLRQV RI PRQRFURWDOLQH ZHUH DQG PJNJ UHVSHFWLYHO\ 0DWWRFNV 0LUDQGD HW DO f DOO RI WKH DQLPDOV GLHG ZLWKLQ GD\V RI WKH OHWKDO GRVDJH ,Q WKLV VWXG\ PHODQRWLV DQG 3 K\ORFHWHV LQJHVWHG RQ DYHUDJH DQG PJNJ RI PRQRFURWDOLQH SHU QLJKW UHVSHFWLYHO\ ZLWKRXW DQ\ PRUWDOLW\ (YHQ WKRXJK WKHVH GRVDJHV DUH ZHOO EHORZ WKH 8AJnV RI WKH PLFH DQG UDWV PHODQRWLV DQG K\ORFHWHV LQJHVWHG WKHVH GRVDJHV RYHU FRQVHFXWLYH QLJKWV +RZHYHU EHIRUH DQ\ GHILQLWH FRQFOXVLRQV FDQ EH GUDZQ UHJDUGLQJ WKH WR[LF VHQVLWLYLWLHV RI WKHVH WZR VSHFLHV RI PLFH WR PRQRFURWDOLQH WKH OLYHUV RI WKH PLFH XVHG LQ WKLV VWXG\ PXVW EH H[DPLQHG KLVWRORJLFDOO\ IRU VLJQV RI 3$ WR[LFLW\ VHH &KDSWHU f 7KLV DQDO\VLV LV LQ SURJUHVV $W WKH HQG RI ([SHULPHQW DOO PLFH ZHUH NLOOHG E\ FHUYLFDO

PAGE 168

GLVORFDWLRQ DQG WKHLU OLYHUV H[WUDFWHG DQG VWRUHG LQ b QHXWUDO EXIIHUHG IRUPDOLQ KDYH VLQFH VHQW WKHVH WLVVXHV WR 'U &KDUOHV 0RQWJRPHU\ D OLYHU SDWKRORJLVW DQG KH ZLOO VHFWLRQ DQG HYDOXDWH WKHP IRU OHVLRQV 5HODWLRQVKLS %HWZHHQ 7DVWH DQG 7R[LF 6HQVLWLYLW\ 'LJLWR[LQ 7KH UHVXOWV IURP ([SHULPHQWV DQG VXJJHVW QR FOHDU UHODWLRQVKLS EHWZHHQ WKH WDVWH DQG WR[LF VHQVLWLYLWLHV LQ DOO VSHFLHV RI PLFH (YHQ WKRXJK DOO VSHFLHV DSSHDUHG WR EH UHODWLYHO\ LQVHQVLWLYH WR GLJLWR[LQ WR[LFLW\ WKH\ YDULHG ZLGHO\ LQ WDVWH VHQVLWLYLW\ %RWK (AB K\ORFHWHV DQG r VXPLFKUDVWL UHMHFWHG GLHWV ZLWK DSSDUHQWO\ KDUPOHVV FRQFHQWUDWLRQV RI GLJLWR[LQ 7KLV VXJJHVWV WKDW WKHVH VSHFLHV GLG QRW EDVH WKHLU IHHGLQJ UHVSRQVHV WR WKH GLJLWR[LQ GLHW RQ SRVW LQJHVWLRQDO IHHGEDFN 5DWKHU LW DSSHDUV WKDW WDVWH ZDV RI RYHUULGLQJ LPSRUWDQFH 8QIRUWXQDWHO\ WKH FRQFHQWUDWLRQV RI GLJLWR[LQ XVHG LQ ([SHULPHQW ZHUH EHORZ (A PHODQRWLV PHDQ WDVWH DYRLGDQFH WKUHVKROG ,W ZRXOG EH LQWHUHVWLQJ WR UHSHDW ([SHULPHQW ZLWK (Br PHODQRWLV H[FHSW XVH KLJKHU FRQFHQWUDWLRQV RI GLJLWR[LQ 6XFK DQ H[SHULPHQW VKRXOG GHWHUPLQH ZKHWKHU (A PHODQRWLV DV ZHOO EDVHV LWV UHVSRQVHV WR GLJLWR[LQ WUHDWHG GLHWV RQ WDVWH RU SRVWLQJHVWLRQDO HIIHFWV 6HYHUDO RWKHU ZRUNHUV WKDW KDYH IRXQG WKDW YHUWHEUDWH DQG LQYHUWHEUDWH SUHGDWRUV DYRLG IRRGV EDVHG RQ WKHLU ELWWHU WDVWH DORQH 5HGZLQJHG EODFNELUGV DYRLGHG D YDULHW\ RI FRPSRXQGV WKDW RFFXUUHG LQ DUWKURSRG VHFUHWLRQV EHFDXVH RI WKH ELWWHU WDVWH
PAGE 169

f 0DQ\ SK\WRSKDJRXV LQVHFW VSHFLHV DSSHDU WR DVVHVV WKH TXDOLW\ RU SRWHQWLDO GDQJHUf RI D IRRGSODQW VLPSO\ E\ WDVWLQJ WKH VXUIDFHV RI LWV OHDYHV %HUQD\V DQG &KDSPDQ &KDSPDQ DQG %HUQD\V f ,Q DGGLWLRQ VHYHUDO VWXGLHV ZLWK VZHHW FRPSRXQGV VXJJHVW WKDW ODERUDWRU\ UDWV DQG GRJV PD\ HDW IRU WDVWH UDWKHU WKDQ IRU FDORULHV XQGHU FHUWDLQ FRQGLWLRQV -DFREV DQG 6KDUPD &DSDOGL HW DO LQ SUDVDf 7KLV GLVFXVVLRQ ZLOO EH HODERUDWHG XSRQ LQ &KDSWHU ,Q FRQFOXVLRQ WKH UHVXOWV RI WKLV FKDSWHU SURYLGH VWURQJ VXSSRUW IRU VHYHUDO K\SRWKHVHV SURSRVHG LQ HDUOLHU FKDSWHUV f WKDW 5 VXPLFKUDVWL DQG K\ORFHWHV DYRLG IHHGLQJ XSRQ RYHUZLQWHULQJ PRQDUFKV EHFDXVH RI WKHLU KLJK VHQVLWLYLW\ WR WKH WDVWH RI &*V f WKDW DOO VSHFLHV RI PLFH DEVRUE &*V SRRUO\ DFURVV WKHLU JDVWURLQWHVWLQDO WUDFW DQG f WKDW LW LV PHODQRWLVn LQVHQVLWLYLW\ WR WKH WDVWH DQG WR[LF HIIHFWV RI WKH PRQDUFKnV &*V WKDW HQDEOHV LW WR WDNH DGYDQWDJH RI RYHUZLQWHULQJ PRQDUFKV 7KH 3BA PHODQRWLV SRSXODWLRQV ZLWKLQ RYHUZLQWHULQJ DUHDV GR QRW DSSHDU WR SRVVHVV VSHFLILF DGDSWDWLRQV IRU HDWLQJ WKH PRQDUFKV &RQWUDU\ WR WKH VXJJHVWLRQ E\ VHYHUDO ZRUNHUV WKDW 3$V SOD\ DQ LPSRUWDQW GHIHQVLYH UROH LQ PRQDUFKV (GJDU HW DO 5RWKVFKLOG DQG 0DUVK (LVQHU .HOOH\ HW DO EXW VHH %URZHU f P\ UHVXOWV VXJJHVW WKDW 3$V GR QRW RFFXU DW VXIILFLHQWO\ KLJK FRQFHQWUDWLRQV LQ RYHUZLQWHULQJ PRQDUFKV WR HIIHFWLYHO\ GHWHU YHUWHEUDWH SUHGDWRUV

PAGE 170

&+$37(5 %(+$9,25$/ $1' (&2/2*,&$/ ,17(5$&7,216 2) )25$*,1* 0(/$127,6 :,7+ 021$5&+ %877(5)/,(6 $V GHPRQVWUDWHG LQ &KDSWHU A PHODQRWLV QHHG QRW FOLPE XS WUHHV RU ORZ YHJHWDWLRQ WR IHHG RQ PRQDUFKV 0DQ\ OLYH EXWWHUIOLHV IDOO WR WKH JURXQG EHFDXVH VWRUPV DQG ELUGV GLVORGJH WKHP DQG PDQ\ GHDG RQHV DOVR DFFXPXODWH RQ WKH JURXQG DV D UHVXOW RI VWDUYDWLRQ GHVLFFDWLRQ DQG LQFRPSOHWH SUHGDWLRQ E\ ELUGV 'XULQJ VWRUPV KXQGUHGV RI WKRXVDQGV PRUH IDOO WR WKH JURXQG &DOYHUW HW DO f 2ZLQJ WR ORZ WHPSHUDWXUHV PDQ\ RI WKH OLYH PRQDUFKV RQ WKH JURXQG LQ ODWH DIWHUQRRQ FDQQRW IO\ LQVWHDG WKH\ FUDZO XS YHJHWDWLRQ RU UHPDLQ LPPRELOL]HG 'HDG PRQDUFKV GHFD\ VORZO\ LQ WKH FROG FOLPDWH -DQXDU\ )HEUXDU\ WHPSHUDWXUH UDQJH WR r& &DOYHUW DQG %URZHU f DQG DFFXPXODWH RQ WKH JURXQG RYHU WKH RYHUZLQWHULQJ SHULRG ,Q WKH DEVHQFH RI SUHFLSLWDWLRQ WKH\ GHVLFFDWH ZLWKLQ K 7KXV WKH JURXQG DQG ORZ YHJHWDWLRQ DUH FRYHUHG ZLWK EXWWHUIOLHV LQ YDU\LQJ VWDWHV RI GHVLFFDWLRQ 7R EHWWHU XQGHUVWDQG KRZ PHODQRWLV IRUDJH RQ RYHUZLQWHULQJ PRQDUFKV H[DPLQHG WKH IROORZLQJ TXHVWLRQV f +RZ PDQ\ PRQDUFKV FDQ DGXOW PLFH NLOO SHU QLJKW" f +RZ GRHV D PRQDUFKnV DFFHVVLELOLW\ LH KHLJKW DERYH JURXQGf LQIOXHQFH LWV ULVN WR PRXVH DWWDFN" f 6LQFH PDQ\ ZHW K\GUDWHGf DQG GU\ GHVLFFDWHGf PRQDUFKV DFFXPXODWH RQ WKH JURXQG DQG LQ ORZ YHJHWDWLRQ GR PLFH VHDUFK VHOHFWLYHO\ IRU ZHW

PAGE 171

PRQDUFKV DQG LI VR ZK\" f ,I RQO\ GU\ PRQDUFKV DUH RQ WKH JURXQG GR PLFH EHJLQ VHDUFKLQJ LQ ORZ YHJHWDWLRQ IRU ZHW RQHV" 0HWKRGV 7UDSSLQJ DQG 0DLQWHQDQFH FROOHFWHG PLFH DQG FRQGXFWHG WKH IHHGLQJ H[SHULPHQWV DW OHDVW P DZD\ IURP WKH PRQDUFK FRORQ\ RQ 6LHUUD &KLQFXD GXULQJ -DQXDU\ 0LFH ZHUH FDSWXUHG LQ 6KHUPDQ OLYH WUDSV [ [ FPf EDLWHG ZLWK UROOHG RDWV DQG SURYLGHG ZLWK D FRPSUHVVHG FRWWRQ EDOO IRU QHVWLQJ PDWHULDO 2QO\ UHSURGXFWLYHO\ LQDFWLYH DGXOWV QRQODFWDWLQJ RU QRQn SUHJQDQW IHPDOHV DQG QRQVFURWDO PDOHVf ZHUH XVHG LQ WKH H[SHULPHQWV ,QGLYLGXDO PLFH ZHUH XVHG RQO\ RQFH 7KH PLFH ZHUH KRXVHG LQGLYLGXDOO\ LQ ZLUH PHVK FDJHV DERXW FP KLJK [ FP LQ GLDPHWHUf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

PAGE 172

LQGLYLGXDOO\ LQ WKHLU FDJHV IRU GD\V ZLWK PRXVH FKRZ DQG ZDWHU DG OLELWXP SULRU WR WHVWLQJ 7KUHH PDOH DQG IHPDOH DGXOW PHODQRWLV ZHUH HDFK RIIHUHG OLYH PRQDUFKV SHU QLJKW IRU FRQVHFXWLYH QLJKWV 1R RWKHU IRRG VRXUFH ZDV DYDLODEOH GXULQJ WHVWLQJ (DFK PRXVHnV ZHLJKW ZDV WDNHQ DW WKH EHJLQQLQJ DQG HQG RI WKH H[SHULPHQW 7KH PRQDUFKV ZHUH SODFHG LQ HDFK PRXVHnV FDJH DW K DQG OHIW WKHUH XQWLO K WKH IROORZLQJ PRUQLQJ (DFK PRXVH ZDV WKHQ UHPRYHG IURP LWV FDJH WR WDOO\ PRQDUFK PRUWDOLW\ WKHQ UHSODFHG DQG IRRG GHSULYHG XQWLO K WKH VDPH GD\ ,Q SUHOLPLQDU\ H[SHULPHQWV PHODQRWLV IHG DOPRVW H[FOXVLYHO\ RQ DEGRPLQDO PDWHULDO WKXV H[FOXGHG WKRUDFLF IHHGLQJ GDWD 0RQDUFK PRUWDOLW\ ZDV JURXSHG LQWR FDWHJRULHV f QRW DWWDFNHG LH QR YLVLEOH GDPDJHf f VDPSOHG LH b RI DEGRPHQ HDWHQf f HDWHQ QRQVHOHFWLYHO\ LH b RI DEGRPHQ HDWHQ LQFOXGLQJ FXWLFOHf DQG f HDWHQ VHOHFWLYHO\ LH b RI FRQWHQWV HDWHQ FXWLFOH UHMHFWHGf ,Q WKLV DQG VXEVHTXHQW H[SHULPHQWV JURXSHG IHHGLQJ FDWHJRULHV WR LQWR QXPEHU DWWDFNHG DQG FDWHJRULHV WR LQWR QXPEHU HDWHQf DVVXPHG WKDW D PRXVH DWWDFN UHVXOWHG LQ GHDWK WR D PRQDUFK ([SHULPHQW $WWDFNV 8QFDJHG 0LFH RQ 0RQDUFKV 3HUFKHG DW 'LIIJUSQW +HLJKWV FRQGXFWHG WKLV H[SHULPHQW LQ D VWDQG RI R\DPHO ILU DERXW P IURP DQ RYHUZLQWHULQJ FRORQ\ $ P WUDQVHFW RI VWDNHV FP WDOO DQG DERXW FP GLDPHWHU P LQWHUVWDNH LQWHUYDOf ZDV VHW 7KUHH ZDGV RI PDVNLQJ WDSH ZHUH VWXFN WR HDFK VWDNH DW DQG FP DERYH WKH JURXQG DQG RQH PRQDUFK ZDV WKHQ VWXFN WR HDFK ZDG RI WDSH ZLWK D FRGH QXPEHU ZULWWHQ RQ ERWK IRUHZLQJV ZLWK LQGHOLEOH LQN

PAGE 173

$QRWKHU PRQDUFK ZDV SXW RQ WKH JURXQG DW WKH VWDNHnV EDVH (DFK PRQDUFK ZDV LPPRELOL]HG E\ VTXHH]LQJ LWV WKRUDFLF PXVFOHV 7KH PLFH FRXOG HDVLO\ UHPRYH PRQDUFKV IURP WKH WDSH EXW PRQDUFKV FRXOG QRW UHPRYH WKHPVHOYHV 0RQDUFKV ZHUH SODFHG RQ WKH WUDQVHFW EHWZHHQ WR K RQ VHSDUDWH QLJKWV EHWZHHQ -DQXDU\ DQG )HEUXDU\ 7R SUHYHQW ELUGV IURP DWWDFNLQJ WKH PRQDUFKV WDOOLHG WKH QXPEHU DWWDFNHG E\ PLFH DW VXQULVH 7ZHQW\RQH OLYH WUDSV ZHUH VHW DORQJ WKH WUDQVHFW SHU VWDNHf GXULQJ WKH QLJKWV RI DQG )HEUXDU\ WR DVFHUWDLQ ZKLFK VSHFLHV RI PLFH ZHUH SUHVHQW ([SHULPHQW ,QIOXHQFH RI 0RQDUFK $FFHVVLELOLW\ RQ 0RXVH $WWDFN 7R H[SORUH WKLV UHODWLRQVKLS IXUWKHU EXW XQGHU PRUH FRQWUROOHG FRQGLWLRQV H[SRVHG LQGLYLGXDO FDSWLYH PLFH WR RI YHUWLFDO GLVWULEXWLRQ SDWWHUQV RI PRQDUFKV XQLIRUP PDMRULW\ KLJK DQG PDMRULW\ ORZf 7KUHH GLIIHUHQW GLVWULEXWLRQ SDWWHUQV ZHUH XVHG EHFDXVH LQ FRORQLHV WKH UHODWLYH QXPEHUV RI PRQDUFKV RQ WKH JURXQG DQG LQ ORZ YHJHWDWLRQ YDULHV LQ WLPH DQG VSDFH 3UHOLPLQDU\ REVHUYDWLRQV LQGLFDWHG WKDW PLFH WHQG WR UHMHFW WKH ELWWHU DQG &*ODGHQ FXWLFOH RI JURXQGHG PRQDUFKV DQG HDW LQWHUQDO WLVVXHV WKXV H[SHFWHG WKHP WR IHHG VHOHFWLYHO\ RQ JURXQGHG PRQDUFKV EXW RZLQJ WR WKH SUHVXPHG GLIILFXOW\ RI IHHGLQJ LQ WKLV PDQQHU ZKLOH JUDVSLQJ D VWDNH H[SHFWHG PLFH WR FDUU\ PRQDUFKV RQ VWDNHV WR WKH JURXQG EHIRUH IHHGLQJ RQ WKHP 7ULDOV ZHUH FRQGXFWHG LQ WKH FHQWHU RI D P [ FP KLJK VKHHW PHWDO HQFORVXUH $ WDUSDXOLQ ZDV SODFHG FP DERYH LW DQG WKH IORRU ZDV EDUH SDFNHG VRLO 7HQ HDFK RI DQG FP KLJK VWDNHV

PAGE 174

FP GLDPHWHU ZHUH VHW LQ D [ DUUD\ ZLWK FP VSDFLQJ DQG ZHUH VXQN LQ WKH VRLO VR DV QRW WR ZREEOH ZKHQ PLFH FOLPEHG RQ WKHP 7KH VSDFLQJ ZDV VXIILFLHQW WR SUHYHQW PLFH IURP MXPSLQJ EHWZHHQ WKH VWDNHV 0RQDUFKV ZHUH VWXFN WR WKH VWDNHV DV LQ ([SHULPHQW DQG DOVR SODFHG RQ WKH JURXQG FP IURP VXUURXQGLQJ VWDNHV )RU HDFK WULDO PRQDUFKV ZHUH GLVWULEXWHG YHUWLFDOO\ LQ RI SDWWHUQV $ % DQG &f ,Q $ PRQDUFK ZDV RQ HDFK VWDNH DQG RQ WKH JURXQG LQ % ZHUH RQ WKH FP VWDNHV SHU VWDNHf DQG WKH RWKHU RQ WKH DQG FP VWDNHV SHU VWDNHf DQG LQ & ZHUH RQ WKH JURXQG DQG WKH RWKHU RQ WKH DQG FP VWDNHV SHU VWDNHf (DFK PRXVH ZDV SODFHG LQ WKH HQFORVXUH DW ZLWK FKRZ DQG ZDWHU G OLELWXP 0RQDUFKV ZHUH SODFHG RQ WKH VWDNHV EHWZHHQ WR K 0RUWDOLW\ ZDV WDOOLHG DV LQ ([SHULPHQW WKH IROORZLQJ PRUQLQJ DW K 7R GHWHUPLQH ZKLFK SHUFKHG EXWWHUIOLHV ZHUH HDWHQ ZKLOH VWXFN WR WKH WDSH DOVR QRWHG ZKHUH HDFK SHUFKHG PRQDUFK ZDV ORFDWHG LH RQ JURXQG RU RQ SHUFKf 6L[ PLFH ZHUH H[SRVHG WR SDWWHUQ $ WR % DQG WR & ([SHULPHQW ,QIOXHQFH RI 0RQDUFK 'HVLFFDWLRQ RQ 0RXVH $WWDFN RIIHUHG FDSWLYH PLFH HTXDO QXPEHUV RI GU\ DQG ZHW IUHVKO\ NLOOHGf PRQDUFKV H[SHFWLQJ WKHP WR SUHIHU WKH ODWWHU DQG WR HDW PDQ\ RI WKHP VHOHFWLYHO\ 0LFH ZHUH WHVWHG LQ WKH VDPH DSSDUDWXV DV GHVFULEHG LQ ([SHULPHQW 7ZR GD\V SULRU WR D IHHGLQJ WULDO WKH WKRUDFHV RI PRQDUFKV ZHUH ILUPO\ VTXHH]HG DQG OHIW LQ D FOHDULQJ IRU KRXUV WR GHVLFFDWH )LYH KRXUV EHIRUH D IHHGLQJ WULDO RWKHU PRQDUFKV ZHUH NLOOHG LQ WKH VDPH PDQQHU EXW SUHYHQWHG IURP GHVLFFDWLQJ

PAGE 175

7KH FDWHJRULHV RI PRQDUFKV ZHUH PDUNHG GLIIHUHQWO\ E\ SODFLQJ D UHG RU JROG VSRW DERXW PP LQ GLDPHWHUf RI QDLO SROLVK RQ HDFK PRQDUFKn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f ZHUH SXW LQ WKH HQFORVXUH LQGLYLGXDOO\ EHWZHHQ WR K DQG KDG DFFHVV WR FKRZ DQG ZDWHU DG OLELWXP $W K GU\ PRQDUFKV ZHUH GLVSHUVHG HYHQO\ RQ WKH JURXQG ZLWKLQ WKH [ VWDNH DUUD\ DQG ZHW PRQDUFKV ZHUH DWWDFKHG WR WKH DQG FP VWDNHV SHU VWDNHf $W K WKH IROORZLQJ PRUQLQJ PRQDUFK UHPDLQV ZHUH WDOOLHG

PAGE 176

6WDWLVWLFDO $QDO\VHV 7R H[DPLQH WKH SDWWHUQ RI DWWDFN E\ WKH ZLOG XQFDJHG PLFH LQ ([SHULPHQW XVHG WKH &RFKUDQ 4 WHVW =DU f 7KLV WHVW ZDV DSSURSULDWH EHFDXVH KDG UHODWHG VDPSOHV SHU VWDNH LH PRQDUFKV DW RU RQ HDFK VWDNHf DQG GLFKRWRPRXV GDWD WUHDWHG HDFK PRXVH DV D UDQGRPL]HG EORFN DQG DQDO\]HG WKH UHVXOWV IURP HDFK QLJKW VHSDUDWHO\ 7R GHWHUPLQH ZKHWKHU PLFH DWWDFNHG DFFHVVLEOH PRQDUFKV GLVSURSRUWLRQDWHO\ LQ ([SHULPHQWV DQG XVHG WKH RQHVDPSOH WDLOf WWHVW =DU f 7KH YDULDQFH IRU WKH REVHUYHG SURSRUWLRQ DWWDFNHG DW HDFK SHUFK KHLJKW ZDV HVWLPDWHG ZLWK WKH UDWLR IRUPXOD IRU FOXVWHU VDPSOHV &RFKUDQ f :LWK WKLV WHVW FRXOG GHWHUPLQH ZKHWKHU PLFH DWWDFNHG PRUH RU OHVV EXWWHUIOLHV WKDQ H[SHFWHG DW HDFK SHUFK KHLJKW 7R PLQLPL]H WKH SUREDELOLW\ RI FRPPLWWLQJ D 7\SH HUURU RZLQJ WR P\ XVH RI PXOWLSOH WWHVWV VHW WKH KLJKHVW VLJQLILFDQFH OHYHO DW 7R GHWHUPLQH ZKHWKHU PLFH IHG GLVSURSRUWLRQDWHO\ RQ ZHW PRQDUFKV LQ ([SHULPHQW FRQGXFWHG VHSDUDWH FKLVTXDUH WHVWV RQ HDFK PRXVH 5HVXOWV ([SHULPHQW ([WHQW RI 0RXVH 3UHGDWLRQ 7KH PHDQ QXPEHU RI PRQDUFKV DWWDFNHGQLJKWPRXVH ZDV 6( UDQJH WR f ZKHUHDV WKH PHDQ QXPEHU HDWHQQLJKWPRXVH ZDV 6( UDQJH WR 7DEOH f 7KHUH ZDV QR DSSDUHQW FKDQJH LQ WKH QXPEHU DWWDFNHG DQG HDWHQQLJKWPRXVH RU LQ WKH ZHLJKWV RI WKH PLFH RYHU WKH GD\V 0HDQ QLJKWO\ IHHGLQJ SDWWHUQV ZHUH WKXV LQGHSHQGHQW RI SUHYLRXV PRQDUFK FRQVXPSWLRQ RYHU D GD\ SHULRG

PAGE 177

7DEOH 0HDQ QXPEHU s 6(f RI ZHW PRQDUFK EXWWHUIOLHV DWWDFNHG DQG HDWHQ E\ FDSWLYH 3 PHODQRWLV SHU QLJKW RYHU FRQVHFXWLYH QLJKWV 1R DWWDFNHG 1LJKW *UDQG 0HDQ sf sf sf sf sf sf sf sf sf sf sf sf sf sf 1R HDWHQ

PAGE 178

([SHULPHQW $WWDFNV EY :LOG 8QFDJHG 0LFH RQ 0RQDUFKV 3HUFKHG DW 'LIIHUHQW +HLJKWV 2XW RI WUDS QLJKWV WUDSV [ QLJKWVf WKH VDPH PLFH 3 PHODQRWLVf ZHUH WUDSSHG DOO QLJKWV 7KXV LW VHHPV OLNHO\ WKDW WKLV VSHFLHV ZDV UHVSRQVLEOH IRU WKH PRQDUFKV DWWDFNHG $OWKRXJK QR FRQFOXVLRQV FDQ EH GUDZQ UHJDUGLQJ WKH QXPEHU RI PRQDUFKV DWWDFNHG E\ HDFK PRXVH FOHDUO\ WKH WRWDO QXPEHU DWWDFNHG LQFUHDVHG RYHU WKH GD\V RI WKLV H[SHULPHQW 7DEOH f 0RUHRYHU RYHU DOO QLJKWV PLFH IHG PRVW KHDYLO\ RQ WKRVH PRQDUFKV RQ WKH JURXQG DQG DW FP &RFKUDQ 4 IRU -DQXDU\ GI RQ WKLV DQG DOO RWKHU QLJKWVf 3 IRU )HEUXDU\ r 3 DQG IRU DQG )HEUXDU\ WR 3 ([SHULPHQW ,QIOXHQFH RI 0RQDUFK $FFHVVLELOLW\ RQ 0RXVH $WWDFN )RU SDWWHUQV $ % DQG & PLFH FRQVLVWHQWO\ DWWDFNHG PRUH PRQDUFKV WKDQ H[SHFWHG RQ WKH JURXQG DQG IHZHU DW KLJKHU SHUFKHV 7DEOH f ,Q 7DEOH SURYLGH D EUHDNGRZQ RI WKH DEGRPLQDO IHHGLQJ SDWWHUQV E\ PLFH GDWD IRU LQGLYLGXDO PLFH ZHUH SRROHGf 0LFH H[SRVHG WR SDWWHUQV $ DQG & IHG VHOHFWLYHO\ RQ PDQ\ RI WKH PRQDUFKV RQ WKH JURXQG EXW UDUHO\ GLG VR ZLWK WKH RQHV RQ VWDNHV &RXQWHU WR SUHGLFWLRQ WKH PDMRULW\ RI SHUFKHG PRQDUFKV HDWHQ E\ PLFH ZHUH HDWHQ RQ WKH VWDNHV IRU SDWWHUQ $ LW ZDV b IRU % LW ZDV b DQG IRU & LW ZDV b FRXOG QRW GHWHUPLQH KRZHYHU ZKHWKHU WKH SUHGDWHG SHUFKHG PRQDUFKV IRXQG RQ WKH JURXQG KDG EHHQ HDWHQ E\ PLFH RQ WKH VWDNHV DQG VXEVHTXHQWO\ GURSSHG RU FDUULHG WR WKH JURXQG DQG WKHQ HDWHQ

PAGE 179

7DEOH 1XPEHU RI ZHW PRQDUFK EXWWHUIOLHV DWWDFNHG E\ ZLOG XQFDJHG PLFH DW GLIIHUHQW KHLJKWV $ P WUDQVHFW RI VWDNHV ZDV UXQ ZLWK PRQDUFKV SHU VWDNH DQG HDFK PRQDUFK VHW DW RI KHLJKWV +HLJKW FPf 1R PRQDUFKV RIIHUHG SHU 1LJKW 1R RI PRQDUFKV DWWDFNHG -DQ )HE )HE )HE )HE 7RWDO 7RWDO

PAGE 180

7DEOH 2EVHUYHG PHDQ s6'f DQG H[SHFWHG SURSRUWLRQV RI ZHW PRQDUFKV DWWDFNHG E\ FDSWLYH PLFH DW GLIIHUHQW KHLJKWV SURSRUWLRQV DUH GHULYHG IURP WKH GDWD LQ 7DEOH f $ WULDO FRQVLVWHG RI RIIHULQJ D PRXVH PRQDUFKV GLVWULEXWHG DPRQJ SHUFKHV LQ RI SRVVLEOH SDWWHUQV $% RU &f 6L[ PLFH ZHUH WHVWHG ZLWK SDWWHUQ $ ZLWK % DQG ZLWK & 7KH REVHUYHG DQG H[SHFWHG SURSRUWLRQ RI PRQDUFKV DWWDFNHG DUH FRPSDUHG ZLWK WKH RQHVDPSOH WDLOf WWHVW r 3 rr 3 'LVWULEXWLRQ 3DWWHUQ RI 0RQDUFKV +HLJKW FPf 3URSRUWLRQ 2EVHUYHG $WWDFNHG ([SHFWHG WYDOXH GI $ 8QLIRUP sf sf r sf 2Q *URXQG sf r % 0DMRULW\ KLJK sf sf sf r & 0DMRULW\ ORZ sf rr sf rr 2Q *URXQG sf r

PAGE 181

7DEOH $EGRPLQDO IHHGLQJ SDWWHUQV E\ FDSWLYH PLFH RQ ZHW PRQDUFKV DW GLIIHUHQW KHLJKWV )HHGLQJ FDWHJRULHV DUH GHVFULEHG LQ WKH WH[W $ WULDO FRQVLVWHG RI RIIHULQJ D PRXVH PRQDUFKV GLVWULEXWHG DPRQJ SHUFKHV LQ RI SRVVLEOH SDWWHUQV $ % RU &f 6L[ PLFH ZHUH WHVWHG ZLWK SDWWHUQ $ ZLWK % DQG ZLWK & 'DWD IRU LQGLYLGXDO PLFH ZLWKLQ HDFK SDWWHUQ ZHUH SRROHG 'LVWULEXWLRQ SDWWHUQ RI PRQDUFKV +HLJKW FPf $EGRPLQDO IHHGLQJ FDWHJRULHV 7RWDO 2IIHUHG 1RW DWWDFNHG (DWHQ QRQ 6DPSOHG VHOHFWLYHO\ (DWHQ VHOHFWLYHO\ $ 8QLIRUP 2Q JURXQG % 0DMRULW\ KLJK & 0DMRULW\ ORZ 2Q JURXQG

PAGE 182

([SHULPHQW ,QIOXHQFH RI 0RQDUFK 'HVLFFDWLRQ RQ 0RXVH $WWDFN 7KH DEGRPLQDO IHHGLQJ SDWWHUQV IRU DOO PLFH SRROHG DUH SUHVHQWHG LQ 7DEOH &KLVTXDUH FRPSDULVRQV SHUIRUPHG RQ HDFK PRXVH VHSDUDWHO\ ZHUH DOO KLJKO\ VLJQLILFDQW FKLVTXDUH YDOXHV UDQJHG IURP GI DOO 3YDOXHV f GHPRQVWUDWLQJ D VWURQJ DVVRFLDWLRQ EHWZHHQ WKH GHJUHH RI GHVLFFDWLRQ DQG WKH IHHGLQJ SDWWHUQV E\ PLFH 0LFH VWURQJO\ SUHIHUUHG ZHW PRQDUFKV DQG DWH WKH PDMRULW\ RI WKHP VHOHFWLYHO\ ([SHULPHQW 5LVN WR /LYH 0RQDUFKV LQ /RZ 9HJHWDWLRQ ZLWK RQO\ 'U\ 0RQDUFKV RQ WKH *URXQG 0LFH DWWDFNHG VLJQLILFDQWO\ IHZHU GU\ PRQDUFKV RQ WKH JURXQG DQG VOLJKWO\ EXW QRW VLJQLILFDQWO\f PRUH ZHW PRQDUFKV RQ WKH VWDNHV WKDQ H[SHFWHG 7DEOH f 0RUHRYHU RQO\ ZHW PRQDUFKV ZHUH IHG XSRQ VHOHFWLYHO\ 7DEOH GDWD IRU LQGLYLGXDO PLFH ZHUH SRROHGf 7KXV WDNHQ WRJHWKHU ZLWK WKH UHVXOWV RI ([SHULPHQWV DQG LW DSSHDUV WKDW ZHW PRQDUFKV SHUFKHG LQ ORZO\LQJ YHJHWDWLRQ KDYH WKH JUHDWHVW ULVN WR PRXVH DWWDFN ZKHQ RQO\ GU\ PRQDUFKV DUH RQ WKH JURXQG 3LV FLLVDLRQ (VWLPDWH RI WKH 1XPEHUV RI 0RQDUFKV .LOOHG E\ 0LFH %\ OLPLWLQJ DGXOW PLFH WR D GLHW RI OLYH PRQDUFKV IRU FRQVHFXWLYH QLJKWV GHWHUPLQHG WKDW LQGLYLGXDOV DWWDFN RQ DYHUDJH ZHW PRQDUFKV SHU QLJKW 0RUHRYHU UHVXOWV IURP &KDSWHU VXJJHVW WKDW f WKH GLHW RI eBr PHODQRWLV LQVLGH FRORQLHV FRQVLVWV HQWLUHO\ RI PRQDUFKV DQG f PHODQRWLV SRSXODWLRQV LQVLGH PRQDUFK FRORQLHV UDQJH EHWZHHQ WR DGXOW PLFH SHU KHFWDUH 7KHUHIRUH LI ZH DVVXPH WKDW IUHH

PAGE 183

7DEOH $EGRPLQDO IHHGLQJ SDWWHUQV GLUHFWHG DW ZHW DQG GU\ PRQDUFK EXWWHUIOLHV E\ LQGLYLGXDO FDSWLYH PHODQRWLV Q f 'DWD IRU LQGLYLGXDO PLFH ZHUH SRROHG $EGRPLQDO IHHGLQJ FDWHJRU\ 6WDWH RI 1RW (DWHQ 1RQHDWHQ 7RWDO PRQDUFK DWWDFNHG VHOHFWLYHO\ VHOHFWLYHO\ 2IIHUHG :HW 'U\

PAGE 184

7DEOH 2EVHUYHG PHDQ s6'f DQG H[SHFWHG SURSRUWLRQ RI ZHW DQG GU\ PRQDUFKV DWWDFNHG E\ PLFH DW GLIIHUHQW KHLJKWV SURSRUWLRQV DUH GHULYHG IURP WKH GDWD LQ 7DEOH f )RU HDFK WULDO PRQDUFKV ZHUH GLYLGHG HYHQO\ DPRQJ WHQ FP VWDNHV DQG WHQ FP VWDNHV DQG GU\ PRQDUFKV ZHUH SODFHG RQ WKH JURXQG 7KH REVHUYHG DQG H[SHFWHG SURSRUWLRQ RI PRQDUFKV DWWDFNHG DUH FRPSDUHG ZLWK WKH RQHVDPSOH WDLOf WWHVW rr 3 +HLJKW FPf 6WDWH RI 0RQDUFKV 3URSRUWLRQ 2EVHUYHG DWWDFNHG ([SHFWHG WYDOXH GI :HW sf :HW sf 2Q *URXQG 'U\ sf rr

PAGE 185

7DEOH $EGRPLQDO IHHGLQJ SDWWHUQV E\ FDSWLYH PLFH RQ ZHW DQG GU\ PRQDUFK EXWWHUIOLHV )RU HDFK WULDO ZHW PRQDUFKV ZHUH GLYLGHG HYHQO\ DPRQJ WHQ FP VWDNHV DQG WHQ FP VWDNHV DQG GU\ PRQDUFKV ZHUH SODFHG RQ WKH JURXQG 'DWD IRU LQGLYLGXDO PLFH ZHUH SRROHG $EGRPLQDO IHHGLQJ FDWHJRULHV +HLJKW FPf 6WDWH RI PRQDUFKV 1RW DWWDFNHG (DWHQ QRQ VHOHFWLYHO\ (DWHQ VHOHFWLYHO\ 7RWDO RIIHUHG :HW :HW 2Q *URXQG 'U\

PAGE 186

UDQJLQJ DGXOW eBA PHODQRWLV NLOO DQ DYHUDJH RI PRQDUFKV SHU QLJKW D SRSXODWLRQ RI WKLV VL]H FRXOG NLOO EHWZHHQ WR OLYH PRQDUFKV SHU QLJKW 7RWDO PRUWDOLW\ LQ D KD FRORQ\ IRU WKH GD\ RYHUZLQWHULQJ VHDVRQ ZRXOG EH DSSUR[LPDWHO\ WR EXWWHUIOLHV RU XVLQJ %URZHU DQG &DOYHUWnV f GHQVLW\ HVWLPDWH RI PLOOLRQ EXWWHUIOLHV SHU KD WR b RI WKH FRORQ\ 7KLV PRUWDOLW\ HVWLPDWH SUREDEO\ XQGHUHVWLPDWHV WRWDO PRUWDOLW\ EHFDXVH WKH QLJKWO\ SUHGDWLRQ UDWHV ZHUH GHWHUPLQHG ZLWK QRQUHSURGXFWLYH FDSWLYH LQGLYLGXDOV 7KH KLJK HQHUJHWLF FRVWV RI ODFWDWLRQ LQ IHPDOHV DQG ORFRPRWLRQ LQ IUHHUDQJLQJ LQGLYLGXDOV 0LOODU 6DGOLHU HW DO 6WHEELQV 3HUULJR f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f 7KHVH UHVXOWV GR QRW QHFHVVDULO\ FRQWUDGLFW PLQH JLYHQ WKDW P\ DFFHVVLELOLW\ FKDOOHQJHV ZHUH

PAGE 187

OLPLWHG )RU LQVWDQFH WDOOHU VWDNHV LQ ([SHULPHQW PLJKW KDYH FDXVHG VZLWFKHV VLPLODU WR WKRVH IRXQG E\ 0RHUPRQG DQG 'HQVORZ 1RQH RI WKH GU\ PRQDUFKV DQG RQO\ D IHZ RI WKH ZHW PRQDUFKV RQ VWDNHV ZHUH HDWHQ VHOHFWLYHO\ ,Q FRQWUDVW PDQ\ RI WKH ZHW PRQDUFKV RQ WKH JURXQG ZHUH HDWHQ LQ WKLV PDQQHU VXJJHVW WKDW PLFH ZHUH XQDEOH WR IHHG VHOHFWLYHO\ RQ GU\ PRQDUFKV EHFDXVH WKH DEGRPLQDO FXWLFOH ERQGV WLJKWO\ WR WKH LQWHUQDO WLVVXHV GXULQJ WKH GHVLFFDWLRQ SURFHVV %URZHU HW DO f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f 0RQDUFK 'HIHQVHV $JDLQVW 0RXVH $WWDFN 7KH PRQDUFKnV VWURQJ WHQGHQF\ WR FUDZO XS YHJHWDWLRQ KDV EHHQ LQWHUSUHWHG DV DQ DGDSWLYH UHVSRQVH IRU DYRLGLQJ IUHH]LQJ JURXQG WHPSHUDWXUHV &DOYHUW DQG %URZHU &DOYHUW DQG &RKHQ f 0\ UHVXOWV LQGLFDWH WKDW WKLV FUDZOLQJ EHKDYLRU LV DOVR DQ HIIHFWLYH

PAGE 188

GHIHQVH DJDLQVW PRXVH DWWDFN FRQFOXGH WKDW RYHU HYROXWLRQDU\ WLPH WKH FRPELQHG PRUWDOLW\ ULVNV RI QRFWXUQDO PRXVH SUHGDWLRQ DQG IUHH]LQJ JURXQG WHPSHUDWXUHV DUH OLNHO\ WR VHOHFW IRU WKH PRQDUFK EXWWHUIOLHVn WHQGHQF\ WR FUDZO XS YHJHWDWLRQ 0RUHRYHU P\ UHVXOWV XQGHUOLQH WKH LPSRUWDQFH RI SURWHFWLQJ WKH XQGHUVWRU\ YHJHWDWLRQ LQ PRQDUFK FRORQLHV WR SURYLGH PRQDUFKV ZLWK DQ HVFDSH UHIXJH DJDLQVW WKHVH PRUWDOLW\ IDFWRUV 0RQDUFKV PD\ SRVVHVV RWKHU GHIHQVHV WKDW DFW LQ FRQMXQFWLRQ ZLWK WKH FUDZOLQJ EHKDYLRU WR UHGXFH WKHLU ULVN WR PRXVH DWWDFN 2QH LV D YDULDWLRQ RQ WKH VHOILVKKHUG HIIHFW +DPLOWRQ f :KHQ D PRXVH HQFRXQWHUV D FOXVWHU RI PRQDUFKV SHUFKHG LQ ORZ YHJHWDWLRQ LW PD\ IHHG PRVW KHDYLO\ RQ WKRVH RFFXS\LQJ WKH ORZ DQG RXWHU SRVLWLRQV LQ WKH FOXVWHU LH WKH PRVW DFFHVVLEOH RQHVf 6XFK SUHGDWLRQ SUHVVXUH ZRXOG IDYRU WKRVH LQGLYLGXDOV EHVW DEOH WR VHFXUH D SRVLWLRQ LQ WKH FHQWUDO RU XSSHU UHJLRQV RI FOXVWHUV ,Q IDFW PRQDUFKV LQ ORZ YHJHWDWLRQ FRPPRQO\ MRFNLHG IRU WKH KLJKHVW SRVLWLRQ LQ FOXVWHUV *OHQGLQQLQJ SHUV REVHUYf 7KH\ PD\ DOVR GHULYH SURWHFWLRQ IURP WKHLU VWRUHG GHIHQVLYH FRPSRXQGV 5HVXOWV LQ &KDSWHUV DQG LQGLFDWH WKDW PHODQRWLV DYRLG PRQDUFKV ZLWK KLJK &* FRQFHQWUDWLRQV IUHTXHQWO\ DVVHVVLQJ &* OHYHOV QRQGHVWUXFWLYHO\ E\ HPSOR\LQJ WDVWH

