Behavioral biology of the striped grass looper, Mocis latipes (Guenée), in north-central Florida

MISSING IMAGE

Material Information

Title:
Behavioral biology of the striped grass looper, Mocis latipes (Guenée), in north-central Florida
Uncontrolled:
Striped grass looper
Mocis latipes
Physical Description:
xiii, 123 leaves : ill. ; 28 cm.
Language:
English
Creator:
Dean, Thomas William, 1949-
Publication Date:

Subjects

Subjects / Keywords:
Moths -- Florida   ( lcsh )
Insects -- Florida   ( lcsh )
Grasses -- Diseases and pests -- Florida   ( lcsh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1985.
Bibliography:
Includes bibliographical references (leaves 113-122).
Statement of Responsibility:
by Thomas William Dean.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 000522354
notis - ACU3981
oclc - 14238954
System ID:
AA00003399:00001

Full Text












BEHAVIORAL BIOLOGY OF THE STRIPED GRASS LOOPER,
Mocis latipes (Guen6e), IN NORTH-CENTRAL FLORIDA










By


THOMAS WILLIAM DEAN


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


1985

































"For the things we have to learn before
we can do them, we learn by doing them."

Aristotle















ACKNOWLEDGEMENTS


I thank the members of my supervisory committee, Dr. D.

H. Habeck, Dr. F. A. Johnson, and Dr. 0. C. Ruelke for their

unequivocal interest in my research program. The willing-

ness with which each has so often set aside the demands

placed on him from other quarters that he might assist me by

kind word, field conference, thoughtful criticism, or pro-

curement of temporary employment, has left me with a clearer

perception of a teacher's proper role. To my mind, teacher

is the most exalted of titles; upon them, it has indeed been

fittingly bestowed.

I would be remiss if I did not here extend an indica-

tion of my gratitude to Dr. D. R. Minnick, for my graduate

study's inception is largely a consequence of his supports,

encouragements, and vision. I am deeply indebted.

I greatly appreciate the time and effort given by Bill

Carpenter of the Entomology and Nematology Department and by

Pat Carlysle of the USDA/ARS Laboratory; in large measure,

the electron microscopy is their handiwork.

No mere word or fine sounding phrase can serve as ample

requital for the forbearance so lovingly given by my wife

Nancy. Without her, both my work and I are blank pages.


iii

















TABLE OF CONTENTS


Page

ACKNOWLEDGEMENTS ........................................ iii

LIST OF TABLES .......................................... vi

LIST OF FIGURES ......................................... ix

ABSTRACT ................................................ xi

INTRODUCTION ............................................ 1

LITERATURE REVIEW.......................................... 5

MATERIALS AND METHODS ....... ............................ 25

Field Experiments................................... 25

Moth Attraction to
Different Host Grasses ........................ 25

Moth Activity in
Callie Giant Bermudagrass ..................... 27

Vertical Movement of Larvae ................... 28

Evaluation of a
Grassland Sampling Device ..................... 31

Laboratory Experiments ................... ............ 36

Chorion Consumption........................... .. 36

Early-Instar Feeding
Surface Preference............................. 37

Stage-Specific Developmental Time.............. 39

RESULTS AND DISCUSSION................................... 41

Field Experiments ................................... 41

Moth Attraction to
Different Host Grasses..... .................... 41











Moth Activity in
Callie Giant Bermudagrass ..................... 52

Vertical Movement of Larvae ................... 66

Evaluation of a
Grassland Sampling Device ..................... 78

Laboratory Experiments............................... 85

Chorion Consumption..................... ..... .. 85

Early-Instar Feeding
Surface Preference.............................. 93

Stage-Specific Developmental Time.............. 100

CONCLUSIONS ............................................. 109

LITERATURE CITED ........................................ 113

BIOGRAPHICAL SKETCH....................................... 123
















LIST OF TABLES


Page


Table 1.





Table 2.





Table 3.




Table 4.





Table 5.




Table 6.





Table 7.


ANOVA of transformed counts of adult
M. latipes visitation to culms of
P. notatum, C. dactylon, and
D. sanguinalis at eight different
times of night...............................

Overall mean abundance of adult
M. latipes per 10-culm cluster
of P. notatum, D. sanguinalis,
or C. dactylon assayed under field
conditions...................................

Time-specific abundance of a field
population of adult M. latipes on
10-culm clusters of either P. notatum,
C. dactylon, or D. sanguinalis ..............

Summary of three-way, split-plot ANOVA
of adult M. latipes abundance in three
different ages of Callie giant bermuda-
grass, C. dactylon, at five different
times of night...............................

Summary statistics of adult M. latipes
abundance in three different ages of
Callie giant bermudagrass, C. dactylon,
at five different times of night ............

Summary statistics of the proportions of
an adult M. latipes population observed
ovipositing in Callie giant bermudagrass,
C. dactylon, plots of two different ages
at five different times of night ............

Summary statistics of the proportion of
an adult M. latipes population observed
in copula in Callie giant bermudagrass,
C. dactylon, plots of two different ages
at five different times of night ............











Table 8.


Table 9.




Table 10.




Table 11.




Table 12.



Table 13.




Table 14.




Table 15.



Table 16.



Table 17.




Table 18.


Summary of two-way, split-plot ANOVA of
the time-specific vertical movements of
fifth and sixth instar M. latipes larvae
in Callie giant bermudagrass, C. dactylon...

Summary of two-way, split-plot ANOVA of
the time-specific vertical movements of
third and fourth instar M. latipes larvae
in Callie giant oermudagrass, C. dactylon...

Time-specific mean height above soil
surface of fifth and sixth instar M.
latipes larvae isolated in caged pots
of Callie giant bermudagrass, C. dactylon...

Time-specific mean height above soil
surface of third and fourth instar M.
latipes larvae isolated in caged pots
of Callie giant bermudagrass, C. dactylon...

Attrition of marked M. latipes larvae
in plots of Callie giant bermudagrass,
C. dactylon..................................

Summary statistics of the proportion
of marked M. latipes larvae present
per plot recaptured per trial by the
pushcart sampling device ....................

Single-classification ANOVA of a
once-repeated measurement of the
chorion consumed by an emerging first
instar M. latipes larva .....................

Two-level nested ANOVA of the chorion
consumed by an emerging first instar
M. latioes larva.............................

Summary statistics of the amount of
chorion consumed by an emerging first
instar A. latipes larva.....................

Point estimates and precisions for the
proportion of M. latipes ova observed
in each of five different consumption
categories ..................................

Age structure of M. latipes larvae
group-cultured in 150 mm petri dishes
supplied with terminal growth of Callie
bermudagrass, C. dactylon ...................


vii











Table 19.





Table 20.





Table 21.




Table 22.




Table 23.


Summary statistics of the feeding
surface preferences of first and
second instar M. latipes larvae
cultured on Callie giant bermudagrass,
C. dactylon, blades .... ..................... 96

Summary statistics for the "error
intervals" accompanying the stage-
specific developmental times of M.
latipes reared in the laboratory on
C. dactylon.................................. 101

Summary statistics for stage-specific
developmental times of M. latipes
reared in the laboratory on Callie
giant bermudagrass, C. dactylon............... 103

Stage-specific partition of derived
developmental duration for M. latipes
reared in the laboratory on Callie
giant bermudagrass, C. dactylon............... 105

Summary statistics of gender-specific
pupal masses obtained for M. latipes
reared in the laboratory on Callie giant
bermudagrass, C. dactylon, blades............ 107


viii
















LIST OF FIGURES


Page


Figure 1.





Figure 2.


Figure 3.


Figure 4.


Figure 5.




Figure 6.






Figure 7.


Figure 8.


Field arrangement of the caged pots
containing Callie giant bermudagrass,
C. dactylon, used to monitor the
vertical movements of M. latipes
larvae .....................................

A pushcart device for sampling M.
latipes larvae in forage grasses ...........

Adjustable features of the pushcart
sampler.....................................

Male Mocis latipes (Guen6e) resting
on a Paspalum notatum Flugge raceme ........

Histogram of adult M. latipes abundance
per night during 6 September to 15
September 1983 in a field of Callie
giant bermudagrass..........................

Least-squares regression functions
describing changes in adult M. latipes
abundance in plots containing different
ages of Callie giant bermudagrass, C.
dactylon, during the period from 6
September to 15 September, 1983 ............

M. latipes mating in Callie giant
bermudagrass, C. dactylon ..................

Deterministic polynomial model
describing the time-specific
vertical movements of fifth and
sixth instar M. latipes larvae
in Callie giant bermudagrass,
C. dactylon, canopy ........................











Figure 9.






Figure 10.




Figure 11.


Figure 12.



Figure 13.


Deterministic polynomial model
describing the time-specific
vertical movements of third and
fourth instar M. latipes larvae
in Callie giant bermudagrass,
C. dactylon, canopy ........................

Photophase resting position of
sixth instar M. latipes larva in
Callie giant bermudagrass, C.
dactylon ...................................

Behavior of M. latipes larvae in
the pushcart sampler 's cage ................

Differences in the degree of chorion
consumed by newly closed M. latipes
larvae .....................................

Electron photomicrograph of a Callie
giant bermudagrass, C. dactylon,
blade ......................................
















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



BEHAVIORAL BIOLOGY OF THE STRIPED GRASS LOOPER,
Mocis latipes (Guende), IN NORTH-CENTRAL FLORIDA



By



Thomas William Dean



August 1985



Chairman: Dr. D. H. Habeck
Major Department: Entomology and Nematology

A field population of adult Mocis latipes (Guenee),

(Lepidoptera: Noctuidae) preferred to visit culms of bahia-

grass, Paspalum notatum Flugge, rather than culms of either

crabgrass, Digitaria sanguinalis (L.) Scop., or bermuda-

grass, Cynodon dactylon (L.) Pers. Nightly moth visitation

peaked at ca. 20:00 hrs EDST and remained substantially

unchanged for the next three hours.

In Callie giant bermudagrass, C. dactylon, adult M.

latipes were more often observed to rest, mate and oviposit

in stands of grass that possessed closed canopies. Moths

usually deposited their eggs singly and demonstrated little










propensity to oviposit on any particular plant part. Ovi-

position occurred from 21:00 hrs to 22:30 hrs. In copulo

pairs were sessile, highly visible, and usually found in

upper portions of bermudagrass canopy. Mating activity

peaked at 01:00 hrs, but occurred from 21:00 hrs to 04:00

hrs.

Larval activity was nocturnal. Fifth and sixth instar

larvae ascended from and returned to basal portions of ber-

mudagrass canopy about an hour earlier than third and fourth

instars. Vertical movement was most pronounced from 19:00

hrs to 21:00 hrs and from 06:00 hrs to 08:00 hrs.

Mark-release-recapture evaluations of a pushcart samp-

ling device indicated a low, but consistent, proportion of

third and fourth instar M. latipes larvae present in

Callie giant bermudagrass may be recaptured when the device

is employed at either 20:00 hrs or 07:00 hrs. Of these two

times, sampler efficiency was higher at 20:00 hrs.

Following hatch, M. latipes larvae consumed more than

one-half of the chorion's spheroidal portion.

In petri dish culture, both first and second instar M.

latipes larvae chose to feed only on upper surfaces of

Callie giant bermudagrass blades. Scanning electron micros-

copy revealed tooth-like trichomes were more abundant on

blade margins and lower surfaces than on upper surfaces.

Stage-specific developmental times for M. latipes

larvae reared in the laboratory on Callie giant bermudagrass

foliage were comparable to those that have been reported


xii











when other grasses were supplied as diet. The average pupal

mass that resulted from larvae having been reared on Callie

giant bermudagrass was noticeably greater than values pre-

viously reported for other diets.


xiii
















INTRODUCTION


Of the four Mocis spp. known to occur in Florida (Ogun-

wolu and Habeck 1979), the striped grass looper, Mocis

latipes (Guenee), (Lepidoptera: Noctuidae), is the species

most likely to achieve economically injurious population

levels in North-Central Florida fields planted to forage

grasses (Genung and Allen 1974). Although dynamics of the

market price for hay and production overhead coupled with

both species and cultivar-specific differentials in

harvestable grass growth necessitates a virtual case-by-case

definition of "injurious level", some indication is given by

noting that M. latipes larval populations evincing densities

from 3 to 8 larvae per m2 have been adjudged detrimental

(Reinert 1975).

M. latipes is present in Florida throughout the year

(Kimball 1965); however, from Lake Okeechobee northward,

population levels are usually low prior to July or August.

Thereafter, larval populations may remain high until early

December (Genung and Allen 1974) or until cold weather

checks the growth of warm-season grasses (which, for the

Gainesville region, can be expected in mid-November).

Although considerable variation may be evidenced in a

series of specimens (Forbes 1960), the imago is usually

grey-brown with a wingspan of about 4 cm. The mesothoracic










wings possess strong antemedial and postmedial lines, with

the latter being only slightly inflected as it approaches

the anal margin. The reniform is usually visible, but may

be obscure; and the subreniform is a fine, fully-closed,

near-perfect circle, a portion of which is contiguous (or

nearly so) with the postmedial line. Moth gender is easily

ascertained: a profuse growth of clothing setae on the meta-

thoracic tibiae is present on males but not females.

M. latipes larvae possess longitudinal stripes that

extend from the adfrontal area to the the anal prolegs;

their coloration is highly variable (Ogunwolu 1974). Since

ventral prolegs are present only on the fifth and sixth

abdominal segments, larval gait is similar to that of the

measuring worms; from whence comes the common name, striped

grass looper.

