Citation
Ecology and behavior of Conomyrma insana

Material Information

Title:
Ecology and behavior of Conomyrma insana
Creator:
Nickerson, James C. Everett, 1942-
Publication Date:
Language:
English
Physical Description:
viii, 105 leaves : ill. ; 28 cm.

Subjects

Subjects / Keywords:
Animal nesting ( jstor )
Ants ( jstor )
Colonies ( jstor )
Eggs ( jstor )
Fire ants ( jstor )
Impact craters ( jstor )
Nymphs ( jstor )
Predation ( jstor )
Soybeans ( jstor )
Species ( jstor )
Ants -- Florida ( lcsh )
Conomyrma insana ( lcsh )
Dissertations, Academic -- Entomology and Nematology -- UF
Entomology and Nematology thesis Ph. D
Solenopsis invicta -- Biological control ( lcsh )
City of Gainesville ( local )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1976.
Bibliography:
Bibliography: leaves 94-100.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by James C. Everett Nickerson.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
This item is presumed in the public domain according to the terms of the Retrospective Dissertation Scanning (RDS) policy, which may be viewed at http://ufdc.ufl.edu/AA00007596/00001. The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator(ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
022388994 ( ALEPH )
02686494 ( OCLC )

Downloads

This item has the following downloads:


Full Text














ECOLOGY AND BEHAVIOR OF CONOMYRMA INSANA


By

JAMES C. EVERETT NICKERSON














A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FGR THE
DEGREE OF DOCTOR OF PHILOSOPHY






UNIVERSITY OF FLORIDA 1976


1

























This dissertation is dedicated to George Davis of Horatio, Arkansas. The debt owed by myself and countless former students of his that he aided can never by fully repaid. Without his help and encouragement, I would not have had the opportunity to embark upon the studies required to begin, much less complete, this work.















TABLE OF CONTENTS



Page

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

ABSTRACT . . . . . . . . . . . . . . vi

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

REVIEW OF LITERATURE... . . . . . . . . . . 3
Introduction . . . . . . . . . . 3
Taxonomy . . . . . . . . . . . 4
Physiology . . . . . . . . . . 7
Predation. . . . . . . . . . . 7
Homoptera Tending. . . . . . . . . 9
Habitat. . . . . . . . . . . . 9
Distribution . . . . . . . . . . 10

CHAPTER

I. DISTRIBUTION AND HABITAT OF Conomyrma insana. . . . 11
Introduction.... . . . . . .. . .11
Methods and Materials. . . . . . . . 11
Results and Discussion . . . . . . . 12
Sunary . . . . . . . . . . . 24

11. COLONY ORGANIZATION AND NUMBERS OF QUEENS . . . 33
Introduction . . . . . . . . . 33
Materials and Methods . . . . . . . 34
Results and Discussion . . . . . . . 35
Conclusions . . . . . o . . . . . 4
Summary. . . . . . . . . . . . 42

III. MIXED NESTS OF Conomyrma insana and C. flavopecta--EVIDENCE
OF TEMPORARY SOCIAL PARASITISM. . . . . . . 43
Introduction . . . . . . . . . . 43
Methods. . . . . . . . . . . . 45
sults. . . . . . . . . . . . 45
Discussion . . . . . . . . . . 43
Surmary. . . . . . . . . . . . 50


iv











Page

IV. PREDATION ON FOUNDING QUEENS OF Solenopsis invicta BY
WORKERS OF Conomyrma insana . . . . . . . 51
Introduction . . . . . . . . . . 51
Methods........... . . .. . .. .. .. .. .. 52
Results and Discussion . . . . . . . 53
Summary. . . . . . . . . . . . 63

V. SOURCES OF CARBOHYDRATES UTILIZED BY Conomyrma insana . 64
Introduction . . . . .. . . . . . 64
Methods. .* . . ... . .. . .. . 65
Results and Discussion . . . . . . . 66
Summary. . . . . . . . . . . . 77

VI. PRESENCE OF S issistilus festinus TENDED BY ANTS: A
FACTOR AFFECTING EGG PREDATION BY ANTS IN SOYBEANS. . 79
Introduction . . . . . . . . . . 79
Methods......... ... .......... 79
Results and Discussion . . . . . . . 81
Conclusions. . . . . . . . . . . 88
Summary. . . . . . . . . . . . 89

PROGNOSIS. . . . . . . . . . . . . . . 91

LITERATURE CITED . . . . . . . . . . . . 94

APPENDICES . . . . . . . . . . . . . . 101

A. GLOSSARY OF SPECIFIC TERMS. . . . . . . . 102

B. PUBLICATIONS. . . . . . . . . . . . 104

BIOGRAPHICAL SKETCH. . . . . . . . . . . . 105


IV










Abstract of Dissertation Presented to the Graduate Ccuncil of the University of Florida in Partial Fulfiliment of the
Requirements for the Degree of Doctor of Philosophy


ECOLOGY AND BEHAVIOR OF CONOMYRMA INSANA By


James C. Everett Nickerson

December, 1976


Chairman: Dr. W. H. Whitcomb
Major Department: Entomology and Nematology


Coromyrma insana (Buckley) enclaves (large numbers of contiguous nests with defineable boundaries) were found at 29 locations on the Tall Timbers Research Station in Leon Co., Fla. The ants were more commonly found in open areas adjacent to heavy vegetation. Nests were absent in cultivated areas, areas of heavy vegetation, and where drainage was poor.

The organization of the colony and number of queens per colony of C. insana and C. flavopecta (M. R. Smith) were investigated using radioisotope tracers and by excavation of nests. Conomyrma insana was polydomous (more than I nest per colony), and had several queens located in a central nest. Conomyrma flavopecta was monodomous (1 nest per colony) with 1 colony queen.

Cononmyrma insana and C. flavopecta were found in the same nest.

Later, clusters of C. insana nests were formed at this location. Based on the evidence, C. insana is a temporary social parasite on C.

flavopecta.


vi









Conomnvrma insana workers are effective predators of fire ant fcunding queens, ScLenopsis invicta Buren. The confrontation behavior of the 2 species was determined. Worker detection of a queen alighting in a C. insana enclave varied from a few seconds to several minutes. The founding queen's reactions to attack by C. insara consisted of 5 basic responses: (1) escape, (2) concealment, (3) cessation of movement, (4) mandibular defense, and (5) rapid flexing or shaking. Less than 3% of the queens observed alighting within a C. insana enclave successfully escaped, and none survived to develop a new colony.

Twenty-six species of Homoptera were tended by Conomyrma insana. The Homoptera were in the families Aphididae (10), Phylloxieridae (1), Coccidae (5), Pseudoccidae (1), Eriococcidae (2), Kermesidae (1), and Membracidae (1). Conomyrma insana also fed from the nectaries of Cassia nictitans L., C. fasciculata Michx., C. obtusifolia L., and Richardia scabra L. Conomyrma insana construct their nests near carbohydrate food sources.

Five ant species, of which C. insana and Solenopsis geninata (F.)

were the most abundant, tended nymphs of the threecornered alfalfa hopper, Spissistilus festinus (Say), in a soybean field. The presence of the ants under these conditions was examined as a factor affecting predation on eggs of the soybean looper, Pseudoplusia includens Walker. Where S. geminata tended the nymphs the percentage of eggs missing after 24 hr was significantly greater from plants with nymphs (77.lX) than from plants without nymphs (36.8%). Where C. insana tended the nymphs, the percentage of eggs missing was significantly greater from plants with nymphs

than from those without nymphs after 4 hr (36.8% vs. 17.5%) and 8 hr


vii









(65.5% vs. 42.9%) but not after 24 hr (90.7% vs. 86.3%). The proportion of eggs missing was greatest from the bottom 1/3 of the plant and least from the top 1/3.


viii


AO















INTRODUCTION


Ants of the dolichoderine genus Conomyrma occur commonly throughout most of the United States. Yet, within this country, comparatively little attention has been directed toward the ecology or biology of the species in this genus. The majority of published reports including these ants consist of a brief mention in a list of ant species collected in an area and/or a notation as to the type of habitat they were found in.

The paucity of attention toward the species of Conomyrma can be partially attributed to the lack of any obvious economic or health problems posed by their presence. Recently, though, accounts have been published on observations of one species, Conomyrma insana (Buckley), attacking and killing newly mated queens of the red imported fire ant, Solenopsis invicta Buren (Fincher and Lund 1967, Markin et al. 1971, and Whitcomb et al. 1972). These reports prompted this investigation of C. insana in northern Florida as the presence of S. invicta does pose economic and health problems.

An examination of the ecological requirements of C. insana was

deemed necessary to evaluate the potential of this species as a candidate biological control agent of the red imported fire ant. In addition, an investigation of predation by C. insana on newly mated fire ant queens and other arthropods was required to evaluate the efficiency of this ant as a biological control agent.

Consequently, my primary objective was to determine and document the principal factors affecting the distribution and populations of C.


*1





2



insana in north Florida. The compilation of this information was considered preparatory to future efforts to increase the populations of C. insana through habitat manipulation and, thereby, increase the effectiveness of the ant as a predator of fire ant queens or other target pests.















REVIEW OF LITERATURE


Introduction

In the continental United States, three species of the genus Conomrma--iolor (Wheeler), flavopecta (M. R. Smith), and insana (Buckley)--were recognized by Snelling (1973). Prior to Snelling's revision, the species of Conomyrma in North America and neighboring areas were lumped as Dorymyrmex pyramicus (Roger) or Conomyrma pyramica (Roger). Both Cole (1957) and Wilson (1957) suggested that C. bicolor was a distinct biological species, based on their independent observations that C. bicolor occurred in intimate sympatry with C. pyramica while remaining distinct in worker color and size.

It is now evident that the Dorymyrmex/Conomyrma pyramicus group

is a complex comprised of several species. In the past, several myrmecologists have attempted to review comprehensively the taxonomic tangle surrounding this complex. The most natable attempts were made by Santschi (1922), Creighton (1950), Kusnezov (1952, 1959), Wilson (1957), and Snelling (1973). For the most part, their attempts have had little effect beyond increasing the confusion. There have been attempts by several authorities, such as McCook (1879), Santschi (1922), M. R. Smith (194e), and Cole (1936, 1957), to separate individual species from the complex, but few used solid morpholigical characters that would withstand critical examination.

Owing to this extreme confusion and until this complex is dealt with as a whole, it is impossible to assign much of the literature references


3





4


correctly to the currently accepted species. Therefore, in the following discourse on the literature, I will deal with the entire complex in

the United States (excluding only those papers dealing strictly with C.

bicolor_). The name used by the author of each article will be quoted

unless I am able to determine that it may be safely and correctly

assigned to one of the species now recognized by Snelling.


Taxonomy

Conomyrma insana was described as Formica insana by S. B. Buckley

in 1866. His description of this ant is reiterated in entirety as

follows:

Worker. Length 0.14 inch--Color black or brownish-black,
smooth and shining throughout; head sub-quadrate, the lateral
margins slightly curved inwards; below oval,and rounded
above; eyes large, subelliptical, and placed on the anterior portion of the head near its lateral margins; antennae long,
filiform; rmandibles large, curved inwards, the truncated
apical ends sharply toothed; head sub-channelled beneath with the anterior and posterior parts deeply depressed; prothorax
about half the width of the head, rounded above; mesothorax
somewhat depressed; metathorax has a rudimentary spine or sharp protuberance on its upper posterior surface, pedicle short, inserted near the base of the anterior part of the abdomen; scale smull, wedge-shaped, and inclined forwards;
abdomen broad-ovate, subacute; legs small, slender, and rather
short.
Female. Length 0.20 inch--Head small, narrower than the
thorax; abdomen broad, oblong and ovate; color black-bronze
with the margins of the segments of the abdomen hyaline;
thorax raised above the head; wings not seen; the remainder
like the worker.
Dwells in the ground but a few inches deep, having numerous holes scattered over a large surface of fifty or more feet
in diameter. It goes now here and now there, without order, from one abode to another, sometimes roving the larvae from one place to another; large colonies dwelling in common. It
is very active and war-like, can conquer most other ants,
boldly seizing the larger species by the legs. If a large ant is so unfortunate as to run among the "crazy ants," he hastens
to escape, but rarely does so without being severely bitten.
Common in Central Texas.





5


The genus Dorymyrmex was established by Mayr (1866). Later, Conoyrma was established by Forel (1913) as a subgenus of Dorymyrmex. Santschi (1922) designated Prenolepis pyramica Roger, as the type species of Conomyrma; Donisthorpe (1943) redesignated the same species as the type. The subgeneric name was rejected by Creighton (1950) on the grounds that Conomy~rma was synonymous with Dorymyrmex, once those species lacking a propodcal tooth were removed to Araucomyrmex, as was done by Gallardo (1916).

Kusnezov (1952) restudied the problem and proposed to divide Dorymyrmex into 2 genera, Dorymyrmex and Conomyrma. The genus Conomyrma was separated from Dorymyrmex in the worker caste by the lack of a psammophore and in the female caste by the presence of a single, versus two, cubital cells in the forewing. Further morphological differences separating the genera were described by Eisner (1957) and Wheeler and Wheeler (1951). Unfortunately, Kusnezov (1952) failed to designate type species for the new subgenera proposed, but he corrected his oversight in a subsequent paper (Kusnezov 1959). The generic name Conomyrma was recognized by Snelling (1973) as a valid genus.

Roger (1863) described Prenolepis pyramica from a single specimen from Corrientes, state of Bahia, Brazil. The name was transferred to Dorymyrmex by Mayr (1866). By 1900, pyramicus was assumed to range from Argentina to the southern United States and over the entire Carribbean area (Forel 1899) and had acquired a number of varieties and subspecies along with an entirely different identity (Snelling 1973).

Norton (1875) suggested that Buckley's Formica insana may prove to be Dorymyrmex pyramicus. McCook (1879) concurred that Norton may have





6


been correct in his suggestion, yet he persisted in calling the ant D. insanus. Forel (1899) placed insana in synonomy to pyramicus. Wheeler (1902) stated that insana was undobutedly a synonym of pyramicus_. Buckley's insana was uniformly black or brownish-black; Roger (1363) stated that pyramicus possessed a yellowish-red head and thorax and brownish gaster. Roger's pyramicus, as described, agreed closely with the description of a new variety, bicolor, by Wheeler (1906).

Creighton (1950) retained insana as a synonym of pyramicus, yet he indicated doubt as to the validity of his decision. He stated that since neither Mayr nor Forel had seen Roger's (1863) types, there was little to indicate that the "typical pyramicus" was the same insect that Roger described or the same form to which that name is applied to in North America. According to Snelling (1973), the ant long referred to as pyrarlJcus in North American literature has a mesonotum which is sharply declivitous behind and is uniformly brownish to blackish with the lower part of the head lighter. This ant clearly is not the same entity as the South American pyramicus described by Roger (1863). The earliest available name for this ant is insana. Based on the evidence, Snelling (1973) recognized insana (Buckley), and gave the synonomy for this name as follows:

insana (Buckley 1866)
= pyr-amicus, sensu Wheeler (1902), Creighton (1950),
not of Roger (1863)
= flavus McCook (1879)
nigra Pergande
= antillana Forel
= smithi Cole (1936)
= brunnea, Kusnezov (1952) = wheeleri Kusnezov (1952)





7


Physiology

The isolation of 2-heptanone (methylpentyl ketone) from the anal glands of Conomyrma pyramica and its role as the alarm pheromone were described by Blum (1969) and Blum and Warter (1966). The substance releases a complex behavioral pattern in the worker ant which includes alarm, attraction, repellancy, and digging. Metcalf and Metcalf (1970) reported that typical alarm behavior by C. pyramica was initiated in the laboratory by exposing the worker ants to isosteres of 2-heptanone (1,1,1-trifluro-methylheptanone, butyl acetate, and butylthiol acetate). They also reported that a remarkable behavioral change was induced by the isostere hexanoyl chloride, which invariably produced a violent combat between worker ants which touched one another.

Blum (1973) reported that the anal glands of dolichoderines are not restricted to biosynthesis of terpenoid natural products. His statement is corrobrated by research by McGurk et al. (1968), which showed that C. pyramicus also produce the monoterpene iridodial.

Vick et al. (1969, 1969) investigated the analysis of volatile compounds and free amino acids in several ant species, including C. pyramica. to determine the potential of the chemicals as taxonomic characters in the study of ant systematics. They concluded that valuable information on ant classification and phylogeny may be obtained.


Predation

Comstock (1879) reported both D. insanus (= C. insana) and D. flavus killing the "cotton-worm." Schwarz (1907) noted that D. pyramicus protected solitary tree cotton from the boll weevil in Cuba. Pierce et al. (1912) andI Hunter and Pierce (1912) listed D. pyramicus var. flavus McCook as an eneily of Alabama argillacea (Hulsner) and Heliothis obsoleta





8


F. in cotton. In addition, Pierce (1912) noted that the ant had been recorded preying on the boll weevil, Anthonomus grandis Boh. Girault (1907) observed D. pyramicus attacking larvae of the lesser peach tree borer and occasionally killing newly emerged moths. Quaintance and denne (1912) stated that the ant D. pyramicus attacked the plum curculio, and that the ant was influential in the destruction of curculio. Van Pelt (1966) reported that foraging workers of D. pyramicus (= C. insana based on Van Pelt's description of the nesting habits) were repeatedly seen carrying insect parts. Whitcomb et al. (1972) stated that the black form of C. pyramicus (= C. insana) is one of the most important predaceous ants in north Florida soybean fields, attacking leaf hoppers, small beetles, and other insects.

Fincher and Lund (1967) noted that D. pyramicus killed some of the fire ant queens, Solenopsis saevissima richteri Forel (= S. invicta Buren) before they had a chance to burrow into the ground after their nuptial flight. Markin et al. (1971) reported that C. pyramicus (= C. insane) were seen dismembering mated females of the imported fire ant and stated that the dolichoderine ant, where it is numerous, may be one of the main predators of fire ant queens. Whitcomb et al. (1972) and Whitcomb (1974) stated that the black form of C. pyramicus (= C. insana) is appar'en:Iy one of the most aggressive ants toward dealate females of S. invicta before they dig into the ground. In another report, Whitcomb et al. (1973) stated that they had observed 368 S. invicta queens killed by C. insana workers in a single afternoon. Hung (1974) reported that he recovered the heads of the queens of 5 species of myrmecine ants from the refuse pile of C. insana (= undescribed species of Conomyrma based on voucher series from his study).





9


Homoptera Tending

Nielsson et al. (1971) reported the association of several aphid species with C. pjramicus, but did not attempt to establish if a mutualistic relationship occurred between the ant and species of aphids listed. In fact, they reported associations with Dactynotus sp., which are not thought to be tended by any ant species. Van Pelt (1966) noted that D. pyramicus (= C. insana based on Van Felt's description of the nesting habits) tended 2 aphid species, Hysteroneura setariae (Thomas) and Pterocoma smithiae (Monell). Research by Odum and Kuenzler (1963), Wiegert et al. (1967), and de la Cruz and Wiegert (1967), using 32P to trace the plant-aphid-ant food chain, demonstrated that D. pyramicus was obtaining honeydew from aphids of the genera Aphis and Mdcrosiphum. Jones (1929) listed 3 aphids, Aphis sp., Chaitophorous negundinus Thos., and C. populela Gill. and Pal., tended by D. pyramicus. Knight (1944) reported that the hemispherical scale, Saisetta hemispherica Targioni, was tended by 0. pyramicus var. niger Perg. on Palay rubber vines. Comstock (1879) reported that D. insanus (= C. insana) feeds upon the sweet exudations of plants, galls, and sweet excretions of aphids.


Habitat

While no habitat studies have been made on D. pyramicus, numerous researchers, including Cole (1940, 1942), Dennis (1938), Smith (1924, 1936), Van Pelt (1958), Wheeler (1905, 1932), and Wilson (1964), have noted that this ant prefers to nest in open sandy areas. Markin et al. (1974) stated that C. insana was common in a part of an open field in Louisi3na. According to Naves (1974), C. insana are more populous on roadways or other open areas with sparse vegetation.





10


Distribution

At present, there is no real means of delineating the range of C. insana. The range of Dorymyrmex/Conomyrma pyramicus is recorded to be as far north as North Dakota and, according to Creighton (1950), occurs in all states except the extreme northeastern United States.

The confirmed records of C. insana are from southern states.

These include Texas (Buckley 1866, Wheeler 1901), Arkansas (Roe 1973), Mississippi (Markin et al. 1971), Florida (Whitcomb et al. 1973), Georgia (Van Pelt 1966), and Louisiana (Markin et al. 1974).














CHAPTER I


DISTRIBUTION AND HABITAT OF Conomyrma insana


Introduction

The nature and abundance of suitable nesting sites are important factors determining the distribution of a species of ant (Letendre and Pilon 1973). In initial investigation of Conomyrma insana, large numbers of nests were found clustered in some areas while other apparently ideal sites were found where the ant was conspicuously absent. According to Naves (1974), C. insana are more populous in dirt roadways in open areas. Markin et al. (1974) mentioned that C. insana was common in portions of open fields in Louisiana. Other than brief notations of the ant occurring in shady areas, there are no definitive studies on the habitat of C. insana.

Therefore, to define the habitat of this ant necessitated locating and comparing sites where the ant was established. Tall Timbers Research Station near Tallahassee, Fia. was chosen for this work because of the diversity of ecosystems represented. In addition, records were available for each site and, on request, the treatment of each location of the station could be maintained.


Methods and Materials

Beginning in 1972, open fields and roadways within the 1133 hectares (2800 acres) of Tall Timbers Research Station near Tallahassee, Fla. were searched for Ccinomnra insana. The locations of all C. insana populations


11









were marked on a map of the station (Fig. 1.1). (The map was prepared by Roy Komarek in July, 1969 and was based on ASC aerial photograph ADV194, 3-22-60.) The sites of C. insana populations were examined for consistencies and variations of features.

The C. insana populations in two fields, referred to as Introduction Field and Persimmon Field, were annually mapped in greater detail. In 1973, a survey map of Introduction Field was prepared by using the compass-traverse survey method (DeMoisy 1949) to establish the field boundaries and locate major landmarks. Permanent markers (.5-m sections of pipe driven into ground) were placed in secure locations on the edge of the field. The boundaries of the C. insana population were located on the prepared map. In 1974, the same procedure was initiated for

Persimmon Field.

The number of nests per m2 within C. insana enclaves was examined in Introduction Field (beginning in 1973) and Persimmon Field (beginning in 1974). The nests in 100 randomly selected m2 quadrats were counted annually in July through 1976.

Results and Discussion

Twenty-nine locally isolated populations of Conomyrma insana were found on Tall Timbers Research Station (Fig. 1.1). At each site, numerous nests were present and apparently interconnected, based on worker traffic, in addition to worker transfer of brood, from nest to nest. The least number of nests found in a single locale was 17, while 2,187 nests were counted in another field. The contiguous nests of the popu!ations of C. insana can only be defined as enclaves as they have definite boundaries within which few other ant species occur.

