PAGE 189

&+$37(5 *(1(5$/ &21&/86,216 ,PSDFW RI 0RXVH 3UHGDWLRQ RQ 2YHUZLQWHULQJ 0RQDUFK %XWWHUIOLHV 7KH RYHUZLQWHULQJ PRQDUFK DJJUHJDWLRQV LQ 0H[LFR SURYLGHG DQ RSSRUWXQLW\ WR LQYHVWLJDWH WKH HIILFDF\ RI D ZHOO NQRZQ FKHPLFDO GHIHQVH V\VWHP DJDLQVW D VXLWH RI VPDOO PDPPDOLDQ SUHGDWRUV 5DWKHU WKDQ EHLQJ DEVROXWH LW OLPLWHG WKH QXPEHU RI PRXVH SUHGDWRUV IURP WR 1HYHUWKHOHVV EHFDXVH VR PDQ\ PHODQRWLV LPPLJUDWH WR WKH FRORQLHV SRSXODWLRQV RI PHODQRWLV PD\ NLOO WR b RI D KD FRORQ\ RYHU WKH HQWLUH RYHUZLQWHULQJ VHDVRQ &KDSWHU f 6LPLODUO\ HYHQ WKRXJK WKH PRQDUFKnV GHIHQVLYH FRPSRXQGV DOVR PD\ UHGXFH WKH QXPEHU RI VSHFLHV RI SRWHQWLDO DYLDQ SUHGDWRUV IURP WR )LQN HW DO $UHOODQR HW DO f WKHVH VSHFLHV DSSHDU WR NLOO DERXW b RI D KD FRORQ\ %URZHU DQG &DOYHUW f 7KXV VSHFLHV RI PRXVH DQG VSHFLHV RI ELUGV PD\ WRJHWKHU NLOO WR b RI D KD FRORQ\ 7KH QXPEHU RI PRQDUFKV NLOOHG E\ PLFH DQG ELUGV VHHPV KLJK ZKHQ FRPSDUHG WR WKH KXJH FRORQLHV RI DSRVHPDWLF DQG FKHPLFDOO\GHIHQGHG ODG\ EHHWOHV LQ WKH 6LHUUD 1HYDGDV ZKLFK DSSDUHQWO\ H[SHULHQFH QR SUHGDWLRQ ZKDWVRHYHU +DJHQ +DJHQ EXW VHH &KDSPDQ HW DO f UHJDUGLQJ SUHGDWLRQ RQ D GLIIHUHQW VSHFLHV LQ WKH 5RFNLHVf +RZHYHU ODUJH DJJUHJDWLRQV RI SDODWDEOH LQVHFWV DSSHDU WR EH IHG XSRQ PXFK PRUH H[WHQVLYHO\ DQG E\ D PXFK JUHDWHU YDULHW\ RI YHUWHEUDWH

PAGE 190

SUHGDWRUV WKDQ DUH WKH PRQDUFK FRORQLHV ([DPSOHV LQFOXGH RSSRUWXQLVWLF SUHGDWLRQ XSRQ FLFDGDV /OR\G DQG '\EDV 6WUHKO DQG :KLWH 6WHZDUG HW DO +DKXV DQG 6PLWK LQ SUHVVf DODWH DQWV 'LDO DQG 9DXJKDQ f J\SV\ PRWK ODUYDH *HUDUGL DQG *ULPP f ODUFK VDZIO\V %XFNQHU DQG 7XUQRFN f SLQH VDZIO\V +ROOLQJ +DQVNL DQG 3DUYLDLQHQ f VSUXFH EXGZRUPV 0RUULV HW DO f EXVK FULFNHWV %DLOH\ DQG 0F&UDH f DQG GHVHUW ORFXVWV 6PLWK DQG 3RSRY +XGOHVWRQ $VKDOO DQG (OOLV :DORII f :K\ OD 3 PHODQRWLV WKH 2QO\ 0RXVH 6SHFLHV WKDW )HHGV RQ WKH 0RQDUFKV" ,Q &KDSWHU WHVWHG WKH K\SRWKHVHV WKDW WKH RWKHU VSHFLHV RI PLFH GLG QRW HDW PRQDUFKV EHFDXVH f WKH PLFURKDELWDW FKDUDFWHULVWLFV HJ XQGHUVWRU\ YHJHWDWLRQf RI RYHUZLQWHULQJ DUHDV VXLWHG RQO\ 3 PHODQRWLV DQG f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eB} PHODQRWLV ZHUH DYHUVH WR PRQDUFKV DV IRRG FRQFOXGH WKDW A PHODQRWLV GLVSOD\HG IHHGLQJ DWWULEXWHV RU PHFKDQLVPV ZKLFK WRJHWKHU HQDEOHG LW WR IHHG H[WHQVLYHO\ RQ PRQDUFKV f LW DWWDFNHG DQG FRQVXPHG OLYH PRQDUFKV UHDGLO\ f LW OHDUQHG UDSLGO\ KRZ WR UHMHFW WKH &* DQG 3$ODGHQ

PAGE 191

FXWLFXODU PDWHULDO DQG IHHG VHOHFWLYHO\ RQ WKH LQWHUQDO FRQWHQWV RI DEGRPHQV DQG f LW ZDV UHODWLYHO\ LQVHQVLWLYH WR WKH WDVWH DQG WR[LF HIIHFWV RI &*V DQG 3$V (YHQ WKRXJK WKH IHHGLQJ H[SHULPHQWV LQ &KDSWHU GR QRW GHPRQVWUDWH GHILQLWLYHO\ WKDW eBr PHODQRWLV LV LQVHQVLWLYH WR WKH WR[LF HIIHFWV RI PRQRFURWDOLQH WKH GHPRJUDSKLF UHVXOWV LQ &KDSWHU VXJJHVW VWURQJO\ WKDW PHODQRWLV ZDV QRW GHELOLWDWHG E\ WKH PRQDUFKnV 3$V ,QGLYLGXDOV LQVLGH FRORQLHV ZHUH DEOH WR EUHHG H[WHQVLYHO\ DQG VXFFHVVIXOO\ ZHDQ ODUJH QXPEHUV RI \RXQJ ZKLOH VXEVLVWLQJ RQ D GLHW ZKLFK VWRPDFK FRQWHQW DQDO\VHV LQGLFDWHG ZDV QHDUO\ b PRQDUFK PDWHULDO IXUWKHU FRQFOXGH WKDW WKH RWKHU VSHFLHV RI PLFH DYRLGHG PRQDUFKV EHFDXVH QHLWKHU SRVVHVVHG DOO RI WKH IHHGLQJ DWWULEXWHV RU PHFKDQLVPV GHVFULEHG DERYH IRU eBA PHODQRWLV 0LFURWXV P VDOYXV ZRXOG QRW DWWDFN DFWLYH PRQDUFKV DQG ZKHQ RIIHUHG SDUDO\]HG DQG LPPRELOH PRQDUFKV GLG QRW OHDUQ WKH VHOHFWLYH IHHGLQJ EHKDYLRU DQG DSSHDUHG WR EH LQWROHUDQW WR WKH WDVWH RI WKH PRQDUFK &*V DQGRU 3$V 7KH WR[LF VHQVLWLYLW\ RI / %/ VDOYXV ZDV QRW H[DPLQHG (YHQ WKRXJK e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

PAGE 192

LWV FRPSOHWH DEVHQFH IURP WKH FRORQLHV VXJJHVW WKDW LW ZDV DYHUVH WR WKH PRQDUFKV DV ZHOO 7KDW WKHVH UHODWLYHO\ FORVHO\UHODWHG VSHFLHV UHVSRQGHG VR GLIIHUHQWO\ WR D VLQJOH VKDUHG FKHPLFDOO\ GHIHQGHG LQVHFW VSHFLHV XQGHUOLQHV WKH LPSRUWDQFH RI VWXG\LQJ DV ODUJH D SHUFHQWDJH RI WKH DVVHPEODJH RI QDWXUDO SUHGDWRUV DV SRVVLEOH ZKHQ HYDOXDWLQJ WKH HIIHFWLYHQHVV RI DQ LQVHFW VSHFLHVn FKHPLFDO GHIHQVH V\VWHP 6LPSO\ FKRRVLQJ WKH PRVW FRQYHQLHQW RU DEXQGDQW SUHGDWRUVf PD\ SURYLGH PLVOHDGLQJ UHVXOWV )RU H[DPSOH KDG RQO\ VWXGLHG %Br VXPLFKUDVWL DQG K\ORFHWHV ZRXOG KDYH FRQFOXGHG WKDW WKH PRQDUFKnV FKHPLFDO GHIHQVH V\VWHP HIIHFWLYHO\ GHWHUUHG DOO VSHFLHV RI PLFH LQ WKH RYHUZLQWHULQJ DUHD $UH 3RSXODWLRQV 4I 3 PHODQRWLV LQ WKH 2YHUZLQWHULQJ $UHDV 6SHFLILFDOO\ $GDSWHG IRU 2YHUFRPLQJ WKH 0RQDUFKnV 'HIHQVLYH &RPSRXQGV" 6HYHUDO OLQHV RI HYLGHQFH SRLQW WR WKH XQOLNHOLKRRG RI VXESRSXODWLRQV RI PHODQRWLV KDYLQJ HYROYHG VSHFLILF JXVWDWRU\ SKDUPDFRORJLFDO DQG EHKDYLRUDO DGDSWDWLRQV IRU GHDOLQJ ZLWK WKH PRQDUFKnV GHIHQVLYH FRPSRXQGV )LUVW LQGLYLGXDOV FROOHFWHG DERXW NLORPHWHUV WR WKH HDVW RI JULG )LJXUH ODf ZHUH MXVW DV LQVHQVLWLYH WR WKH WDVWH DQG WR[LF HIIHFWV RI GLJLWR[LQ D &*Vf DV ZHUH WKRVH FROOHFWHG DERXW P WR WKH VRXWK RI JULG LQ )LJXUH OD DQG E 6HFRQG WKHUH DUH QR DSSDUHQW JHRJUDSKLFDO EDUULHUV WKDW ZRXOG JHQHWLFDOO\ LVRODWH SRSXODWLRQV RI (AB PHODQRWLV RQ WKH ZHVW VLGH RI 6LHUUD &KLQFXD ZKLFK LV ZKHUH WKH PRQDUFKV RYHUZLQWHU &DOYHUW DQG %URZHU f IURP WKRVH RQ WKH HDVW VLGH RU RQ QHDUE\ PDVVLIV $QRQ WRSRJUDSKLF PDSf 7KLUG WKH VDPH SRSXODWLRQV RI PHODQRWLV RQ WKH ZHVW VLGH 6LHUUD &KLQFXD SUREDEO\ GR QRW HQFRXQWHU PRQDUFK FRORQLHV

PAGE 193

HDFK \HDU DV GLVWDQFHV EHWZHHQ FRORQ\ VLWHV RYHU VXFFHVVLYH \HDUV PD\ EH DV JUHDW DV NP &DOYHUW DQG %URZHU f ,W VHHPV PRUH OLNHO\ WKDW WKH FDSDFLW\ RI (A PHODQQI L V WR IHHG XSRQ PRQDUFKV UHIOHFWV SK\ORJHQHWLFDOO\ GHWHUPLQHG f WROHUDQFHV IRU FRSLQJ ZLWK D YDULHW\ RI ELWWHU DQGRU WR[LF IRRGV DQG f IOH[LEOH SUH\ KDQGOLQJ DELOLWLHV 0DUW\ %URZHU HW DO f 7KDW LV 3 PHODQRWLV PD\ KDYH H[DSWHG VHQVX *RXOG DQG 9UED f WKHVH DWWULEXWHV WR IHHGLQJ RQ PRQDUFKV 2WKHU SRVVLEOH H[DPSOHV RI SKDUPDFRORJLFDO DQG EHKDYLRUDO H[DSWDWLRQ LQFOXGH SUHGDWLRQ E\ JURVEHDNV DQG RULROHV RQ RYHUZLQWHULQJ PRQDUFKV %URZHU HW DO f DQG WKH DELOLW\ RI IUXJLYRURXV ELUGV WR LQJHVW DSRVHPDWLF LQVHFWV +HUUHUD f %HQHILWV DQG &RVWV RI %HLQJ 7ROHUDQW WR %LWWHU )RRGV 0\ UHVXOWV GHPRQVWUDWH WKDW E\ WROHUDWLQJ WKH WDVWH RI WKH PRQDUFKnV GHIHQVLYH FRPSRXQGV (A PHODQRWLV ZDV DEOH WR IHHG H[WHQVLYHO\ RQ RYHUZLQWHULQJ PRQDUFKV DQG LQLWLDWH KLJK OHYHOV RI UHSURGXFWLRQ ,Q FRQWUDVW IUHHUDQJLQJ (/ VXPLFKUDVWL DQG (A D K\ORFHWHV DYRLGHG WKH ELWWHUWDVWLQJ PRQDUFKV IHG RQ OHVV FDORULFDOO\ GHQVH IRRGV DQG GLG QRW UHSURGXFH ,I WKHVH WZR VSHFLHV RI PLFH KDG WROHUDWHG WKH ELWWHU WDVWH RI WKH PRQDUFKV WKH\ SUREDEO\ FRXOG KDYH IHG H[WHQVLYHO\ RQ WKH PRQDUFKV DQG LQLWLDWHG ZLQWHU UHSURGXFWLRQ 5HLWKURGRQWRP\V VXPLFKUDVWL DQG eBA K\ORFHWHV GLG QRW DSSHDU WR EH VHQVLWLYH WR WKH WR[LF HIIHFWV RI &*V 7KLV ILQGLQJ LOOXVWUDWHV D FRVW WR UHMHFWLQJ IRRGV EDVHG RQ WDVWH E\ DYRLGLQJ DQ HSKHPHUDO VXSHUDEXQGDQW DQG DSSDUHQWO\ KDUPOHVV IRRG UHVRXUFH LQGLYLGXDO (A K\ORFHWHV DQG ,/ VXPLFKUDVWL DSSHDU WR KDYH PLVVHG DQ RSSRUWXQLW\ WR

PAGE 194

VXEVWDQWLDOO\ LQFUHDVH WKHLU ILWQHVV *LYHQ WKDW WKH DYHUDJH OLIH VSDQ LV D IHZ PRQWKV DW EHVW IRU PRVW PLFH OLYLQJ LQ H[WUHPH HQYLURQPHQWV VXFK DV WKRVH IRXQG DW WKH RYHUZLQWHULQJ VLWHV LQ 0H[LFR %URQVRQ f PDQ\ RI WKH Df§ K\ORFHWHV DQG (r VXPLFKUDVWL SUREDEO\ GLH EHIRUH WKH RQVHW RI VSULQJ ZKHQ PRUH SDODWDEOH IRRG VRXUFHV EHFRPH DEXQGDQW 9 6£QFKH] SHUV FRPPf ,W ZRXOG EH LQWHUHVWLQJ WR GHWHUPLQH ZKHWKHU (r K\ORFHWHV DQG 5 VXPLFKUDVWL DUH PRUH DYHUVH WKDQ (A PHODQRWLV WR WKH WDVWH RI D YDULHW\ RI RWKHU QDWXUDOO\ RFFXUULQJ ELWWHU FRPSRXQGV ERWK KDUPOHVV DQG KDUPIXO RQHVf ,I ( L[ K\ORFHWHV DQG ,/ VXPLFKUDVWL DUH PRUH VHQVLWLYH WR WKHVH ELWWHU FRPSRXQGV LW IROORZV WKDW WKH\ ZRXOG EH PRUH OLNHO\ WKDQ (r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f 6HFRQG (r PHODQRWLV RFFXUV LQ D PXFK JUHDWHU YDULHW\ RI KLJKDOWLWXGH Pf KDELWDWV WKDQ GR WKH RWKHU IRXU VSHFLHV RI PLFH RQ 6LHUUD &KLQFXD *OHQGLQQLQJ XQSXE GDWDf DQG HOVHZKHUH LQ 0H[LFR & *DOLQGR XQSXEO GDWDf 7KLUG WKH QXPEHU RI IRRGV GLVOLNHG E\ KXPDQV LQFUHDVHV LQ GLUHFW SURSRUWLRQ WR WKHLU WDVWH VHQVLWLYLW\ WR TXLQLQH DQG QSURS\OWKLRXUDFLO 3523f ERWK RI ZKLFK DUH H[WUHPHO\ ELWWHU FRPSRXQGV )LVFKHU DQG *ULIILQ )LVFKHU HW DO f

PAGE 195

7KHUH DUH VHYHUDO SRWHQWLDO FRVWV RI D JHQHUDOL]HG WROHUDQFH WRZDUGV ELWWHU IRRGV )LUVW DQ DQLPDO PD\ XQZLWWLQJO\ SRLVRQ LWVHOI 6XFK VHOIDGPLQLVWHUHG WR[LFRVLV KDV EHHQ REVHUYHG LQ VHYHUDO VSHFLHV RI UHSWLOHV ZKLFK DUH KLJKO\ WROHUDQW WR ELWWHU IRRGV 6ZDLQ f 6HFRQG PDQ\ SRLVRQRXV FRPSRXQGV LQ DGGLWLRQ WR 3$V GR QRW LPPHGLDWHO\ VLFNHQ DQLPDOV )UHHODQG DQG -DQ]HQ f DQG WKXV GR QRW SURYLGH WKH SRVWLQJHVWLYH IHHGEDFN QHFHVVDU\ IRU IRUPDWLRQ RI D FRQGLWLRQHG IHHGLQJ DYHUVLRQ ,Q WKLV FDVH VLPSO\ UHMHFWLQJ D IRRG EDVHG RQ LWV ELWWHU WDVWH ZRXOG EH KLJKO\ DGDSWLYH :LWK UHVSHFW WR WKLV GLVVHUWDWLRQ LW LV FRQFHLYDEOH WKDW IHHGLQJ KHDYLO\ RQ PRQDUFKV DFWXDOO\ VKRUWHQHG WKH DYHUDJH OLIH VSDQ RI WKH eB PHODQRWLV EHFDXVH RI WKH GHOD\HG KHSDWRWR[LFLW\ LQGXFHG E\ WKH PRQDUFKnV 3$V 7KLV K\SRWKHVLV FRXOG EH WHVWHG LQGLUHFWO\ E\ FRPSDULQJ IHPDOHV IURP LQVLGH DQG RXWVLGH PRQDUFK FRORQLHV LQ WHUPV RI WKHLU LQFLGHQFH RI 3$ KHSDWRWR[LFLW\ $V KDYH WKH OLYHUV IURP WKH PLFH XVHG LQ WKH VWRPDFK FRQWHQW DQDO\VHV FDQ HDVLO\ GR WKLV +RZHYHU WKH FRVW RI D VKRUWHU OLIH VSDQ PXVW EH ZHLJKHG DJDLQVW WKH LQFUHDVHG UHSURGXFWLYH RXWSXW WKH (r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
PAGE 196

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n ORZ VHQVLWLYLW\ WR WKH WDVWH RI &*V LV FRLQFLGHQWDO DQG WKDW WKH RWKHU VSHFLHV DUH LQ IDFW PXFK PRUH WROHUDQW WKDQ PHODQRWLV WR WKH WDVWH RI D YDULHW\ RI RWKHU ELWWHU FRPSRXQGV 6HFRQG LW ZRXOG EH LQWHUHVWLQJ WR H[SORUH WKH UHODWLRQVKLS EHWZHHQ VWUXFWXUDO DVSHFWV RI &*V HJ SRODULW\f DQG WDVWH UHMHFWLRQ WKUHVKROGV LQ eBA PHODQRWLV K\ORFHWHV 5 VXPLFKUDVWL DQG KXPDQV DV ZHOO %HFDXVH VR PXFK LV NQRZQ DERXW WKH WR[LFRORJ\ RI &*V VXFK D VWXG\ FRXOG VXJJHVW ZKHWKHU WKH PRVW WR[LF &*V DUH LQ IDFW WKH PRVW ELWWHU 7KLUG WKH UROH RI ELWWHU VHQVLWLYLW\ LQ GHWHUPLQLQJ WKH IHHGLQJ SUHIHUHQFHV RWKHU VSHFLHV RI ZLOG DQLPDOV QHHGV WR EH VWXGLHG *HQHUDO TXHVWLRQV PLJKW LQFOXGH f JLYHQ WKH DEXQGDQFH RI ELWWHU FRPSRXQGV LQ SODQW IRRGV DQG ODFN RI VXFK FRPSRXQGV LQ YHUWHEUDWH SUH\ DUH KHUELYRUHV JHQHUDOO\ PRUH WROHUDQW WR ELWWHUQHVV WKDQ FDUQLYRUHV DQG f GR DQLPDOV ZLWK KLJKO\ HIIHFWLYH GHWR[LILFDWLRQ V\VWHPV WHQG WR EH PRUH WROHUDQW WR ELWWHU IRRGV"

PAGE 197

$33(1',; $ 3$,5(' &203$5,6216 2) 7+( 3(5&(17 6,0,/$5,7< 2) 81'(56725< 9(*(7$7,21 21 *5,'6 /2&$7(' ,16,'( $1' 2876,'( 7+( 021$5&+ &2/21,(6 7+( &2'(6 2) 7+( *5,'6 86(' ,1 ($&+ 3$,5(' &203$5,621 $5( 3529,'(' ,1 3$5(17+(6(6 6(( ),*85( )25 $ .(< 72 7+(6( &2'(6 &203$5( ($&+ &2/801 :,7+ 7+( .586.$/:$//,6 7(67 LQVLGH YV LQVLGH RXWVLGH [D RXWVLGH LQVLGH \V RXWVLGH f f f f f f f f f f f f f f f f f f f f f f f f f f f f PHDQ 6(

PAGE 198

$33(1',; % 3(5&(17$*( ,03257$1&( 9$/8(6 3,96f )25 7+( 63(&,(6 2) 81'(56725< 9(*(7$7,21 21 7+( (,*+7 75$33,1* *5,'6 /2&$7(' (,7+(5 ,16,'( 25 2876,'( 021$5&+ &2/21,(6 21 6,(55$ &+,1&8$ 6(( )8*85( )25 7+( /2&$7,21 2) 7+( *5,'6 (;&/8'(' 63(&,(6 :+26( 3,96 :(5( 127 ,1 $7 /($67 21( 2) 7+( *5,'6

PAGE 199

3UR[LPLW\ RI JULG WR FRORQ\ 3ODQW ,QVLGH 2XWVLGH IDPLO\ 6SHFLHV &RPSRVLWDH 6HQHFLR DQJXOLIROLXV '& /DELDWDH 6DWXUHMD PDFURVWHPD %HQWKf %ULT 5RVDFHDH $FDHQD HORQJDWD / &RPSRVLWDH 6WHYLD PRQDUGDHIROLD +%. /DELDWDH 6DOYLD FDUGLQDOLV +%. &RPSRVLWDH 6HQHFLR FDOORVXV 6FK %LS 3RO\SRGLDFHDH $VSOHQLPQ PDQDQWKHV O 2QDJUDFHDH )XFKVLD PLFURSK\OOD +%. 5RVDFHDH $OFKHPLD SURFXPEHQV 5RVH &RPSRVLWDH (XSDWDULXP PDLUHWLDQXP '& &RPSRVLWDH %LGHQV WULSOLQDULD +%. &RPSRVLWDH 9HUEHVLQD RQFDSKRUD 5RE t 6HDW &RPSRVLWDH 6HQFLD EDUEDUDMRKDQQLV '& &RPSRVLWDH $UFKLEDFFKDULV KLHUDFLRLGHV /DELDWDH +HHULQJf %ODNH 6DOYLD HLHJDQV YDKO B B B B /DELDWDH 6DOYLD KHOLDQWKHPLIROLD %HQWK *HUDQLDFHDH *HUDQLXP IHHPDQLL 3H\U 8PEHOOLIHUDH $UUDFDFLD QHOVRQLL $ t & 6FURSKXODU %HQWK t +RRN 3HQVWHPRQ FDPSDQXODWXV &DYf ZLOOG B A B B B B A B B B LDFHDH &RPSRVLWDH (XSDWDULXP SHWLRODUH 0RF B B B B B B B /DELDWDH 6DOYLD JUDFLOLV %HQWK f f fr f f f a &RQWLQXHG

PAGE 200

$33(1',; %f§FRQWLQXHG &DU\RSK\O ODFHDH 6WHOODULD FXVSLGDWD :LOOG f§ f§ a f f &RPSRVLWDH (FLJHURQ JDOHRWWLL *UD\ H[ +HPVO *UHHQH f f§ f§ f§ f§ &RPSRVLWDH &LUVLXP HKUHQJHUJLL $FK DQG %LS &RPSRVLWDH 6HQHFLR DOERQHUYLXV *UHHQP &RPSRVLWDH 6HQHFLR VDQJXLVRUEH '& 3RO\SRGLDFHDH $GLDQWXP VS (ULFDFHDH $UFWRVWDRKYORV GLVFRORU +RRNf '& &RPSRVLWDH 6HQHFLR WROXFDQXV '& &RPSRVLWDH 3LTXHUD SLORVD KEN f f /DELDWDH 6WDFKYV FRFFLQHD -DFT f &ROXPQ 7RWDO 7RWDO QXPEHU RI VSHFLHV LQ HDFK JULG

PAGE 201

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£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£Of $QWL/RFXVW %XOOHWLQ 1R %DLOH\ : DQG $ : 5 0F&UDH 7KH JHQHUDO ELRORJ\ DQG SKHQRORJ\ RI VZDUPLQJ LQ WKH (DVW $IULFDQ WHWWLJRQLLG 5XVSROLD GLIIHUHQV 6HUYLOOHf 2UWKRSWHUDf -RXUQDO RI 1DWXUDO +LVWRU\ %DUQHV & DQG / (OWKHULQJWRQ 'UXJ GRVDJHV LQ ODERUDWRU\ DQLPDOV D KDQGERRN 8QLYHUVLW\ RI &DOLIRUQLD 3UHVV %HUNHOH\ &DOLIRUQLD %DWH6PLWK ( & $WWUDFWDQWV DQG UHSHOOHQWV LQ KLJKHU DQLPDOV 3DJHV LQ % +DUERUQH 3K\WRFKHPLFDO HFRORJ\ SURFHHGLQJV RI WKH SK\WRFKHPLFDO VRFLHW\ V\PSRVLXP $FDGHPLF 3UHVV /RQGRQ %DW]OL 2 1XWULWLRQ 3DJHV LQ 5 + 7DPDULQ %LRORJ\ RI QHZ ZRUOG 0LFURWXV 6SHFLDO 3XEOLFDWLRQ 1R $PHULFDQ 6RFLHW\ RI 0DPPDORJ\ 13

PAGE 202

%HORYVN\ ( 'LHW RSWLPL]DWLRQ LQ D JHQHUDOLVW KHUELYRUH WKH PRRVH 7KHRUHWLFDO 3RSXODWLRQ %LRORJ\ %HQGHOO ) )RRG DV D FRQWURO RI SRSXODWLRQ RI ZKLWHIRRWHG PLFH (VFRP\VFXV OHXFRSXV QRYHERUDFHQVLV )LVFKHUf &DQDGLDQ -RXUQDO RI =RRORJ\ %HQVRQ 6 % DQG $ ( %RUHOO 1RWHV RQ WKH OLIH KLVWRU\ RI WKH UHG WUHH PRXVH 3KHQDFRP\V OQQRLFDQULQD -RXUQDO RI 0DPPDORJ\ %HUQD\V ( $ DQG 5 ) &KDSPDQ 7KH HYROXWLRQ RI GHWHUUHQW UHVSRQVHV LQ SODQWIHHGLQJ LQVHFWV 3DJHV LQ 5 ) &KDSPDQ ( $ %HUQD\V DQG 6WRIIRODQR -U 3HUVSHFWLYHV LQ FKHPRUHFHSWLRQ DQG EHKDYLRU 6SULQJHU9HUODJ 1HZ
PAGE 203

%URZHU / 3 DQG : + &DOYHUW )RUDJLQJ G\QDPLFV RI ELUG SUHGDWRUV RQ RYHUZLQWHULQJ PRQDUFK EXWWHUIOLHV LQ 0H[LFR (YROXWLRQ %URZHU / 3 0 (GPXQGV DQG & 0 0RIILW &DUGHQROLGH FRQWHQW DQG SDODWDELOLW\ RI D SRSXODWLRQ RI 'DQDXV FKU\VLSSQV PRQDUFK EXWWHUIOLHV IURP :HVW $IULFD -RXUQDO RI (QWRPRORJ\ %URZHU / 3 DQG / 6 )LQN $ QDWXUDO WR[LF GHIHQVH V\VWHP FDUGHQROLGHV LQ EXWWHUIOLHV YHUVXV ELUGV $QQDOV RI WKH 1HZ
PAGE 204

&DOYHUW : + DQG $ &RKHQ 7KH DGDSWLYH VLJQLILFDQFH RI FUDZOLQJ XS RQWR IROLDJH IRU WKH VXUYLYDO RI RYHUZLQWHULQJ PRQDUFK EXWWHUIOLHV 'DQDXV SOH[LSSXVf LQ 0H[LFR (FRORJLFDO (QWRPRORJ\ &DOYHUW : + / ( +HGULFN DQG / 3 %URZHU 0RUWDOLW\ RI WKH PRQDUFK EXWWHUIO\ 'DQDXV SOH[LSSXV /f DYLDQ SUHGDWLRQ DW ILYH RYHUZLQWHULQJ VLWHV LQ 0H[LFR 6FLHQFH &DOYHUW : + 6 % 0DOFROP *OHQGLQQLQJ / 3 %URZHU 0 3 =DOXFNL 7 9DQ +RRN % $QGHUVRQ DQG / & 6QRRN LD SUHVV &RQVHUYDWLRQ ELRORJ\ RI PRQDUFK EXWWHUIO\ RYHUZLQWHULQJ VLWHV LQ 0H[LFR 9LGD 6LOYHVWUH 1HRWURSLFDO &DOYHUW Z + : =XFKRZVNL DQG / 3 %URZHU 7KH HIIHFW RI UDLQ VQRZ DQG IUHH]LQJ WHPSHUDWXUHV RQ RYHUZLQWHULQJ PRQDUFK EXWWHUIOLHV LQ 0[LFR %LRWURSLFD &DQHOD 0DQG 9 6£QFKH]& 3DWUQ GHO £UHD GH DFWLYLGDG GH 1HRWRPRGRQ DOVWRQL DOVWRQL 5RGHQWLD &ULFHWLQDHf $QDOHV GHO ,QVWLWXWR GH %LRORJLD GH OD 8QLYHUVLGDG 1DFLRQDO $XWQRPD GH 0[LFR 6HULD =RRORJD f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£XUH GHU VDFKDURVH XQG GHV NRFKVDO]HV HLQLJHU DUWHQ YRQ LQVHNWHQIUHVVHUQ XQ QDJHWLHUHQ $FWD %LRORJLDH ([SHULPHQWDOLV :DUVDZf &OHPHQW & $ 2EVHUYDWLRQV RQ WKH UHDFWLRQ RI &YDQRFLWWD FULVWDWD EURPLD 2EHUKROVHU WR 'DQDXV SOH[LSSXV /LQQDHXV WKH VNLQ IOLFN 6HQLRU +RQRUV 7KHVLV $PKHUVW &ROOHJH

PAGE 205

&RFKUDQ Z 6DPSOLQJ WHFKQLTXHV WKLUG HGLWLRQf -RKQ :LOH\ t 6RQV 1HZ
PAGE 206

)DUORZ $ FRQVLGHUDWLRQ RI WKH WURSKLF G\QDPLFV RI D ODWH &UHWDFHRXV ODUJH GLQRVDXU FRPPXQLW\ 2OGPDQ )RUPDWLRQf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t 6UDWWRQ 1HZ
PAGE 207

*LOEHUW % 6 DQG & .UHEV (IIHFWV RI H[WUD IRRG RQ 3HURPVYFXV DQG &OHWKULRQRP\V SRSXODWLRQV LQ WKH VRXWKHUQ
PAGE 208

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
PAGE 209

-HXQLDX[ & &KLWLQDVH DQ DGGLWLRQ WR WKH OLVW RI K\GURODVHV LQ WKH GLJHVWLYH WUDFW RI YHUWHEUDWHV 1DWXUH .DODW : DQG 3 5R]LQ /HDUQHG VDIHW\ DV D PHFKDQLVP LQ ORQJGHOD\ WDVWHDYHUVLRQ OHDUQLQJ LQ UDWV -RXUQDO RI &RPSDUDWLYH DQG 3K\VLRORJLFDO 3V\FKRORJ\ .HOOH\ 5 % 1 6HLEHU + -RQHV + 6HJDOO / 3 %URZHU 3\UUROL]LGLQH DONDORLGV LQ RYHUZLQWHULQJ PRQDUFK EXWWHUIOLHV 'DQDXV SOH[LSSXVf IURP 0H[LFR ([SHULHQWLD .HQDJ\ 6DOWEXVK OHDYHV H[FLVLRQ RI K\SHUVDOLQH WLVVXH E\ D NDQJDURR UDW 6FLHQFH .HQDJ\ DQG % 0 %DUQHV 6HDVRQDO UHSURGXFWLYH SDWWHUQV LQ IRXU FRH[LVWLQJ URGHQW VSHFLHV IURP WKH &DVFDGH 0RXQWDLQV :DVKLQJWRQ -RXUQDO RI 0DPPDORJ\ .LQJ & 0 7KH UHODWLRQVKLS EHWZHHQ EHHFK 1RWKRIDJXV VSf VHHGIDOO DQG SRSXODWLRQ G\QDPLFV IR PLFH 0XV PXVFXOXVf DQG WKH GHPRSJUDSKLF DQG GLHWDU\ UHVSRQVHV RI VWRDWV 0XVWHOOD HUPLQHDf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f =HLWVFKULIW ILLU 7LHSV\FKRORJLH /DXWHUEDFK ) ,QWHVWLQDO DEVRUSWLRQ DQG VHFUHWLRQ RI FDUGLDF JO\FRVLGHV 3DJHV LQ *UHHII &DUGLDF JO\FRVLGHV 3DUW ,, 3KDUPDFRNLQHWLFV DQG FOLQLFDO SKDUPDFRORJ\ 6SULQJHU9HUODJ 1HZ
PAGE 210

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n&ORVNH\ 5 7 DQG 7 /DMRLH 'HWHUPLQDQWV RI ORFDO GLVWULEXWLRQ DQG DEXQGDQFH LQ ZKLWHIRRWHG PLFH (FRORJ\ 0DOFROP 6 % DQG / 3 %URZHU (YROXWLRQDU\ DQG HFRORJLFDO LPSOLFDWLRQV RI FDUGHQROLGH VHTXHVWUDWLRQ LQ WKH PRQDUFK EXWWHUIO\ ([SHULHQWLD 0DOFROP 6 % % &RFNUHOO DQG / 3 %URZHU &DUGHQROLGH ILQJHUSULQW RI PRQDUFK EXWWHUIOLHV UHDUHG RQ WKH FRPPRQ PLONZHHG $VFOHSLDV V\ULDFD / -RXUQDO RI &KHPLFDO (FRORJ\ 0DUW\ 0 0 'LVSRVLWLRQ DQG PHWDEROLVP RI PLONZHHG $VFOHSLDGDFHDHf FDUGHQROLGHV LQ PRQDUFK EXWWHUIO\ ODUYDH 'DQDXV SOH[LSSXV DQG LQ D SUHGDWRU 3HURP\VFXV PDQLFXODWXV 'LVVHUWDWLRQ 8QLYHUVLW\ RI &DOLIRUQLD 'DYLV 0DVRQ 5 0 5DELQ DQG $ 6WHYHQV &RQGLWLRQHG WDVWH DYHUVLRQV VNLQ VHFUHWLRQV XVHG IRU GHIHQVH E\ WLJHU VDODPDQGHUV $PE\VWRPD WLJULQXP &RSHLD 0DVWHUV $ 5 6 % 0DOFROP DQG / 3 %URZHU 0RQDUFK EXWWHUIO\ 'DQDXV SOH[LSSXVf WKHUPRUHJXODWRU\ EHKDYLRU DQG DGDSWDWLRQV IRU RYHUZLQWHULQJ LQ 0H[LFR (FRORJ\ 0DWWRFNV $ 5 +HSDWRWR[LF HIIHFWV GXH WR S\UUROL]LGLQH DONDORLG 1R[LGHV ;HQRELRWLFD 0DWWRFNV $ 5 $FXWH KHSDWRWR[LFLW\ DQG S\UUROLF PHWDEROLWHV LQ UDWV GRVHG ZLWK S\UUROL]LGLQH DONDORLGV &KHPLFR%LRORJLFDO ,QWHUDFWLRQV 0DWWRFNV $ 5 7R[LFLW\ DQG PHWDEROLVP RI 6HQHFLR DONDORLGV 3DJHV LQ % +DUERUQH 3K\WRFKHPLFDO HFRORJ\ $FDGHPLF 3UHVV 1HZ
PAGE 211

0DWWRFNV $ 5 5HODWLRQ RI VWUXFWXUDO IHDWXUHV WR S\UUROLF PHWDEROLWHV LQ OLYHUV RI UDWV JLYHQ S\UUROL]LGLQH DONDORLGV DQG GHULYDWLYHV &KHPLFR%LRORJLFDO ,QWHUDFWLRQV 0DWWRFNV $ 5 &KHPLVWU\ DQG WR[LFRORJ\ RI S\UUROL]LGLQH DONDORLGV $FDGHPLF 3UHVV 1HZ
PAGE 212

3DWWRQ + DQG 7 & 5XFK 3UHIHUHQFH WKUVKROGV IRU TXLQLQH K\GURFKORULGH LQ FKLPSDQ]HH PRQNH\ DQG UDW -RXUQDO RI &RPSDUDWLYH 3V\FKRORJ\ 3HDUVRQ / 7KH IXQFWLRQ RI PXOWLSOH DQWLSUHGDWRU PHFKDQLVPV LQ DGXOW WLJHU EHHWOHV &ROHSWHUD &LFLQGHOLGDHf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f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
PAGE 213

EXWWHUIOLHV 'DQDXV SOH[LSSXV /f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f WR ELUGV -RXUQDO RI WKH 1HZ
PAGE 214

6WHEELQV / / (QHUJ\ UHTXLUHPHQWV GXULQJ UHSURGXFWLRQ RI 3HURP\VFXV PDQLFXODWXV &DQDGLDQ -RXUQDO RI =RRORJ\ 6WHZDUG 9 % 6PLWK DQG ) 0 6WHSKDQ 5HGZLQJHG EODFNELUG SUHGDWLRQ RQ SHULRGLFDO FLFDGDV &LFDGLGDH 0DDLFLFDGD VSSf ELUG EHKDYLRU DQG FLFDGD UHVSRQVHV 2HFRORJLD 6WLFNHO / ) 3HURP\VFXV UDQJHV DW KLJK DQG ORZ GHQVLWLHV -RXUQDO RI 0DPPDORJ\ 6WRGGDUG 0 DQG 3 $ 6PLWK :RRGPLFH ,$SRULVPQV VYOYDW L FXDf FDQ GLVWLQJXLVK FRQVSHFLILF IURP KHWHURVSHFLILF RGRUV LQ WKH ILHOG -RXUQDO RI &KHPLFDO (FRORJ\ 6WUHKO & ( DQG :KLWH (IIHFWV RI VXSHUDEXQGDQW IRRG RQ EUHHLGQJ VXFFHVV RI EHKDYLRU RI WKH UHGZLQJHG EODFNELUG 2HFRORJLD 6ZDLQ 7 5HSWLOH$QJLRVSHUP &RHYROXWLRQ 3DJHV LQ $ G %HOODLUV DQG & % &R[ 0RUSKRORJ\ DQG ELRORJ\ RI UHSWLOHV /LQQHDQ 6RFLHW\ 6\PSRVLXP 6HULHV 1R 7DLWW 0 7KH HIIHFW RI H[WUD IRRG RQ VPDOO URGHQW SRSXODWLRQV GHHUPLFH 3HURPV\FXV PDQLFXODWXVf -RXUQDO RI $QLPDO (FRORJ\ 7DQ] 5 DQG 3 5 8UTXLOOD &DUGLDF JO\FRVLGHV 3DJHV LQ & 5 &UDLJ DQG 5 ( 6WLW]HO 0RGHUQ SKDUPDFRORJ\ /LWWOH %URZQ DQG &RPSDQ\ %RVWRQ 8UTXKDUW ) $ )RXQG DW ODVW WKH PRQDUFKnV ZLQWHU KRPH 1DWLRQDO *HRJUDSKLF 9DXJKDQ 7 $ DQG 1 &]DSOHZVNL 5HSURGXFWLRQ LQ 6WHSKHQn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

PAGE 215

:DWWV & + 6 (IIHFW RI VXSSOHPHQWDU\ IRRG RQ EUHHGLQJ LQ ZRRGODQG URGHQWV -RXUQDO RI 0DPPDORJ\ :KLWDNHU 2 )RRG RI 0XV PXVFXOXV 3HURP\VFXV PDQLFXODWXV EDLUGL DQG 3HURPYVFXV OHXFRSXV LQ 9LJR &RXQW\ ,QGLDQD -RXUQDO RI 0DPPDORJ\ :KLWPDQ Z 0 6 %OXP DQG & -RQHV &KHPLFDO GHIHQVH LQ 7DHQLRSRGD HTXHV 2UWKRSWHUD $FULGLGDHf UROH RI WKH PHWDWKRUDFLF VHFUHWLRQ $QQDOV RI WKH (QWRPRORJLFDO 6RFLHW\ RI $PHULFD :KLWPDQ : 0 6 %OXP DQG & -RQHV 2OIDFWRULO\ PHGLDWHG DWWDFN VXSSUHVVLRQ LQ WKH VRXWKHUQ JUDVVKRSSHU PRXVH WRZDUG DQ XQSDODWDEOH SUH\ %HKDYLRUDO 3URFHVVHV :KLWWDNHU 5 + &RPPXQLWLHV DQG HFRV\VWHPV 0DFPLOODQ 3XEOLVKLQJ &RPSDQ\ ,QF 1HZ
PAGE 216

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

PAGE 217

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n VU$ -RKQ ) (LVHQEHUJ .DWKDULQH 2UGZD\ 3URIHVVRU RI (FRV\VWHP &RQVHUYDWLRQ ? FHUWLI\ WKDW KDYH UHDG WKLV VWXG\ DQG WKDW LQ P\ RSLQLRQ LW FRQIRUPV WR DFFHSWDEOH VWDQGDUGV RI VFKRODUO\ SUHVHQWDWLRQ DQG LV IXOO\ DGHTXDWH LQ VFRSH DQG TXDOLW\ DV D GLVVHUWDWLRQ IRU WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\

PAGE 218

, FHUWLI\ WKDW KDYH UHDG WKLV VWXG\ DQG WKDW LQ P\ RSLQLRQ LW FRQIRUPV WR DFFHSWDEOH VWDQGDUGV RI VFKRODUO\ SUHVHQWDWLRQ DQG LV IXOO\ DGHTXDWH LQ VFRSH DQG TXDOLW\ DV D GLVVHUWDWLRQ IRU WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ )UDQN 6ODQVN\ -U & ‘ $VVRFLDWH 3URIHVVRU RI (QWRPRORJ\ DQG 1HPDWRORJ\ )UDQN 6ODQVN\ -U 7KLV GLVVHUWDWLRQ ZDV VXEPLWWHG WR WKH *UDGXDWH )DFXOW\ RI WKH 'HSDUWPHQW RI =RRORJ\ LQ WKH &ROOHJH RI /LEHUDO $UWV DQG 6FLHQFHV DQG WR WKH *UDGXDWH 6FKRRO DQG ZDV DFFHSWHG DV SDUWLDO IXOILOOPHQW RI WKH UHTXLUHPHQWV IRU WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ $XJXVW 'HDQ *UDGXDWH 6FKRRO

PAGE 219

81,9(56,7< 2) )/25,'$


UNIVERSITY OF FLORIDA
3 1262 08554 2735


COMPARATIVE RESPONSES OF FIVE SYMPATRIC SPECIES OF MICE
TO OVERWINTERING COLONIES OF MONARCH BUTTERFLIES IN MEXICO
By
JOHN INGERSOLL GLENDINNING
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1989

ACKNOWLEDGEMENT S
I thank the members of my PhD committee, Donald Dewsbury, John
Eisenberg, Lou Guillette and Frank Slansky, for their guidance and
critical evaluation of my ideas, and my major advisor, Lincoln Brower,
for struggling with me through the painful process of learning how to
write, for his insight, expertise and support, and for showing me how
to camp in style and safety in Mexico. I also thank Tonya Van Hook,
Jim Anderson, Alfonso Alonso, Alfredo Arellano, and Bill Calvert for
their camaraderie and help while camping on Sierra Chincua; Alfonso
Alonso for help collecting the data in Chapter 7; Michael Carleton of
the the National Museum of Natural History for identifying the species
of mice; Monarca A.C., Bernardo Villa-R., and Jorge Soberón for
logistical support in Mexico; Hilda Flores and Jose Luis Villaseñor,
Alfonso Alonso, Alfredo Arellano and Ron Kelley for help identifying
the understory vegetation on Sierra Chincua; the Dirección General de
Flora y Fauna Silvestre of SEDUE for a scientific collecting permit in
Mexico; Steve Malcolm and Tonya Van Hook for patiently teaching me the
lipid analysis and spectrophotometric techniques; Mark Yang and Carlos
Martinez del Rio for statistical advice; Daryl Harrison for help with
the illustrations; Steve Malcolm, Carlos Martinez del Rio, Doug Levey,
Linda Fink, and Myron Zalucki for editorial help; Fred Morrison for
help collecting and rearing monarchs on Asclepias syriaca: Mark
Stelljes for the monocrotaline N-oxides; Mary Allen, Alfredo Arellano,
Ü

William Calvert, Carlos Galindo, Victor Sánchez, Mark Stelljes, and
Jacqueline Roy for the unpublished data; and the Dept, of Zoology,
Sigma Xi Society, and Explorer's Club for financial assistance. Last
but not least, I thank Diana Schulmann for her editorial help,
unfailing support and love throughout the greater and lesser moments
of my doctoral program.
i 1 i

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
COMPARATIVE RESPONSES OF FIVE SYMPATRIC SPECIES OF MICE TO
OVERWINTERING COLONIES OF MONARCH BUTTERFLIES IN MEXICO
By
John Ingersoll Glendinning
August 1989
Chairperson: Lincoln P. Brower
Major Department: Zoology
I compared five species of sympatric mice (Peromyscus melanotis.
E-». aaLSCUS hylocetes, Eeithrodontomys sumichrasti. Microtus mexicanus
aalyua and Neotomodon alstoni alstoni) in terms of their ability to
take advantage of overwintering colonies of monarch butterflies
(Danaus plexippus) in Mexico. To eat this superabundant food
resource, the mice must overcome the monarch's chemical defense
system, which consists of bitter-tasting and potentially toxic
cardiac glycosides (CGs) and pyrrolizidine alkaloids (PAs).
Peromyscus melanotis appeared to be the only species of mouse that
breached these defenses: large numbers of £_*. melanotis immigrated to
the colonies, fed on the monarchs, and initiated winter reproduction.
In contrast, the other species of mice appeared to avoid the
colonies. On grids outside the colonies, individuals of all five
species were common but reproductively inactive throughout most of
iv

the winter. To explain these demographic results, I hypothesized
that E^. melanotis 1) was the only species tolerant to the
microhabitat features of the overwintering sites, 2) aggressively
excluded the other species, and 3) was the only species that
tolerated the monarch's defensive compounds. Because the understory
vegetation patterns were not significantly different from those
outside colonies, I rejected the first hypothesis. The second
hypothesis was rejected because resident E^. melanotis were unable to
dominate the other species in captive agonistic encounters. The
third hypothesis was accepted because only E^_ melanotis was able to
1) thrive on a diet of pure monarchs and 2) learn how to reject the
CG-laden cuticle and feed selectively on the low-CG internal tissues.
To determine whether S^. melanotis was uniquely able to overcome the
taste and toxicity of the monarch's defensive compounds, several
further feeding experiments were initiated to compare £_*. melanotis
with hylocetes and sumichrasti. None of the species was
sensitive to the toxic effects of digitoxin (a CG), which suggests
that all could have ingested the monarch's CGs with impunity.
However, E_*. hylocetes and sumichrasti were much more sensitive
than E_*. melanotis to the taste of digitoxin, suggesting that the
divergent demographic responses of the different mouse species can be
explained by their differential responses to the bitter taste of the
CG-laden monarch butterflies. Monocrotaline (a PA) did not seem to
have an influence on the feeding behavior of either species.