Viewed cumulatively, a considerable body of literature

has been put forward and substantial progress made toward

gaining a better understanding of various aspects of M.

latipes' biology, taxonomic position, and pest status;

however, our knowledge of this moth's behavior in North-

Central Florida grasslands has long been in need of enrich-

ment. Recent advances in the development of high-yielding

forage grass varieties for this region make this need both

more evident and urgent; for many of the more promising

varieties achieve peak growth during the part of the year

that M. latipes can be most expected to achieve economically

injurious populational levels (Ruelke 1983).










Without an understanding of how the adults and larvae

of M. latipes behave in both field and plant space, through

time, in the presence of multiple grass species, and on

agriculturally-important grass varieties, efforts directed

toward their populations' management that utilize either

monitoring or sampling methodologies in a predictive

capacity are likely to provide less than optimal results.

This belief is the foundation of the studies here

presented. These include examination of

1. the relationship between time of night and moth

visitation to three of M. latipes' known host plants;

2. the influence of grass age on, and degree of time

specificity for: nightly moth abundance, oviposition, and

mating behavior of M. latipes in Callie giant bermudagrass;

3. the times for which different ages of M. latipes

larvae change their location in Callie giant bermudagrass

canopy;

4. the potential usefulness of a new grassland sampling

device as a tool for making absolute density estimates of

third and fourth instar M. latipes larvae in Callie giant

bermudagrass;

5. the proportion of the chorion that first instar M.

latipes larvae consume during the post-eclosion period;

6. the degree of specificity evidenced by first and

second instar M. latipes larvae for feeding on either upper

or lower surfaces of Callie giant bermudagrass blades; and








4

7. the stage-specific developmental schedule that

results when M. latipes larvae are supplied ad libitum

quantities of Callie giant bermudagrass foliage.
















LITERATURE REVIEW


The literature treating the striped grass looper, Mocis

latipes (Guen6e), (Lepidoptera: Noctuidae: Catocalinae)

spans 133 years, encompasses 2 continents, and mirrors this

moth's believed geographical range.

After examining 28 specimens, Guen6e (1852) placed this

species in the genus Remigia; hence, several entomological

authors of the 19th century refer to this moth as R. latipes

Guenee (Bethune 1869; Moschler 1880, 1886; Gundlach 1881;

Smith 1893; and Swainson 1900). During this era, this

species was incorrectly accorded a position in various other

noctuid genera; notably, Noctua and Ophiusa (Hampson 1913)

and Pelamia, Baratha, and Cauninda (Hodges et al. 1983).

The systematic revision by Hampson (1913) is the earliest

reference placing this moth in the genus Mocis; however,

broad acceptance seems to have been delayed for at least a

decade following Hampson's efforts. Perhaps as the final

holdout, Bissell (1940) refers to this insect as Pelamia

latipes Guenee.

Synonymy at the species level has served to exacerbate

the consternation that expectedly accompanies confusion of

genera. In the past, the moth now recognized as M. latipes

has been designated by a variety of specific epithets; these









include: repanda, punctularis, deliquens, excidens, sub-

tilis, convieniens, collata, detersa, indentata, and dif-

fluens (Hampson 1913). The vast majority of these names

convey little real difficulty; they are principally the

result of Walker's work at the British Museum and have been

resolved since the early part of the 20th century (Hampson

1913). That so many names have been applied to a single

species is evidence of both the difficulties inherent in the

taxonomy of this group and the intra-species diversity of M.

latipes. Passoa (1983) points this out, and Forbes (1960)

clearly enunciates the differences in forewing markings most

often seen in a series of specimens of this highly variable

species. Regarding synonymy, the most enduring difficulty

is the use of repanda, which is (in correct application) the

epithet of a sibling species described by Fabricius in 1794

and has been misapplied as an epithet in synonomy with

latipes (Hodges et al. 1983).

Notwithstanding the early and pointed indications by

Gundlach (1881), Moschler (1880, 1886), and Smith (1893)

that Remigia repanda and Remigia latipes were different

species, the epithet repanda has been a chronic synonym for

latipes (Dyar 1902, Vickery 1924, Watson 1933, Wolcott and

Otero 1936, Labrador 1964). The opening statement in

Kimball's (1965) treatment of Mocis in Florida is a

cautionary note that ". many of the records may be

mixed ." (Kimball 1965, p. 125); and, when reporting

on M. latipes, he states: ". confusion has arisen










because the species at one time went under the name Remigia

repanda (Fabricius) ." (Kimball 1965, p. 125). All of

the literature from the United States that mentions Mocis

(=Remigia) repanda is, in fact, treating M. latipes; for M.

repanda (F.) is not known to occur in this country (Hodges

et al. 1983). Unfortunately, no such clear-cut aid is

available for separating the Central and South American

literature.

In addition to describing the imago, Guenee (1852)

provided the first descriptions of M. latipes larvae and

pupae. Others that have given larval descriptions include

Dyar (1900), Swainson (1900), Jones and Wolcott (1922),

Crumb (1956), Forbes (1960), and Labrador (1964). None of

these works prove sufficient for species-specific determina-

tions of either larval or pupal specimens; this inadequacy

has been addressed by Ogunwolu and Habeck (1979).

The particulars of M. latipes" geographical origin

remain an enigma; past efforts of systematists seem to

suggest (by tacit agreement) that this moth is a native of

the tropical regions (Hampson 1913, Schaus 1940, Forbes

1960, Hodges et al. 1983). Guerne (1852) states that his

specimens were secured from widely separated locales, and

included Labrador and Madagascar. He maintained that M.

latipes occurs throughout central Africa and in both North

and South America, a notion with which Hampson (1913) was in

complete accord. Schaus (1940) pointed out that the

genitalia of the American specimens differed from those of










African specimens, and suggested that the American moths

were a distinct species.

Holland reported that M. latipes' range extends from

northern Canada to Argentina and that the moth keeps to the

eastern side of both the Rocky and Andes Mountains (Holland

1920); subsequent workers concur (Vickery 1924, Forbes 1960,

Reinert 1975). This moth's presence in the cooler regions

of its known range is quite probably the consequence of a

seasonally-mediated dispersal; Forbes has described this

species as one that obtains the limits of its range by

". straggling northward to Canada in the Fall ."

(Forbes 1960, p. 344), thereby offering a possible means of

accounting for Bethune (1869) being able to secure 2

specimens of M. latipes from the northern shore of Lake

Ontario.

For the past 60 years, the reports and investigations

of M. latipes' occurrence in the United States have been

confined to the states of the southeastern region. These

include: Texas (Vickery 1924), Alabama (Blake et al. 1959,

Guyton 1959), South Carolina (Anonymous 1961, Cuthbert 1954,

Nettles 1954), Georgia (Anonymous 1954; Bissell 1940; Beck

1955; Benton and Beck 1957; Dupree 1959; Geiger 1954;

Johnson 1957, 1959a, 1959b; Johnson and Buchanan 1957;

Jordan 1954, 1959; McGill 1957; Morgan 1955), and Florida

(Anonymous 1961, 1972, 1975a, 1975b; Beem 1954; Brinkley

1954; Clinton and Shirar 1965; Denmark 1957; Genung 1964,

1966, 1967, 1968, 1971; Genung et al.- 1976; Jones and Hodges










1953; Jones et al. 1952; Kelscheimer 1952, Kelsheimer et al.

1953; Kimball 1965; Kleyla et al. 1979; Koehler et al. 1977;

Kuitert 1954; Maltby 1954; May 1954; Mead 1965, 1970;

Ogunwolu 1974; Ogunwolu and Habeck 1975, 1979; Reinert 1975;

Strayer 1971; Van Pelt 1954a, 1954b; Ware 1973; Watson 1933;

Wolfenbarger 1955).

M. latipes has long been recognized as an injurious

species in the Caribbean. Contributions to the literature

have been made from Cuba (Babayan et al. 1975, Gundlach

1881), Jamaica (Moschler 1886), Puerto Rico (Capriles and

Ferrer 1973; Jones and Wolcott 1922; Van Dine 1913; Wolcott

1921, 1923, 1948; Wolcott and Martorell 1943, Wolcott and

Otero 1936), Trinidad (Mahadeo 1977, Urich 1910), and, the

Windward and Leeward Islands (Fennah 1947).

Surprisingly little literature is available from the

Central American mainland; however, McGuire and Crandall

(1967) have reported that M. latipes is present as a pest of

agricultural commodities in Mexico, Guatemala, Costa Rica,

El Salvador, Honduras, Nicaragua, and Panama. Passoa (1983)

has confirmed that this moth adversely affects agriculture

in Honduras.

In the past several decades, a considerable body of

literature treating various aspects of M. latipes has

emerged from some of the South American countries, notably,

Colombia (Varela and Calderon 1982), 'Guyana (Bodkin 1914,

Rambajan 1981), Surinam (Moschler 1880, Van Dinther 1971),

Venezuela (Labrador 1.964), Bolivia (Munro 1960), Brazil










(Bastos et al. 1979; Cruz and Santos 1983; Ferreira and

Parra 1984a, 1984b; Hseih 1979; Lara et al. 1977; Lara and

Silviera Neto 1977; Lima and Berti Filho 1984; Lourencao et

al. 1982), Argentina (Costilla et al. 1973, Hayward 1960),

and Uruguay (Biezanko et al. 1957).

Much has been learned about M. latipes' life history

and biology. In his discussion of the larvae, Guenee (1852)

informs us that these, ". s'enterre vers le commence-

ment d'aout ."; and, regarding the imago, ". se

trouve A la fin d'aout ." (Guen6e 1852, p. 315). Hence,

he [Guenee] provides the first record of this species'

seasonal presence [August] and an approximation of its life

cycle's duration [about 1 month]. Dyar (1900) reared Mocis

larvae and briefly mentioned their nocturnal habits. Urich

(1910) noted that Mocis larvae were found singly on grass

blades, and that they dropped to the soil's surface at the

slightest disturbance. Jones (1913) distinguished M.

latipes larval damage from that caused by Spodoptera frugi-

Derda Smith by noting that larvae of the former fed on un-

folded sugarcane leaves, whereas larvae of the latter fed in

the plant's whorl. After viewing the damage caused by M.

latipes in Texas grasslands, Vickery (1924) conducted the

first detailed lab studies of this moth's biology, subse-

quently determining that the larvae were not cannibalistic

and could be readily reared on leaves of Zea mays L. He

found that the eggs are deposited singly, that development

from egg to pupa required 21 days, and that the pupal stage








11

had an average duration of 12 days (Vickery 1924). Report-

ing from the Antilles, Fennah (1947) declared himself in

agreement with Vickery insofar as larval developmental time

was concerned, but noted that pupal duration was only 6

days. In reporting M. latipes' clutch size, Fennah is

virtually alone, stating that, ". eggs are laid on the

leaves in clusters of 40 60 close to the midrib below the

leaf, ." (Fennah 1947, p. 86); only Van Dinther (1954)

has voiced similar findings, maintaining that clutch size

is as large as 8 ova per ovipositional site.

Labrador (1964) found that a typical female produced an

average of 182 eggs in her lifetime, and that eclosion from

the egg required a 3 or 5 day incubation at 250 C or 280 C,

respectively; larval and pupal developmental times were in

agreement with those given by Vickery. Labrador also

measured the adult's lifespan, finding that moths fed a

dilute solution of honey would live from 2 to 17 days, with

the average being 9 days (Labrador 1964). Ogunwolu (1974)

measured the adult longevity of field-caged moths and found

that these will live an average of 11 days. Ogunwolu and

Habeck (1975) reported that M. latipes adults kept in

cardboard containers and fed a 10% sucrose solution produced

94 to 458 eggs per female, and that under these conditions,

the average number of fertile eggs was 277.4 (Ogunwolu and

Habeck 1975). In sharp contrast, the Cuban research team of

Babayan et al. (1975) reported that M. latipes adults

provided a sucrose solution lived up to 49 days and produced








12

from 215 to 611 eggs per female, with the average number of

eggs being 407.1. They report the larval developmental time

as being from 22 to 75 days (Babayan et al. 1975).

Studies of M. latipes reared in environmentally-

regulated growth chambers have been conducted by Ferreira

and Parra (1984a); these researchers detected no significant

differences in larval developmental time when temperature is

regulated to 200, 250 or 300 C; but that temperatures of 350

C or higher produce adverse effects. They conclude that M.

latipes has a developmental threshold of 13.70 C, and that

at this temperature, this species requires 391.30 growth-

degrees to develop from egg to adult (Ferreira and Parra

1984a). They also investigated the effects of photoperiod

on laboratory-reared M. latipes larvae, and found that

larvae reared under a 14:10 (light:dark) regimen yielded

pupae that were significantly different in mass from those

produced from larvae reared under either 10:14 or 12:12

(Ferreira and Parra 1984b).

Reinert (1975) fed M. latipes larvae freshly-cut

terminal stolons of St. Augustinegrass, Stenotaphrum

secundatum (Walt.) Kuntz, and noted that these larvae,

". consumed an average of 442.44 mg of leaf tissue dur-

ing their development with an average conversion ratio of

0.25 ." (Reinert 1975, p. 203). He contends that lar-

val efficiency in converting leaf tissue to body tissue is

greatest in the penultimate larval instar (Reinert 1975).