Tall Timbers Research Station with locations of Conomvrma insana populations indicated. Where populations were found in open fields, the entire field is marked rather than the boundaries of the limited distribution of C. insana within the field. The locations are: (1) Jones Field, 12) Field N-W5, (3) Gate Field, (4) Middle Gate Road, (5) Sutty Field, (6) Butler Field,
(7) Gum Pond Course Road, (8) Field M-8, (9) Field M-9, (10) Gay Field C, (11) Hall Island Drive Road, (12) Gay Field B, (13) Field M-12, (14) Hall Island Course Road, (15) Field L-8 and L-9, (16) Field L-3, (17) Whitcomb Plots Rep 3, (18) Whitcomb Plots Rep 2, (19) Fennel Field, (20) Tripsacum Field, (21) Field L-12, (22) Tripsacum Course Road, (23) Richard Field,
(24) Introduction Field, (25) Persimmon Field, (26) Field L-14,
(27) Pea Field, (28) Field M-15, (29) Stoddard Lab Field.


Fig. 1.1














Scale: 1" 2085
N.
--C. insana Station Roads




5s~ -'


3


8
-9--0 -I
/\ 11 -.

-- -- 34 28I


5 .4
16~ 25
22
117 2 26


0









L EI A M 0 N I A





15


The soil at all C. insana nesting sites was sandy with little clay or loam content. The nest craters were found only in areas fully exposed to sunlight. No nests were found in heavy vegetation. The vegetation at all sites was sparse to absent because of cultivation, wellmaintizined fire lanes, annual burning (spring), sterile soil, roadways, or closely mowed areas. At one location, vegetation was absent owing to a rapidly expanding gulley. The most common plants occurring within the nesting areas of C. insana were Andropogon virginicus L., Digitaria sp., Paspalum sp., Richardia scabra L., Cassia fasciculata Michx., C. nicitans L., Eupatorium capillifolium (Lam.) Small, and Ambrosia

a.-tei-iisiifi-lia L.

Apparently, slope had little effect on the distribution of C. insana nests. Although the majority of the areas where the ant was found were gently sloped (1-6*), at 2 locations the slope was greater than 30'. All sites were good to well drained. Poorly drained areas adjacent to established enclaves of C. insana were not utilized for nesting.

The C. insana enclaves in Introduction Field from 1973 to 1976 are shown in Figs. 1.2-1.5. The boundaries of the enclaves are somewhat stable, but some change is evident from year to year. The major annual variation occurred on the west side of the field. In 1973, this portion of the field had an extremely patchy vegetative cover (ca. 30%) consisting primarily of Paspalum bosianum Flugge, and C. insana nests were numerous (Fig. 1.2). In 1974 and 1975, the vegetation was apparently too rank for C. insana as they no longer nested in this portion of the field (Fig. 1.3 and 1.4). in 1976, C. insana were nesting in a fire lane a!hng the edge of the west side of the field (Fig. 1.5). This fire lane h..d been maintained annually, but in 1976 the vegetation was less profuse than in previous years.
































Fig. 1.2. Conomyrma insana nests (stippled areas) in Introduction Field
during July, 1973.





11


-











AA






A


-9 AL















0_7



40 -Oak tree
10 meters


--Shrub oak


L== =

































Fig. 1.3. Conomyrma insana nests (stippled areas) in Introduction Field
during July, 1974.





19





N














41s

>1














X_ I 3. _I AI A L J
I -A--













--Oak tree
---10 meters
-Shrub oak

































Conomyrma insana nests (stippled areas) in Introduction Field during July, 1975.


Fig. 1.4.












N



















AA






Y_ A










_x. A



.. ... ....







--Oak tree 10 meters
x. --Shrub oak

































Conomyrma insana nests (stippled areas) in Introduction Field
during July, 1976.


Fig. 1.5.





23




N









-A. A





















AL~~ ~ *L it U--L _L tree z. --Shrub oak








The locations of the enclaves of C. insana nests in Persimmon Field changed little from 1974 through 1976 (Fig. 1.6-1.8). Some fluctuations were evident in the boundaries that could be attributed to maintainance practices. This field was annually cultivated and planted with corn, and a fire lane was maintained around the field. In 1974, the fire lane and fallow areas on the west side of the field were kept clean of vegetation, and the C. insana nests were present over this entire cleared area (Fig. 1.6). In 1975 and 1976, this areas was not kept clear, and the vegetation became rank with no soil surface exposed to the sun. No C. insana nests were found in this heavy vegetation (Fig. 1.7 and 1.8).

Conomyrma insana nests were not found in the center of the corn

fields, although they were present in the edges of the fields. The repeated cultivation of the corn field may have destroyed most of the nests and prevented C. insana from occupying this area. As demonstrated in Introduction Field, the locations of the C. insana enclaves remained stable. The ant did not nest in 1 narrow area, ca. 12 m wide, on the northern side nor on the southern side of the field. Both were low, boggy areas in which drainage was poor.

The number of C. insana nests per square meter in both Introduction Field and Persimmon Field remained relatively constant (Table

1.1). The average number of nests per square meter was slightly greater in Persimmon Field than in Introduction Field. The only major difference in treatment between these two fields was that Introduction Field was mowed frequently while Persimmon Field was cultivated.


Summary

Conomyrma insana enclaves (large numbers of contiguous nests with defineable boundaries) were found at 29 locations on the Tall Timbers


24

































Fig. 1.6. Conomyrma insana nests (stippled areas) in Persimmon Field
during July, 1974.


























CCorn Field Ve
































Conomyrma insana nests (stippled areas) in Persimmon Field during July, 1975.


Fig. 1.7.



















Corn Field








Corn Field .0








Ci -
































Conomyrma insana nests (stippled areas) in Persimmon Field during July, 1976.


Fig. 1.8.












.1 ,Z












Srb Pine-[ardwoo


*e Corn Field




a... ___-_-____-_-_*_.





31


Table 1.1.


Number of nests per square meter within C. insana enclaves in Introduction Field and Persimmon Field, Tall Timbers Research Station, Leon Co., Fla.


Year Introduction Field Persimmon Fi go
i+ 95% C.I a x + 95% C.I. al


1973 2.4 + .41

1974 2.7 + .32 3.6 + .32

1975 2.5 + .25 3.5 + .22

1976 3.1 + .76 3.8 + .27


(a)Mean + 95% Confidence Interval.
(b)NO Samples taken.





32



Research Station. The ants were found more commonly in open areas adjacent to heavy vegetation. Nests were absent in heavily cultivated areas, in areas of heavy vegetation, and where drainage was poor.













CHAPTER II


COLONY ORGANIZATION AND NUMBERS OF QUEENS Introduction


Investigations of the ecological distribution of ants are concerned with colonies of ants rather than individuals, since the colony is the unit of social organization. It is difficult to be certain whether each nest structure is independent and whether there is a direct overground or underground connection with neighboring nest structures. In investigations of Concmyrma irsana (Buckley) and C. flavopecta (M. R. Smith) in north Florida, the nest craters of C. insana appeared to be clustered

while the nest craters of C. flavopecta exhibited no tendency toward clustering. Both Buckley (1866) and Van Pelt (1966) hypothesized that a single colony of C. insana (= Dorymyrmex pyramicus (Roger) in Van Pelt's paper) consisted of large numbers of nests, though they presented no conclusive proof. I wished to determined the colony composition of each species more

precisely in order to make advances in the field ecology of C. insana and C. flavopecta and to compare the ecological requirements of the 2 species.

Until the advent of radioactive tracers and development of adequate

techniques, it was generally only possible to theorize that a colony of ants was composed of more than 1 nest. Since related problems with ants have been solved using radioactive tracers, their use was attempted in this situation. Kannowski (1959) introduced 32P mixed with honey into source nests. He concluded that a colony of this.species was composed of 1 or


33





34


more nests, hence polydomous instead of monodemous (single nest). Sanders and Baldwin (1969) tagged carpenter ants, Camponctus spp., with Iridium192 to trace underground trails. Levieux (1971) used 198Au to study the nest structure of 2 Camponotus spp. The biological half-lives of 134Cs in 3 castes of Solenopsis invicta Buren were determined by Chiu et al. (1974).

Materials and Methods

All experiments using radioisotope tracers on Conomyrma were conducted in forest-pasture areas of the Tall Timbers Research Station, Leon Co., Florida during August 1972. The selection of 134Cs for this experiment was based on its relatively short half-life (2.6 yr compared to 30 yr for 137Cs). Six shell vials containing a mixture of 1.5 microcuries of 134Cs and 0.5 g of sugar were prepared and dried with infrared heat. One min counts with a single channel gamma spectrum analyzer, with a 3 in. NaI-Tl crystal adjusted to optimum counting with a calibrated 137Cs source, averaged 45,000 cpm per 0.5 g of sugar for each of the vials containing the prepared mixture.

Four areas of C. insana nest craters and 2 areas of C. flavopecta nest craters were selected. In each locale, 20 craters were marked with numbered stakes; 1 crater was designated as the "source" at which the radioisotope was incorporated into the nest. A diagrammatic map giving distance and orientation of the nests to the source crater was constructed for each area. One shell vial containing the sugar and 134Cs was placed adjacent to the entrance hole of the source crater. The shell vial and crater entrance were covered with an inverted quart glass jar to insure that only ants from the source crater were able to feed on the radioactive sugar. The jar and empty shell vial were removed after 24 hr. At 1, 2, 4, and 10




35


days after incorporation of the radioisotope into the source crater, it

and all marked peripheral nests were sampled. The ant samples were taken with a standard aspirator equipped with a rubber squeeze bulb to avoid inhalation of the isotope. Each sample of ants was labeled and taken to the laboratory where total radioactivity of 134Cs was assayed with a single channel analyzer.

Beginning in 1973, nest of both Conomyrma species were periodically excavated and examined for structure and contents. The presence of workers, brood, and reproductives was noted.


Results and Discussion

Conomyrma insana workers with 134Cs were recovered from 79 of the 80 nests sampled, including the 4 source nests. Workers taken from I nest in replicate A (Fig. 2.1) were not radioactive. The percentage of
134Cs assayed in the workers taken from all sampled nests in replicates A, B, C, and D (Fig. 2.1) was .46, .58, .63, and 1.27, respectively, of ca. 45,000 cpm of 134Cs placed at each source nest. The recovery of radioactive workers from apparently independent nest units indicated food trophyllaxis between individuals other than nest mates, in addition to exchange of individuals among the nests. Therefore, the data indicate that a colony of C. insana is composed of multiple nests, hence polydomous. The probability that only a single colony was present at each locale was obtained by dividing the number of nests with radioactive ants by the number of nests sampled. The results of the analysis are presented in Table 2.1. The averages over the replications can be viewed as tne level of confidence possessed in rejecting the hypothesis that more than a single colony of C. insana was present at each locale.

The greatest distance separating any 2 sampled nests of the same colony































Fig. 2.1. Location of Conomyrma insana craters in replication A, B, C,
and D, showing the relation of sampled and nonsampled craters
to the source crater.





37


C. insana craters:
0 Source
* Non-radioactive
* Radioactive o Not sampled
1 meter


0


S 0
*
0


0


0


0
0


* 0.
0
0


0
0


0

0
0


0


0


0


0


0
* 5


0


0

C


0C 0

0 C.e


0


0


0


* 0


* 0


00 .0


0

0
0
S


0


*


0


0 0


0


0


0


0


S
0


0


0 9


0


S


0


*


0


+


0


S

0


B


0


)


0
0


00e


0
0
0


0


0


0


0




38


Table 2.1. Probability that only a single colony of Conomyrma insane
occurred within each replicate of 20 nests sampled for workers
containing Cesium-134.


Replicate

A

B C D

Average over Reps.


la

0.45 0.55

0.45 1.00

0.613


aProbability (a single colony) =


Days following application

2 a 4

0.75 0.85 0

0.70 0.85 1

0.90 1.00 1

1.00 1.00 1

0.838 0.925 0


l0a

.95 .00 .00 .00

.988


No. of nests with radioactive ants
Total no. of nests





39


was approximately 5.5 m. The finite number of nests comprising a colony was not determined in the experiment.

Cesium-134 was recovered only from the source nest of C. flavoopecta. All samples of ants taken from nest exits other than the source nest registered normal background radioactivity. Thus, C. flavopecta was monodomous as there was no evidence of exchange of individuals or food between the individual nests.

For the purposes of this research, a terrestrial ant "nest" means a series of closely interconnected tunnels and chambers which have either a single opening to the surface or several closely interconnected openings; the entire structure obviously spatially separated from other similar nests. In the genus Conomyrma, each "nest" usually has a single opening plus a vertical or slanted main tunnel with branching tunnels and with lateral chambers located at various distances from the surface.

A total of 27 clustered groups of C. insana nests has been excavated to date. Of these, dealate queens were found in 5 of the clustered groups. On 25 May 1973, 18 nests were excavated in a 20 m2 area. Sixteen dealate queens were located in 1 nest at depths ranging from 4-13 cm. On 15 June 1973, 32 nests were excavated in a 21 m2 area. One nest contained 13 dealate queens located within 5-15 cm of the soil surface. This nest had large masses of new eggs in addition to larvae and pupae in all stages of development. All other nest sites excavated in this series contained larvae and pupae, but no eggs. On 5 April 1973, a series of 18 nests located in an area of 25 m2 was excavated. No queens were recovered from 17 of the nests excavated in the area. In 1 nest, 9 dealate queens were recovered from a single chamber 20-22 cm below the surface. There was no surface indication that this nest differed from others





40


excavated. This was the only nest of those excavated in which eggs were found in addition to brood of all stages. On 18 July 1974, 26 nests were excavated in an area of 29 m2. Fourteen dealate queens were located in 1 nest at depths ranging from 2-17 cm. No queens were found in the other excavated nests. On 11 February 1975, 13 nests were excavated in an area covering 9 m2. A single C. insana queen was found in a chamber 2 cm below the soil surface in 1 nest. Four queens were located in chambers 5-10 cm below the surface in a second nest 1.3 m from the nest in which the single queen was located. No additional queens were located in any of the remaining nests excavated in this location.

All C. insana queens recovered during the excavation of the clustered mounds were dealate and apparently reproducing, based on the enlarged appearance of the gaster. Eggs were not found in any of the excavated nests unless a queen was present. If a queen was not present, only larvae and pupae were noted. Prior to excavation, each crater was .examined for visual differences, such as a larger crater compared to others, indicating the possible presence of queens. No differences were detected.

Three clustered groups of C. flavopecta nests were excavated in

1973. On 26 May, 11 nests located in an area 37 m2 were excavated. In each of 5 of the 11 nests, a single dealate queen was found. On 27 September, 9 nests were excavated in a 36 m2 area. A single queen was found in each of 2 nests. On 9 October, 14 nests were excavated in a 43 m2 area. A single queen was found in each of 2 nests. On 9 October, 14 nests were excavated in a 43 m2 area. A single queen was located in each of 3 nests.





41


Conclusions

Two basic assumptions were made in this research in relation to C. insana: (1) the center of the colony could be selected, and (2) there would be equal distribution of the isotope throughout the colony. The first assumption could be fallacious as no criteria existed upon which to base a decision; therefore, the decision was purely subjective. Possibly, the isotope may have been placed at the edge of the colony. The second assumption also had some invalidity because of possible "storage" individuals in the colony which could accumulate excessive amounts of isotope. In addition, when samples were taken and assayed, the individuals removed were not replaced in the colony, resulting in a net loss of radioactivity, not only by biological half-life (Chiu et al. 1974), but also by removal of radioactive individuals. Nevertheless, the data appear to validate the conclusion that, despite the relatively large numbers of "exit" holes, only 1 colony of C. insana was present in each of the tested populations. Conomyrma insana can thus be assumed to have polydomous nesting habits and more than 1 queen per colony.

Conomyrma flavopecta, in contrast, appears to have monodomous

nesting habits with only 1 queen per colony. Although queens were not found in 24 out of a total of 34 nests excavated, in those 10 nests where a queen was found, there was only a single queen per nest. These data, plus the lack of distribution of 134Cs among the nests, support the hypothesis that there is only 1 queen per nest, and each nest is a distinct colony. An alternate hypothesis that single females could be serving as the single colony queen for a number of nests is not supported by the 134Cs data. This failure to find queens in 24 of the 34 nests can more reasonable be attributed to.escape responses of the





42


queens in the large quantity of soil which must be excavated and searched, or failure to find the true bottom of the nest.

The radioisotope 134Cs appears to be an excellent tagging material for this group of ants. The isotope was recovered in sufficient quantities for assay, and the evidence indicated that it was rapidly and evenly distributed throughout the C. insana colony.


Summary

The organization of the colony and number of queens per colony of Conomyrnia insana (Buckley) and C. flavopecta (M. *R. Smith) were investigated with the aid of the radioisotope cesium-134 and by excavation of nests. Conomyrma insana was determined to be polydomous (more than 1 nest), had a single colony occupying several square meters, and had multiple queens located in a central nest. Conomyrma flavopecta was monodomous (1 nest per colony) and had 1 colony queen.













CHAPTER III


MIXED NESTS OF Conomyrma insana and C. flavopecta-EVIDENCE OF TEMPORARY SOCIAL PARASITISM


Introduction

Two species of the genus Conomyrma are common in Florida. For the

present, these are identified as Conomyrma insana (Buckley) and C. flavopecta (M. R. Smith) by the identification criteria given by Creighton (1940) and the revision by Snelling (1973). The 2 species differ in color, morphology, and biology. Conomyrma insana is nearly uniformly dark, greyish-black, has a broad, straight or slightly concave occipital border, and has a distinct declivity on the mesonotum when seen in profile; while C. flavopecta is light yellowish-brown, with the

head And apex of gaster weakly to moderately infuscated, the occipital border rounded or straight only medially, and the mesonotum in profile evely convex without trace of a declivity. Colony structuring differs markedly in the 2 species. Conomyrma insana exists in enclaves of

multiple nests with much traffic between the nests, transfer of brood, food exchange, and many queens localized together in only 1 of many nests; while in C. flavopecta, there is no worker traffic or food exchange between workers from separate nests and 1 queen in each nest.

Reactions between the 2 species in the field are entirely inimical. When workers of the 2 species meet, the C. insana worker usually holds its ground and may open its mandibles, while the C. flavopecta worker immediately breaks contact and runs rapidly and erratically away. This


43




44


reaction is termed "panic retreat" paralleling the terminology of Wilson and Regnier (1971) who distinguished between "panic alarm" and "aggressive alarm" reactions in ants. In contrast, accidental or other contacts

between C. insana workers never lead to antagonistic or panic retreat reactions. Chance encounters between 2 foraging C. flavopecta workers

can either be non-antagonistic or can initiate panic retreat (both workers sometime display this response) according to whether the meeting is between workers of the same colony or from 2 different colonies.

In all respects, the differences between the 2 taxa are striking,

and the evidence that 2 distinct species are involved is unequivocal. Therefore, the discovery of mixed nests of the 2 species was totally unanticipated. The specific records in Florida for these mixed nests are as follows: Highlands Co., Archbold Biological Station, Lake Placid, 18

April 1973; Polk Co., Winter Haven, Camp's citrus grove, 19 April 1973; Polk Co., Winter Haven, Camp's citrus grove, 13 April 1974; Alachua Co., Gainesville, Depot Road, 30 April 1974; Alachua Co., Gainesville, 15 May 1974 (Wm. F. Buren); Alachua Co., Gainesville, 10 April 1975 (J. F. Carroll); Alachua Co., Gainesville, 23 May 1975 (Wn F. Buren); and Wakulla Co., Medart, 20 June 1975 (D. P. Wojcik). In all mixed nests,

workers of both species were observed participating in normal activities as nest-mates (foraging, casting out soil particles, and moving together in the nest burrows) without antagonism. The mixed C. insana-flavopecta nests were never found within an established enclave of C. insana nests;

rather, they were found only in areas where C. flavopecta occurred. On one occasion, a mixed nest was found ca. 10 m from a C. insana enclave.

These observations and records indicate that temporary social

parasitism might be occurring in these ants. The Conomyrna, in 1 area




45


where a mixed nest was found, were investigated in order to determine if this explanation for the observations could be fact. Temporary social parasitism is well known in ants, especially in the subfamilies Formicinae and Myrmicinae (Brian 1965, Sudd 1967, and Wilson 1971). In the subfamily Dolichoderinae, species of the genus Bothriomyrmex are temporary social parasites on species of Tapinoma (Santschi 1906, 1920). Several species of the genus Azteca are also suspected of being temporary parasites because they have microgynous queens (Wheeler 1910) andy therefore, cannot be expected to employ the usual claustral method of colony establishment.

Methods

The mixed nest of Conomyrma insana-flavopecta, found on 23 May 1975 on the University of Florida Experimental Farm, was excavated on 27 May. The queen, workers, and brood present in this colony were collected and counted. This queen, 25 C. insana workers, 25 C. flavopecta workers, and the brood were maintained in the laboratory.

A 45 x 20 m area was marked off into 1 m squares. A diagrammatic map showing major vegetation and the relative location of all Conomyrma nest craters was constructed on 1-10 June and 26 June. The species of Conomyrma constructing each nest was recorded.


Results

Three hundred twenty-six Conomyrma flavopecta workers and 726 C. insana workers were collected from the mixed nests excavated on 27 May. A single dealate queen of C. insana was captured in this nest. All pupae (ca. 75) mature enough to identify (by head shape) were C. insana




46


rather than C. flavopecta and all callow workers (n=36) were also C. insana. Although the ratio of C. insana to C. flavopecta workers in the original mixed nests was about 7:3, the ratios of C. insana to C. flavopecta in other mixed nests observed at the time of mapping (1-10 June) was about 20:1, judging from the workers appearing at the nest entrance.

Distributional patterns of the craters of the Conomyrma sp. and the mixed nests are shown in Fig. 3.1 and 3.2. The C. insana craters are clustered into 3 localized enclaves, whereas the C. flavopecta craters have a more scattered, but probably not random, spatial arrangement. No Conomyrma nests were found in the area of high, dense grass. While the location and number of craters constructed by C. flavopecta appeared stable, the number of C. insana nests increased from the time the area was first mapped. A new C. insana enclave also developed in the area where mixed nests had previously been located. Only 1 mixed nest could be found on 26 June. This nest had an estimated ratio of 100:1 C. insana to C. flavopecta workers.

The colony taken from the mixed nest was maintained in the laboratory for 1 year. The last individual of the original 25 C. flavopecta workers died 42 days after capture. During this time, the overall mortality rate of the group of C. flavopecta workers appeared due to normal age-related factors. No antagonism between the C. flavopecta and the C. insana workers was detected. Normal colony duties, such as brood tending and food gathering, were equally performed by both. In addition, trophyllactic exchange of food between the 2 species was repeatedly observed. At 6 months, the colony reached a peak population
















0
0


o 0 00




00


0


0


47


0 00 0
0

0


0O
0 0
000t
&0 0 0
00

0 0
00 do-jHeavy grass Original C. insana-flavopecta
0 C. insana-flavopecta
- C. insana
- 0 C. flavopecta
Peach Tree 5 meters


0


Fig. 3.1. Condition of study area 1 June to 10 June 1975, showing (1)
location of original excavated Conomyrna insana-flavopecta nest, (2) additional mixed C. insana-flavopecta nest found
several days after excavation, (3) small, localized enclaves
of C. insana nests, and (4) scattered occurrence of C.
flaVopecta nests.