TABLE OF CONTENTS
Page
ACKNOWLEDGEMENTS ii
ABSTRACT vi
CHAPTERS
1 GENERAL INTRODUCTION 1
Overwintering Biology of Monarch Butterflies in Mexico 3
Known and Potential Predators of Overwintering Monarchs 4
Overview of Dissertation 6
2 DEFENSIVE COMPOUNDS IN THE MONARCH BUTTERFLY 7
Cardiac Glycosides 8
Chemistry and Toxicology 8
Defensive Value 11
Pyrrolizidine Alkaloids 12
Chemistry and Toxicology 12
Defensive Value 16
3 DEMOGRAPHIC RESPONSES TO COLONIES OF MONARCH BUTTERFLIES 19
Methods 20
Results 34
Discussion 66
4 COMPARATIVE FEEDING RESPONSES TO OVERWINTERING MONARCHS 73
Methods 73
Results 82
Discussion 100
5 CONSUMPTION AND AVOIDANCE OF MONARCHS: THE INFLUENCE OF
MONARCH ABUNDANCE AND CARDIAC GLYCOSIDE CONCENTRATION 105
Methods 105
Results 112
Discussion 120

6 COMPARATIVE TASTE AND TOXIC RESPONSES TO CARDIAC GLYCOSIDES
AND PYRROLIZIDINE ALKALOIDS 127
Methods 127
Results 135
Discussion 153
7 BEHAVIORAL AND ECOLOGICAL INTERACTIONS OF FORAGING
P . MELANOTIS WITH OVERWINTERING MONARCHS 163
Methods 164
Results 169
Discussion 175
8 GENERAL CONCLUSIONS 182
Impact of Mouse Predation on Overwintering Monarch
Butterflies 182
Why is Ej. melanotis the only Mouse Species That Feeds on the
Monarchs? 183
Are Popluations of melanotis in the Overwintering Areas
Specifically Adapted for Overcoming the Monarch's Defensive
Compounds 185
Benefits and Costs of Being Tolerant to Bitter Foods 186
APPENDICES
A PAIRED COMPARISONS OF THE PRECENT SIMILARITY OF UNDERSTORY
VEGETATION ON GRIDS LOCATED INSIDE AND OUTSIDE THE MONARCH
COLONIES 190
B PERCENTAGE IMPORTANCE VALUES FOR THE SPECIES OF UNDERSTORY
VEGETATION ON THE EIGHT TRAPPING GRIDS LOCATED INSIDE OR
OUTSIDE A MONARCH COLONY ON SIERRA CHINCUA 191
REFERENCES 194
BIOGRAPHICAL SKETCH
209

CHAPTER 1
GENERAL INTRODUCTION
Arthropods possess a diverse array of chemical defenses that deter,
sicken, or even kill vertebrate predators (Eisner 1970, Duffey 1976,
Blum 1981, Pasteéis et al. 1983). However, the actual effectiveness of
these defenses in naturally occurring predator-prey complexes is
largely unknown because most studies have been conducted in the
laboratory with unnatural predators (Duffey 1976, Pasteéis et al.
1983). The few studies with natural predators have simply described 1)
how a single predator species responds in the laboratory to the
defenses of several prey species (Cyr 1972, Langley 1981, Redford 1984,
Whitman et al. 1985, Whitman et al. 1986), or 2) the extent to which
sympatric species of birds in the field utilize one or several
chemically defended butterfly species (Fink et al. 1983, Collins and
Watson 1983, Sargent 1987). A notable study by Pearson (1985) examined
how effectively lizards, birds, and robber flies could breach the
multiple defenses (chemical, physical, and behavioral) of adult tiger
beetles. However, to my knowledge, no study has compared
experimentally an entire assemblage of sympatric predator species of
the same taxon in terms of their ability to overcome the chemical
defenses of a single, shared prey species.
1

2
Mammals are thought to have evolved a variety of attributes for
dealing with potentially poisonous foods (Eisner 1970, Freeland and
Janzen 1974, Brower 1984) . They include the ability to 1) learn from
their mother which foods are safe (Galef 1977, Vaughan and Czaplewski
1985), 2) detoxify certain compounds (e.g., Freeland and Janzen 1974,
Carlson and Breeze 1984), 3) feed selectively on nutritious tissues and
reject unpalatable and/or potentially toxic parts of a food item
(Benson and Borell 1931, Hamilton 1962, Kenagy 1972, Cyr 1972, Vaughan
and Czaplewski 1985, Brower et al. 1988, Roy and Bergeron unpuhl.
data), 4) avoid foods by developing a conditioned feeding aversion
(CFA) to them (Rzóska 1953, Garcia and Koelling 1966, Brower 1969,
Kalat and Rozin 1973, Robbins 1978, Jacobs and Labows 1979, Daly et al.
1982, Brower and Fink 1985), and 5) avoid all bitter-tasting foods
(Bate-Smith 1972). This latter attribute is derived from the
observation that many mammals are averse to a variety of bitter
compounds at extremely low concentrations (Patton and Ruch 1944,
Carpenter 1956, Chrzanowski 1965, Pressman and Doolittle 1966, Ganchrow
1977, Jacobs et al. 1978). Virtually all poisons of biological origin
are bitter-tasting to humans (Richter 1950, Bate-Smith 1972, Garcia and
Hankins 1975, Brower 1984). Thus a general aversion to bitter foods
may be an effective way to avoid ingesting such compounds.
The relative importance of these five attributes to foraging wild
animals is unknown. This is because, in most cases, each attribute has
been studied in isolation from the others with only one species of
predator. In this dissertation, I will examine the relative importance
of feeding selectively, CFAs, and avoiding bitter-tasting foods in
determining the responses of 5 species of wild mice to the vast

3
overwintering colonies of monarch butterflies (Panana plexippns L.)
that form each year in Mexico.
Overwintering Biology of Monarch Butterflies in Mexico
Each fall, virtually the entire eastern North American population
of monarch butterflies migrates to several, remote, high-altitude
overwintering sites in central Mexico (Brower and Calvert 1985, Calvert
and Brower 1986) . There, within the oyamel fir forests (Abies
religiosa H.B.K.), they form colonies of tens of millions of
\
individuals. The colonies range in size from 0.3 to 3.3 ha, attain
densities of about 10 million per ha, cover the trees and understory
vegetation, and stay in approximately the same location from late
December through late February (Brower and Calvert 1985, Calvert and
Brower 1986) . Although the colonies form on the same mountainsides
each year, they do not always form in the same location; colony sites
are frequently 1 to 2 kilometers from one another among years (Calvert
and Brower 1986, Figure 2-la, b; Glendinning pers. observ.).
Individual butterflies have extensive lipid reserves (mean in January =
35% of total dry weight; Brower 1985). They are vulnerable to
terrestrially foraging predators because low temperatures constrain the
butterflies' predator avoidance behaviors such as flying back to
communal roosts and crawling up vegetation (Calvert and Cohen 1983,
Masters et al. 1988, Alonso et al. in press). As such, these colonies
constitute a potential, superabundant, lipid-rich food bonanza for
mice.
However, overwintering monarchs contain cardiac glycosides (CGs)
and pyrrolizidine alkaloids (PAs), both of which are thought to protect

4
monarchs against most vertebrate predators. I describe the chemistry,
toxicology, and purported defensive value of these compounds in the
next chapter.
Known and Potential Predators of Overwintering Monarchs
Of the 37 species of omnivorous and insectivorous birds present in
the Sierra Chincua overwintering area (Arellano et al. in press) r only
black-headed grosbeaks and black-backed orioles (Calvert et al. 1979,
Fink and Brower 1981, Fink et al., 1983; Brower and Calvert, 1985,
Arellano et al. in press) feed extensively on the monarchs. Whereas
grosbeaks are relatively insensitive to the toxic effects of CGs,
orioles avoid ingesting large amounts of CGs through taste rejection of
the more toxic monarchs (Fink and Brower 1981) . Mixed flocks of these
two species were estimated to kill 4,550 to 34,300 butterflies per day
in a 2.25 ha colony, or about 9% of the colony (Brower and Calvert
1985) . There is some evidence that prolonged consumption of monarchs
caused one or both of these species to accumulate toxic levels of CGs
and possibly PAs (Brower and Calvert 1985, Arellano et al. in press)
Several species of mice may also feed extensively upon the
monarchs. Individual black-eared mice Peromyscus melanotis J.A. Allen
and Chapman, Neotomodon alstoni alstoni Merriam, and Microtus mexicanus
salvus Saussure that were captured near overwintering colonies in
Mexico have been found to eat monarchs to varying degrees while in
captivity (Brower et al. 1985). (The former two species were
erroneously referred to as Peromyscus maniculatus labecula Elliot and
P. spicilegus J.A. Allen, respectively, in Brower et al., 1985; see
Brower et al. 1988 for proper identification). Stomach content

5
analyses and short-term trapping studies suggest that female P.
melanotis inside monarch colonies feed upon monarchs and are larger and
more reproductively active than female conspecifics outside colonies
(Brower et al. 1985). Thus the evidence suggests that melanotis
actually benefits from feeding upon the monarchs.
There are several other animals known or suspected to feed upon
overwintering monarchs. Both cattle and domestic turkeys occasionally
eat the butterflies (Urquhart 1976, A. Arellano unpub1. data'. Calvert
et al. (1979) found scats presumed to be from the hog-nosed skunk
Conepatus mesoleucus nelsoni that contained monarch parts. I also
observed an individual hog-nosed skunk inside a monarch colony during
the night on Sierra Chincua during February 1986, but I could not
determine whether it was feeding on monarchs. Brower et al. (1985)
caught one Sorex saussurei near a monarch colony on Sierra Chincua but
were unable to determine whether it had fed upon monarchs. Between
1985 to 1989, I trapped 34 additional saussurei (5 inside and 29
outside monarch colonies) but did not analyze their stomach contents.
During my field work in the Sierra Chincua overwintering area
(Figure 3-la and b), I discovered several additional species of small
mammals. They included four muroid rodents (Peromyscus aztecus
hylocetes, Reithrodontomys sumichrasti, Reithrodontomys cf. megalotis.
Nalsoai neotomodon goldmani). a squirrel (Sciurus cf. aureogaster). a
shrew (Cryptotis goldmani alticola), a weasel (Mustela frenata), and a
bat (Lasiurus cinereus). However, all species except the former 2 were
extremely rare; < 2 individuals of each were trapped and/or sighted.

6
Overview of Dissertation
The initial goal of this project was to compare the demographic
responses of Peromyscus melanotis J. A. Allen and Chapmen, E_*. aztecus
hylocetes Merriam, Reithrodontomys sumichrasti Saussure, Microtus
meaicanus salvus Saussure, and Neotomodon alstoni alstoni Merriam to
the overwintering monarchs. In Chapter 3, I describe the discovery
that all species but E^. melanotis largely avoided the monarch colonies.
Large numbers of E^. melanotis immigrated into the colonies, fed
extensively on monarchs, and initiated high levels of reproduction. To
explain this finding, I examine three questions: 1) are there unique
plant species compositions in the sites where monarch colonies form
that render these areas suitable only to E^. melanotis: 2) can P.
melanotis aggressively exclude the other species of mice from the
colonies; or 3) is melanotis better able to breach the monarch's
chemical defense system? I rule out the first two hypotheses in
Chapter 3 and devote Chapters 4 through 6 to testing the third
hypothesis by means of an integrated series of feeding experiments with
monarchs and artificial diets. In Chapter 7, I examine the potential
impact of predation by E^. melanotis on the live and dead butterflies
that accumulate on the ground and in low vegetation, and how this
predation is influenced by the accessibility and degree of desiccation
of the monarchs.

CHAPTER 2
DEFENSIVE COMPOUNDS IN OVERWINTERING MONARCH BUTTERFLIES
Adult monarchs contain cardiac glycosides (CGs) that were
sequestered and stored from the species of milkweed plant (Asclepia.q
spp.) that they fed upon as larvae (Brower and Moffit 1974, Roeske et
al. 1976, Malcolm and Brower 1989). Available evidence indicates that
85 to 92% of overwintering monarchs in Mexico fed as larvae on the
common milkweed A. syriaca L. (Seiber et al. 1986, Malcolm and Brower
1989). Whereas freshly-eclosed A. syriaca-reared individuals initially
have relatively high concentrations of CG (mean = 278 |lg/0.1 g), they
lose on average 86% of these CGs during the long migration to Mexico
(mean = 40 |ig/0.1 g) (Malcolm et al. 1989, Malcolm and Brower 1989) .
Approximately 80% of the CGs in A. syriaca-reared monarchs are located
in the cuticle and wings (Brower et al. 1988) .
Overwintering monarchs in Mexico also sequester and store
pyrrolizidine alkaloids (PAs) from PA containing plant families, such
as the Asteraceae (Compositae), Boraginaceae, Apocynaceae, and Fabaceae
(Leguminosae) (Edgar et al. 1976, Brower 1984, Mattocks 1986, Kelley et
al. 1987, M. Stelljes unpubl. data). The alkaloid appears to be
acquired by imbibing nectar both en route to the overwintering sites
(Kelley et al. 1987) and while overwintering (Glendinning and Kelley
unpubl. data). Preliminary evidence indicates that PAs, like CGs, are
concentrated in the cuticular material (M. Stelljes unpubl. data)
7

8
Cardiac Glycosides
Chemistry and Toxicology
All CGs have as their basic structural component C-23 steroid
glycosides (Figure 2-1). The use of the adjective cardiac stems from
the ability of these compounds to increase drastically myocardial
contractile force in vertebrates. Most biologically active CGs have an
-OH group at position C14, and the steroid moiety (rings A-D) is
completely saturated. There is usually an unsaturated five-member
lactone ring at position C17 (ring E); biological activity is abolished
by saturating the lactone or opening its ring structure. The steroid
nucleus and the lactone ring comprise the aglycone or genin portion of
the molecule and confer pharmacological activity to the structure (Moe
and Farah 1975) . There are usually 1 to 4 sugar residues, or
glycosides, bonded to the steroid nucleus by an ether linkage at
position C3.
The potency and duration of activity of CGs are determined
primarily by the polarity of the glycosides and various steroid
substituents (Moe and Farah 1975, Detweiler 1967) . That is, polarity
strongly influences absorption, distribution, metabolism, excretion and
possibly gustation of CGs in vertebrates. As compared to polar CGs,
less polar ones are 1) absorbed more easily across the lipoid membrane
of the intestine (Herman et al. 1962, Lauterbach 1981), 2) bound more
readily by serum albumin and thus rendered unavailable for
pharmacological or toxicological action or metabolism and excretion
(Lukas and deMartino 1969, Moe and Farah 1975), 3) metabolized less
readily in the liver (Detweiler 1967), 4) excreted more slowly because

9
Figure 2-1. General structure of cardiac glycosides.
R = a glycoside.

o
o
A. Aspecioside
B. Uscharidin
Calactin: 0-OH at C31
(digitoxoae) 3 — o
D. Digltoxin
Figure 2-2. Some representative cardiac glycosides.

11
they get caught in the enterohepatic cycle (i.e., absorbed CGs and
their metabolites are excreted in the bile and then reabsorbed; Okita
1967, Detweiler 1967), and 5) taste rejected by birds at higher
concentrations (Brower and Fink 1985) .
The structures of 5 CGs relevant to this dissertation are presented
in Figure 2-2. Aspecioside is the dominant CG in A^. syr-i ar.a L.
(Malcolm et al. 1989), and because 85 to 92% of the monarchs at the
Mexican overwintering sites appear to have fed on this foodplant, it is
also the dominant CG in their bodies (Seiber et al. 1986, Malcolm and
Brower 1989). Calotropin, calactin and uscharidin are the dominant CGs
in monarchs reared on A^. curassavica L. and are much less polar than
aspecioside (Roeske et al. 1976); mice were offered monarchs reared on
both A*, curassavica and A^ syriaca in Chapter 5. Digitoxin is derived
from several Digitalis species (Scrophulariaceae) and is of
intermediate polarity; mice were offered diets treated with digitoxin
in Chapter 6.
Defensive Value
Cardiac glycosides stored in monarch butterflies may deter
predators because of their bitter taste and/or toxic effects. CGs are
intensely bitter-tasting to humans (see references in Brower and
Glazier 1975, Glendinning pers. observ.) and are avoided by blue jays
Cyanocitta cristata at extremely low concentrations in food (Brower and
Fink 1985). Thus taste alone may be sufficient to render monarchs
unpalatable to mice.
Oral dosages of a variety of different types of CGs, including
those in adult monarchs, induce emesis, cardiac failure, neurotoxicity,

12
ataxia, dyspnea, severe tremors, and convulsions in many vertebrate
species (Detweiler 1967, Marty 1983, Brower 1984) . However,
sensitivity to oral dosages of CGs is species-specific: dogs, some
species of birds, and humans are highly sensitive, whereas the rodents
Mus musculus, Rattus norvegicus and Peromyscus maniculatus are not
(Detweiler 1967, Barnes and Eltherington 1973, Tanz and Urquilla 1982,
Marty 1983, Brower and Fink 1985) . The insensitivity of these rodents
appears to be due primarily to poor absorption of CGs across the
gastrointestinal wall (Marty 1983) . Thus, given that the mice studied
in this dissertation are closely related to the 3 listed above, it is
likely that they are similarly insensitive to the toxic effects of oral
dosages of CGs.
Pyrrolizidine Alkaloids
Chemistry and Toxicology
The ester derivative of all toxic PAs and its numbering is shown in
Figure 2-3a. Two five-membered rings share a common nitrogen and
carbon at positions 4 and 8, respectively. Ester linkages can form at
positions 1 and 7. Pyrrolizidine alkaloids are frequently grouped
based on the number and type of ester linkages they have: 1) monoesters
have a single ester (e.g., intermedine, Figure 2-3b), usually at
position 1; 2) open diesters have two esters that do not join one
another; and 3) macrocyclic diesters have 2 esters that join to form a
ring structure (e.g., senecionine, Figure 2-3c).
The nitrogen atom of all PAs readily undergoes oxidation, which
converts the PA from a free amine to an N-oxide. An example of this
oxidation-reduction reaction is presented in Figure 2-4 for the

A. Amino alcohol derivative
of all toxic PAs with numbering.
R = H or OH
B. Intermedlne, an open monoester,
with numbering on the CHj i ester-
linked acid group.
LjO
C. Seneclonlne, a
macrocycllc dlester.
OR CHjOR
0. Reactive Pyrrole.
R - organic acid
or proton.
Figure 2 3. Some representative pyrrolizidine alkaloids and their derivatives.

Free Amine
N-oxide
Figure 2-4. Reaction involved in the conversion of free amine monocrotaline
to its N-oxide form.

15
macrocyclic diester monocrotaline. The bond between the nitrogen and
oxygen atoms in an N-oxide contains an unequally shared electron.
Because this electron tends to reside closer to the oxygen atom, it
gives the oxygen atom a net negative charge (e.g., see Figure 2-4) and
makes the N-oxide more polar than the free amine. Nevertheless, the
free amine and N-oxide forms of the same PA produce the same type of
toxicity. Both forms are commonly found together in a variety of
plants (Mattocks 1986) and insects (Brown 1984), including
overwintering monarchs in Mexico (Kelley et al. 1987).
There are 3 structural attributes essential for toxicity (Mattocks
198 6) : 1) the ring nucleus must be double-bonded between Cl and C2 (see
Figure 2-3a); 2) there must be at least one esterified hydroxyl group
(e.g., between C9 and CIO, Figure 2-3b); and 3) at least one of the
ester-linked side chains must contain a branched carbon chain (e.g.,
between Cll and C14, Figure 2-3-b). However, PAs are not themselves
toxic. They are rendered toxic by a dehydrogenation reaction in the
liver microsomes that normally metabolize alkaloids like PAs into more
polar and excretable derivatives. This reaction converts PAs into less
polar and highly toxic metabolites referred to as reactive pyrroles
(e.g., see Figure 2-3d; Mattocks 1972). Accordingly, the most toxic
PAs are those most readily coverted to reactive pyrroles.
There are two important structural features of toxic PAs that
facilitate breakdown into pyrrolic metabolites (Mattocks 1981) . First,
low polarity PAs are metabolized into reactive pyrroles at faster rates
because they gain better access to the surfaces of microsomes.
Generally, polarity varies as follows in PAs: monoesters > open
diesters > macrocyclic diesters. Second, the type of ester linkage

16
strongly influences the proportions of pyrroles and N-oxides formed
from the free amines during the dehydrogenation reaction. (Recall that
both pyrroles and N-oxides are formed by this same reaction.) The
conformation of monoesters and macrocyclic diesters favors production
of reactive pyrroles, whereas that of open diesters gives relatively
more N-oxides.
Available evidence indicates that monoesters (e.g., intermedine)
are more common than macrocyclic diesters (e.g., senecionine) in
overwintering monarchs in Mexico; no open diesters were reported
(Kelley et al. 1987). However, this study was based on butterflies
collected during December, which would have probably derived their PAs
from nectar sources along the migration routes (Kelley et al. 1987).
More work is needed to determine whether the monarchs modify the
quantity and/or types of PAs in their bodies over the course of an
overwintering season by nectaring on the locally abundant PA containing
plants (e.g., Senecio and Eupatorium spp.; see Appendix 2 for species
names and their relative abundances; Robbins 1982) .
Defensive Value
As with CGs, PAs may deter predators by means of their bitter taste
and/or toxic effects. PAs are bitter-tasting to humans (Boppré 1986,
Glendinning pers. observ.) and their presence in the arctiid moth
Utetheisa ornatrix. ithomiine butterflies, and artificial foods causes
rejection by a variety of invertebrate and vertebrate predators (Eisner
1980, Brown 1984, Boppré 1986). The relevance of these findings to
monarch's PA-defense is uncertain. The PA content in monarchs is
extremely low relative to that of the ithomiines tested (see Kelley et

17
al. 1987, Brown 1984), and thus may not be sufficiently high to deter
predators. Moreover, the findings from the studies with IL*. omatrix
and PA-adulterated diets cannot be evaluated because they were from
unpublished studies that were referred to in secondary sources (Eisner
1980, Boppré 1986). The relationship of bitterness to polarity and
molecular conformation in PAs has not been studied as it has in CGs
(see above).
Oral dosages of PAs are known to cause a variety of toxic effects
primarily in the liver of mammals, apparently because that is where the
reactive pyrroles are formed (Mattocks 1986). Relatively high dosages
produce acute effects and usually kill mice within 3-7 days by causing
severe hemorrhagic necrosis in liver cells. Lower dosages produce more
chronic effects and may take several months to kill mice by inducing
formation of 1) giant liver cells (megalocytes), 2) veno-occlusive
disease of the liver, and/or 3) depressed rates of growth and food
intake (Schoental and Magee 1957, McLean et al. 1964, Hooper 1978,
Goeger et al. 1983, Cheeke and Pierson-Goeger 1983, Mattocks 1986).
Even though large interspecific differences exist among laboratory
animals (mice, rats, guinea pigs, rabbits, and hamsters) in terms of
sensitivity to particular PAs, no species appears to be insensitive to
all or even a large number of PAs (Mattocks 1986).
Chronic PA toxicity does not usually result in death for several
months, and thus it is unclear whether mammals feel sick immediately
after ingesting sublethal doses of PAs. It may be that the action of
PAs is so insidious, that a mammal would not feel sickened until
several weeks after ingesting them. If this is the case, then it would
be unlikely that mammals could develop CFAs after ingesting them

18
(Brower 1984).

CHAPTER 3
DEMOGRAPHIC RESPONSES TO COLONIES OF MONARCH BUTTERFLIES
Numerous workers have documented strong numerical and reproductive
responses by small mammals to areas with heavy seed crops (Jameson
1953, 1955, Gashweiler 1979, Halvorson 1982, King 1983) and dense
concentrations of insects (Holling 1959, Hanski and Parviainen 1985,
Hahus and Smith ill press). When these food bonanzas occurred in
habitats containing several small species of mammals, however, certain
species invariably responded more strongly than others (Halvorson 1982,
King 1983, Holling 1959, Hanski and Parviainen 1985, Hahus and Smith in
press). The factors that enabled particular species to take greater
advantage of the food bonanzas and thereby increase their reproductive
output proportionately are largely unknown. Possible factors include
1) competitive exclusion (Halvorson 1982), 2) greater tolerance to
microhabitat features of the areas with the food resource, and/or 3)
less sensitivity to the behavioral, physical and/or chemical defenses
of the food resource. In this chapter, I compare the demographic and
feeding responses of several species of mice to overwintering colonies
of monarch butterflies (Danaus plexippus L.) in Mexico, and attempt to
explain why one species enjoys exclusive access to the butterflies.
Four questions regarding the response of nearby mouse populations
to the monarch colonies are explored: 1) what is the density and
biomass of monarchs inside colonies available to terrestrially foraging
19

20
mice; 2) what species of mice occur in the overwintering areas and what
is their relative abundance; 3) are winter demographic patterns of
populations of the different species of mice, such as immigration,
breeding activity, density, home range size and age-class distribution,
influenced by the the colonies; 4) how do the diets of mice inside and
outside the colonies compare? In the course of this investigation, I
discovered that £_*. melanotis was the only species that immigrated to
the colonies and fed extensively on the monarchs. To explain this, I
tested two hypotheses. First, were there unique understory plant
species compositions in the sites where monarch colonies formed that
rendered these areas suitable only to Z». melanotis? Second, did P.
melanotis aggressively exclude the other species of mice from the
monarch colonies?
Methods
The Study Areas
I trapped and conducted experiments from January to March, 1985 to
1988, on the slope of the Arroyo La Plancha of the Sierra Chincua
mountain massif in northeastern Michoacán, Mexico (Figure 3-la, Anon.
1976: topographic map). I also trapped on 3 February and 10 March 1986
on the slope of the Arroyo Los Conejos of the nearby Sierra El
Campanario, at 19°00'N and 100°00'W. These 2 sites are among the 5
principal overwintering areas of the eastern population of the monarch
butterfly in the region (Calvert and Brower 1986).
All work was done between 3000 and 3320 m. The vegetation is
montane, boreal, coniferous forest dominated by the oyamel fir (Abies
religiosa H.B.K.). The understory vegetation is described below. The

21
period from late December to early March is the coldest and driest time
of year (Calvert et al. in pnenn), with night-time forest temperatures
on Sierra Chincua usually ranging between 0 to 6°C (Calvert and Brower
1986, Alonso et al. in press). and with total monthly precipitation
averaging less than 23.3 mm (Calvert et al. in press. Anon. 1982).
Amount of Monarch Material Available to Foraging Mice
To estimate the amount of butterfly material on the forest floor
inside a monarch colony during the night, I ran 3 transects each for 2
consecutive days between 15 and 20 January 1988 inside a 0.44 ha colony
(W. Calvert pers. comm.) on Sierra Chincua. The transects were 80 m
long, each with 10 one-meter2 sampling quadrats set at 8 m intervals.
Because the colony was only about 45 m wide, I used a stratified random
sampling scheme (Zar 1984) to choose lengthwise transect locations.
Each morning between 700 and 730 h, the quadrats were cleared of all
monarch remains. Then, between 1900 and 1930 h on the same day, I
tallied the number of butterflies in each quadrat that were either on
the ground or less than 3 cm above it. Monarchs were classified as
live, moribund, cuticle pierced (i.e., sampled but not eaten), abdomen
removed, abdomen deviscerated, or thorax contents eaten. I disregarded
the few monarchs (< 3% of total monarchs tallied) that did not fall
into one of these 6 categories.
For each daily tally, I determined the mean density of each type of
monarch per m2 quadrat per day. I also estimated the mean grams of
tissue per m2 quadrat per day represented by each type of monarch,
using the following steps. First, I knew the approximate amount of
tissue in the abdomen (123 mg; dry weight) and thorax (41 mg; d.w.) of

22
overwintering monarchs (Chapter 4: Table 4-2). Second, I assumed that
all deviscerated abdomens and damaged thoraces still had on average 50%
of the internal tissues remaining (Fink and Brower 1981, Glendinning
pers. observ.). Then, the number of grams of tissue/m^/day in each
damage category was calculated using the equations provided in Table 3-
1.
Mark-recapture Trapping on Sierra Chincua
I compared the numerical responses of the different species of mice
to the monarch colonies and determined whether those mice living inside
the monarch colonies showed higher levels of reproduction. To do this,
mark-recapture studies were run between 20 January and 16 March 1985-
1987, and between 15 and 20 January 1988, in a total of 8 grid
locations that were either inside or at least 120 m to either side of
the colonies' outer edge (henceforth, outside; Figure 3-la and b). All
of the colonies had formed compact aggregations several weeks before X
set grids inside of them. In the 1985-87 seasons, the inside grids
were within the colonies until the last week of February, at which time
the colonies moved at least several hundred meters downhill.
In the 1985-1987 seasons, each grid was trapped one night per week
for 8 consecutive weeks and consisted of 60 trap stations in a 6 x 10
array with 10 m spacing, resulting in a sampling area of 0.6 ha. Two
traps were set on the ground within 1 m of each trapping station during
all seasons, except in 1985, when only 1 trap was set per station.
During the 1988 season, two grids were trapped for 3 consecutive nights
and consisted of 30 trap stations in a 3 x 10 array with 10 m spacing,
resulting in a sampling area of 0.3 ha. As in previous seasons, two

23
Table 3-1. Equations used to calculate the grams of tissue/m2/day in
the different damage categories that were available to terrestrially
foraging mice on the forest floor inside monarch colonies. For example,
to estimate the amount of live monarch tissue available to mice, I
multiplied the mean number of monarchs per m^ that fell to the monarch
floor per day (mean no./m^/day) times the combined dry weights of the
tissue in an overwintering monarchs's abdomen (123 mg) and thorax (41
mg). Based on Fink and Brower (1981) and Glendinning (pars. observ.). I
assumed that deviscerated abdomens and thoraces still had 50% of the
internal tissues remaining.
Damage category
Equation
live:
moribund:
cuticle pierced:
abdomen removed:
abdomen deviscerated:
thorax deviscerated:
(mean no./m^/day)
(mean no./m^/day)
(mean no. /re?/day)
(mean no./m^/day)
(mean no./m^/day)
(mean no./m^/day)
(123 + 41 mg)
(123 + 41 mg)
(123 + 41 mg)
(41 mg)
[ (123 mg) (50%) + (41 mg)]
[(123 mg) + (41 mg)(50%)]

24
traps were set on the ground. To determine whether my terrestrial
trapping regime had missed an arboreal rodent fauna, a third trap was
also secured to the tree trunk nearest each station, assuming one
occurred within 2 m from the station, at a height of 2 m; a total of 43
arboreal traps were set.
Folding Sherman live traps (8 x 9 x 23 cm) were baited with rolled
oats and provided with a compressed cotton ball for nesting material.
Mice were toe-clipped for permanent identification, and species, sex,
age, weight (to the nearest 0.1 g with 100 g Pesóla scales),
reproductive condition, and trap station were recorded. Peromyscus
melanotis age was recorded as juvenile (<15.5 g) or adult (>15.5 g).
Only adults of the other species of mice were captured. Males of all
species were recorded as reproductively active if their testes were
descended (i.e., scrotal) or inactive if their testes were within their
abdominal cavity (i.e., abdominal). Females of all species were
recorded as pregnant if they had swollen abdomens (and/or weighed > 28
g, if they were E^. melanotis), as lactating if they had large nipples,
or as nonbreeding if they had small nipples and no swollen abdomen.
Traps were set between 1700 and 1900 h and checked the following
morning between 700 and 1000 h. Soiled traps were cleaned thoroughly
with a paper towel and water to reduce the influence of residual odors
from the previous occupant on subsequent captures (Mazdzer et al. 1976,
Stoddard and Smith 1984).
Mice were considered resident if they were present on the same grid
for 2 or more continuous weeks. The minimum number alive (MNA)
enumeration technique (Hilborn et al. 1976) was used to estimate both
the total number of mice on each grid each week and the number of weeks

25
each mouse was present on a grid. To validate use of this enumeration
technique, I also determined the minimum trappability of the population
(Krebs and Boonstra 1984) .
Home range sizes were estimated by calculating the area within a
convex polygon drawn around the outermost trap locations where each
mouse was caught. Home ranges were determined only for those mice 1)
caught at least 5 times, 2) trapped at a station on the edge of the
grid < 1 time, and 3) whose home range size did not increase after the
fifth capture.
To determine the mean weights of the 5 common species of mice on
Sierra Chincua, I combined the results of all mark-recapture studies on
Sierra Chincua between 1985 and 1988. If an individual was captured
more than once, I used its mean weight. Weights of pregnant females
were excluded. However, because most breeding females were caught more
than 4 times, I was able to determine their mean weight before and/or
after their pregnancy. Intersexual comparisons within each species
were made with the unpaired (two-tailed) t-test. In this and all
subsequent statistical tests, I tested for normality with the
Kolmogorov-Smirnov test (Zar 1984) . If a significant departure was
found (i.e., P < 0.05), I used a nonparametric test.
To determine the numerical and reproductive responses of the
different species of mice to the monarch colonies, I treated each grid
as an independent sample. Because the population density estimates
among the successive trapping periods in each grid were not independent
(i.e., many of the mice remained on the same grid for more than one
trapping period), I simply describe trends in these data (Hurlbert
1984). However, I used a two factor ANOVA (repeated over time) to

26
determine the effects of colony proximity and time on the appearance of
new juveniles in the grids.
For several analyses, the combined results from all grids inside
colonies (n = 5) were compared with those from all grids outside
colonies (n = 6). Even though each grid should be analyzed separately
(Hurlbert 1984) , I felt justified combining them because all grids had
relatively similar understory vegetation (see below), species
compositions of mice, and trends in the MNA estimates and levels of
reproductive activity over time. Moreover, sample sizes were small in
each grid. Mice from inside and outside colonies were compared in
terms of 1) the percentage of adult females of each species that were
reproductive, 2) the number of weeks that adult mice of each species
were present on the grids, using the Mann-Whitney U test (with the
normal approximation), 3) the home range sizes of male and female P.
melanotis. using the Mann-Whitney U test, 4) the proportion of juvenile
and adult £_*. melanotis that remained on the grids for greater than one
week, and 5) the effects of colony proximity and sex on the body
weights of male and female E^. melanotis, using the two factor ANOVA.
For comparisons 2 and 4, I excluded those mice first trapped during the
last three weeks of the season (i.e., from 23 February to 9 March,
1985-87).
Snap-trapping of mice at Sierra Chincua and Sierra El Campanario
In addition to these mark-recapture studies, I compared the
demographic responses of mice to the monarch colonies on Sierra
Campanario and Sierra Chincua. To do this, I ran 4 parallel transects
on Sierra Campanario on 3 February and on 15 March 1986 in a 2.1 ha

27
monarch colony (W. Calvert pers. comm.) f as well as on Sierra Chincua
on 15 February 1986 in a 0.6 ha colony on grid 7 (Figure 3-lb). During
each night, two of the transects were set inside the colony and the
other two 300 m to either side of the colonies' edges. Each transect
was 80 m long with 10 trapping stations set at 8 m intervals. I also
snap-trapped on 21 and 22 March 1986 in grids 1 and 2 (Figure 3-lb) at
the same stations used for the above-described mark-recapture studies.
I set two Museum Special snap-traps per station and baited each
with rolled oats and peanut butter. Traps were set between 1700 and
1900 h and rechecked the next morning between 700 and 900 h. I
recorded species, sex, age, weight, and reproductive condition and used
the same indices of age and reproductive status as described above for
the mark-recapture study.
The results from Sierra Campanario were used to compare the species
composition and reproductive activity of mice inside and outside of
colonies. Because of small sample sizes, I combined the results of the
two sampling nights when making the reproductive comparisons. These
results were compared to those from the mark-recapture studies on
Sierra Chincua. The stomachs from mice captured on both massifs were
included in the dietary analyses described below.
Stomach Content Analyses
I compared the stomach contents of mice from inside and outside
monarch colonies in terms of 1) the occurrence of monarch material, 2)
dry weight, 3) lipid weight, 4) percent lipid, 5) total cardiac
glycoside (CG) content, and 6) CG concentration. All weight
measurements were made to the nearest 0.0001 g with a Mettler AK 160

28
balance. To get representative stomach samples from naturally foraging
mice, I used the mice caught in snap-traps, to assure that they had
died immediately upon capture. (Food boluses pass through the
digestive tracts of live, captive Peromyscus maniculatus in 2 to 7
hours (Marty 1983).)
Each animal's stomach was removed and preserved.in 50 ml of 95%
ethanol, and then frozen (within 3 weeks) in a domestic freezer until
analyzed in October 1988. To quantify the occurrence of monarch
material, I first removed and discarded all bait; this bait was not
included in the lipid and CG analyses. Next, I placed the contents of
each stomach in a petri dish, while suspended in about 10 ml of 95%
ethanol, and observed 20 fields under lOOx magnification. If I found
no food items in a field, I chose another one. I determined the
presence or absence of monarch material (e.g.. small pieces of
exoskeleton, testes pigment, fat, and tracheal material) in each field,
and then computed the frequency of occurrence of monarch material in
all 20 fields for each stomach sample. Reference samples of monarch
material were made by feeding test mice a diet consisting solely of
monarchs and running their stomach contents through the procedure
described above. The monarch material and ethanol in which it was
stored (henceforth, stomach content sample) were together subjected to
the lipid and CG analyses described below.
Each stomach content sample was dried for 16 h at 60°C in a forced-
draft oven, then cooled under vacuum with dry-rite, and weighed.
Neutral lipids were extracted with petroleum ether (see methods in
Walford 1980, Brower ei ni. in prep.) and CGs from the lean residue
with ethanol (see methods in Brower et al. 1975) . Gravimetric