Conducting investigations in a similar vein, Ogunwolu and










Habeck (1975) noted that of the grass species which they

evaluate, M. latipes larvae both took longer to develop and

had higher early-instar mortality rates when confined to a

St. Augustinegrass diet. They also evaluated larval

performance on non-grass hosts and found that, without

exception, M. latipes larvae die without feeding when

provided a diet of only legume foliage; and that, when given

the choice of both legumes and grasses, the larvae feed only

upon the latter (Ogunwolu and Habeck 1975). More recently,

Cruz and Santos (1983) have determined M. latipes larvae

demonstrate no significant differences in developmental

time, pupal mass, or larval mortality, when reared on

foliage of either corn, Zea mays L., or Johnsongrass,

Sorghum halepense (L.) Pers.; they report observing a 55%

mortality rate in larvae fed an artificial diet (Cruz and

Santos 1983).

Very little of the literature on M. latipes treats the

subjects of populational dynamics or moth behavior in the

field. Silviera Neto et al. (1977) have investigated the

flight activity time of this moth species in Brazil and

report that in an automatic light trap calibrated to

partition nightly catches into 3-hour units, M. latipes is

taken only during the intervals of 18:00-21:00 hours and

21:00-24:00 hours; moreover, moth abundance in the early

evening interval is about twice that obtained for the late

night interval. Lara and Silviera Neto (1977) analyzed

light trap catches for the interval July 1972 to June 1975










and concluded that in Jaboticabal, Brazil M. latipes is

present year-round, and that its populational presence is

bimodal with peaks occurring during the months of June and

December. Also reporting from Brazil, the team of Lara et

al. (1977) studied M. latipes' presence in different locales

and developed a predictive index, ". Indice Constancia

Simultanea ." (Lara et al. 1977, p. 52), with which they

demonstrate capture of a M. latipes adult in one locale is

sufficient to predict its presence in another.

Although most have not made M. latipes' seasonal

presence the focal point of their efforts, numerous authors

have mentioned the time of year pertinent to their observa-

tions. In general, these are indicative of populations that

have caused serious injury to some crop. Such remarks are

scattered widely through the literature; however, in com-

posite, these serve to provide a fairly clear picture of

this moth's seasonal presence in various latitudes of the

Western Hemisphere; e.g., Argentina: Feb., Mar. (Costilla et

al. 1973); Uruguay: Feb., Mar., Apr. (Biezanko et al. 1957);

Brazil: Jan., Feb. (Lourencao et al. 1982); the Amazon Basin

of Brazil: Apr., May (Hseih 1979); Colombia: Sept. (Alvarez

and Sanchez 1981); Venezuela: Oct. (Labrador 1964); British

Guiana: year-round (Bodkin 1914); Guyana: year-round

(Rambajan 1981); Trinidad: Feb., Mar., Apr. (Urich 1910),

Aug. (Mahadeo 1977); Panama: Sep., Oct. (Crumb 1956);

Honduras: Mar., Jun., Jul., Aug., Nov. (Passoa 1983); Puerto

Rico: year-round (Jones 1913; Wolcott 1921, 1948; Wolcott










and Martorell 1943; Wolcott and Otero 1936; Capriles and

Ferrer 1973); Cuba: Sep. (Babayan et al. 1975); Florida:

year-round (Kimball 1965); Texas: Oct. (Vickery 1924);

Alabama: Oct. (Blake et al. 1959, Guyton 1959); Georgia:

Mar., (Johnson 1957); Aug., Sep., Oct. (McGill 1957, Morgan

1955, Geiger 1954); and, South Carolina: Sep., Oct. (Nettles

1954, Cuthbert 1954).

Many authors that have contributed to the literature on

M. latipes have included in their remarks some treatment of

this insect's host plants. In several cases these remarks

have been of a general nature, citing M. latipes on some

undetermined grass, pasture plant, or cover crop (Anon.

1954, Cuthbert 1954, Denmark 1957, May 1954, Passoa 1983,

Swainson 1900, Wolfenbarger 1955). A few authors, notably

Vickery (1924), Labrador (1964), and Tietz (1972), have

iterated the findings of others (and thus to a greater or

lesser extent have provided a host plant list). Most authors

have reported M. latipes only on the plant (or plants) of

immediate interest in their studies; in consequence, reports

of the host plants of M. latipes are scattered widely

throughout the literature. In composite, these reports

provide a reasonably clear picture of this insect's hosts

and indicate that many of the agriculturally important

grasses serve in this capacity. These include: sugarcane,

Saccharum officinarum L., (Biezanko et al. 1957; Capriles

and Ferrer 1973; Costilla et al. 1973; Hayward 1960;

Holloway 1933; Jones 1913; Jones and Wolcott 1922; Labrador










1964; Mahadeo 1977; Tietz 1972; Urich 1910; Van Dine 1913;

Vickery 1924; Wolcott 1921, 1948; Wolcott and Martorell

1943); bermudagrass, Cynodon dactylon (L.) Pers., (Anon.

1973, 1975a, 1975b; Beck 1955; Costilla et al. 1973; Geiger

1954; Johnson 1957, 1959a; Johnson and Buchanan 1957; Morgan

1955; Labrador 1964; Ogunwolu and Habeck 1975; Vickery 1924;

Watson 1933); corn, Zea mays L., (Biezanko et al. 1957, Cruz

and Santos 1983, Costilla et al. 1973, Hayward 1960, Kimball

1965, Labrador 1964, McGill 1957, Passoa 1983, Tietz 1972,

and Vickery 1924); rice, Oryza sativa L., (Biezanko et al.

1957, Hayward 1960, Hseih 1979, Kimball 1965, Labrador 1964,

Munro 1960, Passoa 1983, Rambajan 1981, Van Dinther 1971);

millet, Pennisetum americanum L., ( Benton and Beck 1957,

Blake et al. 1959, Johnson 1959b, McGill 1957, Ogunwolu and

Habeck 1975); bahiagrass, Paspalum notatum Flugge, (Anon.

1972, Beck 1955, Brinkley 1954, Lourencao et al. 1982,

Ogunwolu and Habeck 1975, Van Dinther 1971, Van Pelt 1954b);

Pangola digitgrass, Digitaria decumbens Stent., (Costilla et

al. 1973, Genung 1971, Jones et al. 1952, Jones and Hodges

1953, Kelsheimer et al. 1953, Kuitert 1954, Maltby 1954);

guineagrass, Panicum maximum L., (Capriles and Ferrer 1973,

Labrador 1964, Lourencao et al. 1982, Urich 1910, Wolcott

1948); paragrass, Brachiaria mutica (Forsk.) Stapf.,

(Kelsheimer et al. 1953, Genung 1968, Labrador 1964, Urich

1910); hairy crabgrass, Digitaria sanguinalis (L.) Scop.,

(McGill 1957, Tietz 1972, Watson 1933); Saint Augustine-

grass, Stenotaphrum secundatum (Walt.) Kuntz, (Anon. 1972,








17

Genung 1966, Ogunwolu and Habeck 1975, Reinert 1974); sudan-

grass, Sorghum sudanense (Piper) Stapf., (Genung 1964,

McGill 1957, Ogunwolu and Habeck 1975); grain sorghum,

Sorghum bicolor (L.) Moench, (Costilla et al. 1973, Labrador

1964); wheat, Triticum aestivum L., (Biezanko et al. 1957,

Labrador 1964); oats, Avena sativa L., (Jordan 1954, Mead

1965); maidencane, Panicum hemitomon Schult., (Clinton and

Shirar 1965); jaraguagrass, Hyparrhenia rufa (Stapf.),

(Labrador 1964, Lourencao et al. 1982); dallisgrass,

Paspalum dilatatum Poir., (Dupree 1959); ryegrass, Lolium

multiflorum Lam., (Genung 1967); Salina buffelgrass,

Cenchrus ciliaris L., (Capriles and Ferrer 1973);

Napiergrass, Pennisetum purpureum Schumach., (Costilla et

al. 1973, Wolcott 1948); Johnsongrass, Sorghum halepense

(L.) Pers., (Cruz and Santos 1983); Transvala digitgrass,

Digitaria decumbens Stent., (Ogunwolu and Habeck 1975);

goosegrass, Elusine indica (L.) Gaertn., (Labrador 1964);

dogtoothgrass, Cynodon incompletus Nees., (Ogunwolu and

Habeck 1975); rhodesgrass, Chloris gayana Kunth, (Ogunwolu

and Habeck 1975); multiflowered false rhodesgrass, Tri-

chloris pluriflora Fourn., (Tietz 1972, Vickery 1924);

malojillograss, Panicum barbinode L., (Jones 1913); cari-

maguagrass, Andropogon gayanus L., (Varela and Calderon

1982); and, natalgrass, Tricholaena rosea Nees., (Watson

1933). M. latipes has also been reported by Labrador

(1964), Tietz (1972) and Vickery (1924) on several grasses

which lack common names. These include: Eriochloa punctata










L., Leptochloa neallyi Vasey, Cenchrus viridis Spreng.,

Cenchrus brownii Roem., Panicum faciculatum Swartz, Paspalum

faciculaturm Wihl., and Ioxphorus unisetus L. In his origi-

nal description of the moth, Guenre (1852) noted that M.

latipes larvae fed on blades of Hypericum sp.

In addition to the aforementioned 39 grasses, M.

latipes has been periodically reported to occur on various

non-grasses, including: alfalfa, Medicago sativa L., (Kleyla

et al. 1979, Labrador 1964, Jordan 1959, Costilla et al.,

1973); soybean, Glycine max (L.) Merr., (Bissell 1940,

Costilla et al., 1973); field pea, Pisum sativum L.,

(Labrador 1964, Van Pelt 1954a); beans, Phaseolus vulgaris

L., (Biezanko et al., 1957); cowpea, Vigna sinensis (Endl.),

(Hayward 1960); hairy indigo, Indigofera hirsuta L., (Beem

1954); broadbean, Vicia faba L., (Kimball 1965); sweet

potato, Ipomoea batatas (L.) Lam., (Guyton 1959); turnip,

Brassica campestris L., (Kimball 1965); coffee, Coffea

arabica L., (Labrador 1964); and, Washington palm, Washing-

tonia robusta Wendl., (Ware 1973).

Reports of non-grass host plants of M. latipes have

evoked protestation by some researchers. After noting that

most of these non-grasses are members of the Leguminosae,

Ogunwolu and Habeck (1975) undertook to rear M. latipes

larvae on leaf tissue of several different legume species;

their findings led them to conclude that reports of M.

latipes on non-grasses are probably erroneous if the

determinative criterion of a host plant includes larval










development. It seems quite possible that reports of non-

grass hosts of M. latipes have been the consequence of a

researcher having observed last-instar larvae seeking a

pupation site, or the finding of M. latipes pupae on a non-

grass plant. Wolcott (1948) has reported finding M. latipes

pupae on Ipomoea sp. and Lourencao et al. (1982) have noted

that M. latipes pupae can be found on Bidens pilosa L.

foliage. Similiarly, Ogunwolu and Habeck (1975) have

recorded collecting M. latipes pupae from blackberry, Rubus

cuneifolius Pursh; dogfennel, Eupatorium compositifolium

Walt.; ragweed, Ambrosia artemisifolia L.; and Florida

pusley, Richardia scabra L.

The earliest mention of natural enemies of M. latipes

is probably that of Urich (1910), who observed that the

savannah black bird, Quiscaulus crassirostris Swainson, and

the smooth-billed ani, Crotophaga ani L., would feed on M.

latipes larvae that had been dislodged from sugarcane

foliage by laborers busy weeding the fields. Urich also

mentioned that, ". a number of parasitical flies were

bred from the caterpillars taken from the fields ."

(Urich 1910, p. 161); however, he made no specific determi-

nation of the parasites. Shortly thereafter, Bodkin (1914)

reported the coccinellid, Megilla maculata DeGeer, and the

Demerara robin, Leistes guianensis L., as predators of M.

latipes larvae. Jones and Wolcott (1922) noted that, in

Puerto Rico, 90% of the parasitization of M. latipes larvae

could be attributed to three species of tachinid flies:










Phorocera claripennis Macq., Linnaemyia fulvicauda Walton,

and Helicobia helicis Townsend. In addition, they recorded

the chalcid, Chalcis robusta Cresson; the sarcophagid,

Sarcophaga sternodontis Townsend; and an ichneumonid, Rogas

sp., as parasites of this moth's larvae (Jones and Wolcott

1922). However, Vickery (1924) reported examining large

numbers of Mocis larvae and pupae in Texas but was unable to

rear any parasites from them. Vickery found a single larva

parasitized by Euplectrus sp., but stated that the parasites

died as immatures (Vickery 1924). Fennah (1947) confirmed

the findings of predecessors and added Sarcophaga lambens

Wd. to the list of parasites reared from M. latipes; and,

Wolcott (1948) cited the lizards Anolis cristatellus Dumreril

and Bibron, Anolis krugii Peters, Anolis pulchellus Durmril

and Bibron, and Anolis stratulus Cope as predators. After

conducting crop-content analyses, Genung and Greene (1974)

and Genung et al. (1976) concluded that both the eastern

meadowlark, Sturnella magna argutulla Bangs., and Maynard's

redwing blackbird, Agelaius phoeniceus floridanus Maynard,

are instrumental in reducing populations of M. latipes

larvae in South Florida grasslands. Similiarly, Hodges et

al. (1975) have noted the beneficial effects of blackbirds

in Florida pastures.