-0 0
0 00 '.


Soo 00


o oHeavy grass C. insana-flavopecta
T e*C. ~insana
o o **e -OC_ fl avopecta
0 Peach T ree
0 Mmeters
Cb 00 00 6o00


Fig. 3.2.


Condition of study area on.26 June 1975 showing growth of the C. insana enclaves and appearance of a new enclave of C. insana in the same area as the previously present mixed nests, while the unmixed C. flavopecta nests appear to have remained unchanged.


0




48


of an estimated 800 workers. Altogether, based on counts of dead workers, ca. 4,300 C. insana workers were reared in the colony during the year.

Discussion

The mixed nests of C. insana-flavopecta found Florida and the

localized enclaves of C. insana give presumptive evidence that temporary social parasitism followed by hesmosis (budding to form new nests) was occurring in these ants. In addition, the data indicate that C. insana

was parasitic on C. flavopecta. The modus operandi for the adoption of the parasitic queen, elimination of the host queen, and eventual maturation to a C. insana nest should be similar, in a general way, to numerous other temporary parasitic ant species.

Further evidence of temporary social parasitism may be given by the morphology of the queens. The head of C. insana is large, wider than the thorax, while the gaster is small and slender in alate queens. The head of C. flavopecta is of moderate size, about equal in width to the

thorax, and the gaster is large and voluminous in both alate and dealate queens. The morphological form of the C. insana queens is similar to

the modifications already known for many temporary parasitic ants. According to Wilson (1971), queens of temporary parasitic species are all atypical to a degree. In many species, the head is enlarged and the gaster is reduced in relation to the thorax. In addition, the mandibles may be strongly modified to form falcate, piercing organs; or the queen may be small (microgynous) in relation to the worker and/or may have

various other characterisitics such as modified pilosity or integument. In contrast, females of free living species, which practice the normal




49


claustral method of new colony formation, have voluminous gasters and relatively small heads with mandibles, pilosity, and integument not modified from the norm for the genus.

The existence of incipient nests is additional evidence that a species has the normal claustral type of new colony establishment as temporary parasitic queens have lost the ability to begin a new colony on their own (Emery 1909). Incipient nests of C. flavopecta have been found repeatedly in Florida; whereas, no incipient nests of C. insana have been found.

Apparently, the C. flavopecta queen had already been eliminated when the mixed nest was excavated on 27 May as only a C. insana queen was found, and only brood and callows of C. insana were present. In many temporary parasitic ants, adoption of the parasitic queen by a host colony is difficult and not often successful (Sudd 1967). Mixed colonies of these species and their hosts are rare. Nevertheless, the parasitic species may become locally abundant by the process of colony fission (hesmosis) to form an enclave of closely spaced, interconnected nests (Emery 1909). Further adoption of newly mated parasitic queens, additional to the original parasitic queen, is a necessary step in the growth of such an enclave. Colony "towns" of such ants often have been noted by observers. In the case of the allegheny mound building ant, Formica exsectoides Forel, enclaves containing up to 1,700 individual nests have been recorded (Wheeler 1901). Similarly, enclaves of C. insana that exceed 2,000 nests have been located in Leon Co., Fla.

The clustering of the mixed nests in the study area indicates that the process of hesmosis begins to occur even while the nests are still




50


mixed. Obviously, 1 or more additional C. insana queens were present in these mixed nests, as the new C. insana enclave continued to develop even after the C. insana queen found in the original mixed nest was removed.

This is the first report of temporary social parasitism in the genus Conomyrma. It may be noted that mixed nests of Conomyrma have been recorded (as Dorymyrmex) previously by Wheeler (1901) who gave the

remarks of Forel upon finding mixed nests of ants identified as Dorymyrmex pyramicus var. niger Pergande and D. pyramicus var. flavus McCook near Faison, North Carolina. Forel attributed these mixed nests to a queen of each taxon fortuitously associating together when first founding a new colony. Such a chance association of 2 species has never been confirmed for any ants, although reported in older literature (Forel 1874, Wheeler 1901). In the Forel colonies, all alates found were identified as niger except for 1 male which Forel identified as flavus on the basis of color. In the absence of any positive statement, I assume no dealate queens were found to confirm Forel's hypothesis.


Summary

Conomyrma insana and C. flavopecta were found in the same nest in

Florida. Conomyrma insana queen, brood, and workers were in one nest with C. flavopecta workers. Later, clusters of C. insana nests were formed at locations where the mixed nests were found. Conomyrma insana is a temporary social parasite on C. flavopecta.













CHAPTER IV


PREDATION ON FOUNDING QUEENS OF Solenopsis invicta BY WORKERS OF Conomyrma insana


Introduction

The founding queens of the red imported fire ant, Solenopsis invicta Buren, are especially exposed to predation during the period following the nuptial flight and before entering the soil (Whitcomb et al. 1973). The major predators at this point appear to be its own workers and those of other ant species. Hung (1974), Markin et al. (1971), Whitcomb et al. (1973) have reported the dolichoderine ant Conomyrma insana (Buckley) as one of the more effective ant predators of the S. invicta founding queens. Initial observations indicate that these queens have a great deal of difficulty establishing where C. insana occur in north Florida. Although C. insana is sporadic in distribution in this region, it may be an important factor in limiting local populations of the red imported fire ant.

Competition for preferred nesting sites and foraging territories

appears to be the major factor determining the distribution of many formicine species (Bhatkar et al. 1972, Leston 1973, Van Pelt 1966). Nowhere is this more true than in the aggression of the formicine species toward founding queens. Predation of founding queens alighting in a territory dominated by a given ant species would reduce or eliminate future competition for available resources in favor of the dominant species. Carroll and Janzen (1973) stated that founding queens were


51




52


usually killed by worker ants of the same or other species. This strong statement is supported by research of several workers (Janzen 1967, Levieux 1971, Marikovsky 1961, Pontin 1960). Janzen (1973) presented research on Pseudomyrmex ferruginia F. Smith and described in detail how workers destroyed new queens of their own species. Hung (1974) analyzed the refuse pile of a Conomyrma insana colony and reported the recovery of discarded heads of the queens of 5 species of myrmecines, including those of Solenopsis invicta queens.

The predator-prey relationship of Conomyrma insana workers and the founding queens of Solenopsis invicta is of special interest because of the apparent complexity of the encounters leading to the death of the founding queens or the rare successful escape from predation by the latter. The first objective of the present research was to gain information of the general behavior pattern of a dolichoderine ant worker attacking a myrmecine founding queen. Furthermore, such data are needed on the predator-prey relationship of these 2 species, in particular, to evaluate the effectiveness of C. insana as a predator and its potential in limiting the local increase of S. invicta. Conversely, this information was collected to determine the ability of S. invicta to survive within an area under these high stress conditions.


Methods

This research consisted of field observations of the behavior(s)

of the predator (C. insana workers) and prey (S. invicta founding queens) under natural conditions. These events were observed in detail in Leon county in northern Florida in areas where Conomyrma insana nests were present. Of approximately 2,000 Solenopsis invicta founding queens








observed while under attack, 240 were watched from the time they alighted until they were either destroyed by predators or escaped. When possible, the action was verbally described and recorded on tape as it occurred. Most observations were made on bare ground, either on dirt roadways or in fallow fields. Additional observations were made in mowed fields with sparse vegetation cover. This report is based on observations of queens which alighted naturally; information on queens which had been captured and handled was discarded.

In 1972, the effectiveness of predation by C. insana was determined. Nests of C. insana were examined and excavated to determine the numbers of S. invicta queens taken as prey. In addition, five 3 x 8 m plots were randomly located in an open field. These plots were examined at 1/2 hr intervals from 1400 to 1830 on the dates of heavy S. invicta nuptial flights. Predation of S. invicta queens within these plots was noted and recorded. For 3 consecutive days following nuptial flight, this field was searched for additional S. invicta females on the surface

and for signs of brood chamber construction.


Results and Discussion

A sumaary of the responses of S. invicta queens to attack(s) by C. insana workers is shown in the flow diagram in Fig. 4.1. While the responses of the queen to attack are varied, they can be separated into

artificial categories which recur frequently.

An abridged summary of a single attack sequence by C. insana workers on 1 S. invicta queen is presented in Table 4.1. This was taken from observations made 21 June 1972 of a queen that had alighted at 1626. The nuptial flight began after 2 days of intermittent storms













S. invicto 9 _n__n Q insan Y Token as Prey
C1n=al) by r,. Insona

Aliglit, E sc p e A ttem pts 1

Mandibular Defente (T Flexing and Twisting
Cease Motion N=2' Concealment
Shed Wings & ~(tihr, .E scope





I. PREDATION
----------------- ------------------------------------Random Searc Construct Brood Chamber
for Nesting Site
Cop rood Chamber


Shed Wings

1st Workers Emerge tL SUCCESSFUL NEST-FOUNDING 1st drysE
(.30 days)


Fig. 4.1. Flow diagram of responses of 240 Solenopsis invicta founding
queens to initial contact, pursuit, and attack by Conomyrma
insana workers. The contrasting behavior leading to nest
founding is included.




55


TABLE 4.1. Abridged summary of observation of attack by Conomyrma insana
(Ci) on a founding queen of Solenopsis invicta (Si beginning
at 1625 on 21 June 1972. Leon Co., Fla.



Elapsed
time (min) Observations


0 Si alights, immediately moves under debris and into
soil crevices.

4 Met by 1st Ci; duration of contact momentary. Si
movement rapid and erratic.

9 Si reverted to normal rate of movement.

13 Ci attacked (n=2) and pursued (n=l). Si hid under
leaf.

20 Si left leaf. Ci attacked Si at base of wing (n=1),
head (n=l), and thorax (n=lT? Si fled.

21 Si has traveled 8.5 m from alighting point, wings
intact.

30 Ci attacked (n=l) Si at head. Si fled.

33 Ci attacked (n=9), Si shook Ci loose by rapidly
vexing body.

34 Ci attacked (n=4), Ci clinging to wing, Si shook Ci
Toose. Si climbed blade of grass, attempted to fly, failed.

35 Ci pursuing (n=4), Ci attached (n=3) to Si. All Ci
dislodged from Si.

36 Ci clinging to thorax and wing (n=2). Si dislodged
C11.

38 Ci attacked (n=4), Si motionless (6 sec), fled. Ci
pursued and attacked (n=2).
39 Ci pursuing (n=7) and clinging to Si head (n=l).

40 Si moving rapidly, attempting to break off wings. Ci
Thn=3) pursuing and biting at legs and wings. Si flexed abdomen rapidly, curled into "C" position and bit at Ci. Si had traveled 16.8 m since alighting.


P- #




56


TABLE 4.1--Continued.

Elapsed
time (min) Observations

41 Si knocked left forewing off. Ci cling to Si legs.
Si bit and killed Ci (n=W). Si now moving rapidly, Ci pursuing (n=9). Ci clinging (n=3), Si shook all oh, still moving rapidly. Ci clingingTn=3) and pursuing (n=ll). Si curled into "C" position and bit at Ci, then moved away with Ci in pursuit (n=14).

48 Si still moving with no pursuit.

50 Ci attack (n=2) Si. Si shook off attack.

51 Si under leaf momentarily, then emerged.

52 Si attacked by Ci (n=). Si curled into "C" position
anid bit Ci. Ci~(n=17) pursuing Si. Si shook off pursuit and attack, moved under clod of soil.

53 Si emerged, immediately attacked by Ci (n=). Ci atthcked (n=3). Si dislodged attackers. Ci clinging to wing. Si climbed blade of grass, immediately descended with Ci (n=l) still clinging to wing.

54 Ci (n=) attacked Si at right metathoracic leg. Si
curled into "C" position and bit Ci, forcing it to release. Ci (n=2) attacked, forced away by Si.

55 Ci attacked (n=2), Si curled into "C" position and
bit at Ci. Si moved with Ci still clinging to leg. Ci (n=7Tbiting and pursuing, Si flexing abdomen and thorax, intermixed with assuming "C" position and biting.

58 Ci (n=3) attacking Si. Si shook loose Ci. Additional
Ci (n=8) attacked. Si dislodged all Ci_ by shaking and biting.

60 Si moved forward and went under a leaf.

61 Si emerged from under the leaf and was immediately
attacked by Ci (n=). Si curled into "C" position and bit Ci until it left. Ci (n=3) in pursuit of Si. Ci i=3) caught Si. Si bit at Ci on leg. Additional Ci attacked (=9). Si assumed "C" position and bit Ci clinging to leg but did not dislodge.




57


TABLE 4.1--Continued.

Elapsed
Time (min) Observations


63 Si moving with Ci clinging. Additional Ci in pursuit
i=7) and attackiTng (n=15+) Si moving rapidly while shaking and flexing body. Stepped, assumed "C" position bit and killed Ci (n=). Ci attacking (n=4).

66 Ci (n=3) clinging to Si. Si dislodged Ci amd moved
forward rapidly. Ci (i=4) Ypursuing, Ci~Tn=l) clinging to leg. Si curled into "C" position and bit Ci.

68 Ci (n=12) attacking. Si halted, assumed "C" position,
Et Ci (n=), now moving with Ci continuing attack. Si biting at Ci killed 1.

72 Si nearly subdued (= taken as prey by Ci). Little
response by Ci and only feeble attempts to escape.

75 Ci have severed left prothoracic and both metathoracic
legs.

76 Ci have severed left mesothoracic leg.

77 All legs severed.

80 Ci moving appendageless Si to nest. Si alive and trying
to bite Ci. Distance from alighting point = 53 m.




58


resulting in 18 cm rainfall. The temperature at 1414 was 35 C and the wind velocity varied from 8 to 24 kmh. The alates of S. invicta emerged from the mounds at 1331. The first queen observed returning from the nuptial flight alighted at 1450.


Consistencies and Variations in Behavior

Detection. When a S. invicta queen alighted in an area heavily

populated by C. insana, a period, varying from a few seconds to several minutes, elapsed prior to initial contact by C. insana. During this period, if sufficient, the queen shed her wings and commenced searching for a suitable site to construct the brood chamber. Detection of the S. invicta queen by C. insana workers occurred any time after the queen alighted, depending upon the proximity of the workers. The first contact with the queen was made by a single worker approaching from the front or side with mandibles spread and maintaining only momentary contact with the queen. It is probable that the queen was marked with an identifying compound by the C. insana worker. Following attacks were not comparable to the initial attack in that subsequent attacks were progressively more frequent and prolonged. In later attacks, the workers remained attached to the queen until dislodged. In the first approach, penetration by the mandibles was never observed and contact was made at any portion of the body.

During a short span of time following the initial contact, the

queen was unmolested by C. insana workers. Again, the queen may attempt to remove the wings (if not previously shed), but the usual reaction of the queen was to flee from the point of attack. In most observed cases, this was followed by a halcyon period as evidenced by the less erratic




59


and hurried rate of movement. Subsequently, the queen would renew the search for a suitable location in which to excavate the brood chamber.

Secondary contacts. The time between first contact and subsequent attacks varied from 15 sec to 35 min depending on the local population densities and the degree of surface activity of C. insana. These secondary contacts differed from the first encounter in several respects. The C. insana attacks were progressively more frequent and prolonged; the workers tended to cling to the queen until dislodged. Pursuit of the queen by workers also occurred, although individual workers did not take part continuously. The intensity of pursuit was again dependent upon the local density of C. insana. As the attack progressed, the number of individuals of C. insana workers taking part increased up to the time the queen was taken as prey.

Queen response to attack. The responses of the S. invicta founding queens to the attack by the C. insana workers can be divided into 5 basic reactions: (1) escape, (2) concealment, (3) cessation of all movement, (4) mandibular defense, and (5) rapid flexing or shaking of the body. All or several of these responses were utilized by the queens in attempting to avoid or fight off the attack of C. insana worker(s).

Escape attempts from the attack of C. insana workers occurred in 2 forms: (1) flight and (2) surface movement. If a queen was attacked prior to shedding of the wings, she could avoid predation by further flight. This was accomplished by less than 2% of the queens under observation. The queens, when under attack, crawled up vegetation or any other available elevation in an attempt to gain the height necessary to become airborne. Queens not under attack were seldom observed to





60


attempt flight after landing. The response most often elicited by the attack of C. insana was attempted escape on the ground surface. The distance covered varied with the topography, presence of vegetation, or other natural barriers. The escaping queen usually covered distances of 20 to 80 cm per min; some covered distances of 3 m or more. This response was often temporarily successful in avoiding immediate predation by C. insana workers.

While under attack or when unmolested by C. insana workers, the S. invicta queens were continuously moving under twigs, leaves, or into soil cracks. This appeared to be a concealment response, even though it may be difficult to distinguish from the normal instinct of the queen to commence construction of the brood chamber as soon as possible after landing. Hiding was usually effective in momentarily avoiding predation. If the queen abandoned the protection of the leaf or twig, C. insana workers immediately resumed their attack.

In further attempts to avoid predation, the queen often remained motionless briefly (6-10 sec) when attacked or investigated by a C. insana worker; afterwards, attempted escape or defense methods were used. In a number of observed cases, the cessation of motion was temporarily successful as the C. insana worker either did not commence further attack or broke off the attack and left the vicinity of the female.

The queens also attempted to take a defensive stand against the attack of the C. insana workers by aggressively biting. This was most often seen when the S. invicta queen had failed to dislodge attacking workers and 1 or more had seized an appendage. To use the mandibles effectively, the queen would lie on either side and assume a "C" position




61


with the tip of the abdomen curled ventrally toward the mandibles. In this position, the S. invicta queen could use the mandibles effectively for defense. On a number of occasions, a queen severed an abdomen from

a C. insana worker's thorax at the pedicel or decapitated a worker hanging onto 1 of her legs. The worker's decapitated head often remained attached to the seized appendage. This method of defense was effective against an individual worker, although the queen could not remain in the position long as she was vulnerable to further attack by other C. insana workers.

The S. invicta queens frequently attempted to dislodge clinging C. insana workers by rapid shaking or flexing of the abdomen. This was used by all queens under attack by C. insana workers. The queen was always successful in dislodging attacking workers by this means until she became exhausted or was overcome by the mass attack of the C. insana.

Predation and disposal of queens. When the queens were preyed upon successfully by C. insana workers, the appendages were removed by a combination of tugging and biting at the joints. Appendageless queens were dragged, while still alive, to nests by the workers.

On the day following predation, bits of discarded body parts and

heads of S. invicta queens were found on the periphery of the C. insana nests. After a major nuptial flight of 21 June 1972, a series of 24 nests were excavated on each of 2 successive days. On the 1st day, live appendageless queens were found in all nests. The fewest found in a single colony was 7 and the most was 18. A total of 329 S. invicta queens was recovered from the C. insana nests. An additional 83





62


queen heads were counted from the discarded debris around the colonies that were excavated. On the 2nd day, only 1 C. insana nest had live appendageless S. invicta queens (n=2).


Effectiveness of C. insana as a Predator

Nests of C. insana tend to be concentrated in localized areas. The greater the number of C. insana nests within the area, the more intense the attack on fire ant queens. Mortality of S. invicta queens did not depend on the effectiveness of the bite or toxin of the individual workers, but rather on the ability of C. insana to continue the attack until the S. invicta queens were taken as prey. When the process was interrupted prior to removal of appendages, the queen survived and laid eggs. Where the process was followed from beginning to end, however, only 6 queens out of 240 contacted by C. insana workers managed to escape destruction by any means: 4 by flying away from the C. insana area and 2 by escaping on the ground surface to outside of the foraging range of C. insana.

In 1972, an effort to quantify predation by C. insana was made in five 3 x 8 m plots. On 9 May, 613 postnuptial S. invicta queens were counted in these plots; 60% of these were completely subdued by the C. insana workers, 36.4% were under continuous attack, and 3.6% were momentarily unmolested.

A field at Tall Timbers Research Station was searched for S.

invicta queens for 3 successive days following a nuptial flight. On 10 May, 19 S. invicta queens were being dragged to the C. insana colony by a coordinated effort of workers. Only 4 S. invicta queens were still under attack by C. insana. On 11 and 12 May, no live queens were found




63


even though the area was extensively searched. An identical search was conducted following the nuptial flight on 21 June. On 22 June, 24 S. invicta queens were found that had been subdued by C. insana workers and were being taken into the colony. No queens were found on 23 and 24 June within the C. insana enclave.


Summary

Solenopsis invicta Buren founding queens are especially vulnerable to predation during and following the nuptial flight. Conomyrma insana

(Buckley) worker, are effective predators of the queens in some areas of northern Florida. The respective behaviors that occurred during the

confrontation of the 2 species were studied in the field in Leon county of northern Florida. Detection of a founding queen alighting within a C. insana enclave by workers varied from a few seconds to several minutes. The founding queen's reactions to attack by C. insana consisted of 5 basic responses: (1) escape, (2) concealment, (3) cessation of

movement, (4) mandibular defense, and (5) rapid flexing or shaking. Less than 3% of the queens observed alighting within a C. insana enclave were successful in escaping predation. No S. invicta queens were found that survived predation by C. insana workers long enough to construct a brood chamber and begin a new colony within an area occupied by C. insana.














CHAPTER V


SOURCES OF CARBOHYDRATES UTILIZED BY Conomyrma insana


Introduction

The colonies of the ant Conomyrma insana (Buckley) consist of many clustered nests that form enclaves. Conomyrma insana is omnivorous, however, it is a voracious predator (Van Pelt 1966, Hung 1974, Markin et al. 1971, Whitcomb et al. 1972). Way (1968) suggested that, in terms of food

supply, any species of omnivorous ant, in order to maintain continuously large colonies, had to change foraging grounds periodically as doryline

ant do, or to tend Homoptera, which provide honeydew to carry the ants over periods when animal prey is scarce. The workers of C. insana apparently derive a large portion of the energy required to maintain their colonies from honeydew excreted by homopterans and nectar from extrafloral and floral structures of plants.

Two species of Conomyrma, C. insana and C. flavopecta (M. R. Smith), are currently recognized in the southeastern United States (Snelling 1973). Prior to Snelling's revision, the North American species of Conomyrma

were lumped as Dorymyrmex pyramicus (Roger) or Conomyrma pyramicus. Thus, the information contained in reports of aphid tending by D. or C. pyramicus is difficult to assign correctly to the presently accepted species of Conomyrma. Nielsson et al. (1971) reported the association of several aphid species with C. pyramicus, but did not attempt to establish if a mutualistic relationship occurred with the species of aphids listed. Van


64





65


Pelt (1966) noted that Dorymyrmex pyramicus (= Conomyrma insana based on Van Pelt's description of the nesting habits) tended 2 aphid species. Research by Odum and Kuenzler (1963), Wiegert et al. (1967), and de la Cruz and Wiegert (1967), using 32P to trace the plant-aphid-ant food chain, demonstrated that D. pyramicus obtained honeydew from aphids of the genera Aphis and Macrosiphum. Jones (1929) listed 3 aphids (Aphis sp., Chaitophorous negundinus Thos, and C. populella Gill. and Pal.) tended by D. pyramicus.