29
techniques were used to determine total lipid content per stomach
sample, and spectrophotometric techniques to determine the
concentration (p.g/0.1 g dry weight) and weight of CGs in each stomach
(Brower et al. 1975, Brower et al. 1985).
It is unlikely that digestive enzymes or food items in the stomach
interfered with my lipid and CG determinations. First, the small
intestine is the major site of lipid metabolism and absorption in
rodents; some lipolysis but no absorption occurs in the stomach (Booth
et al. 1961, Harrison and Leat 1975) . Second, CGs are metabolized
poorly if at all in the stomachs of £_*_ melanotis and EL. maniculatns as
well as in that of birds (Marty 1983, Brower et al. 1988, Brower et al.
1985) . Third, virtually all ingested bait, which contained lipids and
also could have interfered with the spectroassay (Brower et al. 1985),
was physically removed from the stomach samples prior to the
quantitative analyses; the oats and peanut butter had formed into a
hard clump, which was easy to separate from the other food.
I used the one-way ANOVA and Scheffé F-test to compare the five
species' stomach samples in terms of 1) the frequency of occurrence of
monarch material, 2) dry weight, 3) lipid weight, 4) percent lipid, 5)
gross CG content, and 6) CG concentration. The percent lipid data were
arc-sine transformed (Zar 1984) for the analysis. I also performed a
two-way ANOVA on the results from melanotis to examine the effects
of colony proximity and sex on the same 5 variables.
Comparison of the Understory Vegetation in the Trapping Grids
Numerous studies suggest that microhabitat features (i.e., species
composition and density of understory vegetation) strongly influence

30
the distribution in space and time of species of mice (e.g., Brown
1964, Hansen and Fleharty 1974, M'Closkey and Lajoie 1975, Huntly and
Inouye 1987) . I thus hypothesized that the preponderance of P.
melanotis inside the colonies may have been influenced by unique plant
species compositions found in the colony sites. To test this, I
described and compared the understory vegetation in the eight trapping
grids, employing standard methods (Risser and Rice 1971, Whittaker
1970, Mueller-Dombois and Ellenberg 1974). Voucher specimens of all
plant species were identified by Hilda Flores and Jose Luis Villaseñor
of the MEXU-herbarium at the Universidad Nacional Autonoma de México.
I used a random stratified sampling scheme to select 12 out of the
total of 45 quadrats per grid; each grid was divided into 4 equal
blocks, and then within each block I randomly selected 3 quadrats.
Meter-squared areas in the center of each quadrat were sampled and the
frequency and percent coverage of each plant species was determined by
visual inspection. Then, I determined the importance percentage of
each species by summing its relative dominance (using the percent
coverage data) and relative frequency and dividing by 2. Finally, I
used the following equation to make pair-wise comparisons between the 8
trapping grids:
Percent similarity = 1 - 0.5 X IPa^-Pb^l
where p is the importance percentage divided by 100, a and b refer to
the two grids being compared, and i to each species.
To determine whether the sites where monarchs form colonies possess
unique microhabitat characteristics, I categorized each percent
similarity value into one of three groups, based on the proximity of
the two grids in the pair-wise comparison to a monarch colony: inside

31
inside, outside vs.. outside, and inside 3£S. outside. I used the
Kruskal-Wallis test to compare the three groups and expected the inside
xs. inside PS values to be consistently higher.
Aggressive Relations Among the Different Mouse Species
I also hypothesized that the preponderance of £_^ melanot-i s inside
the monarch colonies could have been due to E^. melanotis aggressively
excluding the other species of mice. To test this, I studied the
aggressive relations of E_^ melanotis. E^. a^. hylocetes. R. sumichrasti.
and iL. salvus by staging 70 intrasexual, paired interactions between
a reproductively active E^. melanotis in its home range inside a monarch
colony (henceforth, resident) and a reproductively inactive,
conspecific or heterospecific from outside the colony (henceforth,
intruder). I thus had a total of eight pairing combinations: 1
conspecific and 3 heterospecific ones for each sex. Sexually active
residents and inactive intruders were used because this pattern
mimicked my trapping results most closely. I selected mice of
comparable body mass (± 2 g) for the conspecific pairings; this was not
possible for the heterospecific pairings, given the large size
differences (Table 3-4). I predicted that resident E^. melanotis would
dominate intruders. Recently Wolff et al. (1983) showed that dominance
among 2 similarly sized Peromyscus species is site-specific and not
species-specific. That is, an individual is more likely to win an
aggressive encounter with a conspecific or heterospecific of the same
sex when in its own home range.
Paired encounters were staged between 0800 and 1100 h in a clear,
plexi-glass arena (32 x 32 x 48 cm), with an open bottom and top,

32
following the methodology of Wolff et al. (1983) . The animals were
thus on natural substrate. Each trial was carried out at the trap
station where the resident was captured the previous night, and
consisted of a 30 s acclimation and a 5 m interaction period. During
the acclimation period, each mouse was placed in the arena on opposite
sides of a removable cardboard partition. The interaction period was
video-taped and began immediately after the partition was removed.
Because it was difficult to distinguish resident and intruder during
conspecific pairings, I painted a 0.5 cm diameter, red or gold spot of
nail polish on the pelage between each animal’s ears. The mice did not
appear to have been disturbed by the nail polish. The assignment of
gold or red for resident or intruder was determined randomly for each
mouse. This color labelling was not necessary for the heterospecific
pairings.
For each interaction period, I recorded the number of non-offensive
and offensive approaches, attacks and retreats exhibited by the
resident and intruder (Table 3-2). To determine the outcome of the
encounters, I modified slightly the criteria of Wolff et al. (1983):
1) Winner: The animal with the fewest number of retreats.
2) Draw: Both animals had a similar number of retreats (i.e., the
total number by each individual differed by < 2).
3) Ha aggression: Neither animal displayed any aggression towards
the other.
To determine whether resident £_*. melanotis won a greater number of
trials than expected by chance in each of the eight pairing
combinations, I used the binomial test. To compare the number of non¬
offensive approaches, offensive approaches, attacks and retreats

33
Table 3-2. Partial ethogram of behaviors displayed by malanotis. p. a.
hylocetes, sumichrasti and M^. salvus during paired interactions .
Many of the behavioral descriptions are modified from Eisenberg's (1968)
more complete ethogram of Peromyscns.
Behavior
Description
Non-offensive approach Mouse approaches opponent slowly with body
Offensive approach
contours relaxed.
Mouse approaches opponent with its body
stretched out. Its body shows rigidity and
muscular tension as it moves. Frequently, the
tail is rigid and the ears and vibrissae are
extended forward.
Attack
a. Rush
Mouse suddenly accelerates approach towards
opponent, usually following an offensive
approach.
b. Attack leap
Mouse leaps into the air at opponent, striking
it with its limbs and/or body.
c. Lunae
Mouse strikes at a nearby opponent with
forepaws, but keeps feet planted on substrate.
Retreat
a. Escape leap
Attacked mouse flees from attacking mouse with
wild and erratic leaps.
b. Fliaht
Attacked mouse runs away from a rush, attack
c. Flight after fight
leap or lunge.
Mouse flees following a locked fight, which is
when the two animals lock their ventrums
together while rolling on the ground.

34
exhibited by the resident and intruder, I used the Wilcoxon matched-
pairs test (Zar 1984). For this analysis, I combined the results from
both sexes, thereby reducing the number of pairing combinations to
four, because no intersexual difference was apparent.
Results
Density of Monarchs on Forest Floor Inside Colonies
An average of 6.5 monarchs/m^ was added to the forest floor inside
a monarch colony each night (Table 3-3). Perhaps more significant is
that 83% of these monarchs had intact abdomens and thoraces (i.e., were
live, moribund and/or sampled) and thus still possessed the majority of
their tissues and body water. An average total of 0.93 gram (dry
weight) of lipid-rich tissue per m^, or 9,300 g per hectare, was
available to foraging mice. This is an underestimate of the total
amount of monarch material available to mice because I did not include
the dead monarchs that had accumulated on the forest floor prior to
sampling or the live ones roosting on low-lying understory vegetation
(Alonso et al. in press).
Irappability and. ..Reliability of Data
Minimum trappabilities for mice marked on Sierra Chincua between
1985 and 1986 were extremely high: for £^. melanotis. 90% inside and 88%
outside the monarch colonies; for £^_ sumichrasti. 93%; for a.
hylocetes. 89%; for 1L_ alstoni. 91%; and for nu. salvus. 82%.
Trappability estimates for the latter 4 species were from individuals
trapped on grids outside colonies. Thus, the MNA estimates were
considered to be accurate within 5% (Hilborn ££. nl., 197 6) . My

35
Table 3-3. Estimated density of live, moribund and bird-damaged
monarchs (mean no./m^; ± S.E.), and quantity of monarch tissue (mean
no. grams/m2; dry weight) on the forest floor at the beginning of
the night inside a 0.44 ha monarch colony on Sierra Chincua.
Samples were made over six nights between 15 and 20 January 1988.
State of
monarch
Density
(no. per m2)
Tissue (g; d.w.)
per m2
live
2.3
(±0.9)
0.38
moribund
1.1
(±0.3)
0.18
cuticle pierced
1.8
(±0.1)
0.30
abdomen removed
1.1
(±0.2)
0.04
abdomen deviscerated
0.1
(±0.1)
0.02
thorax contents eaten
0.1
(±0.1)
0.02
Total
6.5
0.93

36
treatment of the trapping grids as independent samples was
substantiated because 1) I recorded only one instance of a mouse moving
between trapping grids (an adult male melanotis moved from grid 7
(inside colony) to 8 (outside colony) during 1985), and 2) I removed
(i.e., snap-trapped) nearly the entire population of mice from grid 2
at the end of the 1986 season and thus diminished the influence of
temporal pseudoreplication (Hurlbert 1984) on my results from the same
grid during the 1987 season. Finally, my use of terrestrial trapping
results was justified because no mice were caught in the arboreal traps
set during 1988.
Species Composition of Mice Inside and Outside Monarch Colonies
I marked a total of 634 adult and juvenile mice in my mark-
recapture studies on Sierra Chincua between 1985-88. Of this total,
71% were melanotisr 11% were £_». hylocetesf 8% were R.
aumichrasti, 8% were n*. nu. salvus, 2% were M_*. alatoni. Clearly, e^.
melanotis was the most common species.
All species differed greatly in size, except for E_. a^. hylocetes
and 1L_ alstoni (Table 3-4). Peromyscus melanotis and nu salvus were
the only species with apparent sexual dimorphisms, with the former
having larger females and the latter larger males.
The species composition of mice on grids inside and outside monarch
colonies differed markedly and in a similar manner on both Sierra
Chincua and Sierra Campanario (Table 3-5). At least 90% of the mice
captured on grids inside colonies were E_*. melanotis. whereas only 45 to
69% of those captured on grids outside colonies were E_^ melanotis.
These findings were consistent both among and within seasons.

Table 3-4. Mean body weights (±S.E., in grams) of adult mice of 5 species caught inside and outside
monarch colonies. Mice were weighed during mark-recapture studies on Sierra Chincua between 20
January and 16 March, 1985-1988; an average weight was calculated for those mice caught more than
once. Weights of pregnant female £_*. melanotis were excluded; no pregnant females of the other species
were trapped. However, because most breeding, female £_*. melanotis were caught more than 4 times, I
was able to determine their mean weight before and/or after their pregnancy. Sample sizes are in
parentheses. I compared the weights of both sexes with the unpaired t-test (two-tailed).
Gender
R. a. hylocetes
N. alstoni
M. m
salvus
P. melanotis
R. sumichrasti
male
36.2 ±0.5
37.2 ±1.1
29.1
±0.9
18.8 ±0.1
12.7 ±0.3
(40)
(4)
(28)
(220)
(21)
female
35.9 ±0.9
34.4 ±1.5
26.6
±0.5
21.2 ±0.2
13.6 ±0.4
(30)
(8)
(24)
(227)
(32)
t value -0.26 -1.29 -2.29 11.66 1.64
0.793 0.226 0.026 0.0001 0.107
P value

Table 3-5. Total number of mice (adult and juvenile) of 5 species captured inside and outside monarch
colonies on Sierra Chincua and Sierra Campanario, and the percentage of these mice that were P.
melanotis. I show results from mark-recapture studies on Sierra Chincua between 20 January and 16
March, 1985-88, and from snap-trap studies on Sierra Campanario between 3 February and 10 March, 1986.
Total number
of individual mice
captured
Percent
Year
Grid
R. melanotis
£. a. hvlocetes
R. sumichrasti u
. ¿l*. alstoni
M. m... salvos
R. melanotis
SIERRA CHINCUA
inside colony
1985
7
58
2
1
0
l
93.6
1986
2
99
5
4
0
2
90.0
1987
2
153
2
4
0
li
90.0
5
78
1
3
0
2
92.8
1988
1
39
0
1
0
0
97.5
Total
427
1 3
13
0
1 6
91 . 0
1985
gut,3ide colony
62.5
69.1
b
8
1 D
47
9
6
7
0
1
0
4
1986
1
26
12
9
2
9
44.8
3
36
14
10
7
11
46.2
1987
4
36
15
4
2
8
55.4
1988
2
15
5
4
2
4
50.0
Total
175
5 8
4 0
1 4
36
54.1
LO
co
Continued.

Table 3-5—Continued.
1986 2 Feb.
10 March
Total
1986 2 Feb.
10 March
Total
18 0
37 0
55 0
8 1
15 3
23 4
SIERRA CAMPANARIO
inside colony
0
l
1
outside colony
2
2
0
0
0
0
0
0 100.0
0 97.4
0 98.2
1 66.0
5 60.0
62.1
4
0
6
LO

40
Moreover, they are greatly strengthened because my comparisons on
Sierra Chincua included grids that were inside a colony during some
years and outside a colony during others (see grid 2 during 1986-88,
and grid 1 during 1986 and 1988; Figure 3-lb).
Of the mice captured inside colonies, only 2^. melanotis individuals
established residency; they were trapped on average for 3.7 consecutive
weeks (Table 3-6). All individuals of the other species were trapped
only once inside the colonies, except for one cl salvus that was
caught for 2 consecutive weeks. Thus, even though immigrants of all
five species encountered the monarch colonies, only individual P.
melanotis remained and established residency. This contrasts with the
results from outside the colonies, where individuals of all species
remained on average for at least 2.6 consecutive weeks. In fact, the
mean duration of residence on grids outside versus those inside monarch
colonies was significantly greater for £_*. hylocetes r R. snmir.hrasti
and salvus. but not for melanotis (Table 3-6) .
Female £_*. melanotis trapped inside colonies exhibited the highest
levels of reproduction, both on Sierra Chincua and Sierra Campanario
(Table 3-7). Female £^_ melanotis trapped outside colonies also showed
relatively high reproductive activity compared to the other four
species. However, the majority of the females of all five species
outside the colonies did not initiate breeding until late February and
March. This pattern contrasts sharply with that of female melanotis
inside colonies on Sierra Chincua, which were breeding extensively by
20 January (Figure 3-3a).

Figure 3-1. (A) Location of mark-recapture study grids for 4 years from 1985-88 (black squares
inside hatched area) in relation to topography and drainage patterns on Sierra Chincua (adapted
from Anon., 1981). Contour interval = 100 m. (B) The hatched area has been expanded and
presented separately for all four trapping years. Trapping grids are drawn to scale and each grid
is given a unique number. Solid grid squares indicate which grids were trapped each year. Colony
locations are indicated by the stippled area. Even though a colony formed in grid 7 during 1986,
I did not conduct a mark-recapture study there during that year.

-t
ro
Continued.
19*40

Figure 3-1—continued.

Table 3-6. Mean number of weeks (±S.E.) that adult mice of the 5 species remained on the grids inside
(n = 4) and outside (n = 5) monarch colonies on Sierra Chincua. Only those mice first trapped between
20 January and 23 February, 1985 to 1987, were included in the analyses. Sample sizes are in
parentheses. Comparisons are made between mice from inside and outside colonies with the Mann-Whitney
U test (normal approximation).
Colony
proximity
EL_ melanotis
P. a. hylocetes
R. sumichrasti
alstoni
salvus
inside
3.7 ±0.2
(210)
1.0 ±0.0
(11)
1.0 ±0.0
(13)
1.1 ±0.1
(14)
outside
3.8 ±0.3
(77)
3.2 ±0.3
(45)
2.6 ±0.3
(31)
3.5 ±0.5
(11)
2.7 ±0.4
(29)
Z value*
0.020
3.307
3.670
3.190
P value
0.9838
0.0009
0.0002
“ . “
0.0014
* corrected for ties.

Table 3-7. Percentage of adult females of 5 species that were pregnant and/or lactating. I combined
the data for all females trapped in mark-recapture studies on Sierra Chincua between 20 January and 16
March, 1985-88, and in snap-trap studies on Sierra Campanario on 3 February and 10 March, 1986.
Percentage reproductive (total number captured)
Proximity
to colony
£. melanatia
P. a. hvlocet.es
R. sumichrasti
¿L. alstoni
M. m
. salvus
SIERRA CHINCUA
inside
88.2
(271)
0.0 ( 4)
11.1 ( 9)
-.- (0)
0.0
( 6)
outside
54.5
( 55)
4.2 (24)
5.0 (20)
20.0 (5)
7.1
(14)
SIERRA CAMPANARIO
inside
82.6
( 23)
-.- ( 0)
0.0 ( 1)
-.- (0)
-.-
( 0)
outside
37.5
( 8)
0.0 ( 2)
0.0 ( 1)
-.- (0)
0.0
( 2)
vn

46
Demographic Responses of P. melanotis to the Monarch Colonies
Initial population densities (i.e., those during early January) of
mice on the grids inside and outside the colonies were similar, except
on grid 2 in 1986 and 1987, where female density was already high
(Figure 3-2). However, subsequent population densities increased more
than two-fold on the grids inside colonies, while remaining roughly
constant on the grids outside colonies. Peak densities of both- sexes
reached 50 to 97 mice per ha on grids inside colonies, and only 13 to
32 mice per ha on grids outside colonies. A comparison of the
proportion of adult and juvenile mice that established residency inside
colonies suggests that adult immigration, rather than juvenile
recruitment, was primarily responsible for the increased density over
time (Table 3-8) . In the grids outside of the colonies, juveniles and
adults occurred in similar proportions. Immigration of adults was
consistently female-biased to grids inside colonies and male-biased to
grids outside colonies (Table 3-9; Figure 3-2).
Whereas females inside colonies showed high levels of reproduction
throughout the winter, those outside colonies did not initiate high
levels of reproduction until late February (Figure 3-3a). There were
also many more new juveniles caught inside colonies (Figure 3-3b). A
two factor ANOVA revealed a significant effect of colony proximity (F =
19.21, df = 1, P = 0.0032), time (F = 7.67, df - 7, P = 0.0001), and
colony proximity x time (F = 4.86, df = 7, P = 0.0003) on the number of
juveniles captured. The significant interaction indicates that the
number of new juveniles increased with time on the grids inside but not
on those outside colonies. The large number of juveniles inside

Figure 3-2. Minimum number alive estimates of male and female P.
melanotis present on the 4 trapping grids inside and 5 outside monarch
colonies from 20 January to 16 March, 1985 to 1987. The numbers to the
right of each line indicate the specific trapping grid. Because grid 2
was trapped for two seasons, I distinguish the results from 1986 as 2,
and those from 1987 as 2*.

Minimum Number Alive
48
January
February
I
March
4* a> roen

49
Table 3-8 . Proportion of juvenile and adult me la not is remaining
for greater than one week on 4 grids inside and 5 grids outside
monarch colonies on Sierra Chincua. Only those mice that were first
trapped between 20 January and 23 February, 1985-87, were included.
Total
number
Age-class Number of weeks on grid trapped
one > two
inside
colony
juvenile
0.72
0.28
47
adult
0.24
0.76
210
outside
colony
juvenile
0.26
0.74
27
adult
0.35
0.65
77

50
Table 3-9. Percentage of melanotis that established resideny on
grids inside and outside of monarch colonies that were female.
Year
Inside
colony
Outside colony
Grid
no.
Female
%
Grid
no.
Female
%
1985
7
58.6
6
33.3
8
41.2
1986
2
66.1
1
44.4
3
50.0
1987
2
60.1
4
46.2
5
58.3
Mean ±S.E.
60.8 ±1.8
43.0 ±2.8

Figure 3-3. (A) Mean (±S.E.) percentage of female melanotis in
reproductive condition on 4 grids inside and 5 grids outside monarch
colonies from 20 January to 16 March, 1985 to 1987. Sample sizes are
written alongside each symbol. (B) Mean (±S.E.) number of new juvenile
P. melanotis trapped inside and outside monarch colonies on the same
grids and over the same time period as in A.

Mean (±S.E.) no. of new juveniles Mean (±S.E.) percent reproductive
52
A.
B.

53
colonies strongly suggests that the females were able to successfully
wean their young.
Forty-one £_^ melanotis fit my criteria for use in the home range
size determinations (in m2; mean ±S.E.). Home ranges for males inside
(274.4 ±96.1, n = 7) and outside (425.9 ±89.8, n = 8) colonies did not
differ significantly (Z = 1.46, P = 0.144), nor did those for females
inside (158.9 ±30.9, n = 16) and outside (187.7 ±47.5, n = 10) colonies
(Z = 0.24, P = 0.811]. However, when the results from grids inside and
outside colonies were combined, males had significantly larger home
ranges than did females (355.2 ±66.4 and 170.0 ±25.9, respectively; Z =
2.64, P = 0.0084) .
Nonpregnant, resident females were significantly heavier than
resident males in grids inside and outside colonies (Table 3-10). The
colony proximity x sex interaction was not significant. Neither sex
exhibited evidence of weight loss or deteriorating health over the
winter.
Stomach Content Analyses
Detailed visual examinations indicated that the stomach samples of
E^. melanotis collected inside colonies consisted solely of monarch
material (Figure 3-4). I also encountered variable amounts of monarch
material in the stomach of mice from grids outside colonies, with P.
melanotis having significantly more than the other species, and 1L m.
salvus having significantly less. Live and/or dead monarchs do occur
on the forest floor outside colonies, albeit at a relatively low
density (Glendinning unpub1. observ.).

54
Table 3-10. Comparison of the body weights (mean ±S.E., in
grams) of resident £. melanotis trapped inside and outside
monarch colonies between 27 January and 23 February, 1985-87, on
Sierra Chincua. As mice were caught at least two times, an
average weight was calculated for each mouse. Pregnant female
weights were excluded. However, because most breeding females
were caught more than 4 times, I was able to determine their
mean weight before and/or after their pregnancy. Sample sizes
are in parentheses. The results of a 2-factor ANOVA are
presented.
Proximity
to colony
Males
Females
inside
18.5 ±0.3
(75)
21.2 ±0.3
(110)
outside
17.7 ±0.4
(48)
19.8 ±0.5
(31)
Source of
Variation
df F
-ratio
P
Colony
Proximity
1
8.89
0.0031
Sex
1
45.57
0.0001
Interaction
1
5.11
0.3937

Figure 3-4. Mean (±S.E.) occurrence of monarch material in the stomach contents of 74 L.
melanotis collected inside and 38 £j_ melanotis. 17 sumichrasti. 15 L ái. hyloceteSr 5 Ü,.
alstoni. and 10 ÍL nu. salvus collected outside monarch colonies on Sierra Chincua and Sierra
Campanario. I present results of a one-way ANOVA. Different subscripts (a, b, c, d) indicate
significant differences among means (P < 0.05; Scheffé F-test).

a
P. melanotis R. sumichrasti
â–  inside colony
W outside colony
Kruskal-Wallis H
P = 0.0001
c
P. a. hylocetes
140.7
d
N. a. alstoni
M. m. salvus

57
Even though the dry weights of L. hylocetes, u_._ alstoni and
1L. EU salvus1 stomach contents were significantly higher than those of
£_^ melanotis (collected outside colonies) and sumichrastif all
species had statistically equal amounts of lipid (Table 3-11) .
However, melanotis had a significantly higher percentage of lipid,
CG content and CG concentration than did the other species. Together
with the results in Figure 3-4, these results suggest strongly that on
grids outside colonies, E* melanotis consumed substantially more
monarch material than did the other four species.
Two way ANOVA's on the stomach samples from £_*. melanotis inside and
outside colonies demonstrated significant effects of colony proximity
and sex on dry weight and lipid weight, and of colony proximity on
percentage of lipid (Table 3-12). The same effects were significant
for CG concentration, but in the opposite direction (i.e., males both
inside and outside colonies had higher values). Cardiac glycoside
content was not influenced significantly by colony proximity or sex.
These results suggest that E^. melanotis inside colonies fed exclusively
on monarch material and thus accumulated large quantities of lipid,
amounting to approximately 43% of the total dry mass of their stomach
contents. However^ even though conspecifics outside colonies ingested
less monarch materials, the CG concentration in their stomach samples
was significantly greater. Females both inside and outside colonies
had greater dry weights and lipid weights; this intersexual difference
was greatest for mice from inside colonies.

Table 3-11. Comparison of stomach content samples from 5 mouse species for the variables dry weight,
lipid weight, percent lipid, and CG content and concentration (mean ± S.E.). All mice were collected
outside monarch colonies on Sierra Chincua and Sierra Campanario. Interspecific comparisons are
made with one-way ANOVA and Scheffé F-tests. Subscripts, a and b, indicate significant differences
among species within each column (P < 0.05).
Dry
Lipid
Lipid
CG
weight
weight
as %
CG content
concentration
Species
N
(g)
(g)
dry weight
(|lg/0.1 g; d.w.)
P. melanotis
38
0.14 ±0.02
b
0.03 ±.004
a
25.6 ±1.5 a
105.7 ±10.4
a
100.1 ±14.4 a
E- sumichrasti
17
0.13 ±0.01
b
0.02 ±.002
a
14.8 ±1.2 b
1.7 ± 0.8
b
1.3 ± 0.6 b
P. á.. hvlocetes
15
0.44 ±0.07
a
0.04 ±.007
a
9.8 ±1.2 b
2.5 ± 1.1
b
1.3 ± 0.7 b
N. alstoni
5
0.39 ±0.09
a
0.04 ±.007
a
4.1 ±1.8 b
o.o b
0.0 b
M. nu. salvus
10
0.34 ±0.05
a
0.02 ±.002
a
7.0 ±1.1 b
o.o b
0.0 b
F-ratio
17.78
3.13
23.94
22.21
14.20
df
4
4
4
4
4
P-value
0.0001
0.0191
0.0001
0.0001
0.0001

Table 3-12. Comparison of stomach content samples from male and female P. melanotis collected inside and
outside monarch colonies for the variables dry weight, lipid weight, percent lipid, and CG content and
concentration (mean ± S.E.). Mice were trapped on Sierra Chincua and Sierra Campanario. The effects of
sex and colony proximity on each variable were determined with a two-factor ANOVA. NS P > 0.05, * P <
0.05, ** P < 0.005, P < 0.0005.
Proximity
to colony
Sex
N
Dry
weight
(g)
Lipid
weight
(g)
Lipid
as %
dry weight
CG content
(^g)
CG
concentration
((lg/0.1 g; d.w.)
inside
female
37
0.32 ±0.04
0.14 ±0.02
43.6 ±2.0
126.4 ±14.3
52.5 ± 6.7
male
37
0.20 ±0.02
0.08 ±0.01
42.9 ±1.4
91.3 ± 9.5
62.8 ± 7.3
outside
female
14
0.18 ±0.03
0.05 ±0.007
26.9 ±2.4
96.1 ±14.8
74.4 ±17.9
male
24
0.11 ±0.02
0.03 ±0.005
24.9 ±2.0
111.2 ±14.1
114.8 ±20.3
Source of
Variation
df
F-ratios
Colony proximity
1
8.95 **
19.45 ***
70.31 ***
0.48 NS
-8.75 **
Sex
1
6.82 *
5.53 *
0.46 NS
0.13 NS
-4.14 *
Interaction
1
0.65 NS
1.56 NS
0.12 NS
2.98 NS
1.47 NS

60
Comparison of the Understory Vegetation in the Eight Grids
My results suggested that the understory vegetation on the grids
inside monarch colonies did not differ significantly from that on the
grids outside colonies. The pair-wise, percent similarity values from
inside 2S.. inside comparisons were not significantly higher than those
of outside 5¿s.. outside and inside xs.. outside comparisons (mean ±S.E.
in respective order = 47.5 ±4.0, 40.4 ±6.0 and 48.4 ±3.6; Kruskal-
Wallis H value = 1.135, P = 0.567; Appendix A). The percentage
importance values of the 34 understory plant species encountered on the
8 grids are in Appendix B.
Aggressive Relations Among the Different Mouse Species
The results of the 70 pairings in which a win/loss decision was
made are presented in Table 3-13. The number of wins by resident P-
melanotis was significantly greater than expected by chance in
conspecific pairings, but not in any of the heterospecific pairings.
In fact, resident melanotis lost the vast majority of pairings with
E-*. hylocetes and 1L nu. salvus. whereas the results of their pairings
with fL_ sumichrasti were equivocal (Figure 3-5). In conspecific
pairings, the resident IL melanotis exhibited significantly more non¬
offensive and offensive approaches and attacks (rushes) than did the
intruders (Figure 3-6) . Interactions between IL melanotis and IL a.
hylocetes were the least aggressive; the majority of the former's
retreats followed non-offensive approaches by the latter. Most of EL.
nu salvus' attacks involved lunges following non-offensive approaches
by melanotis. Interactions between ÍL melanotis and IL_ sumichrasti
were the most balanced, with the former exhibiting more offensive

61
Table 3-13. Outcomes of intrasexual, paired trials between a
resident £. melanotis and a visiting conspecific or heterospecific
in which a win/loss decision was made. The winner of a trial had
the fewest retreats. The first animal of each pair is the resident
and the second animal is the visitor. The number of wins by the
resident and visitor were compared with the binomial test.
Species
Sex
No. of
Trials
No.of
Wins
Binomial
Prob.
£. melanotis
F
10
9
0.039
E. melanotis
F
1
E. melanotis
M
10
9
0.039
£. melanotis
M
1
£. melanotis
F
9
3
0.870
B. sumichrasti
F
5
£. melanotis
M
8
5
0.453
R. sumichrasti
M
2
£. melanotis
F
8
1
0.070
cu. aalaus
F
7
£. melanotis
M
8
0
0.008
Hl. nu. salvus
M
8
£. melanotis
F
9
0
0.004
P. a. hylocetes
F
9
£• melanotis
M
8
0
0.016
£. a. hylocetes
M
7

Figure 3-5. Percentage of resident wins, intruder wins, draws, and no aggression outcomes in 70
intra- and interspecific pairs of melanotis (£_*. hlJ , fL_ sumichrasti (B^ 2+), nu. salvus (M.
HL_ s.) , and fL. hylocetes (P. a. h.) . N = number of pairings for both sexes combined.

Percent
0 No Aggression
â–¡ Draws
0 Intruder Wins
â–  Resident Wins

Figure 3-6. Mean (±S.E.) number of non-offensive and offensive approaches, attacks and retreats
exhibited by resident E^. melanotis and intruder conspecifics or heterospecifics (fL*. sumichrasti.
M. m. salvus. or E,. hylocetes) during the 5 minute trials. For each behavior, I compare the
number of times it was exhibited by the resident and intruder with the Wilcoxon matched-pairs
signed-rank test; an asterisk indicates significance at the 0.05 level. Species abbreviations are
as in Figure 5.

Mean No. of Occurrences Mean No. of Occurences
6 0-
P. m.
x P. a. h.
5.0-
Non- Offensive
Offensive Approach
Approach
Attack Retreat
Non- Offensive Attack Retreat
Offensive Approach
Approach

66
approaches and the latter more attack leaps, commonly when the P.
melanotis was oriented towards another object in the arena. These
results suggest that resident melanotis could not prevent
individuals of the other species from foraging and establishing
residency inside monarch colonies.
Discussion
Comparison of the Responses of the Five Mouse Species to the
Monarch Colonies
Even though melanotis was the most common species of mice on
Sierra Chincua and Sierra Campanario, my results suggest that
disproportionately large numbers of this species immigrated into the
colonies and established residency. Individuals of the other species
also immigrated into the colonies, but none of them established
residency, with the exception of one el*. salvus. These results
contrast with those from grids outside colonies, where all five species
of mice commonly established residency. Thus it appears that all
species but melanotis left the monarch colonies soon after
encountering them.
The diets of melanotis inside colonies consisted almost entirely
of monarch material and averaged 43% lipid. Females had more than 2.5
times more lipid than conspecifics and heterospecifics outside
colonies. My data suggest that the energy and nutrients the mice
derived from the monarchs enabled them to initiate high levels of
winter reproduction, confirming the findings of Brower et al. (1985).
Similar responses to natural and experimentally created food abundance
in winter have been reported in other rodent populations (Linduska

67
1942, Watts 1970, Andrzejewski 1975, Gashweiler 1979, Taitt 1981,
Jenson 1982, Eriksson 1984, Briggs 1986). Moreover, diets with
comparable levels of fat (40 to 60%) were found to stimulate
reproduction and reduce infant mortality in laboratory rats (Scheer et
al. 1947, Innami et al. 1973).
Even though the £^_ melanotis and other species outside the colonies
bred during the winter, the vast majority did not begin until late
February and it is notable that a much larger percentage of the female
melanotis initiated reproduction than did females of the other
species. This may be due to supplemental nutrients derived from
monarch material, which augmented their winter diet.
Rodent populations do not normally breed in winter in temperate
and/or high altitude regions (Rintamaa et al. 1976, Millar 1984,
Bronson 1985, Kenagy and Barnes 1988), and previously studied
populations of the same species studied herein in high-altitude regions
of the Mexican Transvolcanic Range are no exception (Villa 1952, Canela
and Sánchez 1984, Robertson 1975, V. Sánchez pers. comm.). a variety
of factors interrelate to produce this pattern, including energy,
nutrients, photoperiod, humidity, temperature and social factors.
However, energetic constraints are thought to play a major role, owing
to the high costs of thermoregulation, searching for scarce food, and
reproduction (Schipp et al. 1963, Sadlier et al. 1973, Stebbins 1977,
Millar 1979, Porter and McClure 1984, Bronson and Perrigo 1987, Perrigo
1987). Apparently, gaining access to monarchs allows melanotis to
overcome these energetic constraints and breed during the winter.

68
Demographic Responses of P. melanotis to the Monarch Colonies
Peak MNA estimates of mouse densities inside colonies were much
higher (50 to 97/ha) than those outside (13 to 32/ha) and those
reported for other Peromyscus populations whose diets had been
supplemented artificially (13 to 59/ha; Fordham 1971, Hansen and Batzli
1978, Gilbert and Krebs 1981, Taitt 1981, Briggs 1986, Young and Stout
1986, Wolff 1986). The unusual densities inside colonies is
attributable almost exclusively to adult immigration. Such high
densities and rates of immigration must have created high levels of
intraspecific aggression, associated with the defense of core areas
against intruders (Watson and Moss 1970, Wolff et al. 1983, Wolff
1985). In this study, I found same-sexed, resident melanotis to be
extremely aggressive towards conspecific intruders (Figure 3-6, Table
3-13).
Even with this aggression, home ranges of E^. melanotis inside
colonies were not significantly smaller than those of conspecifics
outside colonies. This suggests that neither food abundance nor
population density are major factors determining home range size in
these mice. In other studies of £_*. leucopus and maniculatus
populations, workers have drawn both similar (Stickel 1960, Sheppe
1966, Hansen and Batzli 1978, Wolff 1985, Wolff 1986) and contrasting
(Bendell 1959, Taitt 1981) conclusions. Clearly, the factors governing
home range size in E^_ melanotis deserve further study, particularly
since the average home range sizes of both sexes are considerably
smaller than those reported for other Peromyscus species (Taitt 1981,
Wolff 1985, Wolff 1986, Vessey 1987) .

69
In this study and that of Brower s£. al. (1985), immigration to
monarch colonies was female-biased, and the females inside colonies
were significantly larger than the males both inside and outside
colonies. Whereas several workers have reported similar female-biased
responses to high quality habitats in other Peromyscus populations
(Bowers and Smith 1979, Fordham 1971), many other workers have not
found such a response (review in Vessey 1987). The reason for the
preponderance of females inside the colonies is unclear. Because
females had smaller home ranges than did males, they may have been able
to pack themselves more tightly within colonies without extensive home
range overlap. Second, the females may limit male immigration as
suggested for several other Peromyscus species (Metzgar 1971, Bowers
and Smith 1979) and Tamias striatus (Wolfe 1966, Brenner et al. 1978).
Unfortunately, I did not examine the intersexual, agonistic relations
of Ej. melanotis. Third, it is possible that the natural sex-ratio on
Sierra Chincua was female-biased. However, this is unlikely given that
the mean percentage of females (+S.E.) in my 23 captive litters of P.
melanotis (from 15 female and 14 male adult parents) collected on
Sierra Chincua was 50.1 ±14.0.
The high densities of breeding females inside colonies also may
have contributed to the low levels of juvenile recruitment. Other
workers have reported negative correlations between juvenile densities
and those of breeding females in Peromyscus populations, and they
concluded that breeding females had aggressively excluded young mice
from their home ranges (Hansen and Batzli 1978, Galindo and Krebs
1987). Clearly, the role of aggression by breeding females towards

70
conspecific males and juveniles in mediating access to high quality
resources deserves further investigation.
Consumption of Monarchy
I was surprised by the relatively large amounts of monarch material
present in the stomachs of £*. melanotis collected outside colonies.
Brower et. ¿lL. (1985) also found monarch material in the stomachs of P.
melanotis from outside colonies, but did not quantify the amount.
There are two ways in which these mice could have accessed monarch
material: 1) by making nightly forays into the nearby colony; and/or 2)
by scavenging within their own home ranges on dead or moribund
monarchs. Many butterflies die outside colonies because of bird
predation, exhaustion due to inadequate lipid reserves, and dehydration
(Walford 1980, Brower 1985) . The second possibility is most likely
because my trappability estimates in grids outside colonies were so
high and I did not record any mice moving from grids outside to ones
inside colonies. Therefore, there must have been a substantial number
of dead and/or moribund monarchs on the forest floor each night in the
grids outside of the colonies. This implies that all species of mice,
except for £. melanotis. not only avoided the colonies, but also
avoided the dead and/or moribund monarchs in the grids outside of the
colonies.
I can offer two explanations for why the stomach contents of P.
melanotis outside monarch colonies contained significantly higher
concentrations of CGs than did those of conspecifics inside colonies.
First, because the density of monarchs is so much higher inside
colonies, mice foraging there could feed selectively on monarchs having

71
low levels of CGs with limited searching costs. Caged melanotis are
able to distinguish between monarchs with low and high levels of CGs
(Chapters 4 and 5). For those mice outside colonies, the caloric costs
of extensive searching for low CG monarchs would have been much higher
and thus may not have been energetically worthwhile. Second, when
offered hydrated monarchs (i.e., ones that are live, moribund, or
recently dead), captive melanotis most commonly consume the
abdominal material by first discarding the cuticle and then eating the
internal tissues (Chapter 3). Monarch cuticle is known to contain high
concentrations of CGs (Brower et al. 1988). In contrast, when offered
desiccated monarchs, captive £^_ melanotis eat both the cuticle and
contents, apparently because the cuticle became tightly bound to the
internal tissues during the desiccation process (Chapter 7, Brower et
al. 1988) . Therefore, given that many desiccated monarchs occur
outside colonies (Glendinning pers. observ.), and that hydrated ones
are common inside colonies (Table 3-3), mice that were foraging outside
colonies may have ingested higher concentrations of cuticle, and hence
of CGs.
Why is P. melanotis the only species of mice that feeds on the monarch
butterflies?
I tested the hypotheses that the four other species of mice did not
eat monarchs because 1) the microhabitat characteristics (e.g..
understory vegetation) of overwintering areas may have only suited P.
melanotis. or 2) melanotis competitively excluded them. I rejected
the first hypothesis because the understory vegetation patterns of
grids inside colonies were not significantly different from those of
grids outside colonies. I also rejected the second hypothesis because

72
resident EL. melanotis were unable to dominate EL. nu. salvus and EL. a .
hylocetesf and were able to dominate only about half of the R.
sumichrasti. This indicated that resident EL. melanotis could not
prevent any of the species from feeding on the monarchs and
establishing residency inside the colonies.
The results reported herein and in Chapters 4 to 6 suggest that all
species of mice except £L melanotis avoided the colonies because of an
aversion to the monarchs. I examine the feeding responses of P -
malanotis, sumichrasti. EL- 5^. hylocetes and 1L HL. salvus to
overwintering monarchs and their defensive compounds in the next
chapter.