To date, the most complete account of M. latipes

natural enemies in North America comes from Ogunwolu and

Habeck (1975); these researchers evaluated 325 Mocis pupae

and found 6% of them parasitized. Parasites which they were










successful in rearing included the sarcophagids Sarcophaga

sp. and Sarcodexia sternodontis Townsend; the braconids

Apanteles scitulus Riley, Meteorus autographe Muesebeck, and

Microplitis maturus Weed; the chalcids Brachymeria ovata

ovata (Say) and Brachymeria robusta (Cresson); and the

ichneumonids Coccygomimus aequalis (Provancher), Enicospilus

purgatus (Say), Enicospilus arcuatus? (Felt), and Gambus

ultimus (Cresson). They also reported that a tenebrionid

adult, Bothrothes fortis (Casey), and a carabid larva,

Pinacodera sp., prey on Mocis larvae (Ogunwolu and Habeck

1975). Babayan et al. (1975) mention having reared

Euplectromorpha sp., Euplectrus sp., Brachymeria sp., Bracon

sp., and Enicospilus sp. from M. latipes immatures. In

sharp contrast to the findings of virtually all other

researchers, this Cuban team reported parasitism levels in

field-collected M. latipes larvae and pupae to be as high as

78% (Babayan et al. 1975).

Working in Venezuela, Labrador (1964) compiled a list

of 27 arthropod parasites and predators of M. latipes. These

include the sarcophagids Sarcodexia stenodontis Townsend,

Sarcophaga aurea Hall, and Sarcophaga sp.; the tachinids

Linnaemya fulvicuada Walt., Phorocera rusti Aldr., Helicobia

helices Townsend, Achaetoneura aletiae Riley, Achaetoneura

archippivora Will., Archytas marmoratus Townsend, Blondelia

barmigera Coq., Cyrtophoeba sp., Euphorocera claripennis

Macq., and Winthemia sp.; the exoristid Platelloapis sp.;

the braconids Rogas sp. and Bracon sp.; the encyrtids










Copidosoma truncatellum Dalhm and Lytopilus sp.; the

ichneumonids Ophion sp. and Enicospilus sp.; the chalcids

Brachymeria incerta Cresson, Brachymeria robusta Cresson,

Chalcis sp., and Spilochalcis femorata Fabr.; the carabid

Calosoma alternans Fabr.; the vespid Polistes versicolor

Oliv.; and, the sphecid Sceliphrons figulum Fahlb. (Labrador

1964).

Other natural enemies of M. latipes in South America

include the tachinids Patelloa similis (Townsend), and

Euphorocera floridensis Townsend; the ichneumonid Netelia

sp.; and the vespid Polistes canadensis Hayward (Lourencao

et al. 1982). Most recently, the eulophid Horismenus

distinguendus Blanchard has been found to be capable of

completing its life cycle in the laboratory when provided

with M. latipes pupae (Lima and Berti Filho 1984).

The natural enemies of M. latipes have not always

effectively suppressed caterpillar-induced crop damage. In

result, various corrective actions have been recommended or

undertaken over the years. A few of these fall under the

general category of cultural control and include such

measures as flooding rice nursery plantings or burning

infested pasturelands after the larvae have pupated (Bodkin

1914). Early researchers advocated using inorganic com-

pounds such as lead arsenate, copper acetoarsenite, slaked

lime, and sodium fluosilicate to control M. latipes larvae

(Bodkin 1914, Jones and Wolcott 1922, Holloway 1933, Fennah

1947).










Immediately following World War II, several synthetic

organic pesticides became relatively inexpensive and widely

available; such compounds became the focus of a number of

researchers interested in the chemical control of M. latipes

in Florida pastures. Jones et al. (1952) noted that M.

latipes larvae became paralyzed approximately 48 hours after

an application of either DDT or chlordane, and that the same

effect could be achieved within 24 hours by an application

of toxaphene. An application of DDT, formulated as a 5%

dust and distributed by towing a commercial fertilizer

spreader set to deliver 30-35 pounds of dust per acre, was

reported to control M. latipes larvae in a paragrass, B.

mutica, pasture (Kelsheimer 1952, Kelsheimer et al. 1953).

Although DDT has not been used in Florida pasture for nearly

two decades, it is quite likely still used in other parts of

the world; DDT was last cited as a material with which to

control m. latipes by the Colombian research team of

Costilla et al. (1973).

Jones and Hodges (1953) demonstrated that a spray made

from parathion 15% wettable powder and applied at a rate of

0.75 pounds active ingredient per acre was sufficient to

kill M. latipes larvae in Pangola digitgrass, D.

decumbens. Labrador (1964) noted that either 0.25 kg

carbaryl, 0.20 kg trichlorfon, or 0.50 kg toxaphene per 100

liters of water would control M. latipes larvae in a hectare

of Venezuelan pasture. Strayer (1971) recommended carbaryl

(either as a 5% dust or 80% sprayable), toxaphene (as a 10%








24

dust or 40% wettable powder), phosdrin (as a 2% dust or 25%

emulsifiable concentrate), and parathion (as a 2% dust or

15% wettable powder) for control of M. latipes in Florida

pastures. Koehler et al. (1977) concluded that a single

aerial application of permethrin applied at a rate of 0.2

pounds active ingredient per acre will kill at least 84%

of the M. latipes larvae in bermiudagrass, C. dactylon,

within 24 hours and virtually 100% within 72 hours.

Many authors have remarked that M. latipes may be a

serious impediment to the production of some commodity, and

in several instances, graphic accounts of M. latipes"

destructiveness have been published (Bodkin 1914, Holloway

1933, Labrador 1964, Capriles and Ferrer 1973). However,

very few documents provide any comparative measure of this

insect's impact. McGuire and Crandall (1967) developed a

ranking system for the economically injurious pests of

select food crops produced in Central America. They ranked

M. latipes as the 5th most important pest of corn, Z. mays;

the 2nd most important pest of grain sorghum, S. bicolor;

the 6th most important in rice, 0. sativa; and, the premier

pest of pasture grasses (McGuire and Crandall 1967). Martin

et al. (1982) reported that M. latipes inflicted ca. $2.6

million damage to the hay and pasture industry in Georgia

during the 1980 growing season.
















MATERIALS AND METHODS


Field Experiments

Moth Attraction to Different Host Grasses

During the nights of 21-25 September 1983, the attrac-

tion of a field population of adult M. latipes to 1280

sample units was evaluated. Each sample unit contained 10

culms of a treatment grass or was a poultry wire control.

The treatment grasses were: bahiagrass, P. notatura; hairy

crabgrass, D. sanquinalis; and Callie giant bermudagrass, C.

dactylon. Both the bahiagrass and the crabgrass culms had

racemes; the bermudagrass culms were vegetative. All of the

plant material used in the experiment was obtained from

roadsides and fields of the University of Florida Campus

Agronomy Farm; the experiment was conducted in one of its

fields. Hourly measurements of temperature, rainfall, and

relative humidity during the experimental interval were

obtained from the Agronomy Farm weather data facility.

Forty receptacles were constructed; each was formed by

cementing the base of a 7-dram plastic shell vial to the

center of a 20 cim plastic potting saucer's inner surface.

Ten receptacles were fitted with a 12 cm x 45 cm strip of

2.5 cm mesh poultry wire. These 10 receptacles were used

as controls. The other 30 receptacles were partitioned








26

into three equal groups. The receptacles in each group were

subsequently filled with culms of a given treatment grass.

On each afternoon preceding a nightly assay, 125

freshly-cut culms of each grass were collected, sealed in

plastic bags, and transported to the experimental site.

Culms of each grass were arbitrarily selected from the bags

and their basal portions trimmed until all were ca. 45 cm

long. These were grouped, species-specifically, to form 30

bundles of 10 culms. The base of each bundle was wrapped

with a 2.5 cm x 12 cm strip of water-soaked surgical cotton

and placed in a receptacle's water-filled shell vial.

At ca. 18:30 hrs, all receptacles were placed in the

field and arranged in 1 row of 10 blocks. Each block

measured 1 m per side and was 5 m from adjacent blocks.

Every block contained all 3 treatment grasses plus a con-

trol. Random numbers determined the placement of recepta-

cles in the corners of each block.

Once ever hour, from 19:00 hrs to 03:00 hrs, each block

was visited and the number of M. latipes adults resting on

the contents of each receptacle was recorded. Whenever a

moth was seen, its behavior was noted and the resting site

examined for clues of its attractiveness. A clear-lensed,

6-volt, headlamp fitted with a duct tape visor was used to

aid observations.

Data were analyzed for differences in the total number

of moths attracted to each of the grasses and for differ-

ences in average moth abundance per grass at the various










times of night. All tests for significance and confidence

interval constructions were at the (a = 0.05) level.

Moth Activity in Callie Giant Bermudagrass

On 30 June 1983, a 0.11 ha field with an established

stand of Callie giant bermudagrass was chop-harvested to a

stubble of ca. 12 cu and fertilized with 18 kg of a 17-5-10

(N-P205-K20) complete fertilizer. After 24 days of

regrowth, the field was divided into 18 plots. Each plot

measured 4 m x 12 m and was separated from adjacent plots by

a 75 cm width of mowed alleyway. A random numbers table was

used to designate both a harvest date for each plot and each

plot's field location. Plot harvests were conducted on 1

August, 10 August, and 24 August. On each of the harvest

dates, 6 plots were mowed with a garden-tractor-powered

sickle bar set for a stubble height of ca. 5 cm; vegetation

was removed by pitchfork; the plot cleared by manual raking;

and 1.3 kg of 17-5-10 complete fertilizer applied.

During the nights of 6-13 September and 14-15 September

1983, the 18 plots were each inspected 5 times per night for

the presence of adult M. latipes. Nightly plot inspections

commenced at 18:00 hrs, 21:00 hrs, 01:00 hrs, 04:00 hrs, and

07:00 hrs. The plot-to-plot sequence for the observations

initiated at each of these times was kept fixed for all

times and nights; regarding field layout, the visitation

sequence was conducted systematically.

Each plot was examined by entering a 4 m side and

slowly moving through the vegetation in an ill-defined,








28

zig-zag path that roughly approximated the plot's long axis.

At every second or third step, movement through the plot was

paused and the vegetation carefully scanned for M. latipes

adults. A clear-lensed, 6-volt, headlamp fitted with a duct

tape visor aided observation. A multiple-key laboratory

counter was used to simultaneously record the number of

moths resting on foliage, the number of moths ovipositing,

and the number of in copulo moth pairs. Moths observed

flying over the plot were not counted.

Data were analyzed for differences in moth abundance in

the different aged plots, differences in the number of moths

present in the field per night, and differences in observed

moth behavior during the various times of night. Tests for

significance ana confidence interval constructions were at

the (a = 0.05) level.

Vertical Movement of Larvae

During the nights of 21-30 October 1983, the movements

of M. latipes larvae on potted plantings of Callie giant

bermuaagrass cultured under field conditions were studied.

Five days prior to data collection, freshly-dug clumps of

25-day-old grass were transplanted to 20 cm diameter plastic

pots filled with a 1:1:1 mixture of peat, Arredondo fine

sana, and vermiculite. Thirty such pots were prepared.

Each pot was placed in a 20 cm diameter plastic potting

saucer. The saucers were filled with tap water at 48 hour

intervals, and each received ca. 300 ml of plant nutrient

solution 96 hrs post-transplant. Each pot was fitted with a


























































Figure 1.


Field arrangement of the caged pots containing
Callie giant bermudagrass, C. dactylon, used
to monitor the vertical movements of M. latipes
larvae. Screen enclosures are positioned to pro-
vide 44 cm of vertical free space for larval
movement.










20 cm x 48 cm cylindrical cage; these were constructed by

bending a 48 cm x 70 cm rectangle of 16-mesh aluminum screen

into a crude cylinder and stapling the 70 cm edges. A

spring-steel paper clamp fastened the base of a cage to the

rim of a pot. The top of each cage was fitted with a 20 cm

diameter plastic potting saucer held in place by plastic

clothespins. All cages were adjusted on their pots to give

a uniform distance of 44 cm from the soil surface to the

inner surface of the cage top. The caged pots were arranged

in a single row along a roto-tilled alleyway adjacent to a

University of Florida Campus Agronomy Farm field planted to

Callie giant bermudagrass (Figure 1).

During the night of 20 October 1983, a single, field-

collected, 5th or 6th instar M. latipes larva was placed on

the foliage in each pot. Beginning on 21 October, the

position of these larvae was determined by the following

procedure: a caged pot was carefully lifted from its site in

the alleyway and placed on a nearby table, the vegetation

examined until the larva was found, the distance from the

surface of the potting soil to the headcapsule of the larva

was measured (to the nearest centimeter) by reading an

upheld meterstick positioned along the screen, the pot was

returned to its field site, and the process repeated until

all 30 pots had been examined.

Every pot was examined 8 times each night; measurements

of larval position in each pot were initiated at 17:00 hrs,

18:00 hrs, 21:00 hrs, 23:00 hrs, 03:00 hrs, 06:00 hrs, 07:00








31

hrs and 08:00 hrs. Pot examination sequence was systematic

(i.e., fixed order, based on field position).

In a few instances, a final instar larva pupated while

in a pot; these were replaced. Larval replacement was done

in a manner that minimized its impact on data integrity. On

25 October the larva in each pot was replaced with either a

3rd or 4th instar larva, and during 26-30 October the same

procedures and measurements that were applied to the 5th and

6th instar larvae (as described above) were used to record

the behavior of the 3rd and 4th instar larvae.