The objective of this research was to determine the sources of carbohydrates available to C. insana within their foraging area.

Methods

Investigations were conducted at Tall Timbers Research Station,

Leon Co., Florida from July 1972 through August 1975. Eight locations with high populations of Conomyrma insana were selected. The flora within and adjacent to the C. insana enclaves were examined for Homoptera tended by ants and ants feeding from floral and extrafloral nectaries. With Homoptera, determination of tending was based on observing the workers antennuate an individual and subsequently receive a droplet of exudate. The presence of the ant on the same plant as the Homoptera was not accepted as prima facie evidence of ant-aphid association. All new Homoptera and their host plants, and plants with nectaries were collected and identified.

Beginning in 1973, a distribution map was constructed annually of the C. insana nests around a solitary short leaf pine located in an open field. To construct the map, the distance of each nest from the base of the pine was measured. Using the pine tree as a base point, the compass angle of each nest was also determined.




66


Results and Discussion

Homoptera Tended by Conomyrma insana

Sixteen species of aphis (Table 5.1) and 10 other species of Homoptera (Table 5.2) were tended by Conomyrma insana. These homopterans were colonizing plants within the foraging area of the ant. The Homoptera were collected from 13 plants families; these were Pinaceae, Gramineae, Fagaceae, Chenopodiaceae, Amaranthaceae, Rosaceae, Caesalpiniaceae, Fabaceae, Onagraceae, Labiatae, Scrophulariaceae, Rubiaceae, and Compositae.

The aphids Cinara watsoni Tissot, Sipha flava (Forbes), Aphis ossypii Glover, A. craccivora Koch, A. spriaecola Patch, Cerosipha rubifoli Thomas, and Hysteroneura setariae (Thomas) were common in all areas where C. insana occurred. Rhopalosiphum maidis (Fitch) were numerous on Paspalum boscianum Flugge and corn, Zea mays L., in only one area. Another aphid, Myzocallis bella (Walsh), was found only on running oak, Quercus pumila Walter, in one field. The remaining aphids presented in Table

5.1 were rarely collected.

The aphids Dactynotus tissoti (Boudreaux), collected on Solidago sp. (Goldenrod); D. ambrosine (Thomas), collected on Solidago sp. and Pyrrhopappus carolinianus (Walt.) DC.; D. helianthicola Olive, collected on Helianthus augustifolius L.; and D. tutaiae Olive, collected on Ambrosia artemissiifolia L. (Ragweed) were within the foraging areas of C. insana. The ant was often found on the same plant as the aphids, but there was no evidence that these aphid species were tended by C. insana.

Adelges sp. and Toumeyella parvicornis (Ckll.) were both tended by C. insana on short leaf pine. Both were attractive to the ant, and large numbers of foraging workers were constantly present on the pine.








Table 5.1. Aphids tended by Conomyrma insana and the plant host(s) that the aphids were found on. Tall
Timbers Research Station, Leon Co., Fla. 1972-75.



Plant Host
Aphid Species Family Species (Common Name)


Cinara watsoni Tissot Cinara melaina Boudreaux Lachnochaitophorus obscurus (Tissot)


Sipha flava (Forbes)




Aphis coreopsidis (Thomas) Aphis gossypii Glover




Aphis oestlundi Gill Aphis spiraecola Patch


Pinaceae Pinaceae Fagaceae Fagaceae Gramineae Gramineae Gramineae Compositae Caesalpiniaceae Rubiaceae Compositae Onagraceae Chenopodiaceae


Compositae


Pinus echinata Miller (Short leaf pine) Pinus echinata Miller (Short leaf pine) Quercus nigra L. (Water oak) Quercus pumila Walter (Running oak) Eragrostis spectabilis (Pursh.) Steud (Love grass) Digitaria villosa (Walt.) Pers. (Crabgrass) Paspalum setaceum Michx. Bidens bipinnata L. (Beggar ticks) Cassia obtusifolia L. (Sickle-pod) Richardia scabra L. (Mexican clover) Gnaphalium obtusifolium L. (Everlasting) Oenothera sp. (Evening primrose) Chenopodium ambrosioides L. (Mexican tea) Eupatorium capillifolium (Lam.) Small (Dog-fennel)








Table 5.1.--Continued.


Plant Host
Aphid.Species Family Species (Common Name)


Aphis spiraecola Patch Aphis fabae Scop. Aphis craccivora Koch


Aphis sp.


Schizaphis sp. Cerosipha rubifoli Thomas


Compositae Compositae Compositae Caesalpiniaceae Caesalpiniaceae


Fagaceae


Amaranthaceae Rubiaceae Scrophulariaceae Compositae Compositae


Labiatae Rosaceae


Pyrrhopappus carolinianus (Walt.) DC Conyza canadensis (L.) Crong. (Horseweed) Ambrosia artemisiifolia L. (Ragweed) Cassia fasciculata Michx. (Partridge pea) Cassia nictitans L. (Wild sensitive plant) Quercus nigra L. (Water oak) Amaranthus hybridus L. Richardia scabra L. (Mexican clover) Agalinis fasiculata (Ell.) Rat. Ambrosia artemisiifolia L. (Ragweed) Eupatorium capillifolium (Lam.) Small (Dog-fennel) Trichostema dichotomum L. (Blue curls) Rubus cunneifolius Pursh. (Sand blackberry)


ON' CO








Table 5.1.--Continued.


Plant Host
Aphid Species Family Species (Common Name)

Hysteroneura setariae (Thomas) Gramineae Eragrostis spectabilis (Pursh.) Steud. (Love grass)

Gramineae Digitaria violascens Link (Crabgrass)

Gramineae Paspalum boscianum Flugge

Gramineae Paspalum plicatulum Michx.

Gramineae Paspalum setaceum Michx.

Gramineae Cenchrus echinatus L. (Sandspurs)

Gramineae Andropogon virginicus L. (Broom sedge)

Rhopalosiphum maidis (Fitch) Gramineae Paspalum boscianum Flugge

Gramineae Zea mays L. (Corn)

Myzocallis bella (Walsh) Fagaceae Quercus pumila Walter (Running oak)

Neosymdobius lutens Tissot Fagaceae Quercus nigra L. (Water oak)


t~o








Table 5.2. Homoptera, other than aphids, tended by Conomyrma insana and the host plant(s) they were found
on. Tall Timbers Research Station, Leon Co., Fla. 1972-75.



Plant Host
Species Family Species (Common Name)


Phylloxieridae

Adelges sp. Coccidae


Saisettia sp. Pulviaria urbicola Ckll.


Toumeyella parvicornia (Ckll.) Coccus hesperidum Linn.


Coccus sp.

Pseudoccidae

Oracella sp.


Pinaceae


Compositae Compositae


Chenopodiaceae


Pinaceae Fabaceae Fabaceae Compositae


Pinus echinata Miller (Short leaf pine)


Eupatorium capillifolium (Lam.) (Don-Fennel) Pyrropappus carolineaus (Walt.) Chenopodium ambrosiodes L. Pinus echinata Miller (Short leaf pine) Glycine max Merr. (Soybean) Lespedeza angustifolia (Pursh.) Ell. Ambrosia artemisiifolia L. (Ragweed)


Quercus virginiana Miller (Live oak)


-


Fagaceae








Table 5.2.--Continued.


Plant Host
Species Family Species (Common Name)

Eriococcidae

Eriococcus quercus (Comst.) Fagaceae Quercus virginiana Miller (Live oak)

Eriococcus smithii Lobdell Gramineae Andropogon virginicus L. (Broom sedge)

Kermesidae

Kermes sp. Fagaceae Quercus nigra L. (Water oak)

Membracidae

Unidentified sp. Compositae Ambrosia artimisiifolia L. (Ragweed)





72


Coccus hesperidium Linn annually built up heavy infestations in a soybean field. The scale was also tended by Solenopsis invicta Buren on the interior of the field.

The mealybug Oracella sp. was found only on a single live oak tree, uercus virginiana Miller. This population persisted in large numbers on the oak.

The first collected specimens of Eriococcus smithii Lobdell were taken from 7 clumps of broom sedge, Andropogon virginicus L. in a shallow ditch between a forest road and a soybean field. By September 1974, colonized broom sedge clumps had increased to 43, spaced out over approximately 30 m within the shallow ditch.

Kermes sp. were common on water oak, Quercus nigra L., and were

heavily tended by C. insana. A species determination was impossible as they were so heavily parasitized that no mature specimens were found. Nectaries of Plants

Conomyrma insana constantly fed from the extrafloral nectaries of Cassia nictitans L. and C. fasciculata Michx. and to a lesser degree on C. obtusifolia L. Cassia nictitans was the most prevalent plant at most C. insana locales. In cultivated fields, mexican clover, Richardia scabra L., was a common weed. The floral nectaries of this plant were frequently visited by C. insana.


Influence of Carbohydrate Food Source

The influence of a carbohydrate food source on selection of nesting sites by C. insana is illustrated in Figs. 5.1-5.4. In this field, vegetation was composed of scattered clumps of broom sedge, Andropogon virginicus L. (ca. 2.1 clumps per m2), with sparse occurrence of Cassia




73


1 meter


0
S
-


0


0

* 0


0


gO*
0*e


.0


0


0


0
0
0 0


0


S


0


Distribution of Conomyrma insana nests in 1973 around a short leaf pine, Pinus echinata, located in an open field at Tall Timbers Research Station. The short leaf pine was heavily infested with Cinara watsoni.


0


Fig. 5.1.


N


V





74


1 meter


0


0 0


N












Uf


0

0


0


0

0
90*


0


0
0
0


0


0



0

0


0


0


0


0


0


0


0 0


0


0


Distribution of Conomyrma insana nests in 1974 around a short leaf pine, Pinus echinata, located in an open field at Tall Timbers Research Station. The short leaf pine was heavily infested with Cinara watsoni.


0
0


0
0
0


0


0


Fig. 5.2.





75


S


0


1 meter


0
0


0
S


S
0
0


S


..
0 5
* S


0.


* 0
*0


0


0


a


S
0


0


0


0


Fig. 5.3. Distribution of Conomyrma insana nests in 1975 around a
short leaf pine, Pinus echinata, located in an open field at Tall Timbers Research Station. The short leaf pine was
heavily infested with Cinara watsoni.


N


0





76


f


0


0


0


0


1 meter













0








0


0


0


S
*0
* 0


0
is


0
0


0


0


0 .


0


0 0


Distribution of Conomyrma insana nests in 1976 around a short leaf pine, Pinus echinata, located in an open field at Tall Timbers Research Station. The short leaf pine was heavily infested with Cinara watsoni.


N


a


Fig. 5.4.




77


nictitans L. and some composites. The distribution of the plants were uniform with most of the soil surface exposed to sunlight. An occasional short leaf pine, Pinus echinata Miller, was present in the field; an enclave of C. insana surrounded each of these trees. The only attractant in each of the pine trees was the pine aphid, Cinara watsoni. The workers of C. insana, from the nests surrounding the trees, continuously tended these aphids.

Throughout the 4 years of monitoring this population of ants, nests were maintained around the trees as diagrammed in Figs. 5.1-5.4. No other major carbohydrate food sources were near this location, although an occasional group of the aphid Hysterouneura setariae was on broom sedge. Apparently, while the presence of the aphid on the pine tree was instrumental in determining the occurrence of C. insana, the quantity of carbohydrates available from this and other sources was insufficient to maintain a larger colony. This hypothesis is supported, in part, by the recorded increase of the number of nests from 31 in 1973 to 50 in 1976, while there was corresponding increase in the pine from ca. 2 m to 4.5 m in height at the same time. Theoretically, the pine would be capable of supporting a larger population of aphids which, in turn, would support a small increase in the ant colony.

Summary

Twenty-six species of Homoptera were tended by Conomyrma insana

(Buckley). The Homoptera were in the families Aphididae (10), Phylloxieridae (1), Coccidae (5), Pseudoccidae (1), Eriococcidae (3), Kermesidae (1), and Membracidae (1). Conomyrma insana also fed from the nectaries of Cassia nictitans L., C. fasciculata Michx., C. obtusifolia, and Richardia scabra L.





78



Conomyrma insana construct their nests adjacent to carbohydrate food sources. The population size in an area may be proportional to the quantity of carbohydrates available.














CHAPTER VI


PRESENCE OF Spissistilus festinus TENDED BY ANTS:
A FACTOR AFFECTING EGG PREDATION BY ANTS IN SOYBEANS Introduction

Ants are important predators on arthropod pests in agroecosystems (Cook 1904, Pierce et al. 1912, Whitcomb and Bell 1964, Whitcomb et al. 1972). Many predaceous ants supplement their diet by collecting honeydew from Homoptera (Carroll and Janzen 1973). This stationary and renewable food source can be an attractant resulting in locally high ant population densities that may be maintained throughout the year (Finnegan 1974). The presence of Homoptera could consequently influence the effectiveness of ants as predators.

In 1975, ants were found tending nymphs of the threecornered alfalfa hopper, Spissistilus festinus (Say), on soybeans in northern Florida. Since the ants were observed more often on plants with the nymphs than on plants without the nymphs, I hypothesized that predation on soybean pests increased in the presence on nymphs. My objective was to test this hypothesis by examining predation on eggs of the soybean looper, Pseudoplusia includens Walker, under field conditions.


Methods

Experiments were conducted between 15 September and 15 October, 1975 in a 15-ha field of Bragg soybeans in Alachua County, Florida. Initially, the field was surveyed for ant species tending nymphs of


79




80


Spissistilus festinus. Two areas were then selected for experimentation, an 1100-M2 area in the interior of the field in which the nymphs were being tended primarily by Solenopsis geminata (F.) (designated as area 1) and a 45-m2 area at the perimeter of the field in which the nymphs were being tended by Conomyrma insana (Buckley) (designated as area 2). On 3 and 15 October, between 1200 and 1700 hr, nymphs were counted in 10 randomly selected 3.05-m sections of soybean row in the S. geminata area and three 3.05-m sections of soybean row in the C. insana area. On both dates, the ant species present on each plant and the number of soybean plants in each 3.05-m section of soybean row were recorded. An analysis of variance (Steel and Torrie 1960) was performed on the squareroot-transformed counts of nymphs. Data were arranged in a 2 x 2 factorial design (= 2 areas of the field x 2 dates).

In each of the 2 areas of the field, 14 plants were selected, 7

with 1 or more nymphs tended by ants and 7 without nymphs. The absence of nymphs was confirmed by shaking each plant over a 0.58-m2 white cloth. Previous tests had established that nymphs were easily shaken from the soybean plants. A trifoliate leaf on the upper, middle, and lower 1/3 of each plant was marked with a small spot of water-soluble paint. Eggs of the soybean looper, Pseudoplusia includens, were obtained from a laboratory colony maintained by N. C. Leppla, Insect Attractants, Behavior and Basic Biology Research Laboratory, ARS, USDA, Gainesville, Fl. Three eggs were attached with casein glue to each marked leaf (1 on the upper surface of each leaflet). Therefore, 9 eggs (3 at each height on the plants) were attached to each of the 14 plants in an area for a total of 126 eggs/area. Twenty-one of the eggs were at





81


each height on plants with nymphs, and 21 eggs were at each height on

-plants without nymphs. The fraction of eggs missing out of each group of 21 eggs was recorded after the eggs had been in the field for 24 hr. Experiments were repeated during five 24-hr periods. An analysis of variance was performed on the arcsine-transformed data arranged in a 2 x 2 x 3 factorial design (= nymphs present or absent x 2 areas of the field x 3 heights on the plants) replicated on 5 dates.

On 14 October, an additional experiment was conducted in three 45m 2 areas of the field in which the nymphs were being tended by C. insana. Experimental procedures were the same as those used in the earlier experiments except that 1 set of 14 plants (7 with 1 or more nymphs tended by C. insana and 7 without nymphs) was selected in each of the 3 areas of the field. In addition, the eggs were examined after

they had been in the field 4, 8, and 24 hr. An analysis of variance was performed on the arcsine-transformed data (numbers of eggs missing/21 eggs) arranged in a 2 x 3 x 3 factorial design (= nymphs present or absent x 3 elapsed periods of exposure x 3 heights on the plants) replicated in 3 areas of the field.


Results and Discussion

Workers of Solenopsis geminata, Conomyrma insana, C. flavopecta (M. R. Smith), Iridomyrmex pruinosus (Roger), and Pheidole morrisi Forel tended threecornered alfalfa hopper nymphs in the soybean field in which the investigations were conducted. Although tending of S. festinus nymphs by ants in soybeans has not previously been reported, Jordan (1952) observed an unspecified ant species 'tending nymphs in alfalfa. In the experimental field, nests of Pheidole morrisi and Iridomyrmex









pruinosus were uncommon, and workers of each species were observed tending nymphs on only 2 occasions. Conomyrma flavopecta and Solenopsis geminata occurred over most of the interior of the field. Workers of S.

geminata foraged on soybean plants from underground trails radiating from their mounds and opening at the bases of plants on which nymphs were being tended. Three clusters of Conomyrma insana nests were present along I edge of the field. Each enclave covered ca. 45 m2 and extended ca. 3 m into the field. No other nests of C. insana were found in the field.

In the interior of the field (area 1), workers of S. geminata or, occasionally, workers of C. flavopecta were observed tending threecornered alfalfa hopper nymphs. On 3 and 15 October, workers of S. geminata were found on 94.6% of the plants with nymphs and on 15.0% of the plants without nymphs. Workers of C. flavopecta were on the remaining 5.4% of the plants with nymphs and on 0.8% of the plants without nymphs. Within a C. insana enclave (area 2), C. insana was the only ant species observed tending threecornered alfalfa hopper nymphs. On 3 and 15 October, at least 1 C. insana worker was found on every plant with

nymphs. Conomyrma insana workers were present on 83.0% of the plants without nymphs. Conomyrma insana workers and, to a lesser extent, S.

geminate workers were present on plants without nymphs but at lower frequency than on plants with nymphs.

Estimates of the densities of the threecornered alfalfa hopper

nymphs are given in Table 6.1. Although nymphs were found on 50.4% of all plants examined on 3 and 15 October, only a single nymph was present on 77.7% of the plants with nymph(s). Densities of nymphs were not







Table 6.1.


Numbers of Spissistilus festinus nymphs in 2 areas of a Florida soybean field: one area in which nymphs were tended by Solenopsis geminata and another in which nymphs were tended by Conomyn a insana.


% of plants examined with:
Ant Survey 0 1 2 3-5 _Nymphs/plapt Nymphs/ha
Species date nymphs nymphs nymphs nymphs x + 95% CI ka) x + 95% CI (a)


S. geminata Oct. 3 36.0 49.0 11.0 4.0 0.84 + 0.22 111.3 + 32.7
(area 1~T
Oct. 15 67.0 28.0 5.0 0.0 0.38 + 0.20 63.8 + 38.1

C. insana Oct. 3 30.0 43.3 20.0 6.7 1.10 + 0.90 140.7 + 140.6
(area 2)
Oct. 15 56.7 36.7 6.7 0.0 0.50 + 0.25 63.5 + 42.0


(a)mean + 95% confidence interval.





84


significantly different between area 1 and area 2 (p>0.1), but significantly more nymphs were present in both areas on 3 October than on 15 October (p
The percentage of soybean looper eggs missing after 24 hr from plants in area 1 and 2 are presented in Fig. 6.1. A significantly greater percentage of eggs was missing from plants in area 2 (88.6%) than from plants in area 1 (57.0%) (p<0.005). Within area 1, a significantly greater percentage of eggs was missing from plants with nymphs (77.1%) than from plants without nymphs (36.8%) (p<0.005). Within area 2, a greater percentage of eggs also was missing from plants with nymphs (90.8%) than from plants without nymphs (86.3%), but the difference between the 2 sets of plants was not significant (p>0.1).

The percentages of eggs missing after 4, 8, and 24 hr from plants within the boundaries of 3 C. insana enclaves are presented in Fig. 6.2. Significantly greater percentages of eggs were missing from plants with nymphs than from plants without nymphs after 4 hr (36.0% vs. 17.5%) and after 8 hr (65.6% vs. 42.9%) (p>0.005). After 24 hr the percentage of eggs missing from plants with nymphs (96.8%) was not significantly different from the percentage of eggs missing from plants without nymphs (92.8%) (p>O.l).

In the interior of the field, rates of predation were greater on plants with S. festinus nymphs tended by ants than on plants without nymphs. I hypothesize that workers of S. geminata were attracted to specific plants by the presence of the nymphs, and that, at any point in time, more ants were foraging on plnats with nymphs than on plants






















Percentages of Pseudoplusia includens eggs missing after 24 hr exposure at 3 heights on soybean plants with or without Spissistilus festinus nymphs. Plants were in an area where Solenopsis geminata was the dominant ant (area 1) or in an area where Conomyrma insana was the dominant ant (area 2).


Fig. 6.1.








Posit ion on Plant upper 113 middle 1/3 lower 1/3


cn

C01) C-7 Co L&J



Ca


706050

4030

20

10

0


- U ~ -- ~ ~I _________________________________ ~ K.


Nymphs
present


Nymphs absent


901


Nymphs
present


Conomyrma insana


K ~


Nymphs
absent


r~x


K



.5' ~ "4




I


Solenpesis Feminata


'


YI .


E


ZF





S,


I


,,,,. ,,


co 0)





87


1000



80




60


Lii
f4--/*o
0
40
/ 0
S/ 7 T Lower 1/3
Middle 1/3 with nymphs
Upper 1/3
/* v Lower 1/31
0 Middle 1/3 -without nymphs o Upper 1/31



4 8 12 16 20 24
HOURS

Fig. 6.2. Percentages of Pseudoplusia includens eggs missing after 4, 8,
and 24 hr exposure at 3 heights on soybean plants with or without Spissistilus festinus nymphs. Plants were in 3 areas where
Conomyrma insana was the dominant ant.




88


without nymphs. Thus the probability of a prey item being encountered by ants would be higher on plants with nymphs than on plants without nymphs. Within the C. insana enclaves, rates of predation were also greater on plants with nymphs than on plnats without nymphs as long as a supply of eggs remained available to the ants. Supplies of eggs were almost exhausted after 24 hr, however, and predation approached 100% on all plants. The data indicate that, at least on cool autumn days, workers of C. insana patrol most plants within an enclave. Thus, chances are high that, given a 24-hr period, a worker would encounter an egg or other stationary prey item on plants within the boundaries of an enclave regardless of the presence of an attractant such as the nymphs. Throughout the rest of the field, where egg predation was low on plants without nymphs, the presence of nymphs could be an important factor affecting survival of pest insects.

In both predation experiments, the percentages of eggs missing were greatest from the lowest 1/3 of the plant and least from the upper 1/3 of the plant. Differences were statistically significant (p<0.005). This suggests that the ants spent more time foraging on the lower parts of the plants than on the upper parts of the plants or that few ants foraged at the top of the plants. Eggs oviposited on the upper parts of the plants would, therefore, be encountered less frequently by ants than eggs oviposited on the lower parts of the plants and may have a greater probability of survival.