CHAPTER 4
COMPARATIVE FEEDING RESPONSES TO OVERWINTERING MONARCH BUTTERFLIES
To compare the feeding responses of £_*. melanotis. P. a. hylocef.es.
sumichrasti. and H*. HL*. salvus to overwintering monarchs, I conducted
3 experiments with caged individuals. I measured how many monarchs
each species ate and the degree to which they avoided cuticular
material. Monarchs store higher concentrations of CGs in their cuticle
compared to their body contents (Table 4-1; Brower et al. 1988) .
Second, I further explored CG avoidance in melanotis. which feeds
naturally on monarchs, by offering them a choice between male and
female monarchs. Male monarchs have on average 30% lower CG
concentrations than females (Brower and Calvert 1985) . I compared how
many abdomens of each sex were eaten and the extent to which each
sexes' cuticle was avoided. Third, I examined how well all four
species could maintain weight on a diet consisting solely of monarchs.
Methods
Trapping and Maintenance of Mice
I trapped and conducted experiments from 20 January through 28
February 1986 on the slope of the Arroyo La Plancha of the Sierra
Chincua mountain massif (Figure 3-la). Mice were trapped in Sherman
live traps (7.6 x 8.9 x 22.9 cm) at least 800 m from the monarch colony
to reduce the possibility that they had previously encountered a
73

74
monarch colony. Because of a high variance in food consumption in
reproductive mice, particularly females (Sadlier et al. 1973, Stebbins
1977, Millar 1979), only reproductively inactive adults (non-lactating
or non-pregnant females and non-scrotal males) were used in the
experiments. Each mouse was used only once.
The mice were housed and tested individually in wire mesh cages
(about 30 cm high x 25 cm in diameter) set on a tarpaulin-covered table
in the shade. Dacron batting was added for nesting material and pieces
of cardboard were placed between the cages to isolate the mice
visually. For six consecutive nights prior to the feeding experiments,
each mouse was maintained on Purina laboratory chow no. 5001
(henceforth, mouse chow) and water. All of them fed and drank
regularly and either maintained or gained weight.
Monarch Collection
Monarchs were collected with butterfly nets from accessible
clusters on fir branches within the butterfly colony. For Experiments
1 and 3, monarchs were offered to mice without reference to their sex.
Branch clusters were about 58% females during January and February (n =
7 samples, for a total of 1095 butterflies; T. Van Hook unpubl. data).
For Experiment 2, equal numbers of each sex were offered to mice.
Experiment 1: Patterns of Feeding by Four Mouse Species
The feeding responses of the 4 species to monarchs were compared.
Each mouse was weighed at the beginning of the experiment. Then 4
males and 4 females of each species were individually caged and each
was offered 40 monarchs, mouse chow ¿d libitum, and water for two

75
consecutive nights. In this and the next experiment, all mice received
live (and active) monarchs, except for eu. salvus. which received
inactive monarchs (i.e., ones whose thorax had been squeezed firmly)
because they would not approach live ones. At 0900 h all mouse chow
was removed from each mouse's cage. At 1900 h on the same day, the
monarchs and fresh mouse chow were given to each mouse. At 0900 h the
following morning, each mouse was removed from its cage to tally
patterns of feeding damage to dead monarchs, then replaced and deprived
of food until 1900 h that day, at which time the feeding trial was
repeated.
I determined from preliminary feeding trials that mice consumed
variable portions of thoraces, but always ate the cuticle together with
the contents. In contrast, when feeding upon abdomens, they
characteristically ate either 100% of the cuticle and contents (i.e.,
fed non-selectively) or 100% of the contents only and discarded the
cuticle (i.e., fed selectively). Based on these observations, I chose
a. priori 6 categories of monarch damage (Table 4-1) . I assume that
sampling a monarch represents active rejection of it based on taste.
Determination of the amount of monarch tissue, cuticle, and CGs
eaten by each mouse in Experiments 1 and 3 involved several steps.
First, I estimated the amount of tissue, cuticle and CGs in the body
parts of overwintering monarchs (see Table 4-2). For these
estimations, I modified the data from Brower £t. al. (1988) because they
were derived from freshly-eclosed, Asclepias syriaca-reared monarchs,
which have higher CG contents and substantially lower amounts of fat in
their abdomens than monarchs reared on the same food plant that have
migrated to the Mexican overwintering grounds (Brower 1985, Malcolm and

76
Table 4-1. Categories of monarch damage by mice.
1. No visible damage
2. Sampled: <25% of abdomen and/or thorax eaten
3. 25-50% (cuticle and contents) of thorax eaten
4. 51-100% (cuticle and contents) of thorax eaten
5. Abdomen eaten non-selectively (> 25% of cuticle and contents)
6. Abdomen eaten selectively (> 25% of contents; cuticle rejected)

77
Brower 1989) . Second, I assumed that whenever a mouse ate 1) an
abdomen non-selectively, it ate 100% of the cuticle and contents, 2) an
abdomen selectively, it ate 100% of the contents, 3) 25 to 50% of a
thorax, it ate 37.5% of the cuticle and contents, and 4) 51 to 100% of
a thorax, it ate 75% of the cuticle and contents. These assumptions
agreed generally with my preliminary observations. Third, I calculated
the total quantity of tissue, cuticle and CGs eaten by a given mouse
over a given time period with the following equations, which
incorporate data from Table 4-2:
tissue eaten = [(no. abdomens eaten selectively and non-selectively)
(123 mg)] + [(no. thoraces eaten 25-50%) (37.5%) (41 mg)]
+ [(no. thoraces eaten 51-100%)(75%)(41 mg)];
cuticle eaten = [(no. abdomens eaten non-selectively)(11 mg)] +
[(no. thoraces eaten 25-50%) (37.5%) (22 mg)] +
[(no. thoraces eaten 51-100%) (75%) (22 mg) ] ;
CG eaten = (no. abdomens eaten selectively)(10 |lg)] +
[(no. abdomens eaten non-selectively) (10 + 24 ]lg) ] +
[(no. thoraces eaten 25-50%) (37.5%) (10 + 11 (ig) ] +
(no. thoraces eaten 51-100%) (75%) (10 + 11 [ig) ] .
Fourth, I added together the predation records from both nights in
Experiment 1, whereas for Experiment 3, I analyzed each night
separately so as to monitor changes in consumption of monarch tissue,
cuticle and CGs over the 6 nights.

Table 4-2. Mean dry weights and CG contents of the abdominal cuticle and tissue, thoracic cuticle
and tissue, and wings of freshly-eclosed monarchs reared on Asclepias syriaca (from Brower et al.
1988: Table 3). These data, along with the mean dry weights and CG contents (all body parts
combined) of monarchs overwintering in Mexico (Malcolm and Brower 1989), were used to estimate the
dry weights and CG contents of the different body parts in overwintering monarchs. Data from A.
syriaca-reared monarchs are appropriate for these estimations because the evidence suggests that 85
to 92% of the overwintering monarchs in Mexico have fed as larvae on this milkweed (Seiber et al.
1986; Malcolm et al. 1989). I assume that: 1) the dry weights of the cuticle, wings and thoracic
tissue were the same in both types of monarchs, and that only the abdominal tissue weights differed,
owing to increased amounts of fat associated with diapause (Brower 1985); and 2) the distribution of
CGs in the different body parts was the same in both types of monarchs, but that the CG contents were
proportionally lower
in
each body
part in
overwintering
monarchs.
Freshly-eclosed monarchs
Estimated
values
for overwintering monarchs
Dry
weight3
CG content3
Dry
weight
CG
content
CG concentration
Body Part
mg
percent
ug
percent
mg
percent
ng
percent
(Ag/0 . lg d. w.
Abdominal cuticle
ii
6%
91
26%
ii
5%
24
26%
218
Abdominal tisssue
70
37%
37
11%
123
51%
10
11%
8
Thorax cuticle
22
11%
41
12%
22
9%
11
12%
50
Thorax tissue
41
22%
37
11%
41
17%
10
11%
24
Wings
45
24%
145
41%
45
19%
39
41%
87
Continued.

Table 4-2—Continued.
Contents of
abdomen + thorax
111
59%
74
21%
165
68%
20
21%
12
Wings + cuticle
78
41%
277
79%
78
32%
74
79%
95
Sum of all parts
189
100%
351
100%
242b
100%
94b
100%
39
a Values are rounded to the nearest whole value.
b These are means derived from 563 overwintering monarch butterflies at Sierra Chincua
(Malcolm and Brower, 1989).

80
Because the 4 species of mice differed greatly in size (Table 4-
6), I could not compare directly the amounts of monarch tissue, cuticle
or CG eaten. Therefore, I standardized consumption by computing the
0 7
ratio of the weight of tissue, cuticle, or CG eaten to mouse weight • .
A 7
The power function, weight , is derived from a regression of
consumption rate on weight for a wide range of mammals (Farlow 1976,
Peters 1983). Mass-specific consumption values were calculated
separately for each mouse, using their weight at the beginning of the
experiment.
Experiment 2: Predation by P. melanotis in Relation to Butterfly Sex
The previous experiment examined whether either of the species
discriminated between abdominal cuticle and contents. This experiment
addressed a more complex question: When given a choice between monarchs
of differing CG concentrations (in this case males versus females),
will £_*. melanotis eat more low CG ones and feed selectively on more of
the high CG ones? Male monarchs have on average 30% lower CG
concentrations than females (Brower and Calvert 1985). I could not
estimate separately for both sexes the amounts of tissue, cuticle and
CGs eaten because the data were not available; those in Table 4-2 are
from pooled samples of male and female monarchs.
Seven female and 8 male melanotis were each offered 25 female
and 25 male monarchs and mouse chow and water ¿d libitum per night for
two consecutive nights. Five hours before each feeding trial, I
squeezed the thoraces of the female and male monarchs and then placed a
1 mm diameter, red or gold spot of nail polish on each butterfly
abdomen, thorax, head, and wings so as to enable tallying of the parts

81
by sex after the mice dismembered them. The assignment of gold or red
for males or females was determined randomly for each mouse. Because
sexual differences in wing length are negligible and wet weights vary
unpredictabilty (Brower and Calvert 1985), I considered size and weight
unlikely bases for killing an excess of one sex.
Experiment 5: Effects of Long-Term Consumption of Monarchs
In this experiment, I explored whether all 4 species could maintain
or gain weight on a pure diet of monarchs. Each mouse (6 £_». melanotis,
5 £u. sumichrasti. 4 L. 1. hylocetes. and 5 sl_ salvua; sex-ratios
were approximately equal) was pre-exposed to 40 monarchs and mouse chow
ad 1ihitum per night for 2 nights. Then for the next 6 nights, each
mouse received 55 monarchs per night and water ¿d libitum, without
mouse chow. Preliminary feeding trials indicated that all species ate
fewer than this number of monarchs per night. Each mouse's weight was
taken at the beginning and end of the experiment, 8 days later. All
mice were tested between 7 and 21 February.
Statistical Analyses
All statistical tests followed Zar (1984) . In Experiment 1, paired
(2-tailed) t-tests were run separately on each species to compare the
number of abdomens its., thoraces eaten, and the number of abdomens eaten
selectively is., non-selectively. One-way ANOVA's and Scheffé F-tests
were used to compare the species in terms of the mass-specific amounts
of monarch tissue, cuticle or CGs eaten, as well as the ratios of the
amounts of CGs to tissues eaten. I was justified in using the mass-
specific consumption values in the ANOVA's because the coefficient of

82
variation of the scaling variable (i.e., mouse weight) was consistently
less than that of the dependent variable (i.e., amount of tissue,
cuticle or CG consumed)(Anderson and Lydic 1977, Packard and Boardman
1988) . The unpaired t-test was used to compare the percentage of
abdomens eaten selectively by £_*. melanotis and R_*. sumichrasti. Data
were transformed (arcsin Vx) for this and all subsequent statistical
comparisons of percentage data.
Paired t-test comparisons were used in Experiment 2 to compare the
total male and female monarch abdomens eaten (selectively and non-
selectively), as well as the percentage of those abdomens that were
eaten selectively.
Two-factor ANOVA's (repeated on time) were used in Experiment 3 to
determine the effects of species (E_^ melanotis. hylocetes. and £_*.
.sumichrasti) and time on tissue, cuticle, and CG consumption. The same
test was used again to determine the effect of species and time on the
percentage of abdomens eaten selectively by melanotis and
sumichrasti. M. m. salvus was excluded from all statistical analyses
in this experiment.
Results
Experiment 1: Patterns of Feeding by all Four Mouse Species
Peromyscus melanotis. sumichrasti and a^. hylocet,a.a each ate
more than 52% of the monarchs offered to them (Table 4-3) and
significantly more abdomens than thoraces (Figure 4-1). In contrast,
M. m. salvus ate less than 13% of the monarchs offered to them and
virtually identical numbers of abdomens and thoraces. Moreover,

Table 4-3. Comparison of the feeding patterns of all 4 mouse species upon monarch butterflies. Eight
individuals per species were each offered 40 monarchs and mouse chow ¿d libitum per night for two
consecutive nights. For each species, the mean number of monarchs eaten, sampled or left undamaged (±
S.E) by each species are presented. Because mice ate both the abdomen and thorax of some monarchs, the
row totals are all greater than 80.
Mean
no. of monarchs ± S.E. in each feeding category
No visible
Thoraces eaten: Abdomens eaten:
No.
offered/
Species
damage
Sampled 25-50% 51-100% non-select, selectively
mouse
melanotis
14.3
±3
15.4
±4
5.1
±2
0.0
±0
18.8
±1
31.3
±2
80
ÍL. sumichrasti
26.0
±3
10.9
±2
4.5
±1
0.9
±.4
19.3
±2
23.3
±1
80
£j. aztecus
20.5
±1
17.8
±2
8.5
±1
4.8
±1
39.4
±4
2.8
±1
80
1L. salvus
57.6
±5
11.9
±5
4.1
±2
3.8
±1
10.1
±1
0.1
±.l
80
oo
U)

Figure 4-1. Mean number of monarch abdomens and thoraces (+S.E.) eaten by 8 individuals of each
of 4 mouse species. Each mouse received 40 monarchs and mouse chow libitum per night for two
consecutive nights. The significant within species comparisons between the number of abdomens and
thoraces eaten are indicated by asterisks.

Mean number eaten
55 -
50 -
45 -
40 -
35 -
30 -
25 -
20 -
15 -
10 -
P.
melanotis
*
*
R.
sumichrasti
â–¡ Abdomens
H Thoraces
** P < 0.0001 ;df = 7
paired t-test
P. a. M. m.
hylocetes salvus
CO
\J1

86
salvus approached only immobilized monarchs. None of the mice consumed
the wings, head or legs.
The standardized consumption values are shown in Table 4-4.
Peromyscus melanotis and ÍL. sumichrasti consumed significantly more
monarch tissue than did E_*. hylocetes and H*. Eb_ salvus. However,
a. hylocetes ate significantly more cuticle and both E^_ sumichrasti and
P. a. hylocetes ate more CGs than did the other 2 species. To help
interpret these patterns of consumption, I computed for each species
the ratio of the amounts of CGs to tissues eaten. The ratios for
melanotis and Ed. sumichrasti were significantly lower than those for
the other 2 species (Table 4-4).
The large interspecific differences in these ratios can be
explained by the way each mouse species consumed monarch abdomens.
Peromyscus a. hylocetes and Eb. salvus rarely ate abdomens
selectively, whereas E^. melanotis and E^. sumichrasti both ate
significantly more abdomens selectively than non-selectively (Figure 4-
2). Peromyscus melanotis fed selectively on a significantly greater
percentage of abdomens than did E^. sumichrasti (means = 62.2 and 55.1,
respectively; unpaired t-value = 2.89; df = 14; P < 0.013). Thus by
feeding selectively, E^. melanotis and E^. sumichrasti reduced
substantially the CG concentration of the monarch tissue they ingested.
By examining mouse-damaged monarchs and watching mice eat them, I
discovered interspecific differences in the way the four species
extracted abdominal contents. Peromyscus melanotis and E^ sumichrasti
commonly made longitudinal slits down the full length of the abdomen
and then sucked and/or licked out the abdominal contents. In contrast,

Table 4-4. Interspecific comparison of the standardized amounts of monarch tissue, cuticle, and CGs
eaten. The mean ratio of the amounts of CGs to tissue ingested (± S.E.) is also compared among
species. Eight mice from each species were each offered forty monarchs and mouse chow ad libitum
per night for two consecutive nights. Between species comparisons are made with one-way ANOVA.
Different subscripts (a, b and c) indicate significant differences among species within each column
(P < 0.05; Scheffé F-test).
Mouse
species
Standardized amounts eaten (mean ± S.E.)
CGs/tissue
ratio*
Tissue
(g/kg0>7 mouse)
Cuticle
(g/kg0-7 mouse)
CGs
(mg/kg0-7 mouse)
E_*. melanotis
676.9 ±25.9 a
32.0 ±1.8 a
96.4 ±4.2
a
0.14 ±0.01
a
sumichrasti
703.4 ±22.6 a
40.9 ±2.6 a/ b
120.0 ±7.0
b
0.17 ±0.01
a
3^. hylocetes
378.4 ±13.2 b
46.1 ±1.8 b
126.4 ±4.3
b
0.33 ±0.01
b
EL. nú. salvus
120.4 ±15.4 c
18.8 ±2.7 c
46.1 ±5.8
c
0.39 ±0.01
c
F-value
190.4
27.1
44.7
212.3
df
3
3
3
3
P
0.0001
0.0001
0.0001
0.0001
* Determined from individual ratios.

Figure 4-2. Mean number of monarch abdomens (±S.E.) eaten selectively and non-selectively by four
mouse species (n = 8/species). Each mouse received 40 monarchs and mouse chow ad libitum per
night for two consecutive nights.

45 n
P.
melanotis
R.
sumichrasti
**
Non-Selectively
â–¡ Selectively
** P < 0.0004; * P < 0.05
df = 7; paired t-test
P. a.
hyiocetes
M. m.
salvus

90
íL_ hylocetes and nu. salvus usually bit off the end of the abdomen
and pulled the contents out with their teeth.
Experiment 2: Predation by P. melanotis in Relation to Butterfly Sex
Peromyscus melanotis killed (feeding categories 2 -6; Table 4-1)
equal numbers of male and female monarchs, which suggests no initial
difference in the risk of male and female monarchs to mouse attack
(Table 4-5). However, adding together the number of abdomens eaten
selectively and non-selectively reveals that melanotis ate a larger
number of male than female abdomens (means = 28.4 and 22.7,
respectively; paired t-value = 3.198, df = 14; P = 0.0064). Of those
abdomens eaten, melanotis fed selectively on greater percentage of
female abdomens (female and male means = 68.1 and 50.0, respectively;
paired t-value = 10.63; df = 14; P = 0.0001) . These results support
the hypothesis that both the quantity of tissue eaten and the tendency
to feed selectively are influenced by CG concentration.
Experiment 3: Effects of Long-Term Consumption of Monarchs
All R^. melanotis gained weight over the six nights, whereas 60-100%
of the individuals of each of the other species lost weight (Table 4-
6). Microtus eu. salvus were excluded from the statistical analyses
because they ate very few monarchs and all died within four days.
The mass-specific amounts of monarch tissue, cuticle and CGs eaten
nightly by R^ melanotis. hylocetes and sumichrasti are plotted
in Figure 4-3a-c. The two-factor ANOVA's performed separately on
tissue, cuticle and CGs all indicated significant effects of species
(for tissue, F = 25.5, df = 2, P < 0.0001; for cuticle, F = 6.41, df =

Table 4-5. Comparative feeding patterns by melanotis upon male and female monarch butterflies.
Fifteen mice were each offered 25 male and 25 female butterflies together with mouse chow ¿d libitum
per night for two consecutive nights. For each monarch sex, the mean number eaten, sampled or left
undamaged (± S.E) by the 15 mice are presented. Because mice ate both the abdomen and thorax of some
monarchs, the row totals are all greater than 50.
Mean no. of monarchs ± S.E. in each feeding category
Thoraces eaten: Abdomens eaten: No.
Monarch No visible offered/
sex
damage
Sampled
25-50%
51-100%
non-select.
selectively
mouse
male
8.5 ±1
12.8 ±1
1.8 ±.4
3.6 ±1
14.1 ±1
14.3 ±1
VO
50 M
female
8.7 ±2
17.8 ±2
1.4 ±.4
1.3 ±.3
7.3 ±1
15.4 ±1
50

Table 4-6. Initial weights (mean ±S.E.), weight changes and % mortality in P_*. melanotis. R.
sumichrasti. P. a. hylocetes and ÍL- HU. salvus given 55 monarchs per mouse per night for 6 consecutive
nights, without alternative food.
Mouse
Species
N
Initial
weight (g)
Weight
change (g)
% weight
change
% of mice
that lost
weight
%
mortality
£.1. melanotis
6
19.6
(±0.4)
+0.83 (±0.2)
+4.2
0
0
Rj. sumichrasti
5
13.4
(±0.3)
-0.38 (±0.2)
-2.8
60
0
P. a. hvlocetes
4
36.8
(+0.6)
-1.70 (±0.7)
-4.6
75
0
el. salvus
5
30.0
(±0.6)
-4.30 (±0.7)
-14.3
100
100*
★
All died within 4 days.

I
Figure 4-3. Standardized amounts of monarch tissue, cuticle and CGs
eaten by 6 L melanotis. 5 JL. sumichrasti and 4 ^ hylocetes. Each
mouse was given 55 monarchs per night for six consecutive nights, after
a two night pre-exposure to monarchs and mouse chow.

Mean (±S.E.) amount eaten
â– 
94
o P me/onotis
• R sum/chrasti
I

95
2, P = 0.01; and for CGs, F = 7.4, df = 2, P = 0.0081) and of time for
CGs (F = 2.9; df = 5, P = 0.0203). There were not significant effects
of time for tissue (F = 1.0, df = 5, P =0.445) nor cuticle (F = 2.0, df
= 5, P = 0.09). The interaction of species and time was significant
for tissue (F = 3.0, df = 10, P = 0.0044), cuticle (F = 3.5, df = 10, P
= 0.001), and CGs (F = 2.5; df = 10; P = 0.014). These results
demonstrate that: 1) melanotis and sumichrasti consistently ate
more tissue than did 4^. hylocetes; 2) E_._ hylocetes consistently
ate more cuticle and CGs than did the other two species of mice; and 3)
nightly consumption of cuticle and CGs by K,. sumichrasti and a.
hylocetes remained roughly constant over the six nights, whereas that
of melanotis tended to diminish.
As in Experiment 1, E_^ melanotis and sumichrasti often fed upon
abdomens selectively (Figure 4-4). The two-factor ANOVA indicated a
significant effect of time (F = 3.3, df = 5, P = 0.014), but not of
species (F = 2.0, df = 1, P = 0.19) or of the species x time
interaction (F = 1.8, df = 5, P = 0.14), on the nightly rate of
selective feeding by these two species. Because E_^ hylocetes rarely
fed selectively on monarch abdomens (mean = 4.3% ± 2.1 of total
abdomens eaten), it was excluded from Figure 4-4 and the analysis.
Figure 4-4 suggests that only E_^ melanotis increased its nightly rate
of selective feeding with time. This apparent interspecific difference
may not have been detected by the ANOVA because of the large amount of
intraspecific variation and small sample sizes.
The ratio of CGs to tissue ingested remained nearly constant over
the six nights for all three species (Figure 4-5). The two factor
ANOVA indicated a significant effect of species (F = 21.3, df = 2, P <

Figure 4-4. Mean percentage of abdomens (± S.E.) eaten selectively by 6 £_*. melanotis and 5 R.
sumichrasti. Each mouse was given 55 monarchs per night for six consecutive nights, after a two
night pre-exposure to monarchs and mouse chow

Mean (±S.E.)% of abdomens eaten selectively
90
85
80
75
70
65
60
55
50
45
40
o R me/onotis
VO

Figure 4-5. Ratios (mean ± SE) of monarch CGs (mg) to tissues (g) eaten by six melanotis, five
R. sumichrasti. and four L. hylocetes over six consecutive nights. Each mouse received 50
monarchs per night (and no mouse chow), after a two night pre-exposure to monarchs and mouse chow.

Mean ratio of CGs to tissue eaten (mg/g)
p p p p p p o
O — ro cm 1 1 1 1 1 1
66
R. sumichrasti

100
0.0001), but not of time (F = 1.4, df = 5, P = 0.2419) or of the
species x time interaction (F = 1.1, df = 10, P = 0.390). Thus, by-
feeding selectively on abdomens, E^. melanotis and £_*. sumichrasti
reduced substantially the ratios of CGs to tissue in their diet over
the six nights.
Discussion
Patterns of Feeding by Mice
The two species that fed selectively on monarch abdomens, P.
melanotis and B^. sumichrasti. also ate the most monarch tissue on a
weight-specific basis. By discarding the abdominal cuticle, they
substantially reduced ingestion of CGs (see Table 4-2). Peromyscus a.
hylocetes and eu. salvus r on the other hand, rarely discarded
cuticular material and thus ate significantly more cuticle and CGs than
did E^. melanotis and E^. sumichrasti.
There are several possible reasons why bl salvus and E^. a.
hylocetes ate so little monarch material as compared to the other two
species. Even though Microtus species are known to eat insects to
varying degrees (Batzli 1985, Hanski and Parviainen 1985), they tend to
be strongly herbivorous and thus may be less motivated to attack and
eat insects, particularly moving ones, than are more omnivorous species
(Cyr 1972) . In fact, Hi. EL. salvus was the only species that would not
attack active monarchs. Nevertheless, the fact that IL. el. salvus also
refused to eat inactive monarchs suggests that they were also averse to
the monarch's defensive compounds. Peromyscus a. hylocetes. on the
other hand, appears to have a greater predilection for eating insects,
given the strongly omnivorous feeding habits of its genus (Landry 1970

101
Whitaker 1966, Wolff et al. 1985) and the large numbers of monarchs
that individuals ate (Figure 4-1 and Table 4-3). Its aversion to the
monarchs is more likely related to its intolerance to CGs (Chapter 6).
Even though £^_ sumichrasti is also strongly deterred by these
compounds, it limited ingestion of them through its selective feeding
behavior.
It is unclear why only melanotis and E^. sumichrasti commonly
showed the selective feeding behavior. They may have modified normal
fruit- or insect-eating behaviors, as has been suggested for black-
backed orioles, which also feed selectively on monarch abdomens (Brower
and Fink 1985) . However, why would E^. hylocetes and salvus
not have modified feeding behaviors of their own? Moreover, the
selective feeding behavior is not unique to melanotis and R.
sumichrasti. Meadow voles Microtus pennsylvanicus remove the phenolic-
rich bark of Norway pine Pinus resinosa and White spruce Picea glauca
saplings before eating the phenolic-poor cambium (J. Roy and J.M.
Bergeron, unpub1. data). Second, red tree mice Phenacomys longicaudns
peel off and discard the externally located resin ducts of Douglas fir
Pseudotsuga taxifolia needles before eating them (Benson and Borell
1931, Hamilton 1962). Third, chisel-toothed kangaroo rats Dipodomys
microps shave off the hypersaline peripheral tissue of the leaves of
saltbush Atriplex confertifolia so that the internal tissue can be
ingested (Kenagy 1972) . And fourth, when grasshopper mice Onychomys
torridus are offered house crickets Acheta domesticus whose heads have
been treated with a highly bitter solution (quinine-hydrochloride and
ethanol), they first chew off and discard the heads before eating the
bodies (Cyr 1972). Further study of insectivory and food handling

102
behavior in the four species of mice associated with the monarch
colonies will be valuable in explaining the different ways they fed
upon monarchs.
In addition to CG-avoidance, three other reasons may explain why
mice fed selectively on monarchs. They may have been avoiding PAs,
which also appear to be concentrated in the cuticle (M. Stelljes
unpubl.data). Although this remains a distinct possibility, results
presented in the next Chapter demonstrate clearly that both P.
melanotis and sumichrasti feed selectively on monarchs that lack
PAs. Second, because small mammals (< 100 g) appear incapable of
digesting cuticle to any significant extent (Griffith and Gates 1985,
M. Allen unpubl. data, but see Jeuniaux 1961), they could increase
their net energy yield of foraging by avoiding cuticle, and several
foraging models predict animals should avoid indigestible foods (e.g.,
Krebs 1978, Belovsky 1978). Cuticle represents 10% of an overwintering
monarch's abdominal biomass and 27% of its thoracic biomass (Table 4-
2). Third, mice may have found the cuticle too tough and leathery.
However, the results from Experiment 2 fail to support these latter two
explanations. Peromyscus melanotis ate more male abdomens and fed
selectively on a greater percentage of female ones. See Chapter 5 for
a more detailed examination of why these mice eat monarchs selectively.
These results are in contrast to Brower et al. (1985), who rarely
observed selective feeding by E^. melanotis (erroneously referred to as
Ej. maniculatus in their paper; Brower et al. 1988) . This discrepancy
is explored in Chapter 5.

103
Absence of Conditioned Feeding Aversions to Monarch CGs
Peromyscus melanotis, sumichrasti and £_*. hylocetes all
consumed extremely high nightly dosages of monarch CGs in Experiments 1
and 2 (means > 20 and 30 mg/kg mouse, respectively) without developing
any apparent signs of CG toxicity (e.g., ataxia, trembling, rolling
convulsions, hyperexcitability; Marty 1983). The LD5g for an oral
dosage of digitoxin in laboratory mice Mus domesticus r rats Rattus
norvegicus, and cats Felis domestica, is 32.7, 23.8 and 0.30,
respectively (Barnes and Eltherington 1973). The mice in this study
may have been able to tolerate such high dosages because they absorbed
only a small percentage of the monarch CGs they ingested. Cardiac
glycoside absorption may be limited for other reasons. First, the
gastrointestinal tracts of two species of mice closely related to the
species studied herein are highly impermeable to monarch CGs
(Peromyscus maniculatus and Mus domesticus; Marty 1983). Second,
overwintering monarchs in Mexico have a high proportion of high
polarity CGs (Seiber et al. 1986), which are absorbed by M^. domesticus
less readily than are low polarity ones (Lauterbach 1981). Third, some
of the ingested monarch CGs may have remained bound to the cuticle
during digestion and were thus unavailable for absorption (see Clement
1977, Brower et al. 1988) . Further study of the toxicological
sensitivity of these three species to CGs, as well as the
bioavailability of CGs bound in cuticle, will help determine whether
either species could develop CG toxicity, and hence conditioned feeding
aversions, following prolonged consumption of monarchs.

104
Effects of Long-Term Consumption of Monarchs
All species of mice, except for nu. salvus. ate nearly constant
amounts of monarch tissue over the six nights. What differed among
these species was the standardized amount of tissue eaten per night.
Peromyscus hylocetes ate consistently less than did melanotis and
R*. sumichrasti.
These results suggest that three-fourths of the JL,. ¿La. hylocetes
lost weight because they ate insufficient amounts of food. However, it
seems paradoxical that three-fifths of the sumichrasti also lost
weight considering they ate similar standardized amounts of tissue as
did the Ej. melanotis. One possible explanation is that because R.
sumichrasti is highly averse to the taste of CGs (Chapter 6), the
monarch diet may have caused stress (sensu Selye 1983), a condition
known to increase heart rate, blood pressure, oxygen consumption
(Porges 1985) and possibly search activity; all of these factors
elevate energetic demands. Stress also leads to weight loss in a
variety of mammals, apparently owing to reduced food intake and/or a
decrease in digestive efficiency (Klasing 1985) . Under such stress, R.
sumichrasti would have lost weight even with the same mass-specific
energy intake as melanotis.

CHAPTER 5
CONSUMPTION AND AVOIDANCE OF MONARCHS: THE INFLUENCE OF MONARCH
ABUNDANCE AND CARDIAC GLYCOSIDE CONCENTRATION
In the previous Chapter, I found that when offered 50 Mexican
overwintering monarchs (henceforth, Mexican monarchs) per night without
an alternate food, individual £_». melanotis ate on average 40 monarchs,
fed preferentially on low CG monarchs (i.e., they ate more males than
females), and frequently ate abdomens selectively (i.e., rejected the
CG-laden abdominal cuticle and ate the low-CG internal tissues).
However, Brower et al. (1985) reported that when offered < 30 Mexican
monarchs, or 30 monarchs with varying concentrations and kinds of CGs,
individual melanotis ate on average 25 monarchs, did not feed
preferentially on low CG monarchs, and rarely fed selectively on
abdomens. To resolve these discrepant findings, I examined how monarch
consumption patterns by melanotis. ^ hylocetes and
sumichrasti are influenced by 1) varying the number of monarchs offered
during one night and 2) varying CG concentration and polarity when
monarchs are superabundant.
Methods
Experiment 1: Effect of Monarch Abundance on Foraging Patterns of Mice
Feeding trials were conducted from 1 through 28 February 1986
on the slope of the Arroyo La Plancha of the Sierra Chincua mountain
105

106
massif in northeastern Michoacán, Mexico, at approximately 20°N
latitude. This site is one of the principal overwintering areas of the
eastern population of the monarch butterfly (Calvert and Brower 1986).
At an altitude of approximately 3200 m, mice were trapped in Sherman
live traps (7.6 x 8.9 x 22.9 cm) at least 800 m from the monarch colony
to reduce the possibility that they had prior experience eating
monarchs. Because of a high variance in food consumption in
reproductive mice, particularly females (Sadlier et al. 1973, Stebbins
1977, Millar 1979), only reproductively inactive adults (non-lactating
or non-pregnant females and non-scrotal males) were used in the
experiments.
A total of 26 to 35 mice of each species was housed and tested
individually in wire mesh cages (about 30 cm high x 25 cm in diameter)
set on a tarpaulin-covered table in the shade. Dacron batting was
added for nesting material and pieces of cardboard were placed between
the cages to isolate the mice visually. For three consecutive nights
prior to the feeding experiments, each mouse was maintained on Purina
laboratory chow no. 5001 (henceforth, mouse chow) and water. All of
them fed and drank, regularly and either maintained or gained weight.
Monarchs were collected with butterfly nets from accessible
clusters on oyamel fir branches (Abies religiosa H.B.K.) within the
butterfly colony, and were offered in their natural sex-ratios to mice.
I did not distinguish monarch mortality with respect to monarch sex.
Each experimental run lasted for 2 nights. During the first night
at 1900 h, 5 live monarchs and mouse chow ad libitum were placed in
each mouse's cage. All mice consumed nearly all of these monarchs
nonselectively. This pre-exposure period insured that all mice had

107
some prior exposure to monarchs. At 0900 h the following morning, all
remaining monarch material was removed. Then at 1900 h on the same
day, a specified number (see below) of monarchs and fresh mouse chow ad
libitum were given to each mouse. At 0900 h the following morning,
each mouse was removed from its cage to tally patterns of feeding
damage to dead monarchs. The same mouse was never tested more than
once in this and the next experiment. As reported in Chapter 4, all
three species of mice feed primarily on monarch abdominal material;
thoraces and wings are rarely eaten. I included only the two most
commonly exhibited patterns of abdominal consumption: 1) non-selective
feeding (i.e., a mouse eats virtually all of the cuticle and contents),
and 2) selective feeding (i..e, a mouse eats virtually all of the
contents and discards the cuticle). Although mice sampled (i.e., ate <
25%) monarch abdomens and consumed thoracic material to varying degrees
in < 5% of the instances of abdominal consumption, I did not include
these feeding events in the analysis.
Individual mice of each species were exposed on the second night to
10, 20, 30, 40 or 50 monarchs. Between 6 to 9 L melanotis. 5 to 6 P.
SLl. hylocetes and 5 to 6 L sumichrasti were subjected to each density;
mouse sex-ratios were roughly equal for each density. Individual mice
were run through a single trial and then released.
Experiment 2: Effect of CG Concentration on Foraging Patterns of Mice
In the previous experiment, when > 40 monarchs were offered per
night, all 3 species of mice consumed the largest amount of monarch
material and also ate the largest percentage of abdomens selectively.
In this experiment, I held monarch abundance constant at 50 monarchs

108
per night and experimentally varied CG concentration. All 3 species
were compared in terms of how CG concentration influenced 1) the number
of monarch abdomens eaten and 2) selective feeding on monarch abdomens.
In March 1986 I collected 10 melanotis. 10 L. ^ hylocetes. and
10 R*. sumichrasti (five of each sex for each species) in the same area
described above and transported them to the University of Florida in
Gainesville, Florida. They were housed in plastic cages (30 x 12 x 16
cm) and maintained on mouse chow and water ad libitum in a
temperature/humidity controlled room, on a 13 h light: 11 h dark cycle.
I used Fx offspring of these mice (20 melanotis. 20 A*. sumichrasti.
20 L. hylocetes) for testing when they were 90 to 120 days old. As
a result, all mice had similar feeding histories.
Three different categories of monarchs were used in the feeding
experiments: ones collected at the Mexican overwintering sites, ones
reared on Asclepias syriaca L. (henceforth, A^. i*. monarchs) , and ones
reared on A*, curassavica L. (henceforth, A^. SL*. monarchs) . The Mexican
monarchs were collected as in Experiment 1, and then sealed inside two
plastic bags to prevent desiccation, transported on ice to the
University of Florida, and frozen. The A^. s^. monarchs were collected
in the wild as fourth and fifth instar larvae in western Massachusetts
(Hampshire and Franklin Counties) in August and September, 1985 and
1986, fed A^. syriaca leaves from the same fields, and shipped as
chrysalids to the University of Florida where they hatched. The A*,
monarchs were the first generation of adults captured nearby the
University of Florida during September, 1985 and 1986. They were
raised from egg to maturity on A^. curassavica leaves from potted plants

109
originating from southern Florida (Dade County) . The A A and A L.
monarchs were frozen inside two plastic bags 24 h after eclosión.
Even though 85 to 92% of all Mexican monarchs appear to have fed as
larvae on A. syriaca plants (Seiber et al. 1986; Malcolm and Brower
1989), they possess substantially lower CG concentrations than do
monarchs that were either laboratory-reared on A syriaca killed about
24 h after eclosión or wild-reared on A syriaca and collected in
Massachusetts (Table 5-1). This is because monarchs lose CGs during
the long migration to the overwintering sites (Malcolm and Brower
1989). As a result, the Mexican and A monarchs should differ in CG
concentration, but not polarity (Table 5-1). In contrast, laboratory-
reared A A monarchs not only have a higher CG concentration than both
A Sj. and Mexican monarchs, but also different kinds of CGs, as
indicated by polarity differences (Table 5-1; Seiber et al. 1986).
Even though the CGs in A A monarchs also have a greater emetic
potency in blue jays (Roeske et al. 1976: Table 6; Malcolm and Brower
1989: Table 9), these findings do not seem relevant here because mice
do not vomit (Landauer et al. 1985).
Each experimental run lasted for two nights. Immediately before a
trial, a mouse was transferred to a clean cage. At 1900 h either 5
Mexican monarchs, 5 A A monarchs, or 3 A A and 3 A A were placed
in the cage. All mice at least partially consumed all of these
monarchs. Mouse chow and water were available ad libitum. At 0900 h
the following morning, all remaining monarch material and mouse chow
was removed. Then at 1900 h on the same day, either 50 Mexican
monarchs, 50 A A monarchs, or a combination of 25 A A and 25 A A
monarchs were placed in the mouse's cage. The third treatment included

110
Table 5-1. Mean CG concentrations (|fg CG/0.1 g powdered butterfly)
and estimates of relative polarity (Rd values) of monarch
butterflies presumed or known to have fed on either A*. syriaca or A^_
curassavica. Rd values are derived from thin-layer chromatography
and are standardized to digitoxin; higher values indicate lower
polarity. The Rd value for the dominant CG in A. syriaca-reared
monarchs (aspecioside; Seiber et al. 1986, Malcolm et al. 1989) and
the range of Rd values for the 3 common CGs in A^. curassavica-reared
monarchs (uscharidin, calotropin and calactin; Roeske et al. 1976)
are presented. Because 85 - 92% of Mexican overwintering monarchs
fed as larvae on A*, syriaca plants (Seiber et al. 1986, Malcolm and
Brower 1989), I assume that aspecioside is the dominant CG in
Mexican monarchs.
Monarch
Sample
Mean CG
concentration
Rd value
of dominant
CG (s)
Overwintering, Sierra
Chincua, Mexico
40a
0.48b
Wild-reared in MA
on A^ syriaca
234b
0.48b
Laboratory-reared
in MA on L syriaca
27 8c
o
00
tr
Laboratory-reared in
fl on Aj. curassavica
377d
1.3 - 2.le
a Malcolm and Brower 1989.
b Malcolm et al. 1989.
c Brower and Glazier unpub1 data.
d Grand mean of 4 reported mean concentrations: 1) 319, n = 34,
Roeske et al, 1976; 2) 288, 15 males, Brower et al. 1975; 3) 336, n
= 15 females, Brower et al. 1975; and 4) 565, n = 13, S. Malcolm
unpubl. data.
e
Roeske et al. (1976: Figure 10).

Ill
both A^. and A^. fij. monarchs because of a limited number of available
A^_ monarchs. Mice always received the same type of monarchs during
both nights. At 0900 h the following morning, the mouse was removed
from its cage to tally patterns of feeding damage. As in Experiment 1,
only abdominal consumption patterns were recorded. The same number of
mice of each species received the Mexican monarchs (n = 7) , the A*,
monarchs (n = 5) , and the A^_ and A^. monarchs (n = 8) .
To distinguish the remains of A A and A^. monarchs after mice
had dismembered them, I placed a 1 mm diameter, red or gold spot of
nail polish on each butterfly's abdomen, thorax, and wings. The
assignment of gold or red for A^. or A*. L monarchs was determined
randomly for each trial. The nail polish was applied 5 h before each
trial. I also marked the A*. and Mexican monarchs in a similar
manner, except that only one color was used; this was determined
randomly for each mouse.
Statistical Analysis
All statistical tests and transformations followed Zar (1984) . In
Experiment 1, the one-way ANOVA and Scheffé F-test were used to compare
1) the number of abdomens eaten (selectively and non-selectively) by
each species when 50 monarch were offered and 2) the percentage of
abdomens eaten selectively by each species at the highest density.
These and all subsequent data were tested for normality with the
Kolmogorov-Smirnov test; the percentage data were arcsine transformed
prior to the normality test. A nonparametric test was used if the data
were not normally distributed (P < 0.05).

112
In Experiment 2, a one-way ANOVA and Scheffé F-test were performed
separately for each species to determine the effects of monarch type
(i.e., Mexican, A*. s^., or A,, + iu. C* monarchs) on the number of
abdomens eaten. The Kruskal-Wallis one-way ANOVA and Nonparametric
Tukey-type multiple comparison were performed separately for each
species to ascertain the effect of monarch type on the percentage of
abdomens eaten selectively. Wilcoxon signed-rank tests were used to
make intra-specific, paired comparisons of 1) the number of A •s.
versus A^ n*. abdomens eaten (selectively and non-selectively) for all 3
species and 2) the percentage of A^_ versus A.». abdomens eaten
selectively for melanotis.
Results
Experiment 1: Effect of Monarch Abundance on Foraging Patterns of Mice
Monarch abundance strongly influenced the number of abdomens eaten
by each species (Figure 5-1). When 10-20 monarchs were offered, all 3
species ate nearly all of the abdomens. However, when greater numbers
of monarchs were offered, nightly consumption by melanotis and
sumichrasti continued to increase whereas that of 2_*. hylocetes did
not. Nightly consumption levelled off for all 3 species when 40 or
more monarchs were offered. An interspecific comparison of the number
of abdomens eaten when 50 monarchs were offered was highly significant
(F = 6.74; df = 2,19; P = 0.0089); melanotis ate significantly more
abdomens than did £_*. hylocetes (Scheffé F-test; P < 0.05) . None of
the other pair-wise comparisons was significant (P > 0.05).
Monarch abundance also influenced the percentage of abdomens eaten
selectively by melanotis and £^_ sumichrasti. but not £_*. i. tiy.lQcetes.