Hourly measurements of temperature, rainfall, and

relative humidity during the experimental interval were

obtained from the Agronomy Farm's weather data facility.

Data were evaluated for differences in larval position

per time of night, and for differences between larvae of the

two different age groups. All tests for significance and

confidence interval constructions were at the (a = 0.05)

level.

Evaluation of a Grassland Sampling Device

A two-wheeled, manual, pushcart sampler capable of a 50

cm swath was constructed by suspending an open-fronted,

aluminum-framed, screen cage from a crossmember-type, steel

chassis (Figure 2). A complementing series of 3 slots,

spaced at 5 cm intervals, were milled into the leading

margin of each wheel stanchion; when fitted with the axle of

a 68.6 cm diameter bicycle wheel, these slots permitted

adjustment of the cage floor to heights of 5, 10 or 15 cm








32

above the soil surface (Figure 3). The forward edge of the

cage floor was fitted with a removable, multiple-tined,

forward-projecting rake head; tine spacing could be adjusted

by either removing or installing individual tines at 5 cm

intervals along the head's length (Figure 3).

The device was taken to the University of Florida's

Campus Agronomy Farm and a series of tests was conducted to

evaluate machine performance, sturdiness, and the sampling

methodology that might best suit to sample M. latipes larvae

in Callie giant bermudagrass, C. dactylon.

These tests entailed pushing the device through 25 m of

a given grass, noting general machine response, and then

inspecting the contents of the cage. Notes were made on the

outcome of trials for all 3 of the cage's height settings,

both the presence and absence of the rake head, and both 5

cm and 10 cm tine spacings. Larvae captured by the device

were watched in situ, and notes were made on their initial

location, response to the cage, and tendency to escape.

The efficacy of this device as an absolute density

estimator of M. latipes larvae in Callie giant bermudagrass

was evaluated by a mark-release-recapture study. Ten day

old M. latipes larvae that had been greenhouse-reared on

Callie giant bermudagrass were marked with fluorescent

powder and released in 5 field plots of 24 day old Callie

giant bermudagrass. Each plot was 0.5 m wide and 2.0 m

long. At 22:00 hrs on the nights of 8 October through 11

October 1984, 30 marked larvae were released per plot. On





























































Figure 2. A pushcart device for sampling M. latipes larvae
in forage grasses.








































46 4 7 4a '419 '5 0 --' 5 2. 1 SY5
06 1ifflnwmwnww


Figure 3.


Adjustable features of the pushcart sampler.
a) Wheel stanchion with slots milled at 5 cm
intervals; when set as shown, sampler's ground
clearance is 5 cm.
b) Removeable rake head with tine spacing set
at 5 cm; individual tines are threaded to permit
spacing adjustments.








35

each of the 4 release nights, the larvae were marked with a

different color (either Day-Glo pigment no. 15, 16, 18, or

21) and left to move freely through the foliage until the

following 07:00 hrs, whereupon each plot was sampled by

making a single pass with the sampling device and

subsequently counting the number of marked larvae found in

the cage. M.iarked larvae were detected with the aid of a

hand held, 12 volt, 366 nm longwave ultraviolet lamp. After

having been counted, the larvae were removed from the cage

and returned to the plot from which they had been captured.

The plots were left undisturbed until the following 20:00

hrs. at which time the sampling process was repeated. To

account for the influences of larval mortality, emigration,

or loss of markings due to molting; at ca. 30 minutes prior

to both the 07:00 hr and the 20:00 hr sampling sessions, all

of the plots were examined with the ultraviolet lamp and the

number of marked larvae seen in each plot was recorded. All

samples were taken with the chassis set for a cage-floor-to-

ground clearance of 5 cm and with the foliage rake removed.

Data were analyzed for differences in the number of

larvae captured per plot, for differences in the number of

larvae captured per trial, and for differences in the number

of larvae captured at each time of night. All tests for

significance and confidence interval constructions were at

the (a = 0.05) level.










Laboratory Experiments

Chorion Consumption

Between 20:00 hrs and 08:00 hrs on the night of 12

October 1984, a single field-collected moth deposited 86 ova

on the wall of a ca. 400 ml plastic cup and then died. The

dead moth was removed and the ova were incubated in situ at

230 10 C. Eclosion commenced at 21:00 hrs on 16 October

and continued until 03:00 hrs on 17 October, by which time a

total of 82 larvae had emerged (the remaining 4 ova never

hatched). Newly emerged larvae were removed only after they

had crawled to the cup's lid (upon which no ova had been

deposited) or were seen dangling from a silken thread that

they had spun from the lid. After all larvae had been

removed, the cup was carefully cut into several sections and

each was viewed with a microscope.

The post-eclosion chorion remains of 10 randomly

selected M. latipes ova were examined at 45X magnification

with a binocular dissecting microscope. This was replicated

8 times. Each ovum in every replicate was examined twice

and on both occasions was accorded a scalar value that

corresponded to the proportion of the chorion that had been

consumed by an emerging larva. These values ranged from 1

to 5; where, a value of 1 indicated that the spheroidal

portion of a post-eclosion chorion was intact except for a

circular emergence hole that constituted no more than one-

sixth of the sphere's surface; the values 2, 3, and 4

indicated that the emerging larva had consumed at least one-










third, one-half, or two-thirds of the spheroidal portion,

respectively; and the value 5 was assigned in those cases

where essentially all of the chorion excepting the flattened

base (cemented to the ovipositional surface) had been eaten

by the emerging larva.

Data were analyzed for differences in the average

amount of chorion consumed per replicate and for differences

in the relative proportions of the various consumption

categories. All tests for significance were conducted at

the (a = 0.05) level.

Early-Instar Feeuing Surface Preference

Primary stems of Callie giant bermudagrass, Cynodon

dactylon, were collected from the field, sealed in plastic

bags, and transported to the laboratory. To avoid excessive

wilting, all plant material was collected at 22:00 hrs. In

the laboratory, a stem with undamaged terminal growth was

cut in the vicinity of its penultimate node and the basal

portion discarded. The terminal portion was placed in a 25

cm x 150 cm plastic petri dish that had been fitted with a

size 14, Hercules C-5, clarifying filter. When the plant

material was placed in the petri dish, the filter disc was

saturated with ca. 10 ml of tap water. Particular care was

taken to position the grass foliage on the filter disc such

that both the upper and lower surfaces of the blades were

not in contact with either the filter disc or the petri dish

cover ana blade contact with the petri dish wall was

minimized. A total of 6 such dishes were set up.










With the aid of an artist's brush, 10 first instar M.

latipes larvae (ca. 3 hrs old) were placed on each filter

disc. The petri dishes were covered and the larvae observed

until they moved onto the foliage. All dishes were kept at

230 10 C with a 13L:11D photoperiod.

The larvae in each dish were viewed at ca. 12 hour

intervals; and when these began to exhibit signs of having

entered the first pre-molt stage, the number of hours

elapsed since petri dish introduction was recorded and the

larvae were transferred en masse to a second petri dish that

had been prepared as described above. The plant material

upon which they had fed as first instars was retained for

microscope examination.

The upper and lower surface of each grass blade was

inspected at 15X magnification with a binocular, dissecting

microscope and the number of feeding scars found on each

surface was recorded. Records were kept dish-specific. In

due process, the entire procedure was repeated for the 6

petri dishes of resultant second instar larvae.

Data were analyzed for differences in the number of

feeding scars on upper versus lower surfaces of blades, for

differences in the average number of feeding abrasions per

petri dish, and differences in the number of feeding

abrasions produced by first instar larvae versus second

instar larvae. All tests for significance were conducted at

the (a = 0.05) level.










Stage-Specific Developmental Time

Thirty-eight M. latipes ova that had been deposited

between 20:00 hrs and 08:00 hrs on the night of 29 September

1984 by a field-captured moth were incubated in situ at 230

20 C and monitored for their developmental progress.

Eclosion was observed and the time recorded.

The 38 larvae were reared in isolation. Each larva was

placed in a 25 cm x 150 cm plastic petri dish that had been

fitted with a tap-water saturated, size 14, Hercules C-5,

clarifying filter disc. Every larva was supplied ad libitum

quantities of freshly-cut Callie giant bermudagrass, Cynodon

dacytlon, foliage. Frass and plant debris were removed from

petri dishes as needed. Rearing conditions were kept fixed

at 230 20 C with a 13L:11D photoperiod until all larvae

had pupated.

All dishes were inspected at ca. 12 hour intervals (at

noon and midnight 3-4 hrs) and the developmental progress

of each larva was noted and recorded.

Pupal development was monitored in a similar fashion.

Pupae were left undisturbed until cuticular pruinescence was

observed; whereupon, they were removed from their cocoons,

sexed, and weighed. Each pupa was placed in a 30 ml plastic

cup and held for adult emergence.

Adults were killed and spread for voucher specimens;

these were placed in the arthropod collection housed by the

Florida Division of Plant Industry in Gainesville.








40

Data were analyzed for determination of the mean

developmental time for each stage of the life cycle and mean

pupal masses; the latter were determined gender-specifically.

All significance testing was conducted at the (a = 0.05)

level.
















RESULTS AND DISCUSSION


Field ExDeriments

Moth Attraction to Different Host Grasses

Rain fell only once (1.1 mm during 18:50 19:45 hrs on

21 September); mean hourly temperature and relative humidity

were 20.20 0.6' C and 86% 2%, respectively.

In 320 observations of the control receptacles, no

adult M. latipes were seen resting on the poultry wire.

These data were excluded from the remaining analyses.

A chi-square test of the total counts for moth visita-

tion, on a per grass basis, indicated that the visitation

ratio was significantly (X2 = 274.88; df = 2) different than

that hypothesized (Ho: ratio of visitation to the three

grasses is 1:1:1). When the same type of test was applied

to overall counts of moth abundance across nights, the

resultant test statistic proved too low (X2 = 1.33 ; df = 3)

to reject a hypothesized homogeneity of nightly visitations.

Thus the incidence of moth visitation was significantly

different for each of the three grass species, but was more

or less uniform from night to night.

It seems likely the short study period, stable weather

conditions, and single test site reduced experimental vari-

ability and increased the across-night homogeneity of moth










behavior. Undoubtedly, test-grass condition and moth pop-

ulation intensity were key elements as well. Quantita-

tively, both were unknowns; however, assignment of ten culms

per sample unit probably compensated for variations in the

former, and tne latter was of damaging (i.e., sufficient to

merit an application of methomyl to the surrounding research

fields on 26 September) proportions.

In 960 viewings of the receptacles containing the three

grasses, 197 adult M. latipes were seen resting on 153

different sample units. Of these, 175 moths on P. notatum

racemes (135 sample units); 4 moths on D. sanguinalis

racemes (4 sample units); and, 18 moths rested on C.

dactylon culms (14 sample units). Thus, in an overall

context, the number of moths per sample unit was sharply

skewed (i.e., evinced a comparatively high frequency of zero

counts).

Evaluation of test statistic U and its accompanying

standard error [SE(U)], as applicable to count data evincing

high frequencies in the zero class (Elliott 1977), revealed

the negative binomial frequency distribution could not be

rejected (a = 0.3849; U = -0.01038; SE(U) = 0.01079) as a

defining function for these data. Parameter estimation (by

iterative solution when applicable) subsequently yielded:

y = 0.2053, s2 = 0.2738, and k = 0.5511, for the

distribution's mean, sample variance, and descriptive

constant, respectively. Three points merit mention. First,

since the interval delimited by the absolute value of SE(U)










contains the value obtained for U, fit of the indicated

distribution function is accepted at an error level slightly

less than a = 0.05. Second, the value obtained for the

constant (k) lies between zero and one; an indication that

contagion characterizes the count data (hence, the Poisson

function would be a poor choice). Third, the value obtained

for the distribution's mean (y) suggests the "overall

success rate" for observing a moth visiting a sample unit

was "fairly good" (i.e., one-fifth moths per sample unit)

notwithstanding the dilution that resulted from including

the comparatively "unattractive" D. sanguinalis sample

units. The principal utility of such a finding is simple:

the data warrant further, and parametrical, analysis.

Application of Taylor's Power Law to these data, via

the methodology set forth by Elliott (1977), revealed the

mean's dependence upon the variance was adequately described

by: 02 = 2026.0493p5.5934. Subsequent employment of this

expression, coupled with consistent integer elevation of all

original data points, yielded: y (y + l)-l179672 as a

transformational expression capable of rendering the mean

independent of the variance, thus establishing a valid basis

for applying parametrical statistical procedures.

An analysis of variance revealed that three factors

(grass species, time of night, and grass x time interaction)

contributed significantly to the overall experimental vari-

ability Moreover, both the effect due to whole unit blocks

and the subunit block-time interaction were insignificant

















Table 1. ANOVA of transformed counts of adult M. latipes
visitation to culms of P. notatum, C. dactylon,
and D. sanquinalis at eight different times of
night.



Source SS df MS Fcalc


Blocks 2.49035 39 0.06394 0.8338

Grasses 18.72331 2 9.36166 122.0715*

Errora 5.98168 78 0.07669

---------------------------------------------------------
Whole Unit 27.19534 119
---------------------------------------------------------

Times 4.34238 7 0.62034 14.5757*

Grass x Time 5.33850 14 0.38132 8.9596*

Block x Time 12.86814 273 0.04713 1.1075

Errorb 23.23805 546 0.04256



Total 72.98241 959



a Counts transformed by: y (y + l)-l179672.