Conclusions

The threecornered alfalfa hopper is common in soybean fields and can kill young soybean plants (Tugwell et al. 1972, Caviness and Miner









1962, Mueller and Dumas 1975). At the present time, however, the economic impact of the insect on soybean production is unclear. Mueller and Dumas (1975) found that no yield losses were incurred under Arkansas conditions even when 30-50% of the soybean plants in a field were destroyed by S. festinus. Threecornered alfalfa hopper nymphs attract ants to soybean plants. The ants, by preying on pest insects, may protect the soybean plants, thus offsetting damage caused by the feeding of the nymphs. Any beneficial effects of the association of ants with Homoptera should be considered when assessing the economic status of homopteran pests. Further studies of ant-Homoptera relationships and their effects of predation should provide a basis for developing strategies that would increase the effectiveness of ants as predators.

Summary

Five ant species, Solenopsis geminata (F.), Conomyrma insana (Buckley), C. flavopecta (M. R. Smith), Iridomyrmex pruinosus (Roger), and Pheidole morrisi Forel, tended nymphs of the threecornered alfalfa hopper, Spissistilus festinus (Say), on soybeans in a north Florida field. The presence of nymphs tended by ants was examined as a factor affecting predation on eggs of the soybean looper, Pseudoplusia includens Walker. Where S. qeminata tended the nymphs, the percentage of eggs missing after 24 hr from plants with nymphs (77.1%) was significantly greater than the percentage missing from plants without nymphs (36.8%). Where C. insana tended the nymphs, the percentage of eggs missing from plants with nymphs was significantly greater than the percentage missing from plants without nymphs after 4 hr (36.8% vs. 17.5%) and 8 hr (65.6% vs. 42.9%) but not





90



after 24 hr (90.7% vs. 86.3%). In all experiments, the proportion of eggs missing was greatest from the lowest 1/3 of the plant and least from the upper 1/3.















PROGNOSIS


From the foregoing studies, it is obvious that the distribution of Conomyrma insana (Buckley) is severely limited by ecological requisites. The most obvious limiting factors are the availability of suitable nesting sites, food sources, and its temporary host, C. flavopecta (M. R. Smith).

In general, C. insana nests must be located in open areas fully exposed to sunlight. The majority of the sites where C. insana nested were acceptable owing to recent disturbances such as mowing, cultivation, or burning. The ant ceased to nest in sites where normal vegetational succession occurred. This can be accepted as proof of their dependence upon open areas for nesting. Additionally, the ant is excluded from excessively hydric locations.

A further restriction on nesting sites is that a location must include or be adjacent to an adequate food supply. For C. insana this means that areas of heavy vegetation must be nearby, as they are dependent on carbohydrate food sources for the energy required to retain a site. The carbohydrate food can be from plant nectaries or honeydew excreted by Homoptera.

The successful colonization of an area is further limited by the

nature of C. insana's habits. Conomyrma insana is temporarily parasitic on C. flavopecta. Although no analysis is available, it is reasonable to assume that, based on prior studies with other temporary parasitic ants, the percentage of successful parasitism is extremely low (Wilson


91





92


1971). Parasitic species of ants must successfully parasitize the host and produce adequate numbers of offspring to dominate an area. This requires that the ant increase by a process of hesmosis (budding) until large areas are their domain with other ants of the same trophic level excluded. They must then accept additional queens of the same species, as is true with C. insana. Thus, by chance, they must initially parasitize the host species within an area that has the prerequisites of suitable nesting conditions with readily available carbohydrate food. This, in effect, decreases the chances of successful new colonizations. That success is low is demonstrated by the relatively low percentage of viable C. insana colonies as compared to the much greater number of sites that appear to have the requisites of suitable nesting conditions with adequate food sources.

As noted by Van Pelt (1966) and Whitcomb et al. (1972), C. insana is a general predator. The ant is omnivorous and utilizes all available food sources for energy to maintain and expand its colonies.

Conomyrma insana is, without any question, an effective predator

of imported fire ant queens, Solenopsis invicta Buren. A fire ant queen alighting in a C. insana enclave has little chance of surviving. Yet, C. insana appears to be a poor candidate for deliberate use as a controlling agent of fire ants via destroying the queens before they succeed in beginning a new colony. The relatively sparse distribution and the compact enclaves must be considered when assessing the predatory effectiveness of the ant in relation to the total number of fire ant queens attempting to begin new colonies. At best, even though C. insana workers do destroy nearly all fire ant queens landing in their enclaves, only a small fraction of all fire ant queens out of a nuptial flight land where





93


C. insana occur. It would be a being ineffective. As Whitcomb factors must be considered when of fire ant queens. Therefore, populations, it is an important


fallacy, though, to dismiss C. insana as et al. (1973) pointed out, all mortality determining the percentage of survival as C. insana has stable and persistent predator of fire ant queens.




Full Text
REVIEW OF LITERATURE
Introduction
In the continental United States, three species of the genus Cono
myrma--bicolor (Wheeler), flavopecta (M. R. Smith), and insana (Buck-
ley)--were recognized by Snelling (1973). Prior to Snelling's revision,
the species of Conomyrma in North America and neighboring areas were
lumped as Dorymyrmex pyramicus (Roger) or Conomyrma pyramica (Roger).
Both Cole (1957) and Wilson (1957) suggested that bicolor was a
distinct biological species, based on their independent observations
that C_. bicolor occurred in intimate sympatry with C_. pyramica while
remaining distinct in worker color and size.
It is now evident that the Dorymyrmex/Conomyrma pyramicus group
is a complex comprised of several species. In the past, several myrme-
cologists have attempted to review comprehensively the taxonomic tangle
surrounding this complex. The most natable attempts were made by Sants-
chi (1922), Creighton (1950), Kusnezov (1952, 1959), Wilson (1957), and
Snelling (1973). For the most part, their attempts have had little
effect beyond increasing the confusion. There have been attempts by
several authorities, such as McCook (1879), Santschi (1922), M. R. Smith
(194), and Cole (1936, 1957), to separate individual species from the
complex, but few used solid morpholigical characters that would withstand
critical examination.
Gwing to this extreme confusion and until this complex is dealt with
as a whole, it is impossible to assign much of the literature references
3


CHAPTER II
COLONY ORGANIZATION AND NUMBERS OF QUEENS
Introduction
Investigations of the ecological distribution of ants are concerned
with colonies of ants rather than individuals, since the colony is the
unit of social organization. It is difficult to be certain whether each
nest structure is independent and whether there is a direct overground or
underground connection with neighboring nest structures. In investiga
tions of Concmyrma irsana (Buckley) and C_. flavopecta (M. R. Smith) in
north Florida, the nest craters of Ch insana appeared to be clustered
while the nest craters of £. flavopecta exhibited no tendency toward clus
tering. Both Buckley (1366) and Van Pelt (1966) hypothesized that a sin
gle colony of C. insana (= Dorymyrmex pyramicus (Roger) in Van Pelt's paper)
consisted of large numbers of nests, though they presented no conclusive
proof. I wished to determined the colony composition of each species more
precisely in order to make advances in the field ecology of C^. insana and
C. flavopecta and to compare the ecological requirements of the 2 species.
Until the advent of radioactive tracers and development of adequate
techniques, it was generally only possible to theorize that a colony of
ants was composed of more than 1 nest. Since related problems with ants
have been solved using radioactive tracers, their use was attempted in this
situation. Kannowski (1959) introduced P mixed with honey into source
nests. He concluded that a colony of this .species was composed of 1 or
33


P^e
IV.PREDATION ON FOUNDING QUEENS OF Solenopsis invicta BY
WORKERS OF Conomyrma insana .... 51
Introduction 51
Methods 52
Results and Discussion 53
Summary 63
V.SOURCES OF CARBOHYDRATES UTILIZED BY Conomyrma insana 64
Introduction 64
Methods 65
Results and Discussion 66
Summary 77
VI.PRESENCE OF Spissistilus festinus TENDED BY ANTS: A
FACTOR AFFECTING EGG PREDATION BY ANTS IN SOYBEANS. ... 79
Introduction 79
Methods 79
Results and Discussion 81
Conclusions 88
Summary 89
PROGNOSIS 91
LITERATURE CITED 94
APPENDICES 101
A. GLOSSARY OF SPECIFIC TERMS 102
B. PUBLICATIONS 104
BIOGRAPHICAL SKETCH 105
v


60
attempt flight after landing. The response most often elicited by the
attack of C^. insana was attempted escape on the ground surface. The
distance covered varied with the topography, presence of vegetation, or
other natural barriers. The escaping queen usually covered distances
of 20 to 80 cm per min; some covered distances of 3 m or more. This re
sponse was often temporarily successful in avoiding immediate predation
by C^. insana workers.
While under attack or when unmolested by C^. insana workers, the S^.
invicta queens were continuously moving under twigs, leaves, or into
soil cracks. This appeared to be a concealment response, even though
it may be difficult to distinguish from the normal instinct of the queen
to commence construction of the brood chamber as soon as possible after
landing. Hiding was usually effective in momentarily avoiding predation.
If the queen abandoned the protection of the leaf or twig, C^. insana
workers immediately resumed their attack.
In further attempts to avoid predation, the queen often remained
motionless briefly (6-10 sec) when attacked or investigated by a £.
insana worker; afterwards, attempted escape or defense methods were
used. In a number of observed cases, the cessation of motion was tem
porarily successful as the C_. insana worker either did not commence
further attack or broke off the attack and left the vicinity of the
female.
The queens also attempted to take a defensive stand against the
attack of the C_. insana workers by aggressively biting. This was most
often seen when the invicta queen had failed to dislodge attacking
workers and 1 or more had seized an appendage. To use the mandibles
effectively, the queen would lie on either side and assume a "C" position


78
Conomyrma insana construct their nests adjacent to carbohydrate
food sources. The population size in an area may be proportional to
the quantity of carbohydrates available.


63
even though the area was extensively searched. An identical search was
conducted following the nuptial flight on 21 June. On 22 June, 24 S^.
invicta queens were found that had been subdued by £. insana workers and
were being taken into the colony. No queens were found on 23 and 24
June within the £. insana enclave.
Summary
Solenopsis invicta Burn founding queens are especially vulnerable
to predation during and following the nuptial flight. Conomyrma insana
(Buckley) workers are effective predators of the queens in some areas of
northern Florida. The respective behaviors that occurred during the
confrontation of the 2 species were studied in the field in Leen county
of northern Florida. Detection of a founding queen alighting within a
C_. insana enclave by workers varied from a few seconds to several min
utes. The founding queen's reactions to attack by C_. insana consisted
of 5 basic responses: (1) escape, (2) concealment, (3) cessation of
movement, (4) mandibular defense, and (5) rapid flexing or shaking.
Less than 3% of the queens observed alighting within a C^. insana en
clave were successful in escaping predation. No SL invicta queens were
found that survived predation by £. insana workers long enough to con
struct a brood chamber and begin a new colony within an area occupied by
C. insana.


This dissertation is dedicated to George Davis of Horatio, Arkansas.
The debt owed by myself and countless former students of his that he
aided can never by fully repaid. Without his help and encouragement, I
would not have had the opportunity to embark upon the studies required
to begin, much less complete, this work.


INTRODUCTION
Ants of the dolichoderine genus Conomyrma occur commonly through
out most of the United States. Yet, within this country, comparatively
little attention has been directed toward the ecology or biology of the
species in this genus. The majority of published reports including these
ants consist of a brief mention in a list of ant species collected in
an area and/or a notation as to the type of habitat they were found in.
The paucity of attention toward the species of Conomyrma can be
partially attributed to the lack of any obvious economic or health
problems posed by their presence. Recently, though, accounts have been
published on observations of one species, Conomyrma insana (Buckley),
attacking and killing newly mated queens of the red imported fire ant,
Solenopsis invicta Burn (Fincher and Lund 1967, Markin et al. 1971,
and Whitcomb et al. 1972). These reports prompted this investigation
of £. Insana in northern Florida as the presence of S_. invicta does pose
economic and health problems.
An examination of the ecological requirements of C_. insana was
deemed necessary to evaluate the potential of this species as a candi
date biological control agent of the red imported fire ant. In addition,
an investigation of predation by C^ insana on newly mated fire ant
queens and other arthropods was required to evaluate the efficiency of
this ant as a biological control agent.
Consequently, my primary objective was to determine and document
the principal factors affecting the distribution and populations of £.
1


4
correctly to the currently accepted species. Therefore, in the follow
ing discourse on the literature, I will deal with the entire complex in
the United States (excluding only those papers dealing strictly with C_.
bicolor). The name used by the author of each article will be quoted
unless I am able to determine that it may be safely and correctly
assigned to one of the species now recognized by Snelling.
Taxonomy
Conomyrma insana was described as Formica insana by S. B. Buckley
in 1866. His description of this ant is reiterated in entirety as
follows:
Worker. Length 0.14 inch--Color black or brownish-black,
smooth and shining throughout; head sub-quadrate, the lateral
margins slightly curved inwards; below oval, and rounded
above; eyes large, subelliptical, and placed on the anterior
portion of the head near its lateral margins; antennae long,
filiform; mandibles large, curved inwards, the truncated
apical ends sharply toothed; head sub-channelled beneath with
the anterior and posterior parts deeply depressed; prothorax
about half the width of the head, rounded above; mesothorax
somewhat depressed; metathorax, has a rudimentary spine or
sharp protuberance on its upper posterior surface, pedicle
short, inserted near the base of the anterior part of the
abdomen; scale small, wedge-shaped, and inclined forwards;
abdomen broad-ovate, subacute; legs small, slender, and rather
short.
Female. Length 0.20 inch--Head small, narrower than the
thorax; abdomen broad, oblong and ovate; color black-bronze
with the margins of the segments of the abdomen hyaline;
thorax raised above the head; wings not seen; the remainder
like the worker.
Dwells in the ground but a few inches deep, having numer
ous holes scattered over a large surface of fifty or more feet
in diameter. It goes now here and now there, without order,
from one abode to another, sometimes moving the larvae from
one place to another; large colonies dwelling in common. It
is very active and war-like, can conquer most other ants,
boldly seizing the larger species by the legs. If a large ant
is so unfortunate as to run among the "crazy ants," he hastens
to escape, but rarely does so without being severely bitten.
Common in Central Texas.


44
reaction is termed "panic retreat" paralleling the terminology of Wilson
and Regnier (1971) who distinguished between "panic alarm" and "aggres-
- sive alarm" reactions in ants. In contrast, accidental or other contacts
between insana workers never lead to antagonistic or panic retreat
reactions. Chance encounters between 2 foraging C^. flavopecta workers
can either be non-antagonistic or can initiate panic retreat (both
workers sometime display this response) according to whether the meet
ing is between workers of the same colony or from 2 different colonies.
In all respects, the differences between the 2 taxa are striking,
and the evidence that 2 distinct species are involved is unequivocal.
Therefore, the discovery of mixed nests of the 2 species was totally
unanticipated. The specific records in Florida for these mixed nests are
as follows: Highlands Co., Archbold Biological Station, Lake Placid, 18
April 1973; Polk Co., Winter Haven, Camp's citrus grove, 19 April 1973;
Polk Co., Winter Haven, Camp's citrus grove, 13 April 1974; Alachua Co.,
Gainesville, Depot Road, 30 April 1974; Alachua Co., Gainesville, 15 May
1974 (Wm. F. Burn); Alachua Co., Gainesville, 10 April 1975 (J. F.
Carroll); Alachua Co., Gainesville, 23 May 1975 (VJm F. Burn); and
Wakulla Co., Medart, 20 June 1975 (D. P. Wojcik). In all mixed nests,
workers of both species were observed participating in normal activities
as nest-mates (foraging, casting out soil particles, and moving together
in the nest burrows) without antagonism. The mixed C^. insana-flavopecta
nests were never found within an established enclave of £. insana nests;
rather, they were found only in areas where C_. flavopecta occurred. On
one occasion, a mixed nest was found ca. 10 m from a insana enclave.
These observations and records indicate that temporary social
parasitism might be occurring in these ants. The Conomyrma, in 1 area
I


46
rather than C. flavopecta and all callow workers (n=36) were also C_.
insana. Although the ratio of C_. Insana to flavopecta workers in the
original mixed nests was about 7:3, the ratios of £. insana to £. flavo
pecta in other mixed nests observed at the time of mapping (1-10 June)
was about 20:1, judging from the workers appearing at the nest entrance.
Distributional patterns of the craters of the Conomyrma sp. and the
mixed nests are shown in Fig. 3.1 and 3.2. The C. insana craters are
clustered into 3 localized enclaves, whereas the C. flavopecta craters
have a more scattered, but probably not random, spatial arrangement. No
Conomyrma nests were found in the area of high, dense grass. While the
location and number of craters constructed by C^. flavopecta appeared
stable, the number of C_. insana nests increased from the time the area
was first mapped. A new C_. insana enclave also developed in the area
where mixed nests had previously been located. Only 1 mixed nest could
be found on 26 June. This nest had an estimated ratio of 100:1 £.
insana to C^. flavopecta workers.
The colony taken from the mixed nest was maintained in the labora
tory for 1 year. The last individual of the original 25 C. flavopecta
workers died 42 days after capture. During this time, the overall
mortality rate of the group of C^. flavopecta workers appeared due to
normal age-related factors. No antagonism between the C_. flavopecta
and the C^. insana workers was detected. Normal colony duties, such as
brood tending and food gathering, were equally performed by both. In
addition, trophyllactic exchange of food between the 2 species was
repeatedly observed. At 6 months, the colony reached a peak population


24
The locations of the enclaves of £. insana nests in Persimmon Field
changed little from 1974 through 1976 (Fig. 1.6-1.8). Some fluctuations
were evident in the boundaries that could be attributed to maintainance
practices. This field was annually cultivated and planted with corn,
and a fire lane was maintained around the field. In 1974, the fire lane
and fallow areas on the west side of the field were kept clean of vege
tation, and the £. insana nests were present over this entire cleared
area (Fig. 1.6). In 1975 and 1976, this areas was not kept clear, and
the vegetation became rank with no soil surface exposed to the sun. No
IS. insana nests were found in this heavy vegetation (Fig. 1.7 and 1.8).
Conomyrma insana nests were not found in the center of the corn
fields, although they were present in the edges of the fields. The re
peated cultivation of the corn field may have destroyed most of the
nests and prevented C^. insana from occupying this area. As demonstrated
in Introduction Field, the locations of the C^. insana enclaves remained
stable. The ant did not nest in 1 narrow area, ca. 12 m wide, on the
northern side nor on the southern side of the field. Both were low,
boggy areas in which drainage was poor.
The number of C_. insana nests per square meter in both Intro
duction Field and Persimmon Field remained relatively constant (Table
1.1). The average number of nests per square meter was slightly greater
in Persircnon Field than in Introduction Field. The only major differ
ence in treatment between these two fields was that Introduction Field
was mowed frequently while Persimmon Field was cultivated.
Summary
Conomyrma insana enclaves (large numbers of contiguous nests with
defineable boundaries) were found at 29 locations on the Tall Timbers


Conomyrma insana workers are effective predators of fire ant found
ing queens, Sclenopsis invicta Burn. The confrontation behavior of the
2 species was determined. Worker detection of a queen alighting in a £.
insana enclave varied from a few seconds to several minutes. The found
ing queen's reactions tc attack by £. insana consisted of 5 basic respon
ses: (1) escape, (2) concealment, (3) cessation of movement, (4) mandi
bular defense, and (5) rapid flexing or shaking. Less than 3% of the
queens observea alighting within a C^. insana enclave successfully
escaped, and none survived to develop a new colony.
Twenty-six species of Homoptera were tended by Conomyrma insana.
The Homoptera were in the families Aphididae (10), Phylioxieridae (1),
Coccidae (5), Pseudoccidae (1), Eriococcidae (2), Kermesidae (1), and
Membracidae (1). Conomyrma insana also fed from the nectaries of Cassia
nictitans L., £. fasciculata Michx., C_. obtusifolia L., and Richardia
scabra L, Conomyrma insana construct their nests near carbohydrate
food sources,
Five ant species, of which iC. insana and Solenopsis geminata (F.)
were the most abundant, tended nymphs of the threecornered alfalfa hopper,
Spissistilus festir.us (Say), in a soybean field. The presence of the
ants under these conditions was examined as a factor affecting predation
on eggs of the soybean looper, Pseudoplusia includens Walker. Where S^.
geminata tended the nymphs the percentage of eggs missing after 24 hr was
significantly greater from plants with nymphs (77.1%) than from plants
without nymphs (36.8%)* Where £. insana tended the nymphs, the percent
age of eggs missing was significantly greater from plants with nymphs
than from those without nymphs after 4 hr (36.8% vs. 17.5%) and 8 hr
vii


CHAPTER III
MIXED NESTS OF Conomyrma insana and C. flavopecta--
EVIDENCE OF TEMPORARY SOCIAL PARASITISM
Introduction
Two species of the genus Conomyrma are common in Florida. For the
present, these are identified as Conomyrma insana (Buckley) and C^.
flavopecta (M. R. Smith) by the identification criteria given by
Creighton (1940) and the revision by Snelling (1973). The 2 species
differ in color, morphology, and biology. Conomyrma insana is nearly
uniformly dark, greyish-black, has a broad, straight or slightly concave
occipital border, and has a distinct declivity on the mesonotum when
seen in profile; while £. flavopecta is light yellowish-brown, with the
head and apex of gaster weakly to moderately infuscated, the occipital
border rounded or straight only medially, and the mesonotum in profile
evely convex without trace of a declivity. Colony structuring differs
markedly in the 2 species. Conomyrma insana exists in enclaves of
multiple nests with much traffic between the nests, transfer of brood,
food exchange, and many queens localized together in only 1 of many
nests; while in C. flavopecta, there is no worker traffic or food ex
change between workers from separate nests and 1 queen in each nest.
Reactions between the 2 species in the field are entirely inimical.
When v/orkers of the 2 species meet, the C. insana worker usually holds
its ground and may open its mandibles, while the C_. flavopecta worker
immediately breaks contact and runs rapidly and erratically away. This
43


I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
Jr. w. H. Whitcomb, Chairman
Professor of Entomology
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
Dr. Wm. F. Burn
Professor of Entomology
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
Dr. P. S. Callahan
Professor of Entomology


LITERATURE CITED
Bhatkar, A., W. H. Whitcomb, W. F. Burn, P. Callahan, and T. Carlysle.
1972. Confrontation behavior between Lasius neoniger (Hymenoptera:
Formicidae) and the imported fire ant. Environ. Entomol. 1:274-9.
Blum, M. S. 1969. Alarm pheremones. Annu. Rev. Entomol. 19:57-80.
Blum, M. S. 1973. Comparative exocrinology of the formicidae. Proc.
VII Congr. IUSSI, London. 23-40.
Blum, M. S. and S. L. Warter. 1966. Chemical releasers of social beha
vior. VII. The isolation of 2-heptanone from Conomyrma pyramica
and its modus operandi as a releaser of alarm and digging behavior.
Ann. Entomol. Soc. Amer. 59:774-9.
Brian, M. V. 1965. Social insect populations. Academic Press, Inc.
London. 135 p.
Buckley, S. B. 1866. Descriptions of new species of North American
Formicidae. Proc. Entomol. Soc. Philadelphia 6:152-72, 335-49.
Carroll, C. R. and D. H. Janzen. 1973. Ecology of foraging by ants.
Ann. Rev. Ecol. Syst. 4:231-57.
Caviness, C. F. and F. D. Miner. 1962. Effects of stand reduction of
soybeans simulating threecornered alfalfa hopper injury. Agron. J.
54:300-2.
Chiu, Y. J., R. Levy, and H. L. Cromroy. 1974. Biological half-life of
cesium-134 in the red imported fire ant, Solenopsis invicta. Fla.
Entomol. 57:141-4.
Cole, A. C., Jr. 1936. Descriptions of seven new western ants (Hymenop.
Formicidae). Entomol. News 47:118-21.
Cole, A. C., Jr. 1940. A guide to the ants of the great Smoky Mountains
National Park, Tennessee. Amer. Midi. Nat. 24:1-88.
Cole, A. C., Jr. 1942. The ants of Utah. Amer. Midi. Nat. 28:358-88.
Cole, A. C. 1957. Notes on western ants (Hymenoptera:Formicidae). N.
Y. Entomol. Soc. 65:129-31.
94


Fig. 1.4. Conomyrma insana nests (stippled areas) in Introduction Field
during July, 1975.