Figure 5-1. Relationship of the number of abdomens eaten (mean ±S.E.) to the number of monarchs
offered in £*. melanotis, IL_ sumichrasti and E^_ hylocetes ■ The line of equality (y = x) is
presented for comparison. Alternate food (mouse chow) was present ¿d libitum-

Mean no. of abdomens eaten (±S.E.)
0 1 0 20 30 40 50
No. of monarchs offered

115
(Figure 5-2) . The latter species rarely fed selectively on any
abdomens. The relationship of the rate of selective feeding against
monarch abundance was S-shaped for both melanotis and R^_
siimichrasti. The upper plateau of the curves for these two species
differed slightly, but not significantly. An interspecific comparison
of the percentage of abdomens eaten selectively, when 50 monarchs were
offered, was highly significant (F = 38.84; df = 2,19; P = 0.0001); E^_
melanotis and E^. sumichrasti ate significantly higher percentages of
abdomens selectively than hylocetes (Scheffé F-test; P < 0.05) .
The percentage of abdomens eaten by E_^ melanotis and E^. sumichrasti did
not differ significantly (P > 0.05).
Experiment 2: Effect of CG Concentration on Foraging Patterns of Mice
The total number of abdomens eaten by E^. melanotis was not
influenced by monarch type (Table 5-2) . In contrast, IL». sumichrasti
ate significantly more abdomens from Mexican monarchs than they did
from A*. and the combination A*. and A^_ l monarchs. Peromyscus a.
hylocetes ate significantly more abdomens from Mexican and Aj.
monarchs than they did from the combination of A*, i*. and A^.
monarchs. Interspecific comparisons show that E_^ melanotis ate more
abdomens from all 3 treatments than did the other 2 species. These
results suggest that hylocetes and E^. sumichrasti were deterred
by the abdomens with higher levels of CGs, while melanotis was not.
Monarch type significantly influenced the percentage of abdomens
eaten selectively by both £_^ melanotis and E^. sumichrasti. but in
opposite directions (Table 5-3). The former species ate a
significantly higher percentage of the high CG abdomens (i.e., A*. S_*. +

Figure 5-2. Relationship of the percentage of abdomens eaten selectively (i.e.,
cuticle was rejected and the low CG, internal contents eaten; mean fS.E.) to the
monarchs offered in £_*. melanotis. sumichrasti and Ej_ hylocetes ■ Alternate
was present ¿d libitum.
the CG-laden
number of
food (mouse chow)

Mean % of abdomens eaten selectively (±S.E.)
Number of monarchs offered

118
Table 5-2. Mean number of monarch abdomens eaten both selectively
and non-selectively (±S.E.) during 1 night by each species.
Individual mice were offered either 50 Mexican monarchs, 50 A. s.
monarchs, or a combination of 25 A*. and 25 A^. monarchs. No
alternate food was present. The same number of mice of each species
received the Mexican monarchs (n = 7) , the A^. monarchs (n = 5) ,
and the A^. S_i. and A*. SL*. monarchs (n = 8) . Intraspecific comparisons
are made within each column with the one-way ANOVA. Different
subscripts (a, b) indicate significant differences among mean within
each column (Scheffé F-test; P < 0.05).
Type of
monarch Z*. melanotis R*. sumichrasti Z*. 3— hylocetes
Mexican
38.8 ±1.0a
26.0 ±1.4a
24.0 ±3
A â–  S a
38.3 ±l.la
18.4 ±2.2b
17.4 ±1
A. s . + A*. £_>.
35.7 ±1.9a
17.4 ±1.0b
10.1 ±1
F-ratio
1.77
13.06
10.47
df
2,19
2,19
2,19
P
0.2000
0.0004
0.0011

119
Table 5-3. Mean percentage of each type of monarch abdomen eaten
selectively (±S.E.) during 1 night by each species. Individual mice
received either 50 Mexican monarchs, 50 A*. monarchs, or a
combination of 25 A^. A*. and 25 A*. monarchs. No alternate food
was present. The same number of mice of each species received the
Mexican monarchs (n = 7) , the A*, A*. monarchs (n = 5) , and the A*. A-.
and A*. £*. monarchs (n = 8) . Intraspecific comparisons are made
within each column with the Kruskal-Wallis one-way ANOVA. Different
subscripts (a, b) within each column indicate significant
differences among
comparison).
means (P < 0.05;
Nonparametric
Tukey-type multiple
Type of
monarch
2-k melanotis
R. sumirhrasti
P. a. hvlnnetfis
Ma&isaa
45.8 ±2.6a
44.8 ±2.9b
3.3 ±1.5a
A*. A^
42.6 ±2.9a
34.4 ±5.3b
4.0 ±2.6a
A. s . + Aj. o â– 
56.1 ±1.5b
23.2 ±6.0a
o.oa
H-value*
9.80
7.10
4.41
No. of groups
2
2
2
P
0.0074
0.0287
0.1103
* corrected for ties

120
A^. ones) , whereas the latter species ate a significantly lower
percentage of the high CG abdomens selectively. The hylocet.es
rarely fed selectively on either type of abdomen.
For monarch treatment 3, individual mice received equal numbers of
¿L. 5^. versus A^. monarchs. Analysis of the predation records for
this treatment alone demonstrates that all three species fed
preferentially on L monarchs (Table 5-4). However, despite this
apparent interspecific similarity, it is important to note that P.
melanotis ate substantially more of both the A*. 5^. and the A^. c.
abdomens than did the other species. Of the A-. s^. and A^. abdomens
eaten, melanotis ate a significantly greater percentage of the A*.
ones selectively than they did of the A*. ones (mean = 67.2 and
13.0, respectively; Z-value with normal approximation = 2.37; n = 7; P
= 0.0180) . Thus £_*. melanotis not only ate more of the A^. abdomens
than did the other species, but they also fed selectively more often on
them.
Discussion
Influence of Monarch Abundance on Consumption Patterns
The results from Experiment 1 suggest earlier discrepancies in the
numbers of abdomens eaten and the frequency of selective feeding by P.
melanotis may be explained by the numbers of monarchs offered. When <
30 Mexican monarchs were offered, all mice ate the majority of the
monarch abdomens; of those abdomens eaten, most were done so non-
selectively. However, when > 30 monarchs were offered, large species
differences appeared. Nightly consumption by £_*. hylocetes did not
increase, whereas that of sumichrasti and E^. melanotis increased

121
Table 5-4. Mean number of A* a^. and Au £_*. abdomens eaten both
selectively and non-selectively (±S.E.) during 1 night by each
species. Each mouse received simultaneously 25 A. i* and 25 A*. £*.
monarchs. No alternate food was present. Intra-specific, paired
comparisons are made within each column with the wilcoxon signed-
rank test (with normal approximation).
Type of
monarch
£-. melanotis
£L. sumichrasti
P. a. hvlofistss
A^ 2^.
22.1 ±1.2
11.6 ±1.9
8.6 ±0.7
A*, z*.
13.6 ±1.9
1.6 ±0.6
1.6 ±0.4
Z-value
2.38
2.38
2.38
n
7
7
7
p
0.0176
0.0171
0.0180

122
until it plateaued at about 29 and 35 abdomens, respectively.
Moreover, the percentage of monarchs eaten selectively increased
dramatically for £_*. melanotis and sumichrasti. but not for E^. a.
hylocetes• The lower intake of abdomens by L. i, hylocetes cannot be
explained by lower energy demands. As compared to the other 2 species,
it is ^ 1.8 times heavier and eats on average 1.5 times more mouse chow
(Chapter 6) . Thus E^. hylocetes must have been more averse to the
Mexican monarchs than were the other species.
I can offer 2 possible explanations for why the selective feeding
behavior of melanotis and IL. sumichrasti was influenced so strongly
by the number of monarchs offered. As monarchs abundance increases,
the mice may simply become less tolerant of bitter-tasting body parts
(i.e., cuticle). Second, the mice may have to eat a large number of
monarchs (e.g., 30) before they learn how to feed on abdomens in this
manner both quickly and easily. There is some empirical support for
the latter hypothesis. When maintained on a diet of 50 monarchs per
night over 6 nights, the percentage of abdomens eaten selectively by P.
melanotis increased steadily from ca 60% on night 1 to about 80% on
night 6 (Chapter 4). Curiously, in the same experiment, the percentage
of abdomens eaten by R*. sumichrasti did not increase over the 6 nights.
Detailed observational analysis of both species feeding on varying
numbers of monarchs are needed to address these hypotheses.
Influence of CG Concentration and Polarity on Abdominal Consumption by
Mice
Reithrodontomys sumichrasti and E^. i. hylocetes ate slightly more
Mexican than monarchs, whereas E^. melanotis ate virtually

123
identical numbers of both types of monarchs. These findings suggest
that ÍL. sumichrasti and hylocetes. but not melanotis. were
deterred by the higher concentrations of the CGs in L. monarchs.
All 3 species avoided the ¿u. monarchs most strongly. However,
because the CGs in A_i. monarchs are less polar and more concentrated
than those in monarchs, it is difficult to determine whether it
was one or both of these factors that made the A^. monarchs so
unpalatable. Brower and Fink (1985) found that blue jays (Cyanocitta
cristata) were less deterred gustatorially by low than high polarity
CGs. Moreover, when I tasted the A^. and A^. monarchs, they seemed
qualitatively but not quantitatively similar; that is, both types of
monarchs had a similar bitter taste, but the bitterness of A^.
monarchs was more intense. These two findings suggest that it was the
higher concentration of CGs in the A*. monarchs that made them less
palatable to the mice.
Patterns of Selective Feeding
When E_^ melanotis was offered both A_^ and A*. monarchs, it fed
selectively on a significantly greater percentage of the A^.
monarchs. This finding supports the hypothesis (Chapter 4) that P.
melanotis selectively avoids the cuticle of abdomens with high
concentrations of CGs, thereby rendering such abdomens more palatable.
Two alternate explanations for the selective feeding behavior, which
predict that the selective feeding behavior should occur independently
of CG concentration and palatability, are not supported by this
finding. First, because small mammals do not appear to digest cuticle
to any significant extent (Griffith and Gates 1985, M. Allen unpubl.

124
data, but see Jeuniaux 1961), they should avoid ingesting it when
possible and thereby increase their net energy yield of foraging (see
Krebs 1978, Belovsky 1978) . Second, mice avoid cuticle because of its
tough and leathery quality.
Unlike melanotis. EL. sumichrasti fed selectively on a greater
percentage of abdomens from low-CG monarchs (i.e., Mexican and
ones). I suggest that this can be explained by their relatively high
sensitivity to the taste of CGs (Chapter 6). The high CG
concentrations in the A. s. and A*. monarchs may have been so aversive
that they disrupted IL. sumichrasti1s normal foraging behavior, thus
causing it to eat fewer abdomens and to feed selectively less often.
It is likely that the internal tissues of many of these monarchs had
intolerable levels of CGs (Table 4-2 and Figure 6-6). In the next
Chapter, I maintain all three species on a mouse chow diet treated
with a relatively high concentration (54.6 |ig/0.1 g chow) of a CG
(digitoxin). Whereas both EL. melanotis and E^. hylocetes eventually
habituated to the diet's taste and fed and gained weight normally, R.
sumichrasti did not habituate and eventually rejected the diet
altogether.
It is unclear why E^. A*, hylocetes rarely fed selectively on
abdomens. They may possess much less flexible food handling
capabilities than do the other 2 species, which learned the selective
feeding behavior during their first night of exposure to monarchs (see
Cyr 1972) . Peromyscus hylocetes did not acquire this behavior after
feeding on a diet consisting soley of Mexican monarchs for 6
consecutive nights (Chapter 4) . If E_^ 2^. hylocetes had learned to

125
reject the CG-laden cuticle, it is likely that it would have ingested
substantially greater numbers of monarch abdomens.
Potential Influence of PAs on Monarch Consumption
Several authors have suggested that PAs may play an important
defensive role for monarchs (Edgar et al. 1976, Rothschild and Marsh
1978, Eisner 1980, Kelley et al. 1987, but see Brower 1984). However,
no study to my knowledge has shown this to be the case. In fact, the
results presented herein and in Chapter 6 suggest that PAs may play a
minor defensive role. In this chapter, sumichrasti and E^. a.
hylocetes were more strongly deterred by the R*. S_*. monarchs, which
lacked PAs because they had no access to PA sources as larvae and
adults, than they were by the wild-caught Mexican monarchs, which are
known to possess PAs (Kelley et al. 1987) . Nevertheless, it is
conceivable that E_^ melanotis and sumichrasti fed selectively on a
slightly, but not significantly, higher percentage of Mexican than
monarchs because the additional presence of the PAs may have
rendered the cuticle less palatable.
Why do R. sumichrasti and P. a. hylocetes Avoid Eating Overwintering
Monarchs in Msxirn?
In this chapter, neither R«. sumichrasti nor E^. ^ hylocetes
appeared to be strongly averse to Mexican monarchs. However, I suggest
that free-ranging sumichrasti and £_^ hylocetes are much less
likely than captive ones to eat Mexican monarchs, and that free-ranging
individuals of both species are sufficiently deterred by the taste of
the CGs to avoid taking advantage of colonies. This hypothesis is
supported by two lines of evidence. When R^. sumichrasti. £_^

126
hylocetes and 2^. melanotis were fed a diet consisting soley of Mexican
monarchs for 6 nights, 2^. melanotis was the only species able to
maintain weight (Chapter 4). Second, when offered paired choices
between control diets and diets treated with a series of different
concentrations of digitoxin (a type of CG) , both sumichrasti and £_*.
3^. hylocetes avoided the digitoxin diets at concentrations well below
the mean for Mexican monarchs (Chapter 6). In contrast, R^. melanotis
did not reject the digitoxin diets until a concentration above the mean
for overwintering monarchs.

CHAPTER 6
COMPARATIVE RESPONSES TO THE TASTE AND TOXIC EFFECTS OF CARDIAC
GLYCOSIDES AND PYRROLIZIDINE ALKALOIDS
The results in Chapters 4 and 5 suggest strongly that £_ melanof.is
is much more tolerant to the taste of the CGs and/or PAs in
overwintering monarch butterflies in Mexico. In this Chapter, I
further examine this hypothesis by comparing the responses of P.
melanotis. a* hvlocetes and IL. sumichrasti to representative CGs and
PAs in artificial diets. The experiments were designed to determine
the relative importance of post-ingestional versus taste cues in
determining each species' respective responses to the compounds.
Methods and Materials
Experiment 1: Determination of Taste Sensitivity
In this experiment individual mice melanotis. P. a. hylocetes.
and sumichrasti) were subjected to a series of paired-choices
between a control and a tastant diet treated with successively
increasing or decreasing concentrations of a representative CG
(digitoxin) or one of two representative PAs (free-base and N-oxide
monocrotaline). Both forms of monocrotaline were used because the PAs
in overwintering monarchs in Mexico also occur in these two alternate
forms (Kelley et al. 1987). I used digitoxin and monocrotaline as
representative CGs and PAs, respectively, because 1) the particular CGs
127

128
and PAs in overwintering monarchs are not commercially available and 2)
overwintering monarchs have a bitter taste similar to digitoxin and
monocrotaline (Glendinning, pers. observ.). To quantify each species'
taste sensitivity, a modified form of the so called method of limits
was used to determine each species' respective avoidance threshold for
each compound (see difference threshold; Gescheider 1976).
In March 198 6 I trapped 10 E^. melanotis, 10 L. i.. hylocetes. and 10
B-. sumichrasti (five of each sex for each species) < 500 m from the
overwintering colony of monarch butterflies on Sierra Chincua in
Michoacán, México (Calvert and Brower 1986) and transported them to the
University of Florida in Gainesville, Florida. They were housed in
plastic cages (30 x 12 x 16 cm) with dacron batting, and maintained on
Purina Rodent Lab Chow 5001 (henceforth, mouse chow) and water ad
libitum in a temperature/humidity controlled room, on a 13 h light: 11
h dark cycle. I used the F^_ offspring of these mice (60 melanotis.
60 a-, hylocetes and 42 sumichrasti) for testing in Gainesville,
Florida when they were 90 to 120 days old.
I also live-trapped nine adult 2^. melanotis > 2.5 km to the east of
the site of the 1986 monarch colony on 8 and 9 January 1988 (see Figure
2-1). This area was on the eastern slope of Sierra Chincua, where
monarch colonies do not form (Calvert and Brower 1986), and I assume
that populations of mice in this area have not encountered
overwintering monarchs. These mice were housed and maintained in a
manner similar to that described above, except that they were kept in a
tent in the shade next to our campsite on Sierra Chincua. They
experienced natural daily fluctuations in temperature (about 4 to 16°

129
C) and light (about 12 h light: 12 h dark). After testing, all of
these mice were released.
For testing in Gainesville, I rehoused the mice individually in
stainless steel, wire-bottomed cages (25 x 42 x 18 cm) with dacron
batting. These cages had two glass food cups (5 cm tall, 5.5 cm
diameter, with a 3.5 cm diameter hole in the plastic lid) secured to
the front side of the cage, each eight cm from the respective corners.
Spillage was intercepted with two petri dishes (16 cm diameter) placed
underneath each food cup outside the cage. One water bottle was
attached to the cage directly in between the food cups. The tastant
and control diets were presented to the mice in the food cups.
For testing on Sierra Chincua, I rehoused the 9 E^. melanotis
individually in plastic cages identical to those described above,
except that two food cups were secured to the front right and left-hand
corners. Spillage was collected by resting each food cup in a petri
dish (9 cm diameter). A water bottle was placed directly in between
the food cups.
To prepare the tastant diets, I treated Purina Rodent Lab Meal no.
5001 (henceforth, mouse chow meal) with either digitoxin, free amine
monocrotaline or N-oxide monocrotaline. The digitoxin and free amine
monocrotaline were purchased from Sigma Chemical Co., St. Louis, MO,
while the N-oxide monocrotaline was derived from the free amine using
the technique outlined in Schwartz and Blumbergs (1964) and Craig and
Purushothaman (1970). Exact concentrations of these compounds were
established in the tastant diets by 1) dissolving a given quantity of
one of the compounds in 500 ml of 95% ethanol, 2) mixing that solution
thoroughly with 500 g of lab chow meal, and 3) drying the diet for 14

130
hours at 60°C in a forced-draft oven. The control diet was prepared
similarly, except that no tastant was added. All diets were stored in
a refrigerator.
Mice were subjected to a series of consecutive, 48 h, two-choice
preference tests between the tastant and control diets. The
concentration (|ig of compound/0.1 g mouse chow) of digitoxin, free
amine monocrotaline or N-oxide monocrotaline in the tastant diet was
systematically increased or decreased each 48 h in natural log
increments. For example, melanotis received 5 ascending
concentrations of digitoxin: e2 = 7.4, e2 = 20.1, e4 = 54.6, e5 =
148.4, e® = 403.4 (p.g digitoxin/0.1 g mouse chow). The range of
concentrations each mouse received was species- and compound-specific
(see Table 6-1); the ranges were determined during preliminary feeding
trials. Individual mice were exposed to either an ascending or
descending series of one of the compounds, and mice from the same
litter were dispersed roughly evenly among the tastant and control
diets. Individual mice were tested only once.
To familiarize the mice with the mouse chow meal, they received the
control diet in both food cups during the first 48 h period. Between
1600 to 1800 h, approximately 15 g of the control diet was placed in
each food cup. Twenty four h later, the amount of diet consumed from
each food cup was determined gravimetrically (to the nearest 0.1 g).
The food cups were then refilled and their positions in the testing
cage reversed for the next 24 h period. After this and all subsequent
48 h periods, the same procedure was repeated, except that: 1) the
tastant diet was placed in one cup and the control diet in the other;
2) we used clean food cups; and 3) fresh control diet was used. All

Table 6-1. The range of concentrations (|ig compound/0.1 g mouse chow) of digitoxin, free amine
monocrotaline and N-oxide monocrotaline used in the tastant diets. Each mouse received an ascending
(A) or descending (D) series of concentrations of one of the compounds in natural log steps. For
example, £^_ melanotis received the following ascending series of digitoxin concentrations: e2 = 7.4,
e3 = 20.1, e4 = 54.6, e5 = 148.4, and e6 = 403.4 Jig digitoxin/0.1 g mouse chow meal. The ranges of
concentrations used for each compound and species were determined during preliminary feeding trials.
All mice were collected < 500 m from a monarch colony on Sierra Chincua, except for the E_*. melanot i s
designated as (far), which were collected > 2.5 km from the same monarch colony. The number of mice
of each species exposed to each compound and series of concentrations is in parentheses.
Mouse Type of Digitoxin Free amine N-oxide
species series monocrotaline monocrotaline
melanotis A 7.4 -
(10)
D 403.4
(10)
EL. melanotis (far) A 7.4 -
( 9)
£_*. it hylocetes A 0.38
(10)
D 20.1
(10)
A 0.14
( 7)
D 2.7 -
( 7)
403.4
7.4 - 148.4
7.4 -
(10)
(10)
- 7.4
148.4 - 7.4
403.4
(10)
(10)
54.6
—
—
54.6
7.4 - 148.4
7.4 -
(10)
(10)
1.0
148.4 - 7.4
403.4
(10)
(10)
54.6
7.4 - 148.4
7.4 -
( 7)
( 7)
0.14
148.4 - 7.4
403.4
( 7)
< 7)
403.4
M
UÜ
M
- 7.4
t
403.4
- 2.7
403.4
- 2.7
IL. sumichrasti

132
mice received the tastant diet on the left side for the first 24 h. In
the final 48 h period of all descending series, control diet was placed
in both food cups.
For each mouse, I calculated a preference ratio for each 48 h
exposure to a given tastant concentration by dividing the amount of
tastant diet eaten by the total amount of both diets eaten. Equal
consumption of both diets yielded a preference ratio of 0.50, whereas
total avoidance of the tastant diet yielded a preference ratio of 0.00.
I then calculated each species' mean preference ratio for each of the
concentrations of digitoxin and both types of monocrotaline.
The avoidance threshold for both ascending and descending series
were set a. priori as the lowest concentration at which the mean
preference ratio was significantly below 0.50 [P < 0.01; one-sample
(one-tailed) t-test]. All preference ratios were transformed (arcsine
•Vx) before statistical analyses. Then, to estimate each species actual
avoidance threshold for each compound, I took the average of the
ascending and descending thresholds.
I also wanted to assess whether the total amount of mouse chow
eaten (control and tastant diets combined) by mice remained constant as
I varied 1) the concentrations of the 3 compounds in the tastant diets,
and 2) the order of their presentation. To do this, I ran a separate
3-factor repeated measures ANOVA for each species on the total amount
of diet eaten each 48 hour period by compound type, order of
presentation, and concentration (the repeated measure). However,
because each species was exposed to a different range of concentrations
of each compound (Table 6-1) , only the four lowest concentrations were
included. The range of concentrations used in the analysis for 1) 2-*.

133
melanotis. was 7.4 to 148.4 p.g/0.1 g diet for all 3 compounds; 2) P. a .
hylocetes. was 1.0 to 20.1 for digitoxin and 7.4 to 148.4 for both
forms of monocrotaline; and 3) E*. sumichrasti. was 0.14 to 2.7 for
digitoxin and 7.4 to 148.4 for both forms of monocrotaline. The data
from the E*. melanotis collected 2.5 km from the monarch colony were not
included in the analysis. In this and all subsequent ANOVAs, the data
were shown to be normally distributed with the Kolmogorov-Smirnov test
(P > 0.25) .
Experiment 2; Determination of Toxic Sensitivity
In this experiment, I determine for all three species the
physiological consequences of chronically eating diets containing
digitoxin and monocrotaline at concentrations similar to those used in
Experiment 1. Because orally ingested N-oxide monocrotaline is
converted to the free amine form in the small intestine of mice
(Mattocks 1971), I only tested the free amine form of monocrotaline.
Mice were housed and maintained in Gainesville, Florida, under the
same conditions as in Experiment 1. Thirty-eight day old, F]_ offspring
of wild-caught mice (40 melanotis. 41 L. ix hylocetes. and 16 R.
sumichrasti) were used. All three species were tested with digitoxin,
whereas only the former two were tested with free amine monocrotaline.
For testing, mice were rehoused in the same stainless steel cages
described above. One glass food cup and water bottle were secured to
the front of each cage and a petri dish (16 cm diameter) was placed
under each food cup to intercept spillage. The tastant diet was
presented in a single food cup.

134
I prepared the control and tastant diets in the same manner
described above. However, I used only two concentrations of free amine
monocrotaline and digitoxin: 7.4 and 54.6 |ig of either compound per 0.1
g mouse chow meal. These concentrations are similar to those found in
overwintering monarchs (see Figures 6-6a and 6-6b).
Each mouse was given approximately 15 g of the control diet during
the first two nights of testing; preliminary feeding trials
demonstrated that individuals of all 3 species usually ate less than
half this amount of mouse chow meal during a given night. Then, for
each of the next 26 nights each mouse was offered an equivalent amount
of either the control or one of the four tastant diets. I monitored
each mouse's weight to the nearest 0.25 g with a 50 g Pesóla scale
every other day between 1600-1800 h, and food consumption to the
nearest 0.1 g with a Mettler P1000 balance every day during the same
time period. Littermates were distributed roughly evenly among the
different dietary treatments.
Two-way repeated measure ANOVAs were used to test for the effects
of age (in days) and diet type on 1) the weight change of the mice and
2) the amount of diet eaten per day. These tests were performed
separately for each species and compound.
Interspecific comparisons of the nightly dosages (mg compound/kg
mouse) of digitoxin and free amine monocrotaline were made with a one¬
way ANOVA and Scheffé F-tests. To compute these nightly dosage values
I divided the amount of compound ingested each night by a given mouse
(which I could estimate accurately based on the amount of food eaten)
by its weight each night. Since we collected weights every other day,

135
I assumed that each mouse's weight remained constant over the 48 h
periods between weighings.
Results
Experiment 1: Determination of Taste Sensitivity
The avoidance thresholds derived from ascending and descending
series of all three compounds were either the same or within one
natural log step of one another in all cases, except for those from £_*.
á_». hylocetes with N-oxide monocrotaline, which differed by 2 natural
log steps (Table 6-2). Because the responses to the ascending and
descending series were so similar, I 1) determined the mean avoidance
threshold for each compound and species by averaging the means from
both series, and 2) pooled the results from both series and then
plotted each species' mean preference ratio as a function of the
concentration of digitoxin (Figure 6-la), N-oxide monocrotaline (6-lb)
and free amine monocrotaline (6-lc) in the tastant diets. The mean
avoidance thresholds are indicated with arrows on the x-axes these 3
plots.
All three species of mice responded quite differently to the
digitoxin diets. Peromyscus melanotis' mean avoidance threshold was
about 38 times higher than sl^ hylocetes ' , and á_. hylocetes ' was
about 19 times higher than sumichrasti's. The melanotis
collected nearby (< 500 m) and far away (> 2.5 km) from Sierra Chincua
exhibited identical ascending avoidance thresholds for digitoxin (Table
6-2). In contrast to the results with digitoxin, all three species
responded relatively similarly to the N-oxide and free amine
monocrotaline diets. Peromyscus melanotis' mean avoidance threshold

Table 6-2. Avoidance threshold concentrations of digitoxin, free amine monocrotaline and N-oxide
monocrotaline for £_^ melanotis. £_*. hylocetes, and L sumichrasti from ascending and descending
series of presentations. The mean avoidance threshold is the average of the ascending and descending
thresholds. I also present the mean avoidance threshold as a molar concentration. All mice were
collected < 500 m from a monarch colony on Sierra Chincua, except for the IL_ melanotis designated as
(far), which were collected > 2.5 km from the same monarch colony.
Avoidance threshold concentrations Mean
()lg compound per O.lg mouse chow) avoidance
threshold
Compound
Species
Ascending
series
Descending
series
Mean avoidance
threshold
in moles compd.
per g mouse chow
Digitoxin
E^
melanotis
54.6
148.4
101.5
1.3
X
10"6
E^
melanotis (far)
54.6
-.-
E_
^ Rylacetes
2.7
2.7
2.7
3.5
X
10-8
SUmiChraSti
0.14
0.14
0.14
1.8
X
10-9
Free base
E_
melanotis
148.4
54.6
101.5
3.1
X
10-6
monocrotaline
Ej_
i. hylocetes
54.6
54.6
54.6
1.7
X
10“6
R+.
¿ümiChraSti
148.4
54.6
101.5
3.1
X
10-6
N-oxide
E_
melanotis
148.4
54.6
101.5
3.0
X
10-6
monocrotaline
E_
a^. hylocetes
54.6
7.4
31.0
9.1
X
10-7
sumichrasti
54.6
20.1
37.4
1.1
X
10-6

Figure 6-1. Mean preference ratios for different concentrations of
digitoxin, N-oxide monocrotlaine and free amine monocrotaline ((J.g/0.1 g
diet) in 2-^ me.ianotis (£-. HU) , E*. hvlocetes (E^. lu) and
sumichrasti . Each mean was calculated from preference ratios
from both ascending and descending series of presentations. The
species-specific mean avoidance thresholds for each compound are
indicated with arrows on the respective x-axes and also in Table 6-2.
A mean preference ratio of 0.5 (indicated by the dotted line)
represents no preference for the control or tastant diet, while a ratio
of 0.0 represents total rejection of the tastant diet.

Mean preference ratio (±S.E.)
138
o P melonotis
Concentration (¿ig/O.lg diet)

139
for the N-oxide was about 3 times higher than that of the other
species, and its mean avoidance threshold for the free amine was equal
to R^. sumichrasti' s and only about 2 times higher than
hylocetes1.
The results of 3-way ANOVAs performed separately on each species
indicate that the amount of food eaten over each 48 h period was not
influenced significantly by the type of compound, order of
presentation, nor concentration of the compound (Table 6-3). Thus,
within each species, all mice ate a relatively constant and similar
amounts of mouse chow meal throughout the two-choice feeding
experiments (mean no. of grams ingested ±S.E. per 48 h period by E_*.
melanotis = 6.3 ±0.06, E_^ hylocetes = 9.7 ±0.13, and R^_ sumichrasti
= 6.3 ±0.06). What differed between the 48 h periods was the relative
amounts of control 2S.. tastant diet eaten.
Experiment 2: Determination of Toxic Sensitivity
All mice fed relatively normally on the control diets and the diets
treated with digitoxin and free amine monocrotaline, except for 3 of
the E— hylocetes and all 5 of the sumichrasti given the high
digitoxin diet (Table 6-4). These 8 mice ate extremely small amounts
of the diet each night (about 0.5 g) and lost weight quickly. They
were removed from the experiment within 7 days so they would not die,
and were not included in subsequent data analyses.
The effects of the digitoxin diets on growth and consumption rate
on all three species of mice are illustrated in Figures 6-2 and 6-3 and
analyzed statistically in Table 6-5. Addition of digitoxin to the
diets did not influence E^_ melanotis' and E^. hylocetes1 growth rate.

140
Table 6-3. Three-way repeated measure ANOVA of mouse chow eaten per
48 h period by compound (digitoxin, N-oxide monocrotaline and free
amine monocrotaline), order of presentation (ascending or
descending), and concentration of the compound in the tastant diet
(the repeated measure). This analysis was performed separately for
£* melanotis. ^ hylocetes. and sumichrasti.
Source of variation
df
F
P
£_>. melanotis
Compound (A)
2
0.88
0.4188
Presentation order (B)
1
0.21
0.6516
A x B
2
0.02
0.8170
Concentration (C)
3
0.18
0.9116
A x C
6
1.51
0.1772
B x C
3
0.37
0.7759
A x B x C
6
0.28
0.9471
2-l. au. hylocetes
Compound (A)
2
1.12
0.3344
Presentation order (B)
1
1.52
0.2229
A x B
2
1.26
0.2922
Concentration (C)
3
1.64
0.1830
A x C
6
1.58
0.1572
B x C
3
1.16
0.3259
A x B x C
6
0.56
0.7611
R. sumichrasti
Compound (A)
2
1.13
0.3355
Presentation order (B)
1
0.49
0.4908
A x B
2
0.61
0.5503
Concentration (C)
3
1.26
0.2933
A x C
6
0.69
0.6582
B x C
3
1.36
0.2578
A x B x C
6
0.10
0.9965

141
Table 6-4. The number of juvenile mice tested under each dietary
treatment and the percentage of these mice that fed normally and
gained weight on the diets.
Compound
Mouse
Species
Concentration (|ig cmpd.
/0.1 g diet)
0
n
i.O
%
7
n
.4
%
54.6
n %
Digitoxin
£L. melanotis
8
100
8
100
8
100
E-u SL*. hylocetes
8
100
8
100
9
67*
R*. sumichrasti
6
100
5
100
5
0*
Free amine
Ej. melanotis
8
100
8
100
8
100
monocrotaline
Ej. hylocetes
8
100
8
100
9
100
* Three of the E-l. a* hylocete3 and all 5 R*. sumichrasti ate
exceedingly small amounts of the diet and thus lost weight
precipitously. They were removed from the experiment within 7 days.

Figure 6-2. Weight changes over time in melanotis (n =
8/treatment), hylocetes (n = 7-8/treatment), and R*. sumichrasti
(n - 5-6/treatemnt) that were maintained on one of 3 dietary
treatments. The treatments differed only with respect to the
concentration of digitoxin. All mice began the experiment at 40 days
of age.

143
o Control diet

Figure 6-3. Amount of diet eaten over time by melanotis (n =
8/treatment) , R*. hylocetes (n = 7-8/treatment), and R*. sumichrasti
(n * 5-6/treatemnt) that were maintained on one of 3 dietary
treatments. The treatments differed only with respect to the
concentration of digitoxin.

145
o Control diet

146
Table 6-5. Results of 6 two-factor repeated measure ANOVAs on the
effects of diet type and age on the dependent measures 1) weight of
mouse/every other night, and 2) amount of diet eaten/mouse/night.
The diet factor included three concentrations of digitoxin (0.0,
7.4 or 54.6 Mg/0.1 g diet) for Peromyscus melanotis and ^
hvlocetes. and only two (0.0 and 7.4 Mg/0.1 g diet) for R.
sumichrasti. See Table 6-4 for the number of mice used in each
ANOVA and Figures 6-2 and 6-3 for graphical presentation of the
data.
Dependent
measure
Species
Source of
variation
F
df
P
Weight of
Eu. melanotis.
Diet type
0.47
2
0.6292
mouse
Age
179.517
13
0.0001
Interaction
0.72
26
0.8464
P. a. hvlocetes
Diet type
0.18
2
0.8395
Age
393.26
13
0.0001
Interaction
2.53
26
0.0001
Buu sumichrasti
Diet type
23.96
1
0.0009
Age
76.75
13
0.0001
Interaction
6.15
13
0.0001
Amount of
E*. melanotis
Diet type
4.06
2
0.0324
diet eaten
Age
5.31
25
0.0001
Interaction
1.15
50
0.2327
P. a. hvlocetes
Diet type
0.46
2
0.6390
Age
26.34
25
0.0001
Interaction
2.96
50
0.0001
Bul. sumichrasti
Diet type
6.37
1
0.0326
Age
16.21
25
0.0001
Interaction
1.88
25
0.0087

147
The effect of age was significant for both species, and the interaction
of diet type and age was nonsignificant for E_^ melanotis and
significant for E_^ a^. hylocetes. In contrast, addition of digitoxin
greatly diminished sumichrasti's growth rate. There was a
significant effect of diet type and age on weight change; the
interaction of these factors was also significant.
Even though all three species appeared initially deterred from
eating the digitoxin treated diets (see days 39-42 in Figure 6-3), E_^
a^. hylocetes and E^. melanotis eventually began eating equal or even
greater amount of the control and digitoxin-treated diets. In
contrast, sumichrasti appeared to be deterred from eating the
digitoxin-treated diets throughout the entire experiment. It largely
rejected the diet with the high concentration of digitoxin and ate
significantly less of the one with the low concentration of digitoxin
than it did the control diet. This result may explain ¡L». sumichrasti' s
depressed growth rate on the diet with low concentrations of digitoxin.
The effects of the free amine monocrotaline diets on growth and
consumption rate in E^. melanotis and E^. hylocetes are illustrated in
Figures 6-4 and 6-5 and analyzed statistically in Table 6-6. Addition
of free amine monocrotaline had no impact on growth rate in P.
melanotis. but had a dose-dependent, negative impact in P. a.
hylocetes. Nevertheless, diet type had no influence the amount of food
eaten by either species.
Even though £_*. ahylocetes ate more of the digitoxin- and
monocrotaline-treated diets on a given night than did the two smaller
species (see Figures 6-3 and 6-5), and thus also ingested greater
quantities of each compound, their mean nightly dosage of these

Figure 6-4. Weight changes over time in melanotis (n = 8/treatment)
and hylocet.es (n = 8-9/treatment) that were maintained on one of
3 dietary treatments. The treatments differed only with respect to the
concentration of free amine monocrotaline. All mice began the
experiment at 40 days of age.

149
I
° Control diet
i
• 7.4^g monocrotoline/O.lg diet
19
â–  54.6 uq monocrotoline/O.lg diet
18
17
16
_ 15
o*
w 14
0)
Q
P metanotis
£ 13
«y
o 12
*
•; 39
5
c 37
o'
o
s'
5 35
33
31
29
P 0 hytocetes
27
<_,38 42 46 50 54 58 62 66
Age (days)

Figure 6-5. Amount of diet eaten over time by melanotis (n =
8/treatment) and hylocetes (n = 8-9/treatment) that were
maintained on one of 3 dietary treatments. The treatments differed
only with respect to the concentration of free amine monocrotaline.

Mean amount of diet eaten (g)
151
o Control diet
• 7.4monocrotaline/O.I g diet
46
56
62
66

152
Table 6-6. Results of 4 two-factor repeated measure ANOVAs on the
effects of diet and age on the dependent measures: 1) weight of
mouse/every other night, and 2) the amount of diet
eaten/mouse/night. The diet factor included 3 concentrations of
free amine monocrotaline (0.0, 7.4 and 54.6 |ig/0.1 g diet). See
Table 6-4 for the number of mice used in each ANOVA and Figures 6-4
and 6-5 for a graphical presentation of the data.
Dependent
measure
Species
Source of
variation
F
df
P
Weight of
S-l. melanotis
Diet type
0.004
2
0.9958
mouse
Age
185.49
13
0.0001
Interaction
2.03
26
0.0029
P . a. hvloeetes
Diet type
11.95
2
0.0003
Age
175.92
13
0.0001
Interaction
5.35
26
0.0001
Amount of
S-l. melanotis
Diet type
0.001
2
0.9986
diet eaten
Age
13.44
25
0.0001
Interaction
2.45
50
0.0001
P. a. hvlnretes
Diet type
0.78
2
0.4688
Age
17.42
25
0.0001
Interaction
0.86
50
0.7452

153
compounds was significantly less (Table 6-7) . Re-i thmrinntnmys
sumichrasti's nightly dosage of digitoxin in the 7.4 (ig/0.1 g diet was
significantly higher than that of the other two species (P < 0.05;
Scheffé F-test) .
Discussion
Differences in Taste Sensitivities Among the Three Species
The mean avoidance thresholds for digitoxin differed among the 3
species by several orders of magnitude, with £_*. melanotis showing the
highest thresholds. This finding agrees with results from the feeding
experiments in Chapters 4 and 5, where melanotis appeared to be more
tolerant to the CGs in monarchs than were hylocetes and R.
sumichrasti. That the melanotis collected > 2.5 km away from the
Sierra Chincua overwintering area had the same threshold for digitoxin
as ones collected < 500 m from the overwintering area suggests that CG
taste-insensitivity is a trait characteristic of the species rather
than a specific adaptation of the population in the overwintering areas
(see Chapter 8 for elaboration of this hypothesis).
That such closely related species of mice responded so differently
to the taste of digitoxin is not surprising. Studies with laboratory
mice have revealed inter-strain differences in taste sensitivity to
several bitter compounds that are of comparable magnitude to those
reported in this study. The bitter compounds include sucrose octa-
acetate (Warren and Lewis 1970, Lush 1981, Harder et al. 1984), quinine
sulfate (Lush 1984), strychnine (Lush 1982), and the acetates of
raffinose, galactose and 15-lactose (Lush 1986) .

154
Table 6-7. Nightly dosage (mg/kg mouse) and intake (mg) of digitoxin
and free amine monocrotaline (mean ±S.E.) by mice over 26 nights in
diets with one of two concentrations of either compound.
Interspecific comparisons are made within each column, separately for
each compound and concentration, with the one-way ANOVA. Different
subscripts (a, b, c) indicate significant differences among means
within each column for the 3 species comparisons (P < 0.05; Scheffé
F-test). * P < 0.0009
; * * P < 0.0001.
Concentration
in diet Mouse
Compound (|lg/0.1g) species
n
Nightly
dosage
(mg/kg)
Amount
ingested/
night
(mg)
Digitoxin 7.4
P . a . hvlnnet.ps
8
10.2 ±0.5C
0.33 ±0.la
melanotis
8
15.3 ±0.1b
0.24 ±0.1b
ÍL. sumichrasti
5
18.1 ±0.5a
0.20 ±0.1c
F-ratio
df
72.8**
2
-74.4**
2
54.6
hylasatea
7
72.2 ±3.8
2.37 ±.14
£_>. melanotis
8
107.0 ±1.8
1.72 ±.03
F-ratio
78.3*
-22.1*
df
1
1
Monocrotaline 7.4
tolacstaa
8
9.6 ±0.5
0.30 ±.02
E-m. melanotis
8
14.3 ±0.4
0.23 ±.01
F-ratio
df
68.7*
1
-16.9*
1
54.6
P . a . hvlofiet.es
9
74.0 ±2.7
2.18 ±.10
melanotis
8
104.9 ±2.5
1.62 ±.05
F-ratio
df
69.3*
1
-25.1*
1

155
In contrast to digitoxin, the mean avoidance thresholds for
monocrotaline (both free amine and N-oxide forms) did not differ
greatly among species. This indicates that: 1) the divergent responses
these 3 species of mice show to overwintering monarch butterflies is
probably not influenced strongly by their respective taste
sensitivities to PAs; and 2) there is no significant difference in the
gustatory properties of free amine and N-oxide forms of PAs, even
though the later form is slightly more polar, owing to it being a salt.
It is notable that the avoidance thresholds (in molar units) for
digitoxin and both forms of monocrotaline were similar in R^. melanot.is.
but widely divergent in L. hylocetes and R^. sumichrasti (Table 6-2).
Thus digitoxin appears to be more effective than monocrotaline as an
aversive stimulus against R^ i*. hylocetes and R^. sumichrasti. It would
be interesting to determine whether this applies to other kinds of CGs
and PAs.
Ecological Relevance of the Avoidance Thresholds
To place the avoidance thresholds of the three species of mice in
an ecological context, I overlaid each species' avoidance threshold for
digitoxin and monocrotaline on the natural frequency distributions of
CGs and PAs, respectively, as found in Mexican overwintering monarchs
(Figures 6-6a and 6-6b). (Because the avoidance thresholds for the
free amine and N-oxide forms of monocrotaline were so similar, I
averaged them for each species, and present these averages on Figure 6-
6b.) Figure 6-6a indicates that for R^. .a^. hylocetes and R.
sumichrasti. more than 75% of the monarchs have suprathreshold
concentrations of CGs, and for R^. melanotis. only 13% do. In contrast

Figure 6-6. Frequency distribution of the whole body concentration of
(A) CGs and (B) PAs in overwintering monarchs on Sierra Chincua, one of
the primary sites in Mexico. The CG data were derived from Malcolm and
Brower (1989), and the PA data calculated from results in Kelley et al.
(1987) and Brower and Calvert (1985). The mean avoidance thresholds
for digitoxin and monocrotaline in melanotis. P. a. hylocete3. and
E^. sumichrasti are indicated on the respective CG and PA distributions.
Rather than presenting separate mean avoidance thresholds for free
amine and N-oxide monocrotaline, I present averages of these 2
thresholds for each species.