* Values followed by asterisk (*) significant at the
a = 0.05 level.








45

(Table 1). In combination, an insignificant F-statistic for

both the whole unit blocks and the subunit block-time inter-

action indicates that the overall field positioning of the

receptacles had a negligible influence on the observed

variability in moth visitation. More particularly, an in-

significant effect for whole unit blocks is an indication

that the moths displayed no inclination to rest on the

sample units in one block over those in another; whereas,

insignificance of the subunit block-time interaction in-

dicates that the number of moth visitations recorded within

a block at a specific time did not significantly affect the

number of moth visitations recorded for the same block at

either the preceding or following times. This is a critical

point, for if otherwise had been the case, the effect that

time of night has on moth visitation to the three grasses

could not have been evaluated via these data.

Comparison of the mean number of moth visitations per

grass indicate that more moths chose to rest on racemes of

P. notatum, and that moth visitations to D. sanguinalis and

C. dactylon culms were statistically equivalent (Table 2).

These data highlight two notable issues. First, (recalling

no moths landed on the poultry wire controls) it seems clear

that M. latipes adults select their resting sites discrimi-

nately. Second, it seems unlikely that these moths cued

mainly on plant profile, for, a ten culm cluster of P. nota-

tum physically resembles a comparable-sized cluster of D.

sancuinalis culms, yet, the evidence indicates the moths
















Table 2. Overall mean abundancea of adult M. latipes per
10-culm cluster of P. notatuim, D. sanquinalis, or
C. dactylon assayed under field conditions.


Grass Species


P. notatum


C. dactylon


D. sanguinalis


Pleanb, c


(0.236) 0.684


(0.019) 0.967


(0.005) 0.991


95% Confidenced
Interval


0.726 0.641


1.010 0.924


1.034 0.948


a Means are each of n=320 observations, with these the
result of moth counts taken hourly from 19:00 until
03:00 on four different nights.


b Values enclosed by parentheses are detransformed point
estimates.


c Determined from count data transformed as follows:

y (y + 1)-1.79672

d Confidence intervals are constructed about transformed
means.










overwhelmingly opted to visit and rest on culms of the

former rather than those of the latter.

At present, P. notatum is the dominant vegetation of at

least 2.5 million acres of Florida pasture (Johnson et al.

1975), and covers an undocumented (but large) amount of road

right-of-way, airport grounds, playing fields, parks,

scnoolyards, and lawns. In their investigations of adult

velvetDean caterpillar (VBC), Anticarsia aemmatalis Hubner,

Greene et al. (1973) reported finding VBC adults resting on

P. notatum racemes. When these two facts are coupled with

the evidence presented above, a scenario with potentially

profound implications emerges: the acreage sustaining this

grass may well be a reservoir of a substantially attractive

plant species that is tapped by two economically injurious

noctuias. Thus, the presence of P. notatum throughout the

state may be facilitating the migratory behaviors of agri-

culturally important moth species.

M. latipes' response to the sample units was influenced

by time (Table 1). Hartstack et al. (1979) have noted that

sampling and monitoring conducted during peak moth activity

periods may be the superior way to forecast both larval

populations and crop damage. Such efforts presuppose time-

specific Knowledge of moth behavior. Time-specific differ-

tials in adult M_,. latipes abundance per sample unit on each

of the three grasses are presented in Table 3.

Moths were simultaneously present on all three grasses

only during 20:00 hrs (Table 3). In light of the extremely















Table 3. Time-specific abundance of a field population of
adult M. latipes on 10-culm clusters of either P.
notatum, C. dactylon, or D. sanginalis.


P. notatum

Mean Std.Err.

0.827 0.032

0.531 0.045

0.470 0.035

0.543 0.042

0.530 0.037

0.765 0.033

0.907 0.023

0.889 0.024


C. dactylon

Mean Std.Err.

0.982 0.032

0.904 0.045

0.978 0.035

0.907 0.042

0.964 0.037

1.000 0.033

1.000 0.023

1.000 0.024


D. sanguinalis

Mean Std.Err.

1.000 0.032

0.947 0.045

1.000 0.035

1.000 0.042

1.000 0.037

0.982 0.033

1.000 0.023

1.000 0.024


Means are of n=40 observations per grass per time.

Values reported are those obtained from analysis after
transformation of original moth counts.

Transformation given as: y (y + l)-1.79672.


Hour

19:00

20:00

21:00

22:00

23:00

24:00

01:00

02:00










low number of moth visitations to the D. sanguinalis culms,

such a finding could either be the result of coincidence or

could be indicative of some as yet poorly understood aspect

of this moth's behavior. Although the evidence is insuf-

ficient for definitive judgement, the latter interpretation

seems somewhat more likely since moth abundance on the two

minimally visited grasses (i.e., C. dactylon and D. san-

guinalis) evinced similar trends between 19:00 hrs and

21:00 hrs, while moth abundance steadily increased on the P.

notatum during the same interval (Table 3). Undoubtedly,

these time-specific shifts were instrumental in rendering

the subplot interaction term (Table 1) significant.

The time specificity of M. latipes' behavior is better

seen when examination of the data presented in Table 3 is

focused on moth abundance per time on P. notatum. At 19:00

hrs, moths were present in low numbers. Moth population

intensity increased dramatically during the ensuing hour,

and by 20:00 hrs, had essentially peaked. This condition

remained substantially unchanged (i.e., the means were not

significantly different) for the next three hours. At 24:00

hrs, moth abundance per sample unit had declined, and by

01:00 hrs fewer moths were seen resting on P. notatum culms

than had been present during the 19:00 hr count. This con-

dition (very few moths) aptly describes the remainder of the

assay period. Thus, moth activity on P. notatum evinced a

cyclical (unimodal and platykurtic) character nightly during

the period from 19:00 hrs to 01:00 hrs (Table 3).








50

The dramatic increase in the number of moths seen rest-

ing per sample unit of P. notatum that occurred during the

19:00 to 20:00 hr interval has a utilitarian implication:

this time frame will likely prove a poor period to undertake

quantitative assessments of M. latipes populations, for it

combines a relatively short time interval with an expecta-

tion of high sample variance; these two features antagonize

most sampling schemes. Conversely, moth counts taken during

the interval from 20:00 hrs through 23:00 hrs indicate this

time frame was one in which the moth population was both

comparatively high and essentially stable. The implication

here is that this may prove the better time to estimate

adult populations of M. latipes in fields planted to P.

notatum.

None of the M. latipes adults counted during the exper-

imental period were seen to either copulate or oviposit.

Visual inspection of those sites chosen by moths who rested

on sample units of either C. dactylon or D. sanguinalis

yielded no tangible evidence to link the moths' presence to

either of these two hosts. In contrast, in virtually every

instance that a moth was seen resting on a sample unit of P.

notatum, the moth was clinging to the grass inflorescence

racemee), had its proboscis extended, and was probing the

raceme's surface and crevices. In most of the instances

that this activity was observed, the region of the raceme

proved by the moth glistened with smears or droplets of a

clear liquid (Figure 4).
























































Figure 4. Male Mocis latipes (Guen6e) resting on a
Paspalum notatum Flugge raceme. The distal
aspect of the moth's proboscis probes a clear
liquid coating a localized region of the raceme.
The dark blotches on the raceme are floral
parts. The moth in the background is Anticarsia
gemmatalis Hubner.








52

Karr (1976) observed noctuids congregating on roadside

stands of Paspalum virgatum L. growing near Fort Clayton,

Canal Zone. In describing the grass, he stated, "Four of

the inflorescences had open spikelets with a sticky, almost

mucous, and sweet-tasting fluid ." and that, ". by

19:15 moths were being attracted to the inflorescences with

open spikelets ." (Karr 1976, p. 284). Similiarly,

while investigating a stand of P. virgatum growing along a

canal near Cardenas, Mexico; Lloyd (1981, 1984) noted that

predator fireflies and their prey fed on a sticky substance

coating the grass seed heads. The descriptions put forward

by these researchers closely parallel the evidence offered

above, and in large part, serve to corroborate the suspicion

that the behavioral link between the observed high degree of

visitation by adult M. latipes to racemes of P. notatum is

that the moths are utilizing a plant-derived substance as a

food source.

Moth Activity in Callie Giant Bermudagrass

The magnitudes of the F-statistics obtained via an

analysis of variance of the moth count data indicate that

grass age and time of moth counts were the principal sources

of variability in moth abundance (Table 4). Hence, there

were more M. latipes adults present in plots containing

grass of one age than in comparable-sized plots containing a

different age of grass; moreover, more moths were seen at

one time of night than another. This implies that moth

behavior is modified both by host plant age and by short















Table 4. Summary of three-way, split-plot ANOVA of adult
M. latipes abundance in three different ages of
Callie giant bermudagrass, C. dactylon, at five
different times of night.




Source df Fcalc Ftab


Nightsa 7 6.84 2.09

Grass Ageb 2 167.78 3.07

Night x Age 14 2.54 1.75

Errora 120


Timec 4 115.32 2.37

Night x Time 28 7.94 1.48

Age x Time 8 21.14 1.94

Night x Age x Time 56 2.44 1.33

Errorb 480




a Defined as the scotophasic interval commencing at 18:00
hrs and ending at 07:00 hrs; moth counts were made
during this interval for the period 6 September through
13 September 1983, and 14 September to 15 September
1983.
b Above-ground plant growth in a given plot on 6 September
1983 was either A=15, B=25, or C=35 days old.
c Moth counts commenced at 18:00 hrs, 21:00 hrs, 01:00
hrs, 04:00 hrs and 07:00 hrs.








54

term, temporal change. Such information is quite desirable;

for, when elucidated with specifics regarding grass age and

time of night, it provides a framework for an adult moth

monitoring program that is based upon characteristics of

the moths' behavior.

Further examination of these data (Table 4) shows that

a straightforward specification of the'differences in moth

abundance due to effects of either grass age or the time of

night (partitioned per age or per time) is not forthcoming

via paired comparisons between levels within factors, for

factors other than these two also made significant contri-

butions to experimental variability. Moreover, in all cases

but one (the effect of nights), these are interaction terms;

hence, the relationships between main effects (i.e., nights,

grass age, and time) are such that experiment-wide compari-

sons utilizing a common variance prove nonorthogonal. Thus

the value of the ANOVA is twofold: all three of the main ef-

fects merit further scrutiny, and the analysis of each will

be better rendered through partition.

During the period for which moth activity in Callie

giant bermudagrass was measured, a total of 2270 individual

moth sightings were recorded. With respect to nights, this

number yields 283.75 64.78 as the 95% confidence interval

(df = 7) about the expected value for moth abundance per

night. Only one night's moth count (the data obtained for

the night of 14-15 September, 1983) is not contained by this

interval (Fig. 5). Undoubtedly, retention of this night's
















380 r-


369


I-
0


L. 300
0






220


-J

I-
0
I-


308





















I7<


339 341












A"'///,


231


\J f 1 f J *'


1 2 3 4 5 6 7


314


8 9


NIGHT


Figure 5. Histogram of adult M. latipes abundance per
night during 6 September to 15 September 1983
in a field of Callie giant bermudagrass.



a Determined by counting the number of moths seen on the
foliage in 18 different 4 m x 12 m plots of grass; moth
counts in each plot were repeated at 5 different times
(18:00, 21:00, 01:00, 04:00, and 07:00 hrs) each night.


138

DATA


140h-


230













I<


' ff -f / u -#jJ' f a i J1 if --'









moth count in the overall data set substantially affected

the effect due to nights. Conversely, its exclusion (which

would make counts per night more homogeneous) would reduce

the number of degrees of freedom available for subsequent

analysis.

Total moth abundance per night was at one of three

levels; moreover, these were juxtaposed in a decidedly

non-random configuration (Fig. 5). At the onset of data

collection the nightly count was 230 moths (level 2),

followed by a sharp increase that remained essentially

stable for the five ensuing nights (level 3), and thereafter

declined (at a rate virtually identical to that of the

original increase) to a low of 138 moths (level 1) for the

ninth night. Thus the heterogeneity detected by the ANOVA

was of an orderly sort (i.e., more or less normally

distributed about a sustained peak), and is perhaps best

explained as being a natural consequence of emigration,

immigration, natality (in the sense of moths being "born"

from mature pupae), and mortality forces in a field popula-

tion of M. latipes adults.

Every night, significantly more moths were present in

the plots containing the oldest grass than in those having

either of the other two grass ages (Fig. 6); however, moth

presence in the intermediate age grass compared more

favorably with that noted for the oldest grass than it did

for the moth presence detected in the youngest grass (moth

presence in this last was near-inconsequential). These


















0.601-


0.50


6 ----C


I 9


9
a


I.'


0.30F-


0.10o


0.00


I I ~ ~ I I I I


1 2


3 4 5 6 7 8 9


NIGHT








Figure 6. Least-squares regression functions describing
changes in adult M. latipes abundance in plots
containing different ages of Callie giant ber-
mudagrass, C. dactylon, during the period from
6 September to 15 September, 1983. Functions A,
B, and C correspond to moth populations in grass
plots aged 15 to 24, 26 to 35, and 35 to 44 days,
respectively.