10 meters


(65.5% vs. 42.9%) but not after 24 hr (SO.7% vs. 86.3%). The proportion
of eggs missing was greatest from the bottom 1/3 of the plant and least
from the top 1/3.
viii


19
N


57
TABLE 4.1 Continued.
Elapsed
Time (min) Observations
63 Si moving with Ci_ clinging. Additional Ci_ in pursuit
I=7) and attacking (n=15+) Si^ moving rapidly while
shaking and flexing body. Stopped, assumed "C" posi
tion bit and killed £i_ (n=l). Ci_ attacking (n=4).
66 Ci^ (n=3) clinging to Si. Si_ dislodged Ci amd moved
forward rapidly. Ci_ T=4) pursuing, Ci~Tn=l) cling
ing to leg. Si_ curled into "C" position and bit Ci_.
68 Ci^ (n=12) attacking. Sj^ halted, assumed "C" position,
bit Ci_ (n=l), now moving with Ci_ continuing attack.
Si biti ng at Ci_ ki 11 ed 1.
72 Si_ nearly subdued (= taken as prey by Ci_). Little
response by Ci_ and only feeble attempts to escape.
75 Ci_ have severed left prothoracic and both metathoracic
legs.
76 Ci_ have severed left mesothoracic leg.
77 All legs severed.
80 Ci_ moving appendageless Sj_ to nest. Si_ alive and trying
to bite Ci_. Distance from alighting point = 53 m.


10 meters


12
were marked on a map of the station (Fig. 1.1). (The map was prepared
by Roy Komarek in July, 1969 and was based on ASC aerial photograph
ADV194, 3-22-60.) The sites of insana populations were examined for
consistencies and variations of features.
The £. insana populations in two fields, referred to as Introduc
tion Field and Persimmon Field, were annually mapped in greater detail.
In 1973, a survey map of Introduction Field was prepared by using the
compass-traverse survey method (DeMoisy 1949) to establish the field
boundaries and locate major landmarks. Permanent markers (,5-m sections
of pipe driven into ground) were placed in secure locations on the edge
of the field. The boundaries of the £. insana population were located
on the prepared map. In 1974, the same procedure was initiated for
Persimmon Field.
2
The number of nests per m within C_. insana enclaves was examined
in Introduction Field (beginning in 1973) and Persimmon Field (beginning
2
in 1974). The nests in 100 randomly selected m quadrats were counted
annually in July through 1976.
Results and Discussion
Twenty-nine locally isolated populations of Conomyrma insana were
found on Tall Timbers Research Station (Fig. 1.1). At each site, numer
ous nests were present and apparently interconnected, based on worker
traffic, in addition to worker transfer of brood, from nest to nest.
The least number of nests found in a single locale was 17, while 2,187
nests were counted in another field. The contiguous nests of the popu
lations of C^. insana can only be defined as enclaves as they have
definite boundaries within which few other ant species occur.


37
C. insana craters: ^
J
I
o
0 Source
0
Non-radioactive
Radioactive
O
o Not sampled
o
1 meter



o e
e


o



0
o'
o
%




o

0
o
A o
B
o
0
o
c

o e
o
o
e
o



*
o
.
3
O
O


e
e


3
c

0
O C
o D


73
o
c

e
o
o
Fig. 5.1. Distribution of Conomyrma insana nests in 1273 around a
short leaf pine, Pinus echinata, located in an open field
at Tall Timbers Research Station. The short leaf pine was
heavily infested with Cinara watsoni.


Fig. 1.5.
Conomyrma insana nests (stippled areas) in Introduction Field
during July, 1976.


Fig. 4.1. Flow diagram of responses of 240 Solenopsis invicta founding
queens to initial contact, pursuit, and attack by Conomyrma
insana workers. The contrasting behavior leading to nest
founding is included.


97
Marikovsky, P. I. 1961. Material on sexual biology of the ant Formica
rufa L. Insectes Soc. 8:23-30.
Markin, G. P., J. H. Dillier, S. 0. Hill, M. S. Blum, and H. R. Hermann.
1971. Nuptial flight and flight ranges of the imported fire ant,
Solenopsis saevissima richteri (Hymenoptera-.Formicidae). J. Ga.
Entorno!. Soc. 6:145-56.
Markin, G. P., J. O'Neal, and H. L. Collins. 1974. Effects of mirex on
the general ant fauna of a treated area in Louisiana. Environ.
Entomol. 3:895-8.
Mayr, G. 1866. Myrmecologische Beitrage. Sitz. Adak. Wiss. Wien. 63:
484-517.
Metcalf, R. A. and R. L. Metcalf. 1970. Effects of isosteres of 2-
heptanone on the alarm behavior of the ant Conomyrma pyramica.
Ann. Entomol. Soc. Amer. 63:34-5.
Mueller, A. J. and B. A. Dumas. 1975. Effects of stem girdling by the
threecornered alfalfa hopper on soybean yields. J. Econ. Entomol.
63:511-2.
Naves, M. A. 1974. Ecological manipulation affecting ant populations.
(Hymenoptera:Formicidae). M. S. thesis. Univ. Fla., Gainesville
77 p.
Nielsson, R. J., A. P. Bhatkar, and H. A. Denmark. 1971 A preliminary
list of ants associated with aphids in Florida. Fla. Entomol. 54.
245-8.
Norton, E. 1875. Wheeler's Report of Geological Exploration. Zool.
p. 734.
Odum, E. P. and E. J. Kuenzler. 1963. Experimental isolation of food
chains in old-field ecosystem with the use of phosphorus-32. Li
Schultz, Vincent and A. W. Klenent, Jr. (eds.). Radioecology.
Reinhold Pub. Corp., N.Y.
Pierce, W. C., R. A. Cushman, C. E. Hood, and W. D. Hunter. 1912. The
insect enemies of the cotton boll weevil. U. S. Dep. Agr. Bur.
Entomol. Bull. 100. 99 p.
Pontin, A. J. 1960. Field experiments on colony foundation by Lasius
niger (L.) and l. flavus (F.). Insectes Soc. 7:227-30.
Quaintance, A. L., and E. L. Jenne. 1912. The plum curculio. U. S.
Dep. Agr. Bur. Entomol. Bull. 103. 205 p.
Roe, R. A. 1973. A biological study of Solenopsis invicta Burn, the
red imported fire ant, in Arkansas with notes on related species.
M. S. Thesis. Univ. Ark., Fayetteville. 135 p.


81
each height on plants with nymphs, and 21 eggs were at each height on
- plants without nymphs. The fraction of eggs missing out of each group
of 21 eggs was recorded after the eggs had been in the field for 24 hr.
Experiments were repeated during five 24-hr periods. An analysis of
variance was performed on the arcsine-transformed data arranged in a 2
x 2 x 3 factorial design (= nymphs present or absent x 2 areas of the
field x 3 heights on the plants) replicated on 5 dates.
On 14 October, an additional experiment was conducted in three 45-
2
m areas of the field in which the nymphs were being tended by £.
insana. Experimental procedures were the same as those used in the
earlier experiments except that 1 set of 14 plants (7 with 1 or more
nymphs tended by insana and 7 without nymphs) was selected in each of
the 3 areas of the field. In addition, the eggs were examined after
they had been in the field 4, 8, and 24 hr. An analysis of variance was
performed on the arcsine-transformed data (numbers of eggs missing/21
eggs) arranged in a 2 x 3 x 3 factorial design (= nymphs present or
absent x 3 elapsed periods of exposure x 3 heights on the plants)
replicated in 3 areas of the field.
Results and Discussion
Workers of Solenopsis geminata, Conomyrma insana, C. flavopecta (M.
R. Smith), Iridomyrmex pruinosus (Roger), and Pheidole morrisi Forel
tended threecornered alfalfa hopper nymphs in the soybean field in which
the investigations were conducted. Although tending of S.. festinus
nymphs by ants in soybeans has not previously been reported, Jordan
(1952) observed an unspecified ant species tending nymphs in alfalfa.
In the experimental field, nests of Pheidole morrisi and Iridomyrmex
f


CHAPTER IV
PREDATION ON FOUNDING QUEENS OF Solenopsis invicta
BY WORKERS OF Conomyrma insana
Introduction
The founding queens of the red imported fire ant, Solenopsis in
victa Burn, are especially exposed to predation during the period
following the nuptial flight and before entering the soil (Whitcomb et
al. 1973). The major predators at this point appear to be its own
workers and those of other ant species. Hung (1974), Markin et al.
(1971), Whitcomb et al. (1973) have reported the dolichoderine ant
Conomyrma insana (Buckley) as one of the more effective ant predators of
the Su invicta founding queens. Initial observations indicate that
these queens have a great deal of difficulty establishing where C_.
insana occur in north Florida. Although £. insana is sporadic in dis
tribution in this region, it may be an important factor in limiting
local populations of the red imported fire ant.
Competition for preferred nesting sites and foraging territories
appears to be the major factor determining the distribution of many for-
micine species (Bhatkar et al. 1972, Leston 1973, Van Pelt 1966). No
where is this more true than in the aggression of the formicine species
toward founding queens. Predation of founding queens alighting in a
territory dominated by a given ant species would reduce or eliminate
future competition for available resources in favor of the dominant
species. Carroll and Janzen (1973) stated that founding queens were
51


52
usually killed by worker ants of the same or other species. This strong
statement is supported by research of several workers (Janzen 1967,
Levieux 1971, Marikovsky 1961, Pontin 1960). Janzen (1973) presented
research on Pseudomyrmex ferruginia F. Smith and described in detail how
workers destroyed new queens of their own species. Hung (1974) analyzed
the refuse pile of a Conomyrma insana colony and reported the recovery
of discarded heads of the queens of 5 species of myrmecines, including
those of Solenopsis invicta queens.
The predator-prey relationship of Conomyrma insana workers and the
founding queens of Solenopsis invicta is of special interest because of
the apparent complexity of the encounters leading to the death of the
founding queens or the rare successful escape from predation by the
latter. The first objective of the present research was to gain infor
mation of the general behavior pattern of a do!ichoderine ant worker
attacking a myrmecine founding queen. Furthermore, such data are needed
on the predator-prey relationship of these 2 species, in particular, to
evaluate the effectiveness of C^. insana as a predator and its potential
in limiting the local increase of SL invicta. Conversely, this informa
tion was collected to determine the ability of S_. invicta to survive
within an area under these high stress conditions.
Methods
This research consisted of field observations of the behavior(s)
of the predator (£. insana workers) and prey (S_. invicta founding queens)
under natural conditions. These events were observed in detail in Leon
county in northern Florida in areas where Conomyrma insana nests were
present. Of approximately 2,000 Solenopsis invicta founding queens
(


Abstract cf Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
ECOLOGY AND BEHAVIOR OF CONOMYRMA INSANA
By
James C. Everett Nickerson
December, 1976
Chairman: Dr. W. H. Whitcomb
Major Department: Entomology and Nematology
Conomyrma insana (Buckley) enclaves (large numbers of contiguous
nests with defineable boundaries) were found at 29 locations on the
Tall Timbers Research Station in Leon Co., Fla. The ants were more
commonly found in open areas adjacent to heavy vegetation. Nests were
absent in cultivated areas, areas of heavy vegetation, 3nd where
drainage was poor.
The organization of the colony and number of queens per colony of
C. insana and C. flavopecta (M. R. Smith) were investigated using radio
isotope tracers and by excavation of nests. Conomyrma insana was poly-
domous (more than 1 nest per colony), and had several queens located in
a central nest. Conomyrma flavopecta was monodomous (1 nest per colony)
with 1 colony queen.
Conomyrma insana and C. flavopecta were found in the same nest.
Later, clusters of £. insana nests were formed at this location. Based
on the evidence, £. insana is a temporary social parasite on C^.
flavopecta.
vi


Fig. 1.8. Conotnyrma insana nests (stippled areas) in Persimmon Field
during July, 1976.


40
excavated. This was the only nest of those excavated in which eggs were
found in addition to brood of all stages. On 18 July 1974, 26 nests were
2
excavated in an area of 29 m Fourteen deal ate queens were located in 1
nest at depths ranging from 2-17 cm. No queens were found in the other
excavated nests. On 11 February 1975, 13 nests were excavated in an area
2
covering 9m. A single C^. insana queen was found in a chamber 2 cm
below the soil surface in 1 nest. Four queens were located in chambers
5-10 cm below the surface in a second nest 1.3 m from the nest in which
the single queen was located. No additional queens were located in any
of the remaining nests excavated in this location.
All C^. insana queens recovered during the excavation of the clus
tered mounds were dealate and apparently reproducing, based on the en
larged appearance of the gaster. Eggs were not found in any of the
excavated nests unless a queen was present. If a queen was not present,
only larvae and pupae were noted. Prior to excavation, each crater was
examined for visual differences, such as a larger crater compared to
others, indicating the possible presence of queens. No differences were
detected.
Three clustered groups of £. flavopecta nests were excavated in
O
1973. On 26 May, 11 nests located in an area 37 m were excavated. In
each of 5 of the 11 nests, a single dealate queen was found. On 27
p
September, 9 nests were excavated in a 36 m area. A single queen was
found in each of 2 nests. On 9 October, 14 nests were excavated in a 43
2
m area. A single queen was found in each of 2 nests. On 9 October, 14
2
nests were excavated in a 43 m area. A single queen was located in each
of 3 nests.


34
more nests, hence polydomous instead of monodcmous (single nest). Sanders
and Baldwin (1969) tagged carpenter ants, Camponctus spp., with Iridium-
193
192 to trace underground trails. Levieux (1971) used Au to study the
134
nest structure of 2 Camponotus spp. The biological half-lives of Cs in
3 castes of Solenopsis invicta Burn were determined by Chiu et al. (1974).
Materials and Methods
All experiments using radioisotope tracers on Conomyrma were conducted
iri forest-pasture areas of the Tall Timbers Research Station, Leon Co.,
134
Florida during August 1972. The selection of Cs for this experiment
was based on its relatively short half-life (2.6 yr compared to 30 yr for
137Cs). Six shell vials containing a mixture of 1.5 microcuries of 134Cs
and 0.5 g of sugar were prepared and dried with infrared heat. One min
counts with a single channel gamma spectrum analyzer, with a 3 in. Nal-Tl
137
crystal adjusted to optimum counting with a calibrated Cs source, ave
raged 45,000 cpm per 0.5 g of sugar for each of the vials containing the
prepared mixture.
Four areas of £. insana nest craters and 2 areas of C^. flavopecta nest
craters were selected. In each locale, 20 craters were marked with num
bered stakes; 1 crater was designated as the "source" at which the radio
isotope was incorporated into the nest. A diagrammatic map giving distance
and orientation of the nests to the source crater was constructed for each
area. One shell vial containing the sugar and 134Cs was placed adjacent
to the entrance hole of the source crater. The shell vial and crater en
trance were covered with an inverted quart glass jar to insure that only
ants from the source crater were able to feed on the radioactive sugar.
The jar and empty shell vial were removed after 24 hr. At 1, 2, 4, and 10


35
days after incorporation of the radioisotope into the source crater, it
and all marked peripheral nests were sampled. The ant samples were
taken with a standard aspirator equipped with a rubber squeeze bulb to
avoid inhalation of the isotope. Each sample of ants was labeled and
134
taken to the laboratory where total radioactivity of Cs was assayed
with a single channel analyzer.
Beginning in 1973, nest of both Conomyrma species were periodically
excavated and examined for structure and contents. The presence of
workers, brood, and reproductives was noted.
Results and Discussion
134
Conomyrma insana workers with Cs were recovered from 79 of the
80 nests sampled, including the 4 source nests. Workers taken from 1
nest in replicate A (Fig. 2.1) were not radioactive. The percentage of
134
Cs assayed in the workers taken from all sampled nests in replicates
A, B, C, and D (Fig. 2.1) was .46, .58, .63, and 1.27, respectively, of
134
ca. 45,000 cpm of Cs placed at each source nest. The recovery of
radioactive workers from apparently independent nest units indicated
food trophyllaxis between individuals other than nest mates, in addition
to exchange of individuals among the nests. Therefore, the data indicate
that a colony of insana is composed of multiple nests, hence polydo-
mous. The probability that only a single colony was present at each
locale was obtained by dividing the number of nests with radioactive
ants by the number of nests sampled. The results of the analysis are
presented in Table 2.1. The averages over the replications can be
viewed as tne level of confidence possessed in rejecting the hypothesis
that more than a single colony of C^. insana was present at each locale.
The greatest distance separating any 2 sampled nests of the same colony


98
Roger, J. 1863. Die ner aufgefuhrten Gattungen und arten meines
Formiciden-Verzeichnisses. Berliner Entomologische Zeitschrift.
7:131-214.
Sanders, C. J., and W. F. Baldwin. 1969. Iridium-192 as a tag for car
penter arits of the genus Camponotus (Hymenoptera:Formicidae). Can.
Entomol. 101:416-8.
Santschi, F. 1906. A propos des moeurs parasitiques temporaires des
fourmis de genre Bothriomyrmex. Annales de la Societie Entomolo-
gioue de France. 75:363-92.
Santschi, F. 1920. Fourmis du genre Bothriomyrmex Emery. Revue
Zoologizue Africaine. 7:201-24.
Santschi, F. 1922. Myrmecines, Do!ichoderines et autres formicides
neotropiques. Bull. Soc. Sci. Nat. Laussane 54:345-78.
Schwarz, E. A. 1907. Note on Dorymyrmex pyramicus Roger. Proc. Entomol
Soc. Wash. 7:4.
Smith, M. R. 1924. An annotated list of the ants of Mississippi (Hym.).
Entomol. News 35:45-54, 77-85, 121-7.
Smith, M. R. 1936. The ants of Puerto Rico. J. Agr. Univ. Puerto Rico
20:819-75.
Smith, M. R. 1944. Additional ants recorded from Florida, with
descriptions of two new subspecies. Fla. Entomol. 27:14-7.
Snellinq, R. R. 1973. The ant genus Conomyrma in the United States.
(Hymenoptera:Formicidae). Contributions in Science. Los Angeles
Co. Natr. Hist. Mus. 238. 6 p.
Steel, R. G. D. and J. H. Torrie. 1960. Principles and Procedures of
Statistics. McGraw-Hill Book Co., Inc., N.Y. 481 p.
Sudd, J. H. 1967. An introduction to the behaviour of ants. Edward
Arnold Ltd., London 200 p.
Tugwell, P., F. D. Miner, and D. E. Davis. 1972. Threecornered alfalfa
hopper infestations and soybean yields. J. Econ. Entomol. 65:1731-
3.
Van Pelt, A. F. 1966. Activity and density of oldfield ants of the
Savannah River Plant, South Carolina. J. Elisha Mitchell Sci.
Soc. 82:35-43.
Van Pelt, Jr., A. F. 1958. The ecology of the ants of the Welaka Re
serve, Florida (Hymenoptera:Formicidae). Part II. Annotated List.
Amer. Midi. Nat. 59:1-57.
Vick, D., W. A. Drew, J. Young, D. J. McGurk, and E. J. Eisenbraun. 1969
Chemotaxonomic studies of ants: Volatile compounds. Can. Entomol.
101:879-83.


Table 6.1. Numbers of Splsslstilus festinus nymphs in 2 areas of a Florida soybean field: one area in
which nymphs were tended by Solenopsis geminata and another in which nymphs were tended by
Conomyrma insana.
% of plants examined with:
Ant
Species
Survey
date
0
nymphs
i
nymphs
2
nymphs
3-5
nymphs
Nymphs/plant
x + 95% Cl (a)
Nymphs/ha
x + 95% Cl la)
S. geminata
Oct. 3
36.0
49.0
11.0
4.0
0.84 + 0.22
111.3 + 32.7
(area 1)
Oct. 15
67.0
28.0
5.0
0.0
0.38 + 0.20
63.8 + 38.1
C. insana
Oct. 3
30.0
43.3
20.0
6.7
1.10 + 0.90
140.7 + 140.6
(area 2)
Oct. 15
56.7
36.7
6.7
0.0
0.50 + 0.25
63.5 + 42.0
(a)mean + 95% confidence interval.


65
Pelt (1966) noted that Dorymyrmex pyramicus (= Conomyrma insana based
on Van Pelt's description of the nesting habits) tended 2 aphid species.
Research by Odum and Kuenzler (1963), Wiegert et al. (1967), and de la
Cruz and Wiegert (1967), using ^2P to trace the plant-aphid-ant food
chain, demonstrated that £. pyramicus obtained honeydew from aphids of
the genera Aphis and Macrosiphum. Jones (1929) listed 3 aphids (Aphis
sp., Chaitophorous negundinus Thos, and £. populella Gill. and Pal.)
tended by [). pyramicus.
The objective of this research was to determine the sources of
carbohydrates available to £. insana within their foraging area.
Methods
Investigations were conducted at Tall Timbers Research Station,
Leon Co., Florida from July 1972 through August 1975. Eight locations
with high populations of Conomyrma insana were selected. The flora
within and adjacent to the C^. insana enclaves were examined for Homop-
tera tended by ants and ants feeding from floral and extrafloral nectar
ies. With Homoptera, determination of tending was based on observing
the workers antennuate an individual and subsequently receive a droplet
of exudate. The presence of the ant on the same plant as the Homoptera
was not accepted as prima facie evidence of ant-aphid association. All
new Homoptera and their host plants, and plants with nectaries were
collected and identified.
Beginning in 1973, a distribution map was constructed annually of
the Z. insana nests around a solitary short leaf pine located in an open
field. To construct the map, the distance of each nest from the base
of the pine was measured. Using the pine tree as a base point, the
compass angle of each nest was also determined.