% of sample % of sample
35 -\
CM
CO
CO
o
CM
CO
00
o
CM
CD
CO
o
CM
CD
00
O
CM
CO
CD
r*.
00
O)
o
CM
CO
JO
CO
CO
o>
o
CM
CM
CM
CM
CM
O
CM
CO
CO
d
CM
if
d
CM
C*5
CO
f"*
CO
to
CO
CM
CM
â–  CO
00
d
CM
00
o>
o
o
CO
m
CO
00
o>
O
CM
CM
CM
CM
Whole body concentration (pg compound/0.1 g dry weight of monarch)

158
Figure 6-6b indicates that for all three species, less than 25% of the
monarchs have suprathreshold concentrations of PAs. In fact, for P.
melanotis. all monarchs have subthreshold concentrations of PAs.
Extrapolation of these results to free-ranging mice suggests that the
encounter rate of palatable monarchs (i.e., ones with subthreshold
concentrations of PAs and CGs) on the forest floor inside a colony by
2.*. a^. hylocetes and sumichrasti would be less than 1 out of 4,
whereas that by melanotis would be 3 out of 4. The low encounter
rate for £_^ hylocetes and R*. sumichrasti may explain why they do not
feed naturally on overwintering monarchs.
In making these extrapolations, I am assuming that the tastes of
the CGs and PAs in overwintering monarchs are similar to those of
digitoxin and monocrotaline, respectively, in mouse chow meal.
However, there are several factors that may cause the tastes to differ
quantitatively and/or qualitatively. First, because aspecioside (the
dominant CG in overwintering monarchs) is more polar than digitoxin
(Table 4-1), it may be avoided at a lower concentration (Brower and
Fink 1985). Second, CGs bound in monarch cuticle may be tasted more
weakly than ones mixed into mouse chow (Clement 1977). Third, there
may be synergistic interactions between the CGs and PAs in the same
monarch. Such taste 3ynergisms have been found to occur with certain
sweeteners in rats (Smith et al. 1976). These possibilities raise
interesting questions for future research.
Toxic Sensitivity of Three Species to Digitoxin and Monocrotaline
Digitoxin. For the E^. melanotis and E^. 3^. hylocetes that habituated
to the low and high digitoxin diets, there was no effect of dietary

159
treatment on weight change or food intake. This result becomes even
more significant when one compares the nightly dosages of digitoxin
both species experienced under the high CG treatment with the acute
LD5Q values for oral dosages of digitoxin in other species of mammals.
The mean nightly dosages for melanotis and hylnnsts^ was
greater than 70 mg/kg mouse (Table 6-7), whereas the acute LD50 for
laboratory mice, rats, and cats is 33-50, 23.8 and 0.30 mg/kg,
respectively (Barnes and Eltherington 1973, Marty 1983). Thus it
appears that £_^ melanotis and hylocetes, like E^. maniculatus and
laboratory mice (Marty 1983), absorb digitoxin poorly across the
gastrointestinal tract.
However, irrespective of this toxicological insensitivity, 2^. a.
hylocetes did not readily eat the high digitoxin diet. Three of the 9
mice offered this diet refused to eat it and had to be removed from the
experiment. The other 6 mice ate relatively small quantities for the
first few days, but eventually habituated to it. As shown in
Experiment 1, the concentration of digitoxin in this diet, 54.6 (lg/0.1
g chow, was well above their mean avoidance threshold.
Reithrodontomys sumichrasti did not appear to fully habituate to
either of the digitoxin diets. All 5 mice offered the high digitoxin
diet rejected it and had to be removed from the experiment because of
precipitous weight loss. Even though all 5 mice offered the low
digitoxin diet fed and gained weight, they ate significantly less food
than did the controls. This low food intake appears to explain their
slow growth rate, as compared to the control mice. It is unlikely that
a toxic response to the digitoxin depressed their growth rate since no
individuals showed symptoms of CG toxicity (see Chapter 2). The strong

160
aversion IL* sumichrasti displayed towards the high and low digitoxin
diets is perhaps not surprising given their extremely low mean
avoidance threshold for this compound.
Monocrotaline. The high and low monocrotaline diets had no
significant effect on weight gain or food intake in mslanntis.
suggesting that it was not affected adversely by them. In contrast,
the same diets caused a significant, dose-dependent decrease in growth
rate, but no corresponding decrease in food intake, in L. hyloretes.
It follows that the E_^ hylocetes grew more slowly on the
monocrotaline diets because of a toxic effect.
Because the route of administration (e.g., oral versus intra-
peritoneal) does not influence the hepatotoxic action of a given dosage
of a PA (Mattocks 1986), it is possible to compare results from studies
in which different types of administration procedures were used. Acute
LDso's for laboratory rats and mice that received intra-peritoneal
injections of monocrotaline were 154 and 259 mg/kg, respectively
(Mattocks 1972, Miranda et al. 1981); all of the animals died within 7
days of the lethal dosage. In this study, melanotis and P.
hylocetes ingested on average 107 and 72 mg/kg of monocrotaline per
night, respectively, without any mortality. Even though these dosages
are well below the U^g's of the mice and rats, melanotis and
hylocetes ingested these dosages over 28 consecutive nights.
However, before any definite conclusions can be drawn regarding the
toxic sensitivities of these two species of mice to monocrotaline, the
livers of the mice used in this study must be examined histologically
for signs of PA toxicity (see Chapter 2). This analysis is in
progress. At the end of Experiment 2, all mice were killed by cervical

161
dislocation and their livers extracted and stored in 10% neutral
buffered formalin. I have since sent these tissues to Dr. Charles
Montgomery, a liver pathologist, and he will section and evaluate them
for lesions.
Relationship Between Taste and Toxic Sensitivity Digitoxin
The results from Experiments 1 and 2 suggest no clear relationship
between the taste and toxic sensitivities in all 3 species of mice.
Even though all 3 species appeared to be relatively insensitive to
digitoxin toxicity, they varied widely in taste sensitivity. Both E^_
hylocetes and Ü*. sumichrasti rejected diets with apparently harmless
concentrations of digitoxin. This suggests that these 2 species did
not base their feeding responses to the digitoxin diet on post-
ingestional feedback. Rather, it appears that taste was of over-riding
importance. Unfortunately, the concentrations of digitoxin used in
Experiment 2 were below E^. melanotis1 mean taste avoidance threshold.
It would be interesting to repeat Experiment 2 with E_*. melanotis.
except use higher concentrations of digitoxin. Such an experiment
should determine whether E^. melanotis as well bases its responses to
digitoxin treated diets on taste or post-ingestional effects.
Several other workers that have found that vertebrate and
invertebrate predators avoid foods based on their bitter taste alone.
Red-winged blackbirds avoided a variety of compounds that occurred in
arthropod secretions because of the bitter taste (Yang and Rare 1968).
Blue jays reject pieces of bread treated with sub-emetic dosages of CGs
(Brower and Fink 1985). Laboratory rats displayed unlearned taste
aversions to the bitter secretions of Tiger salamanders (Mason et al.

162
1982). Many phytophagous insect species appear to assess the quality
(or potential danger) of a foodplant simply by "tasting" the surfaces
of its leaves (Bernays and Chapman 1987, Chapman and Bernays 1989) . In
addition, several studies with sweet compounds suggest that laboratory
rats and dogs may eat for taste rather than for calories under certain
conditions (Jacobs and Sharma 1969, Capaldi et al. jja prasa). This
discussion will be elaborated upon in Chapter 8.
In conclusion, the results of this chapter provide strong support
for several hypotheses proposed in earlier chapters: 1) that R.
sumichrasti and a^_ hylocetes avoid feeding upon overwintering
monarchs because of their high sensitivity to the taste of CGs; 2) that
all 3 species of mice absorb CGs poorly across their gastrointestinal
tract; and 3) that it is melanotis' insensitivity to the taste and
toxic effects of the monarch's CGs that enables it to take advantage of
overwintering monarchs. The P_^ melanotis populations within
overwintering areas do not appear to possess specific adaptations for
eating the monarchs. Contrary to the suggestion by several workers
that PAs play an important defensive role in monarchs (Edgar et al.
1976, Rothschild and Marsh 1978, Eisner 1980, Kelley et al. 1987, but
see Brower 1984), my results suggest that PAs do not occur at
sufficiently high concentrations in overwintering monarchs to
effectively deter vertebrate predators.

CHAPTER 7
BEHAVIORAL AND ECOLOGICAL INTERACTIONS OF FORAGING MELANOTIS WITH
MONARCH BUTTERFLIES
As I demonstrated in Chapter 3, 2^. melanotis need not climb up
trees or low vegetation to feed on monarchs. Many live butterflies
fall to the ground because storms and birds dislodge them, and many
dead ones also accumulate on the ground as a result of starvation,
desiccation, and incomplete predation by birds. During storms,
hundreds of thousands more fall to the ground (Calvert et al. 1983).
Owing to low temperatures, many of the live monarchs on the ground in
late afternoon cannot fly; instead they crawl up vegetation or remain
immobilized. Dead monarchs decay slowly in the cold climate (January-
February temperature range = 2 to 14°C; Calvert and Brower 1986) and
accumulate on the ground over the overwintering period. In the absence
of precipitation, they desiccate within 48 h. Thus, the ground and low
vegetation are covered with butterflies in varying states of
desiccation.
To better understand how melanotis forage on overwintering
monarchs, I examined the following 4 questions: 1) How many monarchs
can adult mice kill per night? 2) How does a monarch's accessibility
(i.e. height above ground) influence its risk to mouse attack? 3)
Since many wet (hydrated) and dry (desiccated) monarchs accumulate on
the ground and in low vegetation, do mice search selectively for wet
163

164
monarchs, and if so why? 4) If only dry monarchs are on the ground, do
mice begin searching in low vegetation for wet ones?
Methods
Trapping and Maintenance
I collected mice and conducted the feeding experiments at least 800
m away from the monarch colony on Sierra Chincua during January 1986.
Mice were captured in Sherman live traps (7.6 x 8.9 x 22.9 cm) baited
with rolled oats and provided with a compressed cotton ball for nesting
material. Only reproductively inactive adults (nonlactating or non¬
pregnant females and non-scrotal males) were used in the experiments.
Individual mice were used only once.
The mice were housed individually in wire mesh cages (about 30 cm
high x 25 cm in diameter) that were set on a tarpaulin-covered table in
the shade next to my campsite. Dacron batting was added for nesting
material and pieces of cardboard were placed between the cages to
isolate the mice visually. Prior to the feeding experiments, mice were
maintained on a diet of Purina laboratory chow #5001 and water. Only
mice that fed normally were tested. Mice experienced natural daily
fluctuations in temperature and light.
Monarchs were collected from accessible clusters within the colony
with nets.
Experiment 1: Extent of Mouse Predation
In this experiment I estimated the number of wet monarchs EL.
melanot.is could attack and eat per night. Mice were maintained

165
individually in their cages for 2 days with mouse chow and water ad
libitum prior to testing.
Three male and 3 female adult melanotis were each offered 55
live monarchs per night for 6 consecutive nights. No other food source
was available during testing. Each mouse's weight was taken at the
beginning and end of the experiment. The monarchs were placed in each
mouse's cage at 1900 h and left there until 0900 h the following
morning. Each mouse was then removed from its cage to tally monarch
mortality, then replaced and food deprived until 1900 h the same day.
In preliminary experiments melanotis fed almost exclusively on
abdominal material. I thus excluded thoracic feeding data. Monarch
mortality was grouped into 4 categories: 1) not attacked (i.e., no
visible damage); 2) sampled (i.e., < 25% of abdomen eaten); 3) eaten
non-selectively (i.e., > 25% of abdomen eaten, including cuticle); and
4) eaten selectively (i.e., > 25% of contents eaten, cuticle rejected).
In this and subsequent experiments, I grouped feeding categories 2 to 4
into number "attacked" and categories 3 to 4 into number "eaten”. I
assumed that a mouse attack resulted in death to a monarch.
Experiment 2: Attacks Uncaged Mice on Monarchs Perched at
Diffgrpnt Heights
I conducted this experiment in a stand of oyamel fir about 800 m
from an overwintering colony. A 180 m transect of 21 stakes (100 cm
tall and about 1.5 cm diameter; 9 m inter-stake interval) was set.
Three wads of masking tape were stuck to each stake at 100, 50, and 10
cm above the ground and one monarch was then stuck to each wad of tape
with a code number written on both forewings with indelible ink.

166
Another monarch was put on the ground at the stake's base. Each
monarch was immobilized by squeezing its thoracic muscles. The mice
could easily remove monarchs from the tape, but monarchs could not
remove themselves.
Monarchs were placed on the transect between 2100 to 2200 h on 5
separate nights between 30 January and 6 February. To prevent birds
from attacking the monarchs, I tallied the number attacked by mice at
sunrise. Twenty-one live traps were set along the transect (1 per
stake), during the nights of 7, 10 and 13 February, to ascertain which
species of mice were present.
Experiment 3: Influence of Monarch Accessibility on Mouse Attack
To explore this relationship further, but under more controlled
conditions, I exposed individual, captive mice to 1 of 3 vertical
distribution patterns of monarchs (uniform, majority high, and majority
low). Three different distribution patterns were used because, in
colonies, the relative numbers of monarchs on the ground and in low
vegetation varies in time and space.
Preliminary observations indicated that mice tend to reject the
bitter and CG-laden cuticle of grounded monarchs, and eat internal
tissues. I thus expected them to feed selectively on grounded
monarchs, but owing to the presumed difficulty of feeding in this
manner while grasping a stake, I expected mice to carry monarchs on
stakes to the ground before feeding on them.
Trials were conducted in the center of a 2.5 m x 35 cm high sheet
metal enclosure. A tarpaulin was placed 1.5 cm above it, and the floor
was bare, packed soil. Ten each of 10, 25, and 60 cm high stakes, 1.5

167
cm diameter, were set in a 6 x 5 array with 12 cm spacing and were sunk
in the soil so as not to wobble when mice climbed on them. The spacing
was sufficient to prevent mice from jumping between the stakes.
Monarchs were stuck to the stakes as in Experiment 2, and also placed
on the ground 6 cm from surrounding stakes.
For each trial, 40 monarchs were distributed vertically in 1 of 3
patterns (A, B, and C). In A, 1 monarch was on each stake and 10 on
the ground; in B, 20 were on the 60 cm stakes (2 per stake) and the
other 20 on the 10 and 25 cm stakes (1 per stake); and in C, 20 were on
the ground and the other 20 on the 10 and 25 cm stakes (1 per stake).
Each mouse was placed in the enclosure at 1900 with chow and water
¿d libitum. Monarchs were placed on the stakes between 2100 to 2200 h.
Mortality was tallied as in Experiment 1 the following morning at 0900
h. To determine which perched butterflies were eaten while stuck to
the tape, I also noted where each perched monarch was located (i.e., on
ground or on perch). Six mice were exposed to pattern A, 5 to B, and 5
to C.
Experiment 4: Influence of Monarch Desiccation on Mouse Attack
I offered captive mice equal numbers of dry and wet (freshly-
killed) monarchs, expecting them to prefer the latter and to eat many
of them selectively. Mice were tested in the same apparatus as
described in Experiment 1. Two days prior to a feeding trial, the
thoraces of 25 monarchs were firmly squeezed and left in a clearing for
48 hours to desiccate. Five hours before a feeding trial, 25 other
monarchs were killed in the same manner, but prevented from
desiccating.

168
The 2 categories of monarchs were marked differently by placing a
red or gold spot (about 1 mm in diameter) of nail polish on each
monarch's abdomen, thorax, head, and wings about 5 hours prior to a
feeding trial. The assignment of color was determined randomly for
each mouse.
At 2000 h, 25 wet and 25 dry monarchs were placed on the floor of
the mouse cages, and at 0900 h the next morning the monarch remains
were tallied. Four male and 4 female mice were tested.
Experiment 5: Risk to Live Monarchs in Low Vegetation with Only Dry
Monarchs on the Ground
Previous experiments examined the influence of monarch
accessibility or desiccation on mouse attack. In this experiment I
examined the relative importance of these 2 factors by offering mice
dry monarchs on the ground and wet ones on stakes. I expected mice to
feed most heavily and selectively on the less accessible, wet monarchs.
I used the same mouse enclosure described in Experiment 3. The live
and dry monarchs presented to mice were prepared as in Experiment 4,
except that their wings were marked with indelible ink. Mice (3 male
and 1 female) were put in the enclosure individually between 1900 to
2000 h and had access to chow and water ad libitum. At 2200 h, 20 dry
monarchs were dispersed evenly on the ground within the 6x5 stake
array, and 20 wet monarchs were attached to the 25 and 60 cm stakes (1
per stake). At 0900 h the following morning, monarch remains were
tallied.

169
Statistical Analyses
To examine the pattern of attack by the wild, uncaged mice in
Experiment 2, I used the Cochran Q test (Zar 1984). This test was
appropriate because I had 4 related samples per stake (i.e., 4 monarchs
at or on each stake) and dichotomous data. I treated each mouse as a
randomized block and I analyzed the results from each night separately.
To determine whether mice attacked accessible monarchs
disproportionately in Experiments 3 and 5, I used the one-sample (2-
tail) t-test (Zar 1984). The variance for the observed proportion
attacked at each perch height was estimated with the ratio formula for
cluster samples (Cochran 1977). With this test, I could determine
whether mice attacked more or less butterflies than expected at each
perch height. To minimize the probability of committing a Type 1
error, owing to my use of multiple t-tests, I set the highest
significance level at 0.01.
To determine whether mice fed disproportionately on wet monarchs in
Experiment 4, I conducted separate chi-square tests on each mouse.
Results
Experiment 1; Extent of Mouse Predation
The mean number of monarchs attacked/night/mouse was 39.9 (S.E. =
1.1; range = 30 to 52), whereas the mean number eaten/night/mouse was
37.0 (S.E. = 1.0; range = 29 to 51; Table 7-1). There was no apparent
change in the number attacked and eaten/night/mouse or in the weights
of the mice over the 6 days. Mean nightly feeding patterns were thus
independent of previous monarch consumption over a 6-day period.

Table 7-1. Mean number (± S.E.) of wet monarch butterflies attacked and eaten by 6 captive P.
melanotis per night over 6 consecutive nights.
No. attacked
Night
Grand
1
2
3
4
5
6
Mean
40.5
39.8
41.3
38.7
41.3
37.8
39.9
(±3.0)
(±2.6)
(±4.0)
(±2.9)
(±2.1)
(±1.3)
(±1.1)
37.2
34.7
38.3
37.3
39.1
35.2
37.0
(±2.6)
(±1.6)
(±3.8)
(±1.0)
(±2.2)
(±2.2)
(±1.0)
No. eaten

171
Experiment. 2; Attacks bv Wild, Uncaged Mice on Monarchs Perched at
Different Heights
Out of 63 trap nights (21 traps x 3 nights), the same 3 mice (P-
melanotis) were trapped all 3 nights. Thus it seems likely that this
species was responsible for the 125 monarchs attacked. Although no
conclusions can be drawn regarding the number of monarchs attacked by
each mouse, clearly, the total number attacked increased over the 8
days of this experiment (Table 7-2). Moreover, over all 5 nights, mice
fed most heavily on those monarchs on the ground and at 10 cm: Cochran
Q for 30 January = 9.92, df - 3 (on this and all other nights), P <
0.025; for 1 February = 15.81, P < 0.005; and for 2, 5 and 6 February =
17.5 to 51.0, P < 0.001.
Experiment 3: Influence of Monarch Accessibility on Mouse Attack
For patterns A, B and C, mice consistently attacked more monarchs
than expected on the ground and fewer at higher perches (Table 7-3).
In Table 7-4, I provide a breakdown of the abdominal feeding patterns
by mice (data for individual mice were pooled). Mice exposed to
patterns A and C fed selectively on many of the monarchs on the ground,
but rarely did so with the ones on stakes. Counter to prediction, the
majority of perched monarchs eaten by mice were eaten on the stakes;
for pattern A it was 82%, for B it was 84%, and for C it was 68%. I
could not determine, however, whether the predated, perched monarchs
found on the ground had been eaten by mice on the stakes and
subsequently dropped, or carried to the ground and then eaten.

172
Table 7-2. Number of wet monarch butterflies attacked by wild,
uncaged mice at different heights. A 180 m transect of 21 stakes
was run with 4 monarchs per stake and each monarch set at 1 of 4
heights.
Height
(cm)
No.
monarchs
offered
per Night
No.
of monarchs attacked
30 Jan
1 Feb
2 Feb
5 Feb
6 Feb
Total
100
21
0
0
0
0
5
5
50
21
0
1
1
5
10
17
10
21
1
2
3
19
20
45
0
21
4
7
8
19
20
58
Total
84
5
10
12
43
55
125

173
Table 7-3. Observed (mean ±S.D.) and expected proportions of wet
monarchs attacked by captive mice at different heights (proportions
are derived from the data in Table 7-4). A trial consisted of
offering a mouse 40 monarchs distributed among perches in 1 of 3
possible patterns (A,B, or C). Six mice were tested with pattern A,
5 with B, and 5 with C. The observed and expected proportion of
monarchs attacked are compared with the one-sample (2-tail) t-test. *
P < 0.01; ** P < 0.001.
Distribution
Pattern of
Monarchs
Height
(cm)
Proportion
Observed
Attacked
Expected
t-value
df
A Uniform
60
0.14
(±-02)
0.25
-15.8
★ ★
25
0.14
(±.03)
0.25
-10.2
★ ★
5
10
0.35
(±.03)
0.25
12.2
★ ★
On Ground
0.39
(±.02)
0.25
17.3
★ *
B Majority high
60
0.22
(±-02)
0.50
-30.1
★ ★
25
0.24
(±.02)
0.25
1.3
4
10
0.50
(±.02)
0.25
30.5
★ *
C Majority low 25 0.07 (±.01) 0.25 -28.1 **
10 0.15 (±.002) 0.25 -142.9 **
On Ground 0.78 (±.11) 0.50 5.3 *
4

Table 7-4. Abdominal feeding patterns by captive mice on wet monarchs at different heights. Feeding
categories are described in the text. A trial consisted of offering a mouse 40 monarchs distributed
among perches in 1 of 3 possible patterns (A, B, or C). Six mice were tested with pattern A, 5 with
B, and 5 with C. Data for individual mice within each pattern were pooled.
Distribution
pattern of
monarchs
Height
(cm)
Abdominal
feeding categories
Total
Offered
Not
attacked
Eaten non-
Sampled selectively
Eaten
selectively
A Uniform
60
39
4
16
1
60
25
40
2
18
0
60
10
9
8
39
4
60
On ground
3
1
36
20
60
B Majority high
60
77
0
17
5
100
25
29
0
14
7
50
10
7
1
38
4
50
C Majority low
25
43
0
6
1
50
10
34
0
14
2
50
On ground
18
1
43
38
100

175
Experiment 4: Influence of Monarch Desiccation on Mouse Attack
The abdominal feeding patterns for all mice pooled are presented in
Table 7-5. Chi-square comparisons, performed on each mouse separately,
were all highly significant (chi-square values ranged from 12.9-45.0,
df = 2, all P-values < 0.005), demonstrating a strong association
between the degree of desiccation and the feeding patterns by mice.
Mice strongly preferred wet monarchs, and ate the majority of them
selectively.
Experiment 5: Risk: to Live Monarchs in Low Vegetation with only Dry
Monarchs on the Ground
Mice attacked significantly fewer dry monarchs on the ground and
slightly (but not significantly) more wet monarchs on the stakes than
expected (Table 7-6). Moreover, only wet monarchs were fed upon
selectively (Table 7-7; data for individual mice were pooled). Thus,
taken together with the results of Experiments 2 and 3, it appears that
wet monarchs perched in low-lying vegetation have the greatest risk to
mouse attack when only dry monarchs are on the ground.
Pis cusa,ion
Estimate of the Numbers of Monarchs Killed by Mice
By limiting adult mice to a diet of live monarchs for 6 consecutive
nights, I determined that individuals attack on average 39.9 wet
monarchs per night. Moreover, results from Chapter 3 suggest that 1)
the diet of £_*. melanotis inside colonies consists entirely of monarchs,
and 2) melanotis populations inside monarch colonies range between
50 to 97 adult mice per hectare. Therefore, if we assume that free-

176
Table 7-5. Abdominal feeding patterns directed at 25 wet and 25
dry monarch butterflies by individual captive melanotis (n =
8) . Data
for individual mice were pooled.
Abdominal feeding category
State of
Not Eaten Non-eaten Total
monarch
attacked selectively selectively Offered
Wet
16 53 131 200
Dry
125 75 0 200
0
200

177
Table 7-6. Observed (mean ±S.D.) and expected proportion of wet and
dry monarchs attacked by 4 mice at different heights (proportions
are derived from the data in Table 7-7). For each trial, 20
monarchs were divided evenly among ten 25 cm stakes and ten 60 cm
stakes, and 20 dry monarchs were placed on the ground. The observed
and expected proportion of monarchs attacked are compared with the
one-sample
(2-tail)
t-test. ** P < 0.
001.
Height
State
Proportion
of
attacked
(cm)
Monarchs Observed
Expected
t-value
df
60
Wet
0.32 (±.06)
0.25
2.12
25
Wet
0.42 (±.08)
0.25
4.10
3
On Ground
Dry
0.26 (±.03)
0.50
-15.2 **

178
Table 7-7. Abdominal feeding patterns by 4 captive mice on wet and
dry monarch butterflies. For each trial, 20 wet monarchs were
divided evenly among ten 25 cm stakes and ten 60 cm stakes, and 20
dry monarchs were placed on the ground. Data for individual mice
were pooled.
Abdominal feeding categories
Height
(cm)
State of
monarchs
Not
attacked
Eaten non-
selectively
Eaten
selectively
Total
offered
60
Wet
23
12
5
40
25
Wet
18
12
10
40
On Ground
Dry
66
14
0
80

179
ranging, adult £_^ melanotis kill an average of 39.9 monarchs per night,
a population of this size could kill between 1,995 to 3,870 live
monarchs per night. Total mortality in a 1 ha colony for the 135-day
overwintering season would be approximately 270,000 to 522,500
butterflies, or using Brower and Calvert's (1985) density estimate of
10 million butterflies per ha, 2.7 to 5.2% of the colony.
This mortality estimate probably underestimates total mortality
because the nightly predation rates were determined with
nonreproductive, captive individuals. The high energetic costs of
lactation in females and locomotion in free-ranging individuals (Millar
1979, Sadlier et al. 1973, Stebbins 1977, Perrigo 1987) would
undoubtedly elevate nightly consumption of monarchs.
Foraging Behavior of P. melanotis
When given wet monarchs at different heights, melanotis
consistently attacked those closest to the ground. However, when
offered wet monarchs on stakes and dry monarchs on the ground, mice
attacked the former disproportionately. Mice also preferred wet to dry
monarchs when both were on the ground. These results suggest a
dominance of the effect of desiccation over that of accessibility. A
comparable study of the influence of accessibility and food quality on
fruit selection in several Neotropical birds reported the opposite
result. When a preferred fruit was made progressively less accessible,
birds eventually switched preferences and took those initially less-
preferred (Moermond and Denslow 1983) . These results do not
necessarily contradict mine given that my accessibility challenges were

180
limited. For instance, taller stakes in Experiment 5 might have caused
switches similar to those found by Moermond and Denslow.
None of the dry monarchs and only a few of the wet monarchs on
stakes were eaten selectively. In contrast, many of the wet monarchs
on the ground were eaten in this manner. I suggest that mice were
unable to feed selectively on dry monarchs because the abdominal
cuticle bonds tightly to the internal tissues during the desiccation
process (Brower et al. 1988) . Moreover, I suggest that mice rarely fed
selectively on the perched, wet monarchs because they had difficulty
doing so while grasping the stakes. Peromyscus melanotis appear to
prefer wet monarchs on the ground because they can feed on them more
dexterously. Arboreally adapted mice might be able to feed equally
dexterously on food items while on the ground or grasping low
vegetation.
Several explanations may account for why melanotis avoided dry
monarchs. First, they may always search for succulent food items in
their natural habitats because of limited water, especially during the
winter dry season. Second, they may find dry monarchs less palatable,
owing to changes in flavor and/or nutritional value. And third, they
may have an aversion to the bitter, CG-laden, and indigestible cuticle
(Brower et al. 1988).
Monarch Defenses Against Mouse Attack
The monarch's strong tendency to crawl up vegetation has been
interpreted as an adaptive response for avoiding freezing ground
temperatures (Calvert and Brower 1981, Calvert and Cohen 1983). My
results indicate that this crawling behavior is also an effective

181
defense against mouse attack. I conclude that over evolutionary time
the combined mortality risks of nocturnal mouse predation and freezing
ground temperatures are likely to select for the monarch butterflies'
tendency to crawl up vegetation. Moreover, my results underline the
importance of protecting the understory vegetation in monarch colonies
to provide monarchs with an escape refuge against these mortality
factors.
Monarchs may possess other defenses that act in conjunction with
the crawling behavior to reduce their risk to mouse attack. One is a
variation on the "selfish-herd" effect (Hamilton 1971) . When a mouse
encounters a cluster of monarchs perched in low vegetation, it may feed
most heavily on those occupying the low and outer positions in the
cluster (i.e., the most accessible ones). Such predation pressure
would favor those individuals best able to secure a position in the
central or upper regions of clusters. In fact, monarchs in low
vegetation commonly jockied for the highest position in clusters
(Glendinning pers. observ.). They may also derive protection from
their stored defensive compounds. Results in Chapters 4 and 5 indicate
that £_». melanotis avoid monarchs with high CG concentrations,
frequently assessing CG levels nondestructively by employing taste.

CHAPTER 8
GENERAL CONCLUSIONS
Impact of Mouse Predation on Overwintering Monarch Butterflies
The overwintering monarch aggregations in Mexico provided an
opportunity to investigate the efficacy of a well known chemical
defense system against a suite of small mammalian predators. Rather
than being absolute, it limited the number of mouse predators from 5 to
1. Nevertheless, because so many melanotis immigrate to the
colonies, populations of melanotis may kill 2.7 to 5.2% of a 1 ha
colony over the entire overwintering season (Chapter 7). Similarly,
even though the monarch's defensive compounds also may reduce the
number of species of potential avian predators from 37 to 2 (Fink et
al. 1983, Arellano et al. 1989), these 2 species appear to kill about
9% of a 1 ha colony (Brower and Calvert 1985) . Thus, 1 species of
mouse and 2 species of birds may together kill 11.7 to 14.2% of a 1 ha
colony.
The number of monarchs killed by mice and birds seems high when
compared to the huge colonies of aposematic and chemically-defended
lady beetles in the Sierra Nevadas, which apparently experience no
predation whatsoever (Hagen 1962, Hagen 1970, but see Chapman et al.
(1955) regarding predation on a different species in the Rockies).
However, large aggregations of palatable insects appear to be fed upon
much more extensively and by a much greater variety of vertebrate
182

183
predators than are the monarch colonies. Examples include
opportunistic predation upon cicadas (Lloyd and Dybas 1966, Strehl and
White 1986, Steward et al. 1988, Hahus and Smith in press). alate ants
(Dial and Vaughan 1987), gypsy moth larvae (Gerardi and Grimm 1979),
larch sawflys (Buckner and Turnock 1965), pine sawflys (Holling 1959,
Hanski and Parviainen 1985), spruce budworms (Morris et al. 1958), bush
crickets (Bailey and McCrae 1978), and desert locusts (Smith and Popov
1953, Hudleston 1958, Ashall and Ellis 1962, Waloff 1966).
Why.. la P. melanotis the Only Mouse Species that Feeds on the Monarchs?
In Chapter 3, I tested the hypotheses that the 4 other species of
mice did not eat monarchs because 1) the microhabitat characteristics
(e.g.. understory vegetation) of overwintering areas suited only P.
melanotis. and 2) E_^ melanotis aggressively excluded them. I rejected
the first hypothesis because the understory vegetation patterns of
grids inside colonies were not significantly different from those
outside colonies. I rejected the second hypothesis because resident P.
melanotis were unable to dominate HL salvus and hylocetes. and
were able to dominate sumichrasti in only half of the interactions.
Thus it appears unlikely that resident melanotis could prevent any
of these species from feeding on the monarchs and/or establishing
residency inside the colonies.
Rather, my results suggest that all species but E_w melanotis were
averse to monarchs as food. I conclude that melanotis displayed 3
feeding attributes or mechanisms which together enabled it to feed
extensively on monarchs: 1) it attacked and consumed live monarchs
readily, 2) it learned rapidly how to reject the CG and PA-laden

184
cuticular material and feed selectively on the internal contents of
abdomens, and 3) it was relatively insensitive to the taste and toxic
effects of CGs and PAs. Even though the feeding experiments in Chapter
6 do not demonstrate definitively that £_*. melanotis is insensitive to
the toxic effects of monocrotaline, the demographic results in Chapter
3 suggest strongly that E^ melanotis was not debilitated by the
monarch's PAs. Individuals inside colonies were able to breed
extensively and successfully wean large numbers of young while
subsisting on a diet which stomach content analyses indicated was
nearly 100% monarch material.
I further conclude that the other 3 species of mice avoided
monarchs because neither possessed all 3 of the feeding attributes or
mechanisms described above for £_^ melanotis. Microtus m. salvus would
not attack active monarchs, and when offered paralyzed and immobile
monarchs, did not learn the selective feeding behavior and appeared to
be intolerant to the taste of the monarch CGs and/or PAs. The toxic
sensitivity of 1L hl salvus was not examined. Even though £_^ a.
hylocetes and sumichrasti readily attacked live monarchs, and did
not appear to be sickened by them, both species appeared to be
intolerant to the taste of CGs. Because sumichrasti learned the
selective feeding behavior and E^. hylocetes did not, IL. sumichrasti
was able to ingest more monarch material. Nevertheless, even after
discarding the cuticle, the internal tissues of most monarchs still
appeared to have contained intolerable levels of CGs for B-.
sumichrasti. Even though I was not able to examine the feeding
responses of 1L_ alstoni to the monarchs, owing to its low abundance,

185
its complete absence from the colonies suggest that it was averse to
the monarchs as well.
That these relatively closely-related species responded so
differently to a single, shared, chemically defended insect species,
underlines the importance of studying as large a percentage of the
assemblage of natural predators as possible when evaluating the
effectiveness of an insect species' chemical defense system. Simply
choosing the most convenient or abundant predator(s) may provide
misleading results. For example, had I only studied B_*. sumichrasti and
£_t hylocetes. I would have concluded that the monarch's chemical
defense system effectively deterred all species of mice in the
overwintering area.
Are Populations of P. melanotis in the Overwintering Areas Specifically
Adapted for Overcoming the Monarch's Defensive Compounds?
Several lines of evidence point to the unlikelihood of
subpopulations of melanotis having evolved specific gustatory,
pharmacological, and behavioral adaptations for dealing with the
monarch's defensive compounds. First, individuals collected about 4
kilometers to the east of grid 3 (Figure 3-la) were just as insensitive
to the taste and toxic effects of digitoxin (a CGs) as were those
collected about 800 m to the south of grid 3 in Figure 3-la and b.
Second, there are no apparent geographical barriers that would
genetically isolate populations of melanotis on the west side of
Sierra Chincua, which is where the monarchs overwinter (Calvert and
Brower 1986), from those on the east side or on nearby massifs (Anon.
1976: topographic map). Third, the same populations of E_^ melanotis on
the west side. Sierra Chincua probably do not encounter monarch colonies

186
each year as distances between colony sites over successive years may
be as great as 2 km (Calvert and Brower 198 6) .
It seems more likely that the capacity of E^. melannf. -i s to feed upon
monarchs reflects phylogenetically determined 1) tolerances for coping
with a variety of bitter and/or toxic foods, and 2) flexible prey
handling abilities (Marty 1983, Brower et al. 1985). That is, P.
melanotis may have exapted (sensu Gould and Vrba 1982) these 2
attributes to feeding on monarchs. Other possible examples of
pharmacological and behavioral exaptation include predation by
grosbeaks and orioles on overwintering monarchs (Brower et al. 1988)
and the ability of frugivorous birds to ingest aposematic insects
(Herrera 1985).
Benefits and Costs of Being Tolerant to Bitter Foods
My results demonstrate that by tolerating the taste of the
monarch's defensive compounds, E^. melanotis was able to feed
extensively on overwintering monarchs and initiate high levels of
reproduction. In contrast, free-ranging EL. sumichrasti and E^. a.
hylocetes avoided the bitter-tasting monarchs, fed on less calorically
dense foods, and did not reproduce. If these two species of mice had
tolerated the bitter taste of the monarchs, they probably could have
fed extensively on the monarchs and initiated winter reproduction.
Reithrodontomys sumichrasti and £_^ hylocetes did not appear to be
sensitive to the toxic effects of CGs. This finding illustrates a cost
to rejecting foods based on taste: by avoiding an ephemeral,
superabundant, and apparently harmless food resource, individual £_*.
hylocetes and IL. sumichrasti appear to have missed an opportunity to

187
substantially increase their fitness. Given that the average life span
is a few months at best for most mice living in extreme environments
such as those found at the overwintering sites in Mexico (Bronson
1985), many of the £^_ a— hylocetes and E*. sumichrasti probably die
before the onset of spring, when more palatable food sources become
abundant (V. Sánchez pers. comm.).
It would be interesting to determine whether E* hylocetes and R.
sumichrasti are more averse than E^. melanotis to the taste of a variety
of other naturally occurring bitter compounds (both harmless and
harmful ones). If E ix hylocetes and IL. sumichrasti are more
sensitive to these bitter compounds, it follows that they would be more
likely than E* melanotis to naturally reject harmless, bitter-tasting
foods in nature. As a consequence of their generalized bitter
sensitivity, E^. A^. hylocetes and E^. sumichrasti may have more
restricted diets and thus occupy a narrower range of habitats than P.
melanotis. This reasoning is supported by 3 observations. First, food
habits appear to influence habitat selection in several Peromyscus
species; that is, species with more diverse and flexible food habits
tend to occupy a greater range of habitats (see Drickamer 1976 and
references therein). Second, E*. melanotis occurs in a much greater
variety of high-altitude (> 2800m) habitats than do the other four
species of mice on Sierra Chincua (Glendinning unpub1. data) and
elsewhere in Mexico (C. Galindo unpubl. data). Third, the number of
foods disliked by humans increases in direct proportion to their taste
sensitivity to quinine and 6-n-propylthiouracil (PROP), both of which
are extremely bitter compounds (Fischer and Griffin 1963, Fischer et
al. 1965) .

188
There are several potential costs of a generalized tolerance
towards bitter foods. First, an animal may unwittingly poison itself.
Such self-administered toxicosis has been observed in several species
of reptiles which are highly tolerant to bitter foods (Swain 1976).
Second, many poisonous compounds, in addition to PAs, do not
immediately sicken animals (Freeland and Janzen 1974), and thus do not
provide the post-ingestive feedback necessary for formation of a
conditioned feeding aversion. In this case, simply rejecting a food
based on its bitter taste would be highly adaptive. With respect to
this dissertation, it is conceivable that feeding heavily on monarchs
actually shortened the average life span of the £_ melanotis because of
the delayed hepatotoxicity induced by the monarch's PAs. This
hypothesis could be tested indirectly by comparing females from inside
and outside monarch colonies in terms of their incidence of PA
hepatotoxicity. As I have the livers from the mice used in the stomach
content analyses, I can easily do this. However, the cost of a shorter
life span must be weighed against the increased reproductive output the
E-*. melanotis achieve by taking advantage of the monarch butterflies.
In conclusion, my results illustrate how a physiological
constraint, bitter sensitivity, can profoundly influence the way in
which an assemblage of predator species responds to a potential food
resource. Although free-ranging E^. hylocetes and sumichrasti
were capable of feeding upon overwintering monarchs without immediately
becoming sick, they did not. This appears to have been due to their
inability to habituate to the bitter taste of the CGs. With the
notable exception of several studies (Yang and Kare 1968, Swain 1976,
Jacobs et al. 1978, and Mason et al. 1982), few workers have examined

189
the importance of the bitter taste of foods in controlling intake by
free-ranging vertebrates. Instead, most workers interested in the
control of food intake appear to assume that post-ingestional feedback
takes precedence over taste in the control of food selection, and do
not bother testing this assumption.
There are several potentially fruitful areas of future research.
First, I intend to determine whether melanot.is is less sensitive
than hylocetes and sumichrasti to the taste of a variety of
bitter foods. It is conceivable that Z^. melanotis' low sensitivity to
the taste of CGs is coincidental, and that the other 2 species are in
fact much more tolerant than melanotis to the taste of a variety of
other bitter compounds. Second, it would be interesting to explore the
relationship between structural aspects of CGs (e.g., polarity) and
taste rejection thresholds in £_^ melanotis. hylocetes. R.
sumichrasti and humans as well. Because so much is known about the
toxicology of CGs, such a study could suggest whether the most toxic
CGs are in fact the most bitter. Third, the role of bitter sensitivity
in determining the feeding preferences other species of wild animals
needs to be studied. General questions might include 1) given the
abundance of bitter compounds in plant foods, and lack of such
compounds in vertebrate prey, are herbivores generally more tolerant to
bitterness than carnivores and 2) do animals with highly effective
detoxification systems tend to be more tolerant to bitter foods?

APPENDIX A
PAIRED COMPARISONS OF THE PERCENT SIMILARITY OF UNDERSTORY VEGETATION
ON GRIDS LOCATED INSIDE AND OUTSIDE THE MONARCH COLONIES. THE CODES OF
THE GRIDS USED IN EACH PAIRED COMPARISON ARE PROVIDED IN PARENTHESES;
SEE FIGURE 1 FOR A KEY TO THESE CODES. I COMPARE EACH COLUMN WITH THE
KRUSKAL-WALLIS TEST.
inside vs.
inside
outside xa.
outside
inside ys..
outside
65 (1,2)
31 (3,4)
60
(1,3)
38 (1,5)
31 (3,6)
33
(1,4)
46 (1,7)
22 (3,8)
42
(1, 6)
50 (2,5)
50 (4,6)
29
(1,8)
46 (2,7)
60 (4,8)
59
(2,3)
40 (5,7)
50 (6,8)
47
(2,4)
41
(2, 6)
35
(2,8)
35
(5,3)
70
(5, 4)
48
(5, 6)
62
(5,8)
34
(7,3)
42
(7,4)
76
(7, 6)
60
(7,3)
mean
47.5
40.4
48 .
. 4
S.E.
4.0
6.0
3.
6
190

APPENDIX B
PERCENTAGE IMPORTANCE VALUES (PIVS) FOR THE SPECIES OF UNDERSTORY
VEGETATION ON THE EIGHT TRAPPING GRIDS LOCATED EITHER INSIDE OR
OUTSIDE MONARCH COLONIES ON SIERRA CHINCUA. SEE FUGURE 1 FOR THE
LOCATION OF THE GRIDS. I EXCLUDED SPECIES WHOSE PIVS WERE NOT >
1.0 IN AT LEAST ONE OF THE GRIDS.