O I m


_-.D


m I m m m N n m


v


0.40[-










differences are the reason that the F-statistic for the

effect due to grass age (Table 4) proved significant.

However, neither the moth population per night nor the pro-

portion of moths present per grass age remained static for

the experimental period; a fact indicated by the signifi-

cant value of the night x age F-statistic (Table 4). The

data presented in Figure 6 illustrate this phenomenon.

In those plots containing the oldest grass (age C), the

proportion of the total moths present each night declined in

a curvilinear (Y = 0.6161 0.0443 Loge X) manner; but the

proportional presence of moths in the youngest grass (age A)

evinced a curvilinear (Y = 0.0023 + e0.4302 X) increase

through time. Moth presence in the intermediate aged grass

(age B) remained essentially unchanged (Y = 0.4078 + X) from

night to night (Fig. 6).

The mechanism underlying the observed differences in

moth presence in the different ages of grass is a conse-

quence of related biological phenomena acting in concert.

Several points support this position. First, since all of

the plots were in the same field and in close proximity

(neighboring plots were separated by a 75 cm alleyway), dif-

ferences in moth presence due to differential accessibility

of plots seem unlikely. Second, the plant growth in plots

of different ages was substantially different. The plots

containing the youngest grass (age A) had very few primary

stems higher than ca. 25 cm, and in consequence, lacked a

closed plant canopy (i.e., the soil surface was readily








59

visible). In contrast, plots containing either the interme-

diate or the oldest grass (ages B and C, respectively) had

primary growth at least 45 cm high and well-closed canopies

(i.e., less than 5% of the soil surface readily visible).

Third, to successfully produce progeny, an adult moth must

display some measure of gregariousness. In result, the

presence of one moth in a given plot enhances the likeli-

hood of observing a second nearby. Concatenation of these

points yields a scenario that helps to explain the observed

differential in moth presence set forth in Figure 6.

By having more above-ground plant growth, the plots

with the two older grass stands provided the moths with a

greater diversity of resting sites, ovipositional sites and

sneltered areas; and, moths that entered these plots in

response to these physical conditions rendered them more

attractive to sibling moths, particularly those seeking

mates.

Within nights, moth abundance per time was decidedly

non-uniform. Regardless of grass plot age, more moths were

present at 21:00 hrs; this was the only time of night that

moth presence in the youngest grass was statistically dif-

ferent from zero (Table 5). The time of night determined as

the period of peak abundance also proved to be both the only

time of night that M. latipes adults were observed ovipos-

iting (Table 6) and the earliest time of night that in copu-

lo pairs were observed in plots containing the two older

grass ages (Table 7).
















Table 5. Summary statistics of adult M. latipes abundancea
in three different ages of Callie giant bermuda-
grass, C. dactylon, at five different times of
night.


Grass Aqeb


A

Mean Std.Err.

0.13 0.087

0.83 0.172

0.10 0.104

0.10 0.104

0.04 0.042


B

Mean Std.Err.

3.75 0.539

7.63 0.643

4.38 0.459

2.29 0.328

1.60 0.226


C

Mean Std.Err.

4.69 0.488

9.19 0.543

6.40 0.557

4.00 0.453

2.17 0.363


a Determined
foliage in


by counting the number of moths seen on the
4 m x 12 m plots of grass.


b Grass ages A, B, and C correspond to above-ground grass
plant regrowth aged 15 to 24, 26 to 35, and 35 to 44 days,
respectively.

Means and their accompanying standard errors are each of
n=48 observations.


Time

18:00

21:00

01:00

04:00

07:00
















Table 6. Summary statistics of the proportions of an
adult M. latipes population observed ovipositing
in Callie giant bermudagrass, C. dactylon, plots
of two different ages at five different times of
night.


Grass Age B


(26 to 35 days of regrowth)


Std. Err.


0.000

0.024

0.000

0.000

0.000


Precisionb


10.43


Grass Age C


(35 to 44 days of regrowth)


Std. Err.


0.000

0.018

0.000

0.000

0.000


Precisionb


8.57


a The number of ovipositing moths seen in each of
forty-eight plots divided by the total number of moths
seen per plot. Each plot measured 4 m x 12 m.
b (sy/y) x 100%.


Time

18:00

21:00

01:00

04:00

07:00


Mean

0.00

0.23

0.00

0.00

0.00


Time

18:00

21:00

01:00

04:00

07:00


Mean

0.00

0.21

0.00

0.00

0.00

















Table 7. Summary statistics of the proportionsa of an
adult M. latipes population observed in copula in
Callie giant bermudagrass, C. dactylon, plots of
two different ages at five different times of
night.


Grass Age B


(26 to 35 days of regrowth)


Std. Err.


0.000

0.017

0.041

0.052

0.000


Precisionb


24.29

13.23

26.00


Grass Age C (35 to 44 days of regrowth)


Std. Err.


0.000

0.030

0.047

0.039

0.000


Precisionb


15.79

20.43

24.38


a The number of in copula moths seen in each of
forty-eight plots divided by the total number of moths
seen per plot. Each plot measured 4 m x 12 m.

b (sy/2) x 100%.


Time

18:00

21:00

01:00

04:00

07:00


Mean

0.00

0.07

0.31

0.20

0.00


Time

18:00

21:00

01:00

04:00

07:00


Mean

0.00

0.19

0.23

0.16

0.00










Viewed overall, the time specificity evident in these

data (moth abundance per time of night, ovipositional activ-

ity, and incidence of mating) strengthens the interpretation

put forward earlier: manifest differences in moth abundance

in plots of different ages is substantially influenced by

factors that are consequences of moth presence per se.

When the data for ovipositional activity are considered sep-

arately, the role of the host comes into sharper focus:

moths oviposited only in the plots containing either of the

two older grasses; moreover, the proportion of moths per

plot doing so were statistically identical for the two grass

ages (Table 6). Apparently, when given the choice between

ovipositing on relatively new growth (that lacks a closed

canopy) and older growth (that affords more cover), M.

latipes females will overwhelmingly opt for the latter. Al-

though it will likely be a rare event that a bermudagrass

field in a production situation will have more than one age

of grass in it at any given time, the information put for-

ward here is valuable in two regards. First, untended grass

stands adjacent to production areas that lack closed canopy

may well possess attributes conducive to moth population

buildup (e.g., be old enough to have closed canopy), with

subsequent encroachment of the production site a likely

short-term option. Second, field scouting/sampling activi-

ties conducted in fields whose above ground plant growth is

less than 3 weeks old will likely be of marginal value.



























































Figure 7. M. latipes mating in Callie giant bermudagrass,
C. dactylon. Orientation is female superior.








65

In contrast to the pointedly time specific character of

ovipositional activity, M. latipes adults were observed in

copulo over a period of several hours each night (Table 7).

Unlike ovipositional activity (where the female is in near-

continuous motion, flying and hovering in the interstices of

the middle and lower portions of the plant canopy, pausing

only momentarily to deposit an egg (most often individually,

but occasionally two) on a grass stem, sheath, ligule, or

blade), copulation is a sessile act usually accomplished in

the upper one-third of the grass canopy (Figure 7). In re-

sult, mating; moths are easily detected by visual inspection.

This raises two points that may ultimately prove of value in

a monitoring program. First, sessile character and high

visibility are desirable attributes if moth monitoring is to

be accomplished by visual inspection; assuming this method

is chosen, counting only the number of in copulo pairs per

unit area may yield population estimates that are less vari-

able than those obtained when all moths per unit area are

counted. Second, these data (Table 7) indicate that after

21:00 hrs, a substantial portion of the moth population is

engaged in mating; such moths are likely to be relatively

unresponsive to pheromone trapping. Hence, if pheromone

baited traps are to be the basis of a monitoring program,

counts taken after 21:00 hrs will probably yield data that

underestimate the actual populational intensity.










Vertical Movement of Larvae

Analysis of the data obtained for fifth and sixth in-

star larvae revealed that both a significant effect due to

nights and a significant night x time interaction were the

consequences of variation induced by the inclusion of

measurements taken on the night 24-25 October, and that an

evaluation of the effect of time of night on larval movement

could be better evaluated when based on the 480 measurements

taken during the nights of 21-22 and 22-23 October (Table

8). Similar analysis of the data obtained for the third

and fourth instar larvae yielded comparable findings (Table

9). In result, further analysis for each of the two data

sets was applied to those measurements obtained during the

nights for which homogeneity (with respect to nights) could

not be rejected (i.e., n=480 per larval age group).

Fifth and sixth instar M. latipes larvae occupied

different portions of the grass canopy at different times of

night (Table 10). Measurements taken at 17:00 hrs indicate

that these larvae rest (oriented such that the headcapsule

is directed upward) on the basal portions of primary stems.

During the following hour, the larvae initiate an upward

movement; and by 21:00 hrs, have moved to the canopy's mid-

dle stratum (ca. 21 cm above the soil surface) and remain

in this location for the ensuing next nine hours. Measure-

ments of larval position at 07:00 hrs revealed this time to

be a period of flux; and that by 08:00 hrs, the larvae had

returned to the same portion of the grass canopy that they
















Table 8. Summary of two-way, split-plot ANOVA of the time-
specific vertical movements of fifth and sixth
instar M. latipes larvae in Callie giant bermuda-
grass, C. dactylon.



Source df Fcalc Ftab


Nights 2 3.265 3.11

Errora 87

Times 7 96.937 2.76

Night x Time 14 5.675 1.71

Errorb 609





Source df Fcalc Ftab


Nightstt 1 0.013 4.00

Errora 58

Times 7 69.355 2.01

Night x Time 7 0.657 2.01

Errorb 406



t Scotophase beginning 21, 23 and 24 October, 1983.

Scotophase beginning 21 and 23 October, 1983.

Hours at which measurements were commenced (17:00,
18:00, 21:00, 23:00, 03:00, 06:00, 07:00, and 08:00).
















Table 9. Summary of two-way, split-plot ANOVA of the time-
specific vertical movements of third and fourth
instar M. latipes larvae in Callie giant bermuda-
grass, C. dactylon.



Source df Fcalc Ftab


Nights 2 0.889 3.11

Errora 87

Times* 7 241.498 2.76

Night x Time 14 1.968 1.71

Errorb 609





Source df Fcalc Ftab


Nightstt 1 0.180 4.00

Errora 58

Times* 7 134.095 2.01

Night x Time 7 1.190 2.01

Errorb 406



t Scotophase beginning 26, 27 and 28 October, 1983.

9t Scotophase beginning 27 and 28 October, 1983.

Hours at which measurements were commenced (17:00,
18:00, 21:00, 23:00, 03:00, 06:00, 07:00, and 08:00).










had occupied on the previous evening at 17:00 hrs. Thus,

vertical movements of fifth and sixth instar larvae evinced

substantial unimodal, curvilinear, character per scotophase

(Table 10).

Regression analysis (larval height as a function of

time of night) of these data revealed that a second-order

polynomial, Y = -203.921 + 0.184X 3.714x10-5X2, could not

be rejected as an appropriate (tcalc for 80, 1, and 82 was

-5.69, 6.08, and -6.05, respectively; Fcalc = 18.484, df =

2,5; R2 = 0.881) deterministic model for the vertical move-

ments of fifth and sixth instar M. latipes in Callie giant

bermtudagrass (Fig. 8).

Like their older counterparts, third and fourth instar

M. latipes larvae occupied different portions of the grass

canopy at different times of night. However, the vertical

movements of these larvae, although similar in kind (i.e.,

upward movement in the early hours of scotophase and down-

ward movement in the final hours of scotophase), proved

unique in three regards. First, the measurements taken

during 17:00 hrs and 18:00 hrs indicate that these younger

larvae remained essentially immobile during this period.

Second, they inhabited the middle portion (ca. 25 cm above

the soil surface) of the grass canopy during the interval

from 21:00 hrs to 07:00 hrs (which is at least an hour

longer than the amount of time the older larvae occupied the

same portion). However, the measurements obtained for the

third and fourth instar larvae at 08:00 hrs were not
















Table 10. Time-specific mean height above soil surface of
fifth and sixth instar M. latipes larvae iso-
lated in caged pots of Callie giant bermudagrass,
C. dactylon.



Mean
Time Height Std. Err. Precision n


17:00 3.98 0.244 6.13 60

18:00 6.42 0.804 12.52 60

21:00 20.35 1.297 6.37 60

23:00 23.45 1.528 6.52 60

03:00 20.55 1.508 7.34 60

06:00 20.85 1.399 6.71 60

07:00 8.95 1.125 12.57 60

08:00 3.58 0.271 7.57 60



a Distance, in centimeters, from soil surface to larval
head capsule.
b (s-/i) x 100%.



















Table 11. Time-specific mean heighta above soil surface of
third and fourth instar M. latipes larvae iso-
lated in caged pots of Callie giant bermudagrass,
C. dactylon.



Mean
Time Height Std. Err. Precision n


17:00 2.22 0.148 6.67 60

18:00 2.47 0.169 6.84 60

21:00 24.70 1.279 5.18 60

23:00 27.05 1.328 4.91 60

03:00 26.60 1.404 5.28 60

06:00 24.87 1.520 6.11 60

07:00 19.10 1.775 9.29 60

08:00 4.43 0.509 11.49 60


a Distance, in centimeters, from
head capsule.
b (sI/D) x 100%.


soil surface to larval








72




30-

O

o 1
0
LL
M -


S0 0 20


>
0




o 10



z
F-



16:00 20:00 24:00 04:00 08:00

TIME OF NIGHT






Figure 8. Deterministic polynomial model describing the
time-specific vertical movements of fifth and
sixth instar M. latipes larvae in Callie giant
bermudagrass, C. dactylon, canopy. Brackets
about means indicate 95% confidence intervals.


a Y = -203.921 + 0.184X 3.714xlO5X2.