66
Results and Discussion
Homoptera Tended by Conomyrma insana
Sixteen species of aphis (Table 5.1) and 10 other species of Homop
tera (Table 5.2) viere tended by Conomyrma insana. These homopterans
were colonizing plants within the foraging area of the ant. The Homop
tera were collected from 13 plants families; these were Pinaceae, Gra-
mineae, Fagaceae, Chenopodiaceae, Amaranthaceae, Rosaceae, Caesalpinia-
ceae, Fabaceae, Onagraceae, Labiatae, Scrophulariaceae, Rubiaceae, and
Compositae.
The aphids Cinara watsoni Tissot, Sipha flava (Forbes), Aphis gossy-
pii Glover, A. craccivora Koch, A. spriaecola Patch, Cerosipha rubifoli
Thomas, and Hysteroneura setariae (Thomas) were common in all areas where
£. insana occurred. Rhopalosiphum maidis (Fitch) were numerous on Paspa-
lum boscianum Flugge and corn, Zea mays L., in only one area. Another
aphid, Myzocallis bella (Walsh), was found only on running oak, Quercus
pumila Walter, in one field. The remaining aphids presented in Table
5.1 were rarely collected.
The aphids Dactynotus tissoti (Boudreaux), collected on Solidaqo sp.
(Goldenrod); ]3. ambrosine (Thomas), collected on Solidago sp. and Pyrrho-
pappus carolinianus (Walt.) DC.; ID- helianthicola Olive, collected on He-
lianthus augustifolius L.; and CL tutaiae Olive, collected on Ambrosia
artemissiifolia L. (Ragweed) were within the foraging areas of £. insana.
The ant was often found on the same plant as the aphids, but there v/as
no evidence that these aphid species were tended by £. insana.
Adelges sp. and Toumeyella parvicornis (Ckll.) were both tended by
£. insana on short leaf pine. Both were attractive to the ant, and large
numbers of foraging workers were constantly present on the pine.


5
The genus Dorymyrmex was established by Mayr (1866). Later, Cono-
tryrma was established by Forel (1913) as a subgenus of Dorymyrmex.
Santschi (1922) designated Preo!epis pyramica Roger, as the type species
of Conomyrma; Donisthorpe (1943) redesignated the same species as the
type. The subgeneric name was rejected by Creighton (1950) on the grounds
that Conomyrma was synonymous with Dorymyrmex, once those species lacking
a propoceal tooth were removed to Araucomyrmex, as was done by Gallardo
(1916).
Kusnezov (1952) restudied the problem and proposed to divide Dory
myrmex into 2 genera, Dorymyrmex and Conomyrma. The genus Conomyrma was
separated from Dorymyrmex in the worker caste by the lack of a psammo-
phore and in the female caste by the presence of a single, versus two,
cubical cells in the forewing. Further morphological differences sepa
rating the genera were described by Eisner (1957) and Wheeler and Wheel
er (1951). Unfortunately, Kusnezov (1952) failed to designate type
species for the new subgenera proposed, but he corrected his oversight
in a subsequent paper (Kusnezov 1959). The generic name Conomyrma was
recognized by Snelling (1973) as a valid genus.
Roger (1863) described Prenolepis pyramica from a single specimen
from Corrientes, state of Bahia, Brazil. The name was transferred to
Dorymyrmex by Mayr (1866). By 1900, pyramicus was assumed to range from
Argentina to the southern United States and over the entire Carribbean
area (Forel 1899) and had acquired a number of varieties and subspecies
along with an entirely different identity (Snelling 1973).
Norton (1875) suggested that Buckley's Formica insana may prove to
be Dorymyrmex pyramicus. McCook (1879) concurred that Norton may have


APPENDICES


23
N


CHAPTER I
DISTRIBUTION AND HABITAT OF Conomyrma insana
Introduction
The nature and abundance of suitable nesting sites are important
factors determining the distribution of a species of ant (Letendre and
Piln 1973). In initial investigation of Conomyrma insana, large num
bers of nests were found clustered in some areas while other apparently
ideal sites were found where the ant was conspicuously absent. Accord
ing to Naves (1974), C^. insana are more populous in dirt roadways in
A
open areas. Markin et al. (1974) mentioned that C_. insana was common
in portions of open fields in Louisiana. Other than brief notations
of the ant occurring in shady areas, there are no definitive studies
on the habitat of C_. insana.
Therefore, to define the habitat of this ant necessitated locating
and comparing sites where the ant was established. Tall Timbers Re
search Station near Tallahassee, Fla. was chosen for this work because
of the diversity of ecosystems represented. In addition, records were
available for each site and, on request, the treatment of each location
of the station could be maintained.
Methods and Materials
Beginning in 1972, open fields and roadways within the 1133 hectares
(2800 acres) of Tall Timbers Research Station near Tallahassee, Fla. were
searched for Conomyrma insana. The locations of all C_. insana populations
11


ECOLOGY AND BEHAVIOR OF CONOMYRMA INSANA
By
JAMES C. EVERETT NICKERSON
A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY
*
UNIVERSITY OF FLORIDA
1976


Fig. 1.2. Conoroyrma insana nests (stippled areas) in Introduction Field
during July, 1973.


Table 5.2.Continued.
Plant Host
Species
Family
Species (Common Name)
Eriococcidae
Eriococcus quercus
Eriococcus smithii
Kermesidae
(Comst.)
Lobdell
Fagaceae
Gramineae
Quercus virginiana Miller (Live oak)
Andropogon virginicus L. (Broom sedge)
Kermes sp.
Membracidae
Fagaceae
Quercus nigra L. (Water oak)
Unidentified sp.
Compositae
Ambrosia artimisiifolia L. (Ragweed)


39
was approximately 5.5 m. The finite number of nests comprising a colony
was not determined in the experiment.
Cesium-134 was recovered only from the source nest of £. flavo-
pecta. All samples of ants taken from nest exits other than the source
nest registered normal background radioactivity. Thus, £. flavopecta was
monodomous as there was no evidence of exchange of individuals or food
between the individual nests.
For the purposes of this research, a terrestrial ant "nest" means a
series of closely interconnected tunnels and chambers which have either
a single opening to the surface or several closely interconnected open
ings; the entire structure obviously spatially separated from other
similar nests. In the genus Conomyrma, each "nest" usually has a single
opening plus a vertical or slanted main tunnel with branching tunnels and
with lateral chambers located at various distances from the surface.
A total of 27 clustered groups of C. insana nests has been excavated
to date. Of these, dealate queens were found in 5 of the clustered
2
groups. On 25 May 1973, 18 nests were excavated in a 20 m area. Sixteen
dealate queens were located in 1 nest at depths ranging from 4-13 cm. On
2
15 June 1973, 32 nests were excavated in a 21 m area. One nest con
tained 13 dealate queens located within 5-15 cm of the soil surface.
This nest had large masses of new eggs in addition to larvae and pupae in
all stages of development. All other nest sites excavated in this series
contained larvae and pupae, but no eggs. On 5 April 1973, a series of 18
2
nests located in an area of 25 m was excavated. No queens were recov
ered from 17 of the nests excavated in the area. In 1 nest, 9 dealate
queens were recovered from a single chamber 20-22 cm below the surface.
There was no surface indication that this nest differed from others


PROGNOSIS
From the foregoing studies, it is obvious that the distribution of
Conomyrma insana (Buckley) is severely limited by ecological requisites.
The most obvious limiting factors are the availability of suitable nest
ing sites, food sources, and its temporary host, £. flavopecta (M. R.
Smith).
In general, C_. insana nests must be located in open areas fully ex
posed to sunlight. The majority of the sites where C^. insana nested
were acceptable owing to recent disturbances such as mowing, cultivation,
or burning. The ant ceased to nest in sites where normal vegetational
succession occurred. This can be accepted as proof of their dependence
upon open areas for nesting. Additionally, the ant is excluded from
excessively hydric locations.
A further restriction on nesting sites is that a location must in
clude or be adjacent to an adequate food supply. For £. insana this
means that areas of heavy vegetation must be nearby, as they are depen
dent on carbohydrate food sources for the energy required to retain a
site. The carbohydrate food can be from plant nectaries or honeydew
excreted by Homoptera.
The successful colonization of an area is further limited by the
nature of C. insana's habits. Conomyrma insana is temporarily parasitic
on C. flavopecta. Although no analysis is available, it is reasonable
to assume that, based on prior studies with other temporary parasitic
ants, the percentage of successful parasitism is extremely low (Wilson
91


8
F. in cotton. In addition, Pierce (1912) noted that the ant had been
recorded preying on the boll weevil, Anthonomus grandis Boh. Girault
(19G7) observed jL pvramicus attacking larvae of the lesser peach tree
borer and occasionally killing newly emerged moths. Quaintance and
Jenne (1912) stated that the ant JD. pyramicus attacked the plum curculio,
and that the ant was influential in the destruction of curculio. Van
Pelt (1966) reported that foraging workers of j). pyramicus (= C_. insana
based on Van Pelt's description of the nesting habits) were repeatedly
seen carrying insect parts. Whitcomb et al. (1972) stated that the
black form of C^. pyramicus (= insana) is one of the most important
predaceous ants in north Florida soybean fields, attacking leaf hoppers,
small beetles, and other insects.
Fincher and Lund (1967) noted that D. pyramicus killed some of the
fire ant queens, Solenopsis saevissima richteri Forel (= S^. invicta
Burn) before they had a chance to burrow into the ground after their
nuptial flight. Markin et al. (1971) reported that C^. pyramicus (=
insana) were seen dismembering mated females of the imported fire ant
and stated that the dolichoderine ant, where it is numerous, may be one
of the main predators of fire ant queens. Whitcomb et al. (1972) and
Whitcomb (1974) stated that the black form of C_. pyramicus (= C^. insana)
is apparently one of the most aggressive ants toward deal ate females of
S. invicta before they dig into the ground. In another report, Whitcomb
et al. (1973) stated that they had observed 368 S_. invicta queens killed
£ insana workers in a single afternoon. Hung (1974) reported that
he recovered the heads of the queens of 5 species of myrmecine ants from
the refuse pile of £. Insana (= undescribed species of Conomyrma based on
voucher series from his study).


82
pruinosus were uncommon, and workers of each species were observed
tending nymphs on only 2 occasions. Conomyrma flavopecta and Solenopsis
geminata occurred over most of the interior of the field. Workers of S_.
geminata foraged on soybean plants from underground trails radiating
from their mounds and opening at the bases of plants on which nymphs
were being tended. Three clusters of Conomyrma insana nests were pre-
2
sent along 1 edge of the field. Each enclave covered ca. 45 m and
extended ca. 3 m into the field. No other nests of £. insana were found
in the field.
In the interior of the field (area 1), workers of S_. geminata or,
occasionally, workers of flavopecta were observed tending three-
cornered alfalfa hopper nymphs. On 3 and 15 October, workers of
geminata were found on 94.6% of the plants with nymphs and on 15.0% of
the plants without nymphs. Workers of flavopecta were on the remain
ing 5.4% of the plants with nymphs and on 0.8% of the plants without
nymphs. Within a C^. insana enclave (area 2), C^. insana was the only ant
species observed tending threecornered alfalfa hopper nymphs. On 3 and
15 October, at least 1 insana worker was found on every plant with
nymphs. Conomyrma insana workers were present on 83.0% of the plants
without nymphs. Conomyrma insana workers and, to a lesser extent, S_.
geminata workers were present on plants without nymphs but at lower
frequency than on plants with nymphs.
Estimates of the densities of the threecornered alfalfa hopper
nymphs are given in Table 6.1. Although nymphs were found on 50.4% of
all plants examined on 3 and 15 October, only a single nymph was present
on 77.7% of the plants with nymph(s). Densities of nymphs were not


77
nictitans L. and some composites. The distribution of the plants were
uniform with most of the soil surface exposed to sunlight. An occasional
short leaf pine, Pinus echinata Miller, was present in the field; an en
clave of C.. insana surrounded each of these trees. The only attractant
in each of the pine trees was the pine aphid, Cinara watsoni. The work
ers of C^. insana, from the nests surrounding the trees, continuously
tended these aphids.
Throughout the 4 years of monitoring this population of ants, nests
were maintained around the trees as diagrammed in Figs. 5.1-5.4. No
other major carbohydrate food sources were near this location, although
an occasional group of the aphid Hysterouneura setariae was on broom
sedge. Apparently, while the presence of the aphid on the pine tree was
instrumental in determining the occurrence of £. insana, the quantity of
carbohydrates available from this and other sources was insufficient to
maintain a larger colony. This hypothesis is supported, in part, by the
recorded increase of the number of nests from 31 in 1973 to 50 in 1976,
while there was corresponding increase in the pine from ca. 2 m to 4.5
m in height at the same time. Theoretically, the pine would be capable
of supporting a larger population of aphids which, in turn, would
support a small increase in the ant colony.
Summary
Twenty-six species of Homoptera were tended by Conomyrma insana
(Buckley). The Homoptera were in the families Aphididae (10), Phyllox-
ieridae (1), Coccidae (5), Pseudoccidae (1), Eriococcidae (3), Kermes-
idae (1), and Membracidae (1). Conomyrma insana also fed from the
nectaries of Cassia nictitans L., C. fasciculata Michx., C. obtusifolia,
and Richardia scabra L.


14


APPENDIX A
GLOSSARY OF SPECIFIC TERMS
Alate. Winged, having wings.
Brood chamber. Chamber constructed for purpose of rearing brood.
Callow. Immature; green. Freshly pupated adult ant prior to darkening.
Claustral. Form of colony establishment in which a freshly inseminated
queen ant digs and seals brood chamber, lays eggs, and feeds first
developing brood from stored energy.
Contiguous. Adjoining; near, though not in contact.
Deal ate. Wingless; a term applied to formerly winged females or queen
ants that have shed their wings.
Declivity. Moderately steep. Declivity of mesonotum; the inclined
surface of the mesonotum.
Enclave. An enclosed tract or territory with definite boundaries.
Epinotum. That part of the thorax behind the mesoepinotal suture; in
worker ants the term is loosely applied, since the region morpho
logically is composed of two segments, an anterior metathorax and
a posterior epinotum.
Extrafloral nectaries. Plant structures not associated with flowers that
produce a sweet secretion or nectar relished by ants as food.
Floral nectaries. Certain parts of flowers that secrete a sweet liquid
or nectar relished by ants for food.
Gaster. That portion of the abdomen behind the petiole in ants with a
single segmented petiole, and behind the postpetiole in ants with
a two-segmented petiole.
Gyne. Woman; female. In ants, refers to the reproductive form, or
queen ant.
Hesmosis. Process of new colonies forming by "budding" or "colony
fission" from parent colony.
Honeydew. A sweetish alimentary excretion produced by certain insects
such as plant lice, mealybugs, and scales, and greatly relished as
food by ants.
102


38
Table 2.1. Probability that only a single colony of Conomyrma insana
occurred within each replicate of 20 nests sampled for workers
containing Cesium-134.
Days following application
Replicate
Ia
2a
4a
10a
A
0.45
0.75
0.85
0.95
B
0.55
0.70
0.85
1.00
C
0.45
0.90
1.00
1.00
D
1.00
1.00
1.00
1.00
Average over Reps.
0.613
0.838
0.925
0.988
Probability (a single colony) =
of nests with
Total no.
radioactive
of nests
ants


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EBHB7AZ8L_WZ9OW2 INGEST_TIME 2015-04-14T18:47:31Z PACKAGE AA00029848_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES


58
resulting in 18 cm rainfall. The temperature at 1414 was 35 C and the
v/ind velocity varied from 8 to 24 kmh. The alates of SL invicta emerged
from the mounds at 1331. The first queen observed returning from the
nuptial flight alighted at 1450.
Consistencies and Variations in Behavior
Detection. When a S. invicta queen alighted in an area heavily
populated by £. insana, a period, varying from a few seconds to several
minutes, elapsed prior to initial contact by (1. insana. During this
period, if sufficient, the queen shed her wings and commenced searching
for a suitable site to construct the brood chamber. Detection of the S^.
invicta queen by C_. insana workers occurred any time after the queen
alighted, depending upon the proximity of the workers. The first con
tact with the queen was made by a single worker approaching from the
front or side with mandibles spread and maintaining only momentary con
tact with the queen. It is probable that the queen was marked with an
identifying compound by the £. insana worker. Following attacks were
not comparable to the initial attack in that subsequent attacks were
progressively more frequent and prolonged. In later attacks, the workers
remained attached to the queen until dislodged. In the first approach,
penetration by the mandibles was never observed and contact was made at
any portion of the body.
During a short span of time following the initial contact, the
queen was unmolested by C^. insana workers. Again, the queen may attempt
to remove the wings (if not previously shed), but the usual reaction of
the queen was to flee from the point of attack. In most observed cases,
this was followed by a halcyon period as evidenced by the less erratic


Fig. 1.3. Conomyrma insana nests (stippled areas) in Introduction Field
during July, 1974.


72
Coccus hesperidium linn annually built up heavy infestations in a
soybean field. The scale was also tended by Solenopsis invicta Burn on
the interior of the field.
The mealybug Oracella sp. was found only on a single live oak tree,
Quercus virginiana Miller. This population persisted in large numbers
on the oak.
The first collected specimens of Eriococcus smithii Lobdell were
taken from 7 clumps of broom sedge, Andropogon virginicus L. in a shal
low ditch between a forest road and a soybean field. By September 1974,
colonized broom sedge clumps had increased to 43, spaced out over
approximately 30 m within the shallow ditch.
Kermes sp. were common on water oak, Quercus nigra !_., and were
heavily tended by £. insana. A species determination was impossible as
they were so heavily parasitized that no mature specimens were found.
Nectaries of Plants
Conomyrma insana constantly fed from the extrafloral nectaries of
Cassia nictitans L. and £. fasciculata Michx. and to a lesser degree on
C_. obtusifolia L. Cassia nictitans was the most prevalent plant at most
C^. insana locales. In cultivated fields, mexican clover, Richardia sca-
bra L., was a common weed. The floral nectaries of this plant were
frequently visited by £. insana.
Influence of Carbohydrate Food Source
The influence of a carbohydrate food source on selection of nesting
sites by C^. insana is illustrated in Figs. 5.1-5.4. In this field, vege
tation was composed of scattered clumps of broom sedge, Andropogon vir-
o
ginictis L. (ca. 2.1 clumps per nr), with sparse occurrence of Cassia


2
insana in north Florida. The compilation of this information was con
sidered preparatory to future efforts to increase the populations of £.
insana through habitat manipulation and, thereby, increase the
effectiveness of the ant as a predator of fire ant queens or other
target pests.


ECOLOGY AND BEHAVIOR OF CONOMYRMA INSANA
By
JAMES C. EVERETT NICKERSON
A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY
*
UNIVERSITY OF FLORIDA
1976

This dissertation is dedicated to George Davis of Horatio, Arkansas.
The debt owed by myself and countless former students of his that he
aided can never by fully repaid. Without his help and encouragement, I
would not have had the opportunity to embark upon the studies required
to begin, much less complete, this work.

TABLE OF CONTENTS
ACKNOWLEDGEMENTS
Page
i i i
ABSTRACT Vi
INTRODUCTION 1
REVIEW OF LITERATURE 3
Introduction 3
Taxonomy 4
Physiology 7
Predation 7
Homoptera Tending 9
Habitat 9
Distribution 10
CHAPTER
I. DISTRIBUTION AND HABITAT OF Conomyrrr.a insana II
Introduction 11
Methods and Materials 11
Results and Discussion 12
Summary 24
II. COLONY ORGANIZATION AND NUMBERS OF QUEENS 33
Introduction 33
Materials and Methods 34
Results and Discussion 35
Conclusions 41
Summary 42
III. MIXED NESTS OF Conomyrma insana and C. flavopectaEVIDENCE
OF TEMPORARY SOCIAL PARASITISM 43
Introduction 43
Methods 45
Results 45
Discussion 43
Summary 50
iv

P^e
IV.PREDATION ON FOUNDING QUEENS OF Solenopsis invicta BY
WORKERS OF Conomyrma insana .... 51
Introduction 51
Methods 52
Results and Discussion 53
Summary 63
V.SOURCES OF CARBOHYDRATES UTILIZED BY Conomyrma insana 64
Introduction 64
Methods 65
Results and Discussion 66
Summary 77
VI.PRESENCE OF Spissistilus festinus TENDED BY ANTS: A
FACTOR AFFECTING EGG PREDATION BY ANTS IN SOYBEANS. ... 79
Introduction 79
Methods 79
Results and Discussion 81
Conclusions 88
Summary 89
PROGNOSIS 91
LITERATURE CITED 94
APPENDICES 101
A. GLOSSARY OF SPECIFIC TERMS 102
B. PUBLICATIONS 104
BIOGRAPHICAL SKETCH 105
v

Abstract cf Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
ECOLOGY AND BEHAVIOR OF CONOMYRMA INSANA
By
James C. Everett Nickerson
December, 1976
Chairman: Dr. W. H. Whitcomb
Major Department: Entomology and Nematology
Conomyrma insana (Buckley) enclaves (large numbers of contiguous
nests with defineable boundaries) were found at 29 locations on the
Tall Timbers Research Station in Leon Co., Fla. The ants were more
commonly found in open areas adjacent to heavy vegetation. Nests were
absent in cultivated areas, areas of heavy vegetation, 3nd where
drainage was poor.
The organization of the colony and number of queens per colony of
C. insana and C. flavopecta (M. R. Smith) were investigated using radio
isotope tracers and by excavation of nests. Conomyrma insana was poly-
domous (more than 1 nest per colony), and had several queens located in
a central nest. Conomyrma flavopecta was monodomous (1 nest per colony)
with 1 colony queen.
Conomyrma insana and C. flavopecta were found in the same nest.
Later, clusters of £. insana nests were formed at this location. Based
on the evidence, £. insana is a temporary social parasite on C^.
flavopecta.
vi

Conomyrma insana workers are effective predators of fire ant found
ing queens, Sclenopsis invicta Burn. The confrontation behavior of the
2 species was determined. Worker detection of a queen alighting in a £.
insana enclave varied from a few seconds to several minutes. The found
ing queen's reactions tc attack by £. insana consisted of 5 basic respon
ses: (1) escape, (2) concealment, (3) cessation of movement, (4) mandi
bular defense, and (5) rapid flexing or shaking. Less than 3% of the
queens observea alighting within a C^. insana enclave successfully
escaped, and none survived to develop a new colony.
Twenty-six species of Homoptera were tended by Conomyrma insana.
The Homoptera were in the families Aphididae (10), Phylioxieridae (1),
Coccidae (5), Pseudoccidae (1), Eriococcidae (2), Kermesidae (1), and
Membracidae (1). Conomyrma insana also fed from the nectaries of Cassia
nictitans L., £. fasciculata Michx., C_. obtusifolia L., and Richardia
scabra L, Conomyrma insana construct their nests near carbohydrate
food sources,
Five ant species, of which iC. insana and Solenopsis geminata (F.)
were the most abundant, tended nymphs of the threecornered alfalfa hopper,
Spissistilus festir.us (Say), in a soybean field. The presence of the
ants under these conditions was examined as a factor affecting predation
on eggs of the soybean looper, Pseudoplusia includens Walker. Where S^.
geminata tended the nymphs the percentage of eggs missing after 24 hr was
significantly greater from plants with nymphs (77.1%) than from plants
without nymphs (36.8%)* Where £. insana tended the nymphs, the percent
age of eggs missing was significantly greater from plants with nymphs
than from those without nymphs after 4 hr (36.8% vs. 17.5%) and 8 hr
vii

(65.5% vs. 42.9%) but not after 24 hr (SO.7% vs. 86.3%). The proportion
of eggs missing was greatest from the bottom 1/3 of the plant and least
from the top 1/3.
viii

INTRODUCTION
Ants of the dolichoderine genus Conomyrma occur commonly through
out most of the United States. Yet, within this country, comparatively
little attention has been directed toward the ecology or biology of the
species in this genus. The majority of published reports including these
ants consist of a brief mention in a list of ant species collected in
an area and/or a notation as to the type of habitat they were found in.
The paucity of attention toward the species of Conomyrma can be
partially attributed to the lack of any obvious economic or health
problems posed by their presence. Recently, though, accounts have been
published on observations of one species, Conomyrma insana (Buckley),
attacking and killing newly mated queens of the red imported fire ant,
Solenopsis invicta Burn (Fincher and Lund 1967, Markin et al. 1971,
and Whitcomb et al. 1972). These reports prompted this investigation
of £. Insana in northern Florida as the presence of S_. invicta does pose
economic and health problems.
An examination of the ecological requirements of C_. insana was
deemed necessary to evaluate the potential of this species as a candi
date biological control agent of the red imported fire ant. In addition,
an investigation of predation by C^ insana on newly mated fire ant
queens and other arthropods was required to evaluate the efficiency of
this ant as a biological control agent.
Consequently, my primary objective was to determine and document
the principal factors affecting the distribution and populations of £.
1

2
insana in north Florida. The compilation of this information was con
sidered preparatory to future efforts to increase the populations of £.
insana through habitat manipulation and, thereby, increase the
effectiveness of the ant as a predator of fire ant queens or other
target pests.

REVIEW OF LITERATURE
Introduction
In the continental United States, three species of the genus Cono
myrma--bicolor (Wheeler), flavopecta (M. R. Smith), and insana (Buck-
ley)--were recognized by Snelling (1973). Prior to Snelling's revision,
the species of Conomyrma in North America and neighboring areas were
lumped as Dorymyrmex pyramicus (Roger) or Conomyrma pyramica (Roger).
Both Cole (1957) and Wilson (1957) suggested that bicolor was a
distinct biological species, based on their independent observations
that C_. bicolor occurred in intimate sympatry with C_. pyramica while
remaining distinct in worker color and size.
It is now evident that the Dorymyrmex/Conomyrma pyramicus group
is a complex comprised of several species. In the past, several myrme-
cologists have attempted to review comprehensively the taxonomic tangle
surrounding this complex. The most natable attempts were made by Sants-
chi (1922), Creighton (1950), Kusnezov (1952, 1959), Wilson (1957), and
Snelling (1973). For the most part, their attempts have had little
effect beyond increasing the confusion. There have been attempts by
several authorities, such as McCook (1879), Santschi (1922), M. R. Smith
(194), and Cole (1936, 1957), to separate individual species from the
complex, but few used solid morpholigical characters that would withstand
critical examination.
Gwing to this extreme confusion and until this complex is dealt with
as a whole, it is impossible to assign much of the literature references
3

4
correctly to the currently accepted species. Therefore, in the follow
ing discourse on the literature, I will deal with the entire complex in
the United States (excluding only those papers dealing strictly with C_.
bicolor). The name used by the author of each article will be quoted
unless I am able to determine that it may be safely and correctly
assigned to one of the species now recognized by Snelling.
Taxonomy
Conomyrma insana was described as Formica insana by S. B. Buckley
in 1866. His description of this ant is reiterated in entirety as
follows:
Worker. Length 0.14 inch--Color black or brownish-black,
smooth and shining throughout; head sub-quadrate, the lateral
margins slightly curved inwards; below oval, and rounded
above; eyes large, subelliptical, and placed on the anterior
portion of the head near its lateral margins; antennae long,
filiform; mandibles large, curved inwards, the truncated
apical ends sharply toothed; head sub-channelled beneath with
the anterior and posterior parts deeply depressed; prothorax
about half the width of the head, rounded above; mesothorax
somewhat depressed; metathorax, has a rudimentary spine or
sharp protuberance on its upper posterior surface, pedicle
short, inserted near the base of the anterior part of the
abdomen; scale small, wedge-shaped, and inclined forwards;
abdomen broad-ovate, subacute; legs small, slender, and rather
short.
Female. Length 0.20 inch--Head small, narrower than the
thorax; abdomen broad, oblong and ovate; color black-bronze
with the margins of the segments of the abdomen hyaline;
thorax raised above the head; wings not seen; the remainder
like the worker.
Dwells in the ground but a few inches deep, having numer
ous holes scattered over a large surface of fifty or more feet
in diameter. It goes now here and now there, without order,
from one abode to another, sometimes moving the larvae from
one place to another; large colonies dwelling in common. It
is very active and war-like, can conquer most other ants,
boldly seizing the larger species by the legs. If a large ant
is so unfortunate as to run among the "crazy ants," he hastens
to escape, but rarely does so without being severely bitten.
Common in Central Texas.

5
The genus Dorymyrmex was established by Mayr (1866). Later, Cono-
tryrma was established by Forel (1913) as a subgenus of Dorymyrmex.
Santschi (1922) designated Preo!epis pyramica Roger, as the type species
of Conomyrma; Donisthorpe (1943) redesignated the same species as the
type. The subgeneric name was rejected by Creighton (1950) on the grounds
that Conomyrma was synonymous with Dorymyrmex, once those species lacking
a propoceal tooth were removed to Araucomyrmex, as was done by Gallardo
(1916).
Kusnezov (1952) restudied the problem and proposed to divide Dory
myrmex into 2 genera, Dorymyrmex and Conomyrma. The genus Conomyrma was
separated from Dorymyrmex in the worker caste by the lack of a psammo-
phore and in the female caste by the presence of a single, versus two,
cubical cells in the forewing. Further morphological differences sepa
rating the genera were described by Eisner (1957) and Wheeler and Wheel
er (1951). Unfortunately, Kusnezov (1952) failed to designate type
species for the new subgenera proposed, but he corrected his oversight
in a subsequent paper (Kusnezov 1959). The generic name Conomyrma was
recognized by Snelling (1973) as a valid genus.
Roger (1863) described Prenolepis pyramica from a single specimen
from Corrientes, state of Bahia, Brazil. The name was transferred to
Dorymyrmex by Mayr (1866). By 1900, pyramicus was assumed to range from
Argentina to the southern United States and over the entire Carribbean
area (Forel 1899) and had acquired a number of varieties and subspecies
along with an entirely different identity (Snelling 1973).
Norton (1875) suggested that Buckley's Formica insana may prove to
be Dorymyrmex pyramicus. McCook (1879) concurred that Norton may have

6
beer, correct in his suggestion, yet he persisted in calling the ant £.
insanus. Forel (1899) placed insana in synonomy to pyramicus. Wheeler
(1902) stated that insana was undobutedly a synonym of pyramicus.
Buckley's insana was uniformly black or brownish-black; Roger (1363)
stated that pyramicus possessed a yellowish-red head and thorax and
brownish gaster. Roger's pyramicus, as described, agreed closely with
the description of a new variety, bicolor, by Wheeler (1906).
Creighton (1950) retained insana as a synonym of pyramicus, yet he
indicated doubt as to the validity of his decision. He stated that
since neither Mayr nor Forel had seen Roger's (1863) types, there was
little to indicate that the "typical pyramicus" was the same insect that
Roger described or the same form to which that name is applied to in
North America. According to Snelling (1973), the ant long referred to
as pyramicus in North American literature has a mesonotum which is
sharply declivitous behind and is uniformly brownish to blackish with
the lower part of the head lighter. This ant clearly is not the same
entity as the South American pyramicus described by Roger (1863). The
earliest available name for this ant is insana. Based on the evidence,
Snelling (1973) recognized insana (Buckley), and gave the synonomy for
this name as follows:
insana (Buckley 1866)
= pyramicus, sensu Wheeler (1902), Creighton (1950),
not of Roger (1863)
= flavus McCook (1879)
= nigra Pergande
= antillana Forel
= smithi Cole (1936)
= brunea, Kusnezov (1952)
= wheeleri Kusnezov (1952)

7
Physiology
The isolation of 2-heptanone (methylpentyl ketone) from the anal
glands of Conomyrma pyramica and its role as the alarm pheromone were
described by Blum (1969) and Blum and Warter (1966). The substance
releases a complex behavioral pattern in the worker ant which includes
alarm, attraction, repellancy, and digging. Metcalf and Metcalf (1970)
reported that typical alarm behavior by £. pyramica was initiated in
the laboratory by exposing the worker ants to isosteres of 2-heptanone
(1,1,1-trifluro-methylheptanone, butyl acetate, and butylthiol acetate).
They also reported that a remarkable behavioral change was induced by
the isostere hexanoyl chloride, which invariably produced a violent
combat between worker ants which touched one another.
Blum (1973) reported that the anal glands of do!ichoderines are not
restricted to biosynthesis of terpenoid natural products. His state
ment is corrcbrated by research by McGurk et al. (1968), which showed
that C. pyramicus also produce the monoterpene iridodial.
Vick et al. (1969, 1969) investigated the analysis of volatile com
pounds and free amino acids in several ant species, including CL pyrami-
ca. to determine the potential of the chemicals as taxonomic characters
in the study of ant systematics. They concluded that valuable
information on ant classification and phylogeny may be obtained.
Predation
Comstock (1879) reported both D.. insanos (= CL insana) and D_. flavus
killing the "cotton-worm." Schwarz (1907) noted that CL pyramicus pro
tected solitary tree cotton from the boll weevil in Cuba. Pierce et al.
(1912) and Hunter and Pierce (1912) listed CL pyramicus var. flavus Mc-
Coo'< as an enemy of Alabama argil lacea (Hulsner) and Heliothis obsoleta


8
F. in cotton. In addition, Pierce (1912) noted that the ant had been
recorded preying on the boll weevil, Anthonomus grandis Boh. Girault
(19G7) observed jL pvramicus attacking larvae of the lesser peach tree
borer and occasionally killing newly emerged moths. Quaintance and
Jenne (1912) stated that the ant JD. pyramicus attacked the plum curculio,
and that the ant was influential in the destruction of curculio. Van
Pelt (1966) reported that foraging workers of j). pyramicus (= C_. insana
based on Van Pelt's description of the nesting habits) were repeatedly
seen carrying insect parts. Whitcomb et al. (1972) stated that the
black form of C^. pyramicus (= insana) is one of the most important
predaceous ants in north Florida soybean fields, attacking leaf hoppers,
small beetles, and other insects.
Fincher and Lund (1967) noted that D. pyramicus killed some of the
fire ant queens, Solenopsis saevissima richteri Forel (= S^. invicta
Burn) before they had a chance to burrow into the ground after their
nuptial flight. Markin et al. (1971) reported that C^. pyramicus (=
insana) were seen dismembering mated females of the imported fire ant
and stated that the dolichoderine ant, where it is numerous, may be one
of the main predators of fire ant queens. Whitcomb et al. (1972) and
Whitcomb (1974) stated that the black form of C_. pyramicus (= C^. insana)
is apparently one of the most aggressive ants toward deal ate females of
S. invicta before they dig into the ground. In another report, Whitcomb
et al. (1973) stated that they had observed 368 S_. invicta queens killed
£ insana workers in a single afternoon. Hung (1974) reported that
he recovered the heads of the queens of 5 species of myrmecine ants from
the refuse pile of £. Insana (= undescribed species of Conomyrma based on
voucher series from his study).

9
Homoptera Tending
Nielsson et al. (1971) reported the association of several aphid
species with £. pyramicus, but did not attempt to establish if a mutu
al istic relationship occurred between the ant and species of aphids
listed. In fact, they reported associations with Dactynotus sp., which
are not thought to be tended by any ant species. Van Pelt (1966) noted
that £. pyramicus (= £. insana based on Van Felt's description of the
nesting habits) tended 2 aphid species, Hysteroneura setariae (Thomas)
and Pteroconuna smithiae (Monel!). Research by Odum and Kuenzler (1963),
Wiegert et al. (1967), and de la Cruz and Wiegert (1967), using ^P to
trace the plant-aphid-ant food chain, demonstrated that £. pyramicus
was obtaining honeydew from aphids of the genera Aphis and Macrosiphum.
Jones (1929) listed 3 aphids, Aphis sp., Chaitophorous negundinus Thos.,
and C. populella Gill, and Pal., tended by £. pyramicus. Knight (1944)
reported that the hemispherical scale, Saisetta hemispherica Targioni,
was tended by £. pyramicus var. niger Perg. on Palay rubber vines.
Comstock (1879) reported that ID. insanus (= £. insana) feeds upon the
sweet exudations of plants, galls, and sweet excretions of aphids.
Habitat
While no habitat studies have been made on ID. pyramicus, numerous
researchers, including Cole (1940, 1942), Dennis (1938), Smith (1924,
1936), Van Pelt (1958), Wheeler (1905, 1932), and Wilson (1964), have
noted that this ant prefers to nest in open sandy areas. Markin et al.
(1974) stated that £. insana was common in a part of an open field in
Louisiana. According to Naves (1974), £. insana are more populous on
roadways or other open areas with sparse vegetation.

10
Distribution
At present, there is no real means of delineating the range of
insana. The range of Dorymyrmex/Conomyrma pyramicus is recorded to be
as far north as North Dakota and, according to Creighton (1950), occurs
in all states except the extreme northeastern United States.
The confirmed records of insana are from southern states.
These include Texas (Buckley 1866, Wheeler 1901), Arkansas (Roe 1973),
Mississippi (Markin et al. 1971), Florida (Whitcomb et al. 1973), Georgia
(Van Pelt 1966), and Louisiana (Markin et al. 1974).

CHAPTER I
DISTRIBUTION AND HABITAT OF Conomyrma insana
Introduction
The nature and abundance of suitable nesting sites are important
factors determining the distribution of a species of ant (Letendre and
Piln 1973). In initial investigation of Conomyrma insana, large num
bers of nests were found clustered in some areas while other apparently
ideal sites were found where the ant was conspicuously absent. Accord
ing to Naves (1974), C^. insana are more populous in dirt roadways in
A
open areas. Markin et al. (1974) mentioned that C_. insana was common
in portions of open fields in Louisiana. Other than brief notations
of the ant occurring in shady areas, there are no definitive studies
on the habitat of C_. insana.
Therefore, to define the habitat of this ant necessitated locating
and comparing sites where the ant was established. Tall Timbers Re
search Station near Tallahassee, Fla. was chosen for this work because
of the diversity of ecosystems represented. In addition, records were
available for each site and, on request, the treatment of each location
of the station could be maintained.
Methods and Materials
Beginning in 1972, open fields and roadways within the 1133 hectares
(2800 acres) of Tall Timbers Research Station near Tallahassee, Fla. were
searched for Conomyrma insana. The locations of all C_. insana populations
11

12
were marked on a map of the station (Fig. 1.1). (The map was prepared
by Roy Komarek in July, 1969 and was based on ASC aerial photograph
ADV194, 3-22-60.) The sites of insana populations were examined for
consistencies and variations of features.
The £. insana populations in two fields, referred to as Introduc
tion Field and Persimmon Field, were annually mapped in greater detail.
In 1973, a survey map of Introduction Field was prepared by using the
compass-traverse survey method (DeMoisy 1949) to establish the field
boundaries and locate major landmarks. Permanent markers (,5-m sections
of pipe driven into ground) were placed in secure locations on the edge
of the field. The boundaries of the £. insana population were located
on the prepared map. In 1974, the same procedure was initiated for
Persimmon Field.
2
The number of nests per m within C_. insana enclaves was examined
in Introduction Field (beginning in 1973) and Persimmon Field (beginning
2
in 1974). The nests in 100 randomly selected m quadrats were counted
annually in July through 1976.
Results and Discussion
Twenty-nine locally isolated populations of Conomyrma insana were
found on Tall Timbers Research Station (Fig. 1.1). At each site, numer
ous nests were present and apparently interconnected, based on worker
traffic, in addition to worker transfer of brood, from nest to nest.
The least number of nests found in a single locale was 17, while 2,187
nests were counted in another field. The contiguous nests of the popu
lations of C^. insana can only be defined as enclaves as they have
definite boundaries within which few other ant species occur.

Fig. 1.1 Tall Timbers Research Station with locations of Conomyrma in
sana populations indicated. Where populations were found in
open fields, the entire field is marked rather than the bound
aries of the limited distribution of C. insana within the field.
The locations are: (1) Jones Field, JZ) Field N-W5, (3) Gate
Field, (4) Middle Gate Road, (5) Sutty Field, (6) Butler Field,
(7) Gum Pond Course Road, (8) Field M-8, (9) Field M-9, (10) Gay
Field C, (11) Hall Island Drive Road, (12) Gay Field B, (13)
Field M-12, (14) Hall Island Course Road, (15) Field L-8 and
L-9, (16) Field L-3, (17) Whitcomb Plots Rep 3, (18) Whitcomb
Plots Rep 2, (19) Fennel Field, (20) Tripsacum Field, (21)
Field 1.-12, (22) Tripsacum Course Road, (23) Richard Field,
(24) Introduction Field, (25) Persimmon Field, (26) Field L-14,
(27) Pea Field, (28) Field M-15, (29) Stoddard Lab Field.

14

15
The soil at all £. insana nesting sites v/as sandy with little clay
or loam content. The nest craters were found only in areas fully ex
posed to sunlight. No nests were found in heavy vegetation. The vege
tation at all sites v/as sparse to absent because of cultivation, well-
maintained fire lanes, annual burning (spring), sterile soil, roadways,
or closely mowed areas. At one location, vegetation was absent owing
to a rapidly expanding gulley. The most common plants occurring within
the nesting areas of C_. insana were Andropogon vlrqinicus L., Di gl tari a
sp., Pas pal urn sp., Richardia scabra L., Cassia fasciculata Michx., C_.
nictltans L., Eupatorium capiHi folium (Lam.) Small, and Ambrosia
artemisiifalla L.
Apparently, slope had little effect on the distribution of C^. insana
nests. Although the majority of the areas where the ant was found were
gently sloped (1-6"), at 2 locations the slope was greater than 30*.
All sites were good to well drained. Poorly drained areas adjacent to
established enclaves of C^. insana were not utilized for nesting.
The C_. insana enclaves in Introduction Field from 1973 to 1976 are
shown in Figs. 1.2-1.5. The boundaries of the enclaves are somewhat
stable, but some change is evident from year to year. The major annual
variation occurred on the v/est side of the field. In 1973, this portion
of the field had an extremely patchy vegetative cover (ca. 30%) consist
ing primarily of Paspalum bosianum Flugge, and £. insana nests were
numerous (Fig. 1.2). In 1974 and 1975, the vegetation was apparently
too rank for C. insana as they no longer nested in this portion of the
field (Fig. 1.3 and 1.4). In 1976, C^. insana were nesting in a fire lane
along the edge of the west side of the field (Fig. 1.5). This fire lane
had been maintained annually, but in 1976 the vegetation was less profuse
than in previous years.

Fig. 1.2. Conoroyrma insana nests (stippled areas) in Introduction Field
during July, 1973.

17
N

Fig. 1.3. Conomyrma insana nests (stippled areas) in Introduction Field
during July, 1974.

19
N

Fig. 1.4. Conomyrma insana nests (stippled areas) in Introduction Field
during July, 1975.

21
N

Fig. 1.5.
Conomyrma insana nests (stippled areas) in Introduction Field
during July, 1976.

23
N

24
The locations of the enclaves of £. insana nests in Persimmon Field
changed little from 1974 through 1976 (Fig. 1.6-1.8). Some fluctuations
were evident in the boundaries that could be attributed to maintainance
practices. This field was annually cultivated and planted with corn,
and a fire lane was maintained around the field. In 1974, the fire lane
and fallow areas on the west side of the field were kept clean of vege
tation, and the £. insana nests were present over this entire cleared
area (Fig. 1.6). In 1975 and 1976, this areas was not kept clear, and
the vegetation became rank with no soil surface exposed to the sun. No
IS. insana nests were found in this heavy vegetation (Fig. 1.7 and 1.8).
Conomyrma insana nests were not found in the center of the corn
fields, although they were present in the edges of the fields. The re
peated cultivation of the corn field may have destroyed most of the
nests and prevented C^. insana from occupying this area. As demonstrated
in Introduction Field, the locations of the C^. insana enclaves remained
stable. The ant did not nest in 1 narrow area, ca. 12 m wide, on the
northern side nor on the southern side of the field. Both were low,
boggy areas in which drainage was poor.
The number of C_. insana nests per square meter in both Intro
duction Field and Persimmon Field remained relatively constant (Table
1.1). The average number of nests per square meter was slightly greater
in Persircnon Field than in Introduction Field. The only major differ
ence in treatment between these two fields was that Introduction Field
was mowed frequently while Persimmon Field was cultivated.
Summary
Conomyrma insana enclaves (large numbers of contiguous nests with
defineable boundaries) were found at 29 locations on the Tall Timbers

Fig. i.5. Conomyrma insana nests (stippled areas) in Persimmon Field
during July, 1974.

10 meters

Fig. 1.7. Conomyrma insana nests (stippled areas) in Persimmon Field
during July, 1975.

10 meters

Fig. 1.8. Conotnyrma insana nests (stippled areas) in Persimmon Field
during July, 1976.

10 meters

31
Table 1.1. Number of nests per square meter within £. insana enclaves
in Introduction Field and Persimmon Field, Tall Timbers
Research Station, Leon Co., Fla.
Year
Introduction Field
x + 95% C.I.(a)
Persimmon Field
x + 95% C.I.'a'
1973
2.4 + .41
-(b)
1974
2.7 + .32
3.6 + .32
1975
2.5 + .25
3.5 + .22
1976
3.1 + .76
3.8 + .27
[a(Mean + 95% Confidence Interval.
^b'No Samples taken.

32
Research Station. The ants were found more commonly in open areas adja
cent to heavy vegetation. Nests were absent in heavily cultivated areas,
in areas of heavy vegetation, and where drainage was poor.

CHAPTER II
COLONY ORGANIZATION AND NUMBERS OF QUEENS
Introduction
Investigations of the ecological distribution of ants are concerned
with colonies of ants rather than individuals, since the colony is the
unit of social organization. It is difficult to be certain whether each
nest structure is independent and whether there is a direct overground or
underground connection with neighboring nest structures. In investiga
tions of Concmyrma irsana (Buckley) and C_. flavopecta (M. R. Smith) in
north Florida, the nest craters of Ch insana appeared to be clustered
while the nest craters of £. flavopecta exhibited no tendency toward clus
tering. Both Buckley (1366) and Van Pelt (1966) hypothesized that a sin
gle colony of C. insana (= Dorymyrmex pyramicus (Roger) in Van Pelt's paper)
consisted of large numbers of nests, though they presented no conclusive
proof. I wished to determined the colony composition of each species more
precisely in order to make advances in the field ecology of C^. insana and
C. flavopecta and to compare the ecological requirements of the 2 species.
Until the advent of radioactive tracers and development of adequate