Proximity of
grid to
colony
Plant
Inside
Outside
family
Species
7
2
5
1
6
8
4
3
Compositae
Senecio angulifolius D.C.
9.6
11.9
2.3
7.4
7.8
3.9
1.1
4.9
Labiatae
Satureja macrostema (Benth.) Briq.
8.4
20.1
17.2
10.5
7.5
8.1
20.8
14.7
Rosaceae
Acaena elongata L.
8.3
11.4
2.2
9.5
6.2
1.9
9.2
22.2
Compositae
Stevia monardaefolia H.B.K.
4.7
6.2
7.7
8.7
5.2
6.1
7.2
0.9
Labiatae
Salvia cardinalis H.B.K.
2.1
10.8
3.7
13.2
3.0
4.2
3.4
8.7
Compositae
Senecio callosus Sch. Bip.
5.4
2.0
1.9
8.7
5.3
3.0
2.1
Polypodiaceae
Asplenimn nuananthes l.
0.6
5.8
2.3
1.7
1.6
0.8
1.4
Onagraceae
Fuchsia microphylla H.B.K.
6.6
2.6
18.4
4.0
1.6
4.6
Rosaceae
Alchemi11a procumbens Rose.
2.5
8.2
6.7
10.5
3.7
13.2
Compositae
Eupatarium mairetianum d.c.
1.5
1.4
21.2
10.7
12.1
13.7
Compositae
Bidens triplinaria H.B.K.
0.7
0.4
0.8
0.6
0.7
3.5
Compositae
Verbesina oncaphora Rob & Seat
11.5
13.3
6.9
14.3
6.8
Compositae
Sencia barbara-johannis D.C.
3.9
2.0
1.2
7.7
5.5
Compositae
Archibaccharis hieracioides
1.6
2.5
1.5
1.1
0.7
Labiatae
(Heering.) Blake
Salvia alegaos vahl.
0.6
_ _
5.7
_ _
2.9
11.0
7.7
...
Labiatae
Salvia helianthemifolia Benth.
10.1
3.9
3.7
4.9
Geraniaceae
Geranium feemanii Peyr
2.3
0.3
2.0
2.2
Umbelliferae
Arracacia nelsonii a. & c.
4.1
9.9
4.8
14.3
Scrophular-
Benth. & Hook
Penstemon aampa.aulatus (Cav.) willd.
4.9
_ ^ _
_ _
_ ^ _
4.4
1.1
1.0
_ _
iaceae
Compositae
Eupatorium petiolare Moc.
2.5
_ _
_ _
_ _
4.5
0.8
0.4
_
Labiatae
Salvia gracilis Benth.
9.3
“ . “
“ . “
.
8.2
13.9
” . ~
Continued.

APPENDIX B—continued.
Caryophyl-
laceae
Stellaria cuspidata Willd.
— . —
2.8
~ . -
0.5
- . -
“ . -
1.5
Compositae
Ecigeron galeottii (Gray ex. Hemsl.
Greene.
)
2.2
5.0
— . “
— . —
4.7
Compositae
Cirsium ehrengergii Ach. and Bip.
0.8
1.5
6.6
Compositae
Senecio albonervius Greenm.
0.7
1.6
4.7
Compositae
Senecio sanguisorbe D.C.
2.3
2.5
Polypodiaceae
Adiantum sp.
0.6
1.3
Ericaceae
Arctostaohvlos discolor (Hook) D.C.
4.1
0.8
Compositae
Senecio tolucanus D.C.
2.4
10.2
Compositae
Piquería pilosa h.b.k.
“ . —
1.2
1.2
Labiatae
Stachvs coccinea Jacq.
•
1.0
Column Total
100.0
100.0
100.0
100.0
100.0
99.9
100.0
100.0
Total number
of species in each grid
22
19
16
17
21
19
16
14

REFERENCES
Alonso-M., A., J. I. Glendinning and L. P. Brower, in press. The
influence of temperature on crawling, shivering and flying in
overwintering monarch butterflies in México, in S. B. Malcolm and
M. P. Zalucki. Biology and conservation of the monarch butterfly.
Contributions in Science, Natural History Museum of Los Angeles
County, Los Angeles.
Anderson, D. E. and R. Lydic. 1977. On the effect of using ratios in
the analysis of variance. Biobehavioral Reviews 1: 225-229.
Andrzejewski, R. 1975. Supplementary food and the winter dynamics of
bank vole populations. Acta Theriologica 20: 23-40.
Anon. 1976. Denetal Carta Topographica, México y Michoacán. Angangeo
E14A26, Escala 1:50,000. San Antonio de Abad 124, México 8, D.F.
Cetenal.
Anon. 1982. Normales climatológicas: periodo/1941-1970. S.A.R.H.,
México, D.F.
Arellano-G., A., J. I. Glendinning, J. B. Anderson and L. P. Brower,
in press. Interspecific comparisons of the foraging dynamics of
black-backed orioles and black-headed grosbeaks on overwintering
monarch butterflies in Mexico, in S. B. Malcolm and M. P.
Zalucki. Biology and conservation of the monarch butterfly.
Contributions in Science, Natural History Museum of Los Angeles
County, Los Angeles.
Ashall, C. and P. E. Ellis. 1962. Studies on numbers and mortality in
field populations of the desert locust (Schistocerca gregaria
Forskál). Anti-Locust Bulletin No. 38.
Bailey, W. J. and A. W. R. McCrae. 1978. The general biology and
phenology of swarming in the East African tettigoniid Ruspolia
differens (Serville) (Orthoptera). Journal of Natural History 12:
259-288.
Barnes, C. D. and L. G. Eltherington. 1973. Drug dosages in
laboratory animals: a handbook. University of California Press,
Berkeley, California.
Bate-Smith, E. C. 1972. Attractants and repellents in higher animals.
Pages 45-56 in J. B. Harborne. Phytochemical ecology: proceedings
of the phytochemical society symposium. Academic Press, London.
Batzli, G. 0. 1985. Nutrition. Pages 779-806 in R. H. Tamarin.
Biology of new world Microtus. Special Publication No. 8,
American Society of Mammalogy, N.P.
194

195
Belovsky, G. E. 1978. Diet optimization in a generalist herbivore:
the moose. Theoretical Population Biology 14: 105-134.
Bendell, J. F. 1959. Food as a control of population of white-footed
mice, Escomyscus leucopus noveboracensis (Fischer). Canadian
Journal of Zoology 37: 173-209.
Benson, S. B. and A. E. Borell. 1931. Notes on the life history of
the red tree mouse, Phenacomys lnnoicanrina. Journal of Mammalogy
12:226-233.
Bernays, E. A. and R. F. Chapman. 1987. The evolution of deterrent
responses in plant-feeding insects. Pages 159-173 in R. F.
Chapman, E. A. Bernays and J. G. Stoffolano Jr. Perspectives in
chemoreception and behavior. Springer-Verlag, New York.
Blum, M. S. 1981. Chemical defenses of arthropods. Academic Press,
New York.
Booth, C. C., A. E. Read and E. Jones. 1961. Studies on the site of
fat absorption. I. The sites of absorption of increasing doses of
^-^I-labelled triolein in the rat. Gut 2: 23-31.
Boppré, M. 1986. Insects pharmacophagously utilizing defensive plant
compounds (pyrrolizidine alkaloids). Naturwissenschaften 73: 17-
26.
Bowers, M. A. and H. D. Smith. 1979. Differential habitat utlization
by sexes of the deermouse, Peromyscus maniculatus. Ecology 60:
869-875.
Brenner, F. J., C. P. Gaetano, S. W. Mauser and D. L. Belowich. 1978.
Body weight and social interactions as factors in determing
dominance in captive eastern chipmunks Tamias striatus. Animal
Behaviour 26: 432-437.
Briggs, J. M. 1986. Supplemental food and two islands of Peromsycus
leucopus. Journal of Mammalogy 67: 474-480.
Bronson, F. H. 1985. Mammalian reproduction: and ecological
perspective. Biology of Reproduction 32: 1-26.
Bronson, F. H. and G. Perrigo. 1987. Seasonal regulation in muroid
rodents. American Zoologist 27: 929-940.
Brower, L. P. 1969. Ecological chemistry. Scientific American 220:
22-29.
Brower, L. P. 1984. Chemical defense in butterflies. Pages 109-134 in
R. I. Vane-Wright and P. R. Ackery. The biology of butterflies.
Academic Press, London.
Brower, L. P. 1985. New perspectives on the migration biology of the
monarch butterfly, Danaus plexippus L. Pages 748-785 in M. A.
Rankin. Migration: mechanisms and adaptive significance.
Contributions in Marine Science, Supplement, University of Texas.

Brower, L. P. and W. H. Calvert. 1985. Foraging dynamics of bird
predators on overwintering monarch butterflies in Mexico.
Evolution 39: 852-868.
Brower, L. P., M. Edmunds and C. M. Moffit. 1975. Cardenolide content
and palatability of a population of Danaus chrysippns monarch
butterflies from West Africa. Journal of Entomology 49: 183-196.
Brower, L. P. and L. S. Fink. 1985. A natural toxic defense system:
cardenolides in butterflies versus birds. Annals of the New York
Academy of Science 443: 171-186.
Brower, L. P. and S. C. Glazier. 1975. Localization of heart poisons
in the monarch butterfly. Science 188: 19-25.
Brower, L. P., E. Horner, M. M. Marty, C. M. Moffit and B. Villa-R.
1985. Mice (Peromsyscus maniculatus. spicilegus. and Microtus
mexicanus) as predators of overwintering monarch butterflies
(Danaus plexippus) in Mexico. Biotropica 17: 89-99.
Brower, L. P., B. E. Horner, M. M. Marty, C. M. Moffit and B. Villa-R.
1988. Errata. Biotropica 20: 125.
Brower, L. P. and C. M. Moffit. 1974. Palatability dynamics of
cardenolides in the monarch butterfly. Nature 249: 280-283.
Brower, L. P., C. J. Nelson, L. S. Fink, J. N. Seiber and C. Bond.
1988. Exaptation as an alternative to coevolution in the
cardenolide-based chemical defense of monarch butteflies (Danaus
plexippus L.) against avian predators. Pages 447-476 ia K. C.
Spencer. Chemical mediation of coevolution. Academic Press, New
York.
Brown, K. S. 1984. Adult-obtained pyrrolizidine alkaloids defend
ithomiine butterflies against a predator. Nature 309: 707-709.
Brown, K. S. 1984. Chemical ecology of dehydropyrrolizidine alkaloids
in adult Ithomiinae (Lepidoptera: Nymphalidae). Revista
Brasileira de Biologia 44: 435-460.
Brown, L. N. 1964. Ecology of three species of Peromvscus from
southern Missouri. Journal of Mammalogy 45: 189-202.
Buckner, C. H. and W. J. Turnock. 1965. Avian predation on the larch
sawfly, Pristiphora erichsonii (Htg.), (Hymenoptera:
Tenthredinidae). Ecology 46: 223-236.
Calvert, W. H. and L. P. Brower. 1981. The importance of forest cover
for the survival of overwintering monarch butterflies (Danaus
plexippus. Danaidae). Journal of the Lepidopterists' Society 35:
216-225.
Calvert, W. H. and L. P. Brower. 1986. The location of monarch
buttefly (Danaus plexippus L.) overwintering colonies in Mexico in
relation to topography and climate. Journal of the
Lepidopterists' Society 40: 164-187.

197
Calvert, W. H. and J. A. Cohen. 1983. The adaptive significance of
crawling up onto foliage for the survival of overwintering monarch
butterflies (Danaus plexippus) in Mexico. Ecological Entomology
8: 471-474.
Calvert, W. H., L. E. Hedrick and L. P. Brower. 1979. Mortality of
the monarch butterfly (Danaus plexippus L.): avian predation at
five overwintering sites in Mexico. Science 204: 847-851.
Calvert, W. H., S. B. Malcolm, J. I. Glendinning, L. P. Brower, M. P.
Zalucki, T. Van Hook, J. B. Anderson and L. C. Snook, ia press.
Conservation biology of monarch butterfly overwintering sites in
Mexico. Vida Silvestre Neotropical.
Calvert, w. H., W. Zuchowski and L. P. Brower. 1983. The effect of
rain, snow and freezing temperatures on overwintering monarch
butterflies in México. Biotropica 15: 42-47.
Canela, M.and V. Sánchez-C. 1984. Patrón del área de actividad de
Neotomodon alstoni alstoni (Rodentia: Cricetinae). Anales del
Instituto de Biologia de la Universidad Nacional Autónoma de
México 55(Seria Zoología 2): 285-306.
Capaldi, E. D., R. J. Pulley and J. D. Sheffer. in press. Hunger
and the learning of flavor preference, in R. C. Bolles.
Hedonics of taste. Lawrence Erlbaum Associates, Publishers,
Hillsdale, New Jersey.
Carlson, J. R. and R. G. Breeze. 1984. Ruminal metabolism of plant
toxins with emphasis on indolic compounds. Journal of Animal
Science 58: 1040-1049.
Carpenter, J. A. 1956. Species differences in taste preference.
Journal of Comparative and Physiological Psychology 49: 139-144.
Chapman, J. A., J. L. Romer and J. Stark. 1955. Ladybird beetles and
army cutworm adults as food for grizzly bears in Montana. Ecology
36: 156-58.
Chapman, R. F. and E. A. Bernays. 1989. Insect behavior at the leaf
surface and learning as aspects of hostplant selection.
Experientia 45: 215-222.
Cheeke, P. R. and M. L. Pierson-Goeger. 1983. Toxicity of Senecio
iacobea and pyrrolizidine alkaloids in various laboratory animals
and avian species. Toxicology Letters 18: 343-349.
Chrzanowski, Z. 1965. Die auswahlschwelle, die abweisschwelle und die
geschmackstoleranz von wasserlósungen des chininhydrochlorids, der
zitronenssáure, der sacharose und des kochsalzes einiger arten von
insektenfressern un nagetieren. Acta Biologiae Experimentalis
(Warsaw) 25: 37-58.
Clement, C. A. 1977. Observations on the reaction of Cvanocitta
cristata bromia Oberholser to Danaus plexippus Linnaeus: the skin
flick. Senior Honors Thesis. Amherst College.

198
Cochran, w. G. 1977. Sampling techniques (third edition). John Wiley
& Sons, New York.
Collins, C. T. and A. Watson. 1983. Field observations of bird
predation on Neotropical moths. Biotropica 15: 53-60.
Compton, S. G. and D. A. Jones. 1985. An investigatin of the
responses of herbivores to cyanogenesis in Lotus cominilatns L.
Biological Journal of the Linnean Society 26: 21-38.
Craig, J. C. and K. K. Purushothaman. 1970. An improved preparation
of tertiary amine N-oxides. The Journal of Organic Chemistry 35:
1721-1722.
Cyr, M. A. 1972. Predatory behavior of the grasshopper mouse
(Qnychomys). Dissertation. University of California, Los
Angeles.
Daly, M., J. Rauschengerger and P. Behrends. 1982. Food aversion
learning in kangaroo rats: a specialist-generalist comparison.
Animal Learning and Behavior 10: 314-320.
Detweiler, D. K. 1967. Comparative pharmacology of cardiac
glycosides. Federation Proceedings 26: 1119-1124.
Dial, K. P. and T. A. Vaughan. 1987. Opportunistic predation on alate
termites in Kenya. Biotropica 19: 185-187.
Drickamer, L. C. 1976. Hypotheses linking food habits and habitat
selection in Peromyscus. Journal of Mammalogy 57: 763-766.
Duffey, S. S. 1976. Arthropod allomones: chemical effronteries and
antagonists. Pages 323-394 in D. White. Proceedings of XV
International Congress of Entomology. International Congress of
Entomology, Washington, D.C.
Edgar, J. A., P. A. Cockrum and J. L. Frahn. 1976. Pyrrolizidine
alkaloids in Danaus plexippus L. and Danaus chrysippus L.
Experientia 32: 1535-1536.
Eisenberg, J. F. 1968. Behavior patterns. Pages 451-490 in J. A.
King. Biology of Peromyscus (Rodentia). American Society of
Mammalogists, N.P.
Eisner, T. 1970. Chemical defense against predation in arthropods.
Pages 157-217 in E. Sondheimer and J. B. Simeone. Chemical
Ecology. Academic Press, New York.
Eisner, T. 1980. Chemistry, defense, and survival: case studies and
selected topics. Pages 847-878 in M. Locke and D. S. Smith.
Insect biology in the future. Academic Press, New York.
Eriksson, M. 1984. Winter breeding in three rodent species, the bank
vole Clethrionomys pi a reolns - the yellow-necked mouse Anndsmiis
flavicollis and the wood mouse Apodemus svlvaticus in southern
Sweden. Holarctic Ecology 7: 428-429.

199
Farlow, J. 0. 1976. A consideration of the trophic dynamics of a late
Cretaceous large dinosaur community (Oldman Formation). Ecology
54: 841-857.
Fink, L. S. and L. P. Brower. 1981. Predation of unpalatable
butterflies. Nature 291: 7-13.
Fink, L. S., L. S. Brower, R. B. Waide and P. R. Spitzer. 1983.
Overwintering monarch butterflies as food for insectivorous birds
in Mexico. Biotropica 15: 151-153.
Fischer, R. and F. Griffin. 1963. Quinine dimorphism: a cardinal
determinant of taste sensitivity. Nature 200: 342-347.
Fischer, R., F. Griffin and B. Pasamanick. 1965. The perception of
taste: some psychophysiological, pathophysiological,
pharmacological and clinical aspects. Pages 129-163 in P. H. Hoch
and J. Zubin. Psychopathology of perception. Gruñe & Sratton,
New York.
Fordham, R. A. 1971. Field populations of deermice with supplemental
food. Ecology 52: 138-146.
Freeland, w. J. and D. H. Janzen. 1974. Strategies in herbivory by
mammals: the role of plant secondary compounds. The American
Naturalist 108: 889-894.
Galef, B. G. 1977. Mechanisms for the social transmission of acquired
food preferences for adult to weanling rats. Pages 123-148 in L.
M. Barker. Learning mechanisms in food selection. Baylor
University Press, Waco, Texas.
Galindo, C. and C. J. Krebs. 1987. Population regulation in deer
mice: the role of females. Journal of Animal Ecology 56: 11-23.
Ganchrow, J. R. 1977. Consummatory responses to taste stimuli in the
hedgehog (Erinaceus europaeusl. Physiology and Behavior 18: 447-
453.
Garcia, J. and W. G. Hankins. 1975. The evolution of bitter and the
acquisition of toxiphobia. Pages 39-45 in D. A. Denton and J. P.
Coghlan. Olfaction and taste. V. Proceedings of the 5th
international symposium in Melbourne, Australia. Academic Press,
New York.
Garcia, J. and R. A. Koelling. 1966. The relation of cue to
consequence in avoidance learning. Psychonomic Science 4: 123-
124.
Gashweiler, J. S. 1979. Deer mouse reproduction and its relationship
to tree seed crop. American Midland Naturalist 102: 95-104.
Gerardi, M. H. and J. K. Grimm. 1979. The history, biology, damage
and control of the gypsy moth, Porthetria dispar (L.). Associated
University Presses, Inc., Cranbury, New Jersey.
Gescheider, G. A. 1976. Psychophysics: method and theory. Lawrence
Erlbaum Associates, Publishers, Hillsdale, New Jersey.

200
Gilbert, B. S. and C. J. Krebs. 1981. Effects of extra food on
Peromsvcus and Clethrionomys populations in the southern Yukon.
Oecologia 51: 326-331.
Goeger, D. E., P. R. Cheeke, H. S. Ramsdell, S. S. Nicholson and D.R.
Buhler. 1983. Comparison of the toxicities of Senecio iacobea.
Senecio vulgaris and Senecio glabellus in rats. Toxicology
Letters 15: 19-23.
Gould, S. J. and E. S. Vrba. 1982. Exaptation—a missing term in the
science of form. Paleobiology 8: 4-15.
Griffith, L. A. and J. E. Gates. 1985. Food habits of cave-dwelling
bats in the central Appalachians. Journal of Mammalogy 66: 451-
460.
Hagen, K. S. 1962. Biology and ecology of predaceous coccinelidae.
Annual Review of Entomology 7: 289-326.
Hagen, K. S. 1970. Following the ladybug home. National Geographic
137: 543-553.
Hahus, S. and K. Smith, in press. Effect of periodical cicadas
(Maaicicada) on food habits of Blarina. Peromyscus and Microtus.
Journal of Mammalogy.
Halvorson, C. H. 1982. Rodent occurrence, habitat disturbance, and
seed fall in a larch-fir forest. Ecology 63: 423-433.
Hamilton, W. D. 1971. Geometry for the selfish herd. Journal of
Theoretical Biology 31: 295-311.
Hamilton, W. J. 1962. Reproductive adaptations of the red tree mouse.
Journal of Mammalogy 43: 486-504.
Hansen, C. M. and E. D. Fleharty. 1974. Structural ecological
parameters of a population of Peromyscus maniculatus in west-
central Kansas. The Southwestern Naturalist 19: 293-303.
Hansen, L. and G. 0. Batzli. 1978. The influence of food availability
on the white footed mouse: populations in isolated woodlots.
Canadian Journal of Zoology 56: 2530-2541.
Hanski, I. and P. Parviainen. 1985. Cocoon predation by small
mammals, and pine sawfly population dynamics. Oikos 45: 125-136.
Harder, D. B., G. Whitney, P. Frye, J. C. Smith and M. E. Rashotte.
1984. Strain differences among mice in taste psychophysics of
sucrose octa-acetate. Chemical Senses 9: 311-323.
Harrison, F. A. and W. M. F. Leat. 1975. Digestion and absorption of
lipids in non-ruminant and ruminant animals: a comparison.
Proceedings of the Nutrition Society 34: 203-210.
Herrera, C. M. 1985. Aposematic insects as six-legged fruits:
incidental short-circuiting of their defenses by frugivorous
birds. The American Naturalist 126: 286-293.

201
Herman, R. G., R. J. Parker, F. G. Henderson and K. K. Chen. 1962.
Intestinal absorption of cardiac steroids. Proceedings of the
Society of Experimental Biology and Medicine 109: 646-649.
Hilborn, R., J. A. Redfield and C. J. Krebs. 1976. On the reliability
of enumeration for mark and recapture census of voles. Canadian
Journal of Zoology 54: 1019-1024.
Holling, C. S. 1959. The components of predation as revealed by a
study of small-mammal predation of the European pine sawfly.
Canadian Entomologist 91: 293-320.
Hooper, P. T. 1978. Pyrrolizidine alkaloid poisoning - pathology with
particular reference to differences in animal and plant species.
Pages 161-176 in R. F. Keeler, K. R. Van Kampen and L. F. James.
Effects of poisonous plants on livestock. Academic Press, New
York.
Hudleston, J. A. 1958. Some notes on the effects of bird predators on
hopper bands of the desert locust. Entomologist's Monthly
Magazine 94: 210-214.
Huntly, N. and R. S. Inouye. 1987. Small mammal populations of an
old-field chronosequence: successional patterns and associations
with vegetation. Journal of Mammalogy 68: 739-745.
Hurlbert, S. H. 1984. Pseudoreplication and the design of ecological
field experiments. Ecological Monographs 54: 187-211.
Innami, S., M. G. Yang, O. Mickelsen and H. D. Hafs. 1973. The
influence of high-fat diets on estrous cycles, sperm production
and fertitlity of rats. Proceedings of the Society of
Experimental Biology and Medicine 143: 63-68.
Jacobs, H. L. and K. N. Sharma. 1969. Taste versus calories: sensory
and metabolic signals in the control of food intake. Annals of
the New York Academy of Sciences 157: 1084-1125.
Jacobs, W. W., G. K. Beauchamp and M. R. Kare. 1978. Progress in
animal flavor research. Pages 1-20 ia R. W. Bullard. Flavor
chemistry of animal foods. ACS Symposium Series, No. 67, American
Chemical Society, Washington, D.C.
Jacobs, W. W. and J. N. Labows. 1979. Conditioned aversion, bitter
taste and the avoidance of natural toxicants in wild guinea pigs.
Physiology and Behavior 22: 173-178.
Jameson, E. W. J. 1953. Reproduction of deer mice (Peromyscus
manicnlatus and Ej. boylii) in the Sierra Nevada, California.
Journal of Mammalogy 34: 44-58.
Jameson, E. W. J. 1955. Some factors affecting fluctuations of
Microbus and Peromvscus. Journal of Mammalogy 36: 206-209.
Jenson, T. S. 1982. Seed production and outbreaks of non-cyclic
rodent populations in deciduous forests. Oecologia 54: 184-192.

202
Jeuniaux, C. 1961. Chitinase: an addition to the list of hydrolases
in the digestive tract of vertebrates. Nature 192: 135-136.
Kalat, J. W. and P. Rozin. 1973. "Learned safety" as a mechanism in
long-delay taste-aversion learning in rats. Journal of
Comparative and Physiological Psychology 83: 198-207.
Kelley, R. B., J. N. Seiber, H. J. Jones, H. J. Segall, L. P. Brower.
1987. Pyrrolizidine alkaloids in overwintering monarch
butterflies (Danaus plexippus) from Mexico. Experientia 43: 943-
946.
Kenagy, G. J. 1972. Saltbush leaves: excision of hypersaline tissue
by a kangaroo rat. Science 178: 1094-1096.
Kenagy, G. J. and B. M. Barnes. 1988. Seasonal reproductive patterns
in four coexisting rodent species from the Cascade Mountains,
Washington. Journal of Mammalogy 69: 274-292.
King, C. M. 1983. The relationship between beech (Nothofagus sp.)
seedfall and population dynamics fo mice (Mus musculus), and the
demopgraphic and dietary responses of stoats (Mustella erminea),
in three New Zealand forests. Journal of Animal Ecology 52: 141-
166.
Klasing, K. C. 1985. Influence of stress on protein metabolism.
Pages 269-280 in G. P. Moberg. Animal stress. American
Physiological Society, Bethesda, Maryland.
Krebs, C. J. and R. Boonstra. 1984. Trappability estimates for mark-
recapture data. Canadian Journal of Zoology 62: 2440-2444.
Krebs, J. R. 1978. Optimal foraging: decision rules for predators.
Pages 23-63 in J. R. Krebs and N. B. Davies. Behavioral ecology:
an evolutionary approach. Sinauer Associates, Sunderland,
Massachusetts.
Landauer, M. R., R. L. Balster and L. S. Harris. 1985. Attenuation of
cyclophosphamide-induced taste aversions in mice by
prochlorperazine, A9-tetrahydrocannabinol, nabilone and
levonantradol. Pharmacology, Biochemistry and Behavior 23: 259-
266.
Landry, S. 0., Jr. 1970. The Rodentia as omnivores. Quarterly Review
of Biology 45: 351-372.
Langley, W. 1981. The effect of prey defenses on the attack behavior
of the southern grasshopper mouse (Onychomys torridus).
Zeitschrift fiir Tiepsychologie 56: 115-127.
Lauterbach, F. 1981. Intestinal absorption and secretion of cardiac
glycosides. Pages 105-139 in K. Greeff. Cardiac glycosides, Part
II: Pharmacokinetics and clinical pharmacology. Springer-Verlag,
New York.
Linduska, J. P. 1942. Winter rodent populations in field-shocked
corn. Journal of Wildlife Management 6: 353-363.

203
Lloyd, M. and H. S. Dybas. 1966. The periodical cicada problem. I.
Population ecology. Evolution 20: 133-149.
Lukas, D. S. and A. G. deMartino. 1969. Binding of digitoxin and some
related cardenolides to human plasma proteins. Journal of
Clinical Investigations 48: 1041-1053.
Lush, I. E. 1981. The genetics of tasting in mice. I. Sucrose octa-
acetate. Genetical Research, Cambridge 38: 93-95.
Lush, I. E. 1982. The genetics of tasting in mice. II. Strychnine.
Chemical Senses 7: 93-98.
Lush, I. E. 1984. The genetics of tasting in mice. III. Quinine.
Genetical Research, Cambridge 44: 151-160.
Lush, I. E. 1986. The genetics of tasting in mice. IV. The acetates
of raffinóse, galactose and B-lactose. Genetical Research,
Cambridge 47: 117-123.
M'Closkey, R. T. and D. T. Lajoie. 1975. Determinants of local
distribution and abundance in white-footed mice. Ecology 56: 467-
472.
Malcolm, S. B. and L. P. Brower. 1989. Evolutionary and ecological
implications of cardenolide sequestration in the monarch
butterfly. Experientia 45: 284-294.
Malcolm, S. B., B. J. Cockrell and L. P. Brower. 1989. Cardenolide
fingerprint of monarch butterflies reared on the common milkweed,
Asclepias syriaca L. Journal of Chemical Ecology 15: 819-853.
Marty, M. M. 1983. Disposition and metabolism of milkweed
(Asclepiadaceae) cardenolides in monarch butterfly larvae, Danaus
plexippus, and in a predator, Peromyscus maniculatus.
Dissertation. University of California, Davis.
Mason, J. R., M. D. Rabin, and D. A. Stevens. 1982. Conditioned taste
aversions: skin secretions used for defense by tiger salamanders,
Ambystoma tigrinum. Copeia 1982: 667-671.
Masters, A. R., S. B. Malcolm and L. P. Brower. 1988. Monarch
butterfly (Danaus plexippus) thermoregulatory behavior and
adaptations for overwintering in Mexico. Ecology 69: 458-467.
Mattocks, A. R. 1971. Hepatotoxic effects due to pyrrolizidine
alkaloid N-oxides. Xenobiotica 1: 563-565.
Mattocks, A. R. 1972. Acute hepatotoxicity and pyrrolic metabolites
in rats dosed with pyrrolizidine alkaloids. Chemico-Biological
Interactions 5: 227-242.
Mattocks, A. R. 1972. Toxicity and metabolism of Senecio alkaloids.
Pages 179-200 in J. B. Harborne. Phytochemical ecology. Academic
Press, New York, New York.

204
Mattocks, A. R. 1981. Relation of structural features to pyrrolic
metabolites in livers of rats given pyrrolizidine alkaloids and
derivatives. Chemico-Biological Interactions 35: 301-310.
Mattocks, A. R. 1986. Chemistry and toxicology of pyrrolizidine
alkaloids. Academic Press, New York.
Mazdzer, E., M. R. Capone and L. C. Drickamer. 1976. Conspecific
odors and trappability of deer mice (Peromynnis lsucnmia
noveboracenis). Journal of Mammalogy 57: 607-609.
McLean, E., G. Bras and P. Gyorgy. 1964. Veno-occlusive lesions in
livers of rats fed Crotalaria fulva. British Journal of
Experimental Pathology 45: 242-247.
Metzgar, L. H. 1971. Behavioral population regulation in the
woodmouse, Peromyscus leucopus. The American Midland Naturalist
86: 434-448.
Millar, J. S. 1979. Energetics of lactation in Peromvscus
maniculatus. Canadian Journal of Zoology 57: 1015-1019.
Millar, J. S. 1984. Reproduction and survival of Peromsycus in
seasonal environments. Pages 253-266 in J. F. Merrit. Winter
ecology of small mammals. Special Publication No. 10, Carnegie
Museum of Natural History, Pittsburgh, Pennsylvania.
Miranda, C. L., R. L. Reed, P. R. Cheeke and D. R. Buhler. 1981.
Protective effects of butylated hydroxyanisole against the acute
toxicity of monocrotaline in mice. Toxicology and Applied
Pharmacology 59: 424-430.
Moe, G. K. and A. E. Farah. 1975. Digitalis and allied cardiac
glycosides. Chapter 31 in L. S. Goodman and A. Gilman. The
pharmacological basis of therapeutics. Macmillan, New York.
Moermond, T. C. and J. C. Denslow. 1983. Fruit choice in neotropical
birds: fruit type and accessibility. Journal of Animal Ecology
52: 407-20.
Morris, R. F., W. F. Cheshire, C. A. Miller and D. G. Mott. 1958. The
numerical response of avian and mammalian predators during a
gradation of the spruce budworm. Ecology 39: 487-494.
Mueller-Dombois, D. and H. Ellenberg. 1974. Aims and methods in
vegetation ecology. John Wiley and Sons, New York.
Okita, G. T. 1967. Species differences in duration of action of
cardiac glycosides. Federation Proceedings 26: 1125-1130.
Packard, G. C. and T. C. Boardman. 1988. The misuse of ratios,
indices and percentages in ecophysiological research.
Physiological Zoology 61: 1-9.
Pasteéis, J. M., J. Grégoire-C. and M. Rowell-Rahier. 1983. The
chemical ecology of defense in arthropods. Annual Review of
Entomology 28: 263-289.

205
Patton, H. D. and T. C. Ruch. 1944. Preference thrsholds for quinine
hydrochloride in chimpanzee, monkey and rat. Journal of
Comparative Psychology 37: 35-49.
Pearson, D. L. 1985. The function of multiple anti-predator
mechanisms in adult tiger beetles (Coleóptera: Cicindelidae).
Ecological Entomology 10: 65-72.
Perrigo, G. 1987. Breeding and feeding strategies in deer mice and
house mice when females are challenged to work for their food.
Animal Behaviour 35: 1298-1316.
Peters, R. H. 1983. The ecological implications of body size.
Cambridge University Press, Cambridge, Massachusetts.
Porges, S. W. 1985. Spontaneous oscillations in heart rate: potential
index of stress. Pages 97-112 in G. P. Moberg. Animal stress.
American Physiological Society, Bethesda, Maryland.
Porter, W. P. and P. McClure. 1984. Climate effects growth and
reproduction potential in Sigmodon hispidus and Peromyscus
maniculatus. Pages 173-182 in J. F. Merrit. Winter ecology of
small mammals. Special Publication No. 10, Carnegie Museum of
Natural History, Pittsburgh, Pennsylvania.
Pressman, T. G. and J. H. Doolittle. 1966. Taste preferences in the
Virginia opossum. Psychological Reports 18: 875-878.
Redford, K. H. 1984. Mammalian predation on termites: tests with the
burrowing mouse (Oxvmvcterus roberti) and its prey. Oecologia 65:
145-152.
Richter, C. P. 1950. Taste and solubility of toxic compounds in
poisoning of rats and man. Journal of Comparative and
Physiological Psychology 43: 358-374.
Rintamaa, D. L., P. A. Mazur and S. H. Vessey. 1976. Reproduction
during two annual cycles in a poulation of Peromyscus maniculatus
noveboracensis. Journal of Mammalogy 57: 593-595.
Risser, P. G. and E. L. Rice. 1971. Phytosocial analysis of Oklahoma
upland forest species. Ecology 52: 940-945.
Robbins, D. J. 1982. The pyrrolizidine alkaloids. Pages 115-203 in
W. Herz, H. Grisebach and G. W. Kirby. Progress in the chemistry
of organic natural products, Volume 41. Springer-Verlag, New
York.
Robbins, R. J. 1978. Poison-based taste aversion learning in deer
mice (Peromyscus maniculatus bairdi). Journal of Comparative and
Physiological Psychology 92: 642-650.
Robertson, P. B. 1975. Reproduction and community structure of
rodents over a transect in southern México. Dissertation.
University of Kansas.
Roeske, C. N., J. N. Seiber,- L. P. Brower and C. M. Moffit. 1976.
Milkweed cardenolides and their comparative processing by monarch

206
butterflies (Danaus plexippus L.) . Recent Advances in
Phytochemistry 10: 93-167.
Rothschild, M. and N. Marsh. 1978. Some peculiar aspects of
Danaid/plant relationships. Entomologia Experimentalis et
Applicata 24: 437-450.
Rzóska, J. 1953. Bait shyness, a study in rat behavior. Animal
Behaviour 1: 128-135.
Sadlier, R. M. F. S., K. D. Casperson and J. Harling. 1973. Intake
and requirements of energy and protein for the breeding of wild
deermice, Peromyscus maniculatus. Journal of Reproduction and
Fertility, Supplement 19: 237-252.
Sargent, T. D. 1987. On the relative acceptability of the typical and
melanic morphs of Panthea pallescens McDunnough (Lepidoptera:
Noctuidae) to birds. Journal of the New York Entomological
Society 95: 495-503.
Scheer, B. T., D. F. Soule, M. Fields and H. J. Deuel Jr. 1947. The
effect of fat level of the diet on general nutrition. Journal of
Nutrition 33: 583-592.
Schipp, E., K. Kieth, R. L. Hughes and K. Meyers. 1963. Reproduction
in a free-living population of domestic rabbits, Orvctolagus
cuniculus (L.), on a sub-antarctic island. Nature 200: 858-860.
Schoental, R. and P. N. Magee. 1957. Chronic liver changes in rats
after a single dose of lasiocarpine, a pyrrolizidine (Senecio)
alkaloid. Journal of Pathology and Bacteriology 74: 305-319.
Schwartz, N. N. and J. H. Blumbergs. 1964. Epoxidations with m-
chloroperbenzoic acid. The Journal of Organic Chemistry 29: 1976-
1978.
Seiber, J. N., L. P. Brower, S. M. Lee, M. M. McChesney, H. T. A.
Cheung, C. J. Nelson and T. R. Watson. 1986. Cardenolide
connection between overwintering monarch butterflies from Mexico
and their larval food plant, Asclepias syriaca. Journal of
Chemical Ecology 12: 1157-1170.
Selye, H. 1983. The stress concept: Past, present and future. Pages
1-20 in C. L. Cooper. Stress research. John Wiley and Sons, New
York.
Sheppe, W. 1966. Determinants of home range in the deer mouse,
Peromyscus leucoous. Proceedings of the California Academy of
Sciences 34: 377-418.
Smith, J. C., D. P. Williams and S. Short Jue. 1976. Rapid oral
mixing of glucose and saccharin by rats. Science 191: 304-305.
Smith, K. D. and G. B. Popov. 1953. On birds attacking desert locust
swarms in Eritrea. Entomologist 86: 3-7.

207
Stebbins, L. L. 1977. Energy requirements during reproduction of
Peromyscus maniculatus. Canadian Journal of Zoology 55: 1701-
1704.
Steward, V. B., K. G. Smith and F. M. Stephan. 1988. Red-winged
blackbird predation on periodical cicadas (Cicadidae: Maaicicada
spp.): bird behavior and cicada responses. Oecologia 76: 348-352.
Stickel, L. F. 1960. Peromyscus ranges at high and low densities.
Journal of Mammalogy 41: 433-441.
Stoddard, D. M. and P. A. Smith. 1984. Woodmice IAandamiis svlvatjcua)
can distinguish conspecific from heterospecific odors in the
field. Journal of Chemical Ecology 10: 923-484.
Strehl, C. E. and J. White. 1986. Effects of superabundant food on
breeidng success of behavior of the red-winged black-bird.
Oecologia 70: 178-186.
Swain, T. 1976. Reptile-Angiosperm Co-evolution. Pages 107-122 in A.
d. Bellairs and C. B. Cox. Morphology and biology of reptiles.
Linnean Society Symposium Series, No. 3.
Taitt, M. J. 1981. The effect of extra food on small rodent
populations: I. deermice (Peromsycus maniculatus). Journal of
Animal Ecology 50: 111-124.
Tanz, R. D. and P. R. Urquilla. 1982. Cardiac glycosides. Pages 269-
281 in C. R. Craig and R. E. Stitzel. Modern pharmacology.
Little, Brown and Company, Boston.
Urquhart, F. A. 1976. Found at last: the monarch's winter home.
National Geographic 150: 160-173.
Vaughan, T. A. and N. J. Czaplewski. 1985. Reproduction in Stephen's
woodrat: the wages of folivory. Journal of Mammalogy 66: 429-443.
Vessey, S. V. 1987. Long-term population trends in white-footed mice
and the impact of supplemental food and shelter. American
Zoologist 27: 879-890.
Villa-R., B. 1952. Mamiferos silvestres del Valle de México. Anales
del Instituto de Biologia 23: 269-492.
Walford, P. S. H. T., Amherst, MA. 1980. Lipids in the life cycle of
the monarch butterfly, Danaus plexipous. Honors thesis. Amherst
College, Amherst, MA.
Waloff, Z. 1966. The upsurges and recessions of the desert locust
plague: an historical survey. Anti-Locust Memoirs 8: 1-100.
Warren, R. P. and R. C. Lewis. 1970. Taste polymorphism in mice
involving a bitter sugar derivative. Nature 227: 77-78.
Watson, A. and R. Moss. 1970. Dominance, spacing behavior, and
aggression in relation to population limitation in vertebrates.
Pages 167-218 in A. Watson. Animal populations in relation to
their food resources. Blackwell Scientific Publications, Oxford.

208
Watts, C. H. S. 1970. Effect of supplementary food on breeding in
woodland rodents. Journal of Mammalogy 51: 169-171.
Whitaker, J. O. 1966. Food of Mus musculus. Peromyscus maniculatus
bairdi and Peromvscus leucopus in Vigo County, Indiana. Journal
of Mammalogy 47: 473-486.
Whitman, D. w., M. S. Blum and C. G. Jones. 1985. Chemical defense in
Taeniopoda eques (Orthoptera: Acrididae): role of the metathoracic
secretion. Annals of the Entomological Society of America 78:
451-455.
Whitman, D. W., M. S. Blum and C. G. Jones. 1986. Olfactorily
mediated attack suppression in the southern grasshopper mouse
toward an unpalatable prey. Behavioral Processes 13: 77-83.
Whittaker, R. H. 1970. Communities and ecosystems. Macmillan
Publishing Company, Inc., New York.
Wolfe, J. L. 1966. Agonistic behavior and dominance relationship of
the eastern chipmunk, Tamias striatus. American Midland
Naturalist 76: 190-200.
Wolff, J. 0. 1985. The effects of density, food and interspecific
interference on home range size in Peromvscus leucopus and
Peromvscus maniculatus. Canadian Journal of Zoology 63: 2657-
2662.
Wolff, J. 0. 1986. The effects of food on midsummer demography of
white-footed mice, Peromsycus leucopus. Canadian Journal of
Zoology 64: 855-858.
Wolff, J. O., R. D. Dueser and K. S. Berry. 1985. Food habits of
sympatric Peromsycus leucopus and Peromyscus maniculatus. Journal
of Mammalogy 66: 795-798.
Wolff, J. O., M. H. Freeberg and R. D. Deuser. 1983. Interspecific
territoriality in two sympatric species of Peromyscus (Rodentia:
Cricetidae). Behavioral Ecology and Sociobiology 12: 237-242.
Yang, R. S. H. and M. R. Rare. 1968. Taste response of a bird to
constituents of arthropod defensive secretions. 61: 781-782.
Young, B. L. and J. Stout. 1986. Effects of extra food on small
rodents in a south temperate zone habitat: demographic responses.
Canadian Journal of Zoology 64: 1211-1217.
Zar, J. H. 1984. Biostatistical analysis.
Englewood Cliffs, New Jersey.
Prentice-Hall, Inc.,

BIOGRAPHICAL SKETCH
John Ingersoll Glendinning was born on 11 December 1959 in Bryn
Mawr, Pennsylvania. At the age of 10, he and his family escaped from
this stilted suburb of Philadephia and moved to Aptos, California.
There, amidst the Coast Ranges he developed an interest in natural
history. Three years later, he returned to Philadelphia, but this time
to the "inner-city," where he attended Germantown Friends School
through twelfth grade. In response to the strictness of his Quaker
high school, he decided to attend Hampshire College, where rules and
regulations were conspicuously absent. While at Hampshire he met Diana
Schulmann and discovered Aikido, both of which are still major passions
in his life. Immediately after receiving his Bachelor of Arts degree
at Hampshire in 1982, he entered the zoology graduate program at the
University of Florida. He married Diana Schulmann in April 1989. He
plans to continue his work in feeding ecology at The Florida State
University, where he will be the Lloyd M. Beidler Postdoctoral Fellow
for the next two years.
209

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and qualit
Doctor of Philosophy.
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
.-""A !'
v -.
John F. Eisenberg
Katharine Ordway Professor of
Ecosystem Conservation \
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
Frank Slansky, Jr. C, / â– 
Associate Professor of Entomology and
Nematology
Frank Slansky, Jr.
This dissertation was submitted to the Graduate Faculty of the
Department of Zoology in the College of Liberal Arts and Sciences and
to the Graduate School and was accepted as partial fulfillment of the
requirements for the degree of Doctor of Philosophy.
August 1989
Dean, Graduate School

UNIVERSITY OF FLORIDA
3 1262 08554 2735