O


CC 30




O3
-J





0








10 0








16:00 20:00 24:00 04:00 08:00

TIME OF NIGHT






Figure 9. Deterministic polynomial modela describing the
time-specific vertical movements of third and
fourth instar M. latipes larvae in Callie giant
bermudagrass, C. dactylon, canopy. Brackets
about means indicate 95% confidence intervals.


a Y = -266.591 + 0.235X 4.642xlO-52.








74

statistically different from those that described the loca-

tion of the older larvae for the same period. This points

up the third unique feature of the data obtained for the

younger larvae: these larvae required less than an hour to

move from the center of the canopy to its basal portion

(Table 11).

Regression analysis (larval height as a function of

time of night) of these data revealed that a second-order

polynomial, Y = -266.591 + 0.235X 4.642x10-5X2, could not

be rejected as an appropriate (tcalc for B0, 61, and B2 was

-5.31, 5.53, and -5.40, respectively; Fcalc = 16.477, df =

2,5; R2 = 0.868) deterministic model for the vertical move-

ments of third and fourth instar M. latipes in Callie giant

bermudagrass (Fig. 9).

The overall structure of both models is quite similar

(e.g., both are curvilinear, both are second-order,and both

have peaks that roughly correspond to 01:00 hrs); however,

not only do they differ, but the differences they evince

suggest biological interpretations that may both help to

further understanding of causal relationships existing in

grassland agroecosystems and aid the implementation of a

sampling program that focuses on third or fourth instar

larvae.

A comparative inspection of these data (Figures 8 and

9) reveals that the rate at whicn larvae both ascend and

descend is unique for each of the two larval age groups;

this is the principal difference between the two models.








75

The movement of fifth and sixth instar larvae from the basal

portion of the canopy to its central region was initiated at

least an hour earlier than comparable movements by the third

and fourth instars, yet both groups were first detected in

the central canopy at the same time. The decreased slope

evidenced in the model for fifth and sixth instar movements

reflects this difference. Similarly, the fifth and sixth

instar larvae initiated a return to the lower portion of the

canopy about an hour earlier than did their younger counter-

parts; in result, the slope of the descending phase was les-

sened. Although these phenomena occurred during times of

change in light intensity (the times for civil sunset and

sunrise during the experimental period were 19:35 hrs and

06:50 hrs, respectively), it seems unlikely that larvae of

one age group are better equipped to detect these changes;

hence, though a correlation is evidenced, causation is re-

jected in lieu of two more biologically-oriented factors:

larval appetite, and predator avoidance.

Fifth and sixth instar M. latipes larvae consume sig-

nificantly more grass foliage than do third and fourth in-

stars (Reinert 1975). Since the larvae remain in the lower

portion of the grass canopy (where green foliage is compara-

tively scarce) during photophase (Fig. 10), the daylight in-

terval is a period of larval fasting. Hence, it is possible

that the comparative reduction in the slope of the ascending

phase evidenced in the model of the older larval group's

movements is a consequence of these larvae responding to










hunger, and that this reaches a level sufficient to evoke

movement toward green foliage at a time earlier than that

evidenced for younger larvae. Since foodstuff conversion

efficiency is greater in younger larvae (Reinert 1975), it

does not seem unreasonable to suspect that the older larvae

are prompted more readily by appetite.

Differences evidenced in the rates at which larvae of

the two groups return to the lower portion of the canopy may

have been due to age-specific responses of larvae to preda-

tor pressure. During September and October 1983, a large

(ca. 500) flock of boat-tailed grackles, Quiscaulus major

(Vieillot.), visited the experimental site every morning at

dawn and were often seen searching and feeding in the grass

fields. Inspection of those areas from which the birds had

recently departed invariably revealed that the population of

M. latipes larvae had been substantially reduced by the

birds' presence.

Linkage of this predator-prey relationship to larval

behavior can be established if a single aspect of bird

behavior is assumed operative: when hunting, the grackle

relies only on its vision. Given this, it follows that a

bird would more readily detect a large larva inhabiting the

middle portion of Callie giant bermudagrass canopy (a region

where most of the vegetation is green rather than brown, and

the predominantly brown markings of M. latipes larvae are

less effective as a cryptic device) than it would a small

larva. Alternatively, large larvae that were immobile and


























































Figure 10. Photophase resting position of sixth instar
M. latipes larva in Callie giant bermudagrass,
C. dactylon. Photograph taken at 16:30 hrs.








78

closely appressed to the brown stems and leaf sheaths found

near the soil surface (Fig. 10) would be more likely to go

unnoticed by the larvae-hunting birds. Thus, a plausible

explanation for the earlier downward movements (that con-

tribute to the lessened slope of the model's defending

phase) of the fifth and sixth instar larvae is that this

behavior is a response to selection pressure exerted by

avian predators, and that those large larvae that return to

the lower portion of the grass canopy before the birds

arrive are more likely to reach maturation.

Finally, the data presented in Figure 9 may prove use-

ful in designing a sampling program that focuses on behav-

ioral characteristics of third and fourth instar larvae.

Examination of these data indicates that the larvae of this

age group are most mobile during the intervals 18:30 hrs to

20:30 hrs and 07:00 hrs to 08:00 hrs. If a larva in motion

is more readily dislodged from grass foliage than a larva at

rest, and the sampling method employed relies on larval dis-

lodgement; then sampling conducted at these two times can be

expected to yield a higher proportion of an existing larval

population than would comparable efforts conducted at other

times.

Evaluation of a Grassland Sampling Device

Preliminary tests revealed several noteworthy points.

First, the rake head worsened machine performance; when set

for either 5 or 10 cm tine spacings, attachment of the rake

head caused the sampler's forward progress through the





























































Figure 11. Behavior of M. latipes larvae in the pushcart
sampler's cage.








80

Callie giant bermudagrass to proceed so jerkily that larvae

trapped by the cage were often bounced out. Second, larval

catch was inversely proportional to cage ground clearance

setting, with virtually no larval capture occurring when the

ground clearance was set at 15 cm. Third, examination of

the cage floor immediately following a sample trial showed

that the vast majority of larvae captured were lying on the

anterior one-fifth of the floor's surface, and that within

less than a minute most of these larvae had uncurled and

were crawling about on the screen floor. Captured larvae

were often seen to move toward the cage's open front; how-

ever, upon encountering the ca. 2 cm wide aluminum frame,

these larvae were invariably seen to "rear-up" as if in

search of some structure upon which to climb. Upon finding

nothing there, larvae so engaged usually redirected their

movements and proceeded toward the cage's walls (rarely,

toward the interior). Finally, captured larvae evinced a

pronounced negative geotaxis. Within five minutes after

their capture, virtually all larvae in the cage had moved to

the cage walls, climbed half of the cage height (ca. 25 cm),

and had come to rest oriented such that their head capsules

were uppermost (Fig. 11).

Analysis of the counts that were made immediately prior

to tests of sampler efficacy revealed that the number of

marked larvae present in each 1 m2 plot declined through

time (Table 12). Daylight inspections revealed that the

soil surface in each plot bore numerous bird tracks and a








81







Table 12. Attrition of marked M. latipes larvae in plotsa
of Callie giant bermudagrass, C. dactylon.


Meanc number of
marked larvae present per


A

30.0

17.0


4.4

1.8

1.2

0.6


B

30.0

15.8

8.6

7.0

2.6

1.2


C

30.0

10.4

5.8

1.2

0. 4

0.0


treatment


D

30.0

10.0

5.4

3.8

2.2

1.0


a Rectangular (0.5 m
24-day-old grass.


x 2.0 m)


field plantings containing


D Marked larvae initially released at 22:00 hrs.

c Means are each of r=5 replications per treatment per
time; each replication initially contained m=30 marked
larvae.
d Treatments A, B, C, and D consisted of an application
of Dayglo fluorescent pigment No. 15, 16, 18, and 21,
respectively.
e Mean number of larvae observed per time on an
exoeriment-wide basis; each mean is of n=20
observations.


Hours
elapsed
since
releaseD


Totale

30.0

13.3

6.0

3.5

1.6

0.7

















Table 13. Summary statistics of the proportion of marked M.
latipes larvae present per plot recaptured per
trial by the pushcart sampling device. Data of
trials conducted at two different times.


Recapture conducted at 9

Treatment Meanb

A 0.0733

B 0.0933

C 0.0667

D 0.0800

Total 0.0783

Recapture conducted at 2


Treatment

A

B

C

0)

Total


Meanb

0.1263

0.0762

0.1104

0.1603

0.1183


hours post-release:

Std.Err. Precisionc

0.0125 17.01

0.0245 26.23

0.0183 27.38

0.0170 21.23

0.0081 11.25

2 hours post-release:

Std.Err. Precisionc

0.0163 12.89

0.0226 29.61

0.0235 21.29

0.0281 17.53

0.0175 9.94


a Treatments A, B, C, and D consisted of an application
of Dayglo fluorescent pigment No. 15, 16, 18, and 21,
respectively in n=5 plots per treatment.
b Number of larvae captured divided by the number of larvae
present per plot per time.
c (s2/2) x 100%.


Time

07:00

07:00

07:00

07:00

07:00



Time

20:00

20:00

20:00

20:00

20:00










few feathers; thus, bird predation was probably the chief

cause of reductions in the number of marked larvae per plot.

Due to the magnitude of larval losses, only those data

obtained for sampler trials conducted at 9 and 22 hours

post-release (these times corresponded to 07:00 and 20:00

hours, respectively) were considered for further analysis.

As evidenced by the comparable magnitudes of the

standard errors accompanying the mean proportion of larvae

recaptured per treatment, treatment means within times were

statistically equivalent (Table 13). This time-specific

homogeneity has a twofold importance. First, it is an indi-

cation of consistency, as regards both sampler efficacy and

larval behavior, from plot to plot per time; and second, it

is supportive of comparisons between times being made on the

basis of estimates derived from data pooled per time.

A two-tailed, t-test evaluation for the difference

between two sample means from populations having different

variances revealed that the proportion of marked larvae per

plot captured by sampler trials conducted at 20:00 hrs was

significantly (tcalc = -2.7197, dfadj = 35, a/2 = .025)

different than the proportion captured at 07:00 hrs. Con-

struction of a 95% confidence interval about the difference

yielded 0.0399 0.0288. As indicated by the statistics

presented for totals per time (Table 13), the higher propor-

tion was obtained for trials conducted at 20:00 hrs.

Judging from these data, the pushcart sampler yielded

only slightly better results at one time than another; more-








84

over, at even the better of the two times, the device caught

only slightly more than 10% of the larvae present. Taken

alone, such a finding would seem adequate grounds to dismiss

the device as one having little or no potential. Two points

argue against such a conclusion. First, the higher level of

larval capture occurred at a time (22 hrs post-release) when

the average larval populational density was relatively low

(i.e., 13.3 larvae per m2; Table 12). This implies that the

sampling device is functional when larval populations are at

subinjurious densities. Second, the amount of error term

contained in each of the two time-specific estimates of mean

larval recapture was quite low (11.25% and 9.94% for total

effort at 07:00 and 20:00 hrs, respectively; Table 13).

Thus, even though the number of larvae captured per trial

was low, results from trial to trial were consistent.










Laboratory Experiments

Chorion Consumption

Single-classification analysis revealed that the M.

latipes ova examined were significantly non-uniform (Table

14) in the amount of spheroidal portion of the chorion found

consumed by an emerging first instar larva. The logical

interpretation is that different larvae consume differing

amounts of chorion at (or just after) hatching. If this

interpretation is in fact correct, then the findings

previously reported by Van Dinther (1954), who maintained

that M. latipes larvae eat the chorion, as well as those put

forward by Ogunwolu and Habeck (1975), who found the chorion

was not consumed by the newly-hatched larva, are both (in

some measure) valid claims (q.v., Figure 12).

The seemingly contradictory findings of previous

research efforts could have been due to a group-specific

(rather than an egg-specific) phenomenon. Although a good

starting point, analysis by single classification admits no

provision for determination of how much of the observed

variability is due to differences among the eggs themselves

as opposed to differences that are due to the random effect

of an egg having been a member of a particular group. This

difficulty is overcome through partition of the variability

detected by single classification into component-specific

variation; such a partition (via a nested analysis of

variance: repeated measures in eggs; eggs within groups) is

presented in Table 15.

























































Figure 12. Differences in the degree of chorion consumed by
newly closed M. latipes larvae.

















Table 14. Single-classification ANOVA of a once-repeated
measurement of the chorion consumed by an
emerging first instar M. latipes larva.


SOURCE df SS MS F

Among Eggs 79 344.775 4.364 29.095

Within Eggs 80 12.000 0.150

Total 159 356.775


F-statistic significant at the a = .05 level.









Table 15. Two-level nested ANOVA of the chorion consumed
by an emerging first instar M. latipes larva.



SOURCE df SS MS F


Among Groupsa

Among Eggs
within Groups

Error

Total


80

159


49.275

295.500


12.000

356.775


7.039

4.104


0.150


1.715

27.361


F-statistic significant at the a = .05 level.
a Each group contained n=10 M. latipes ova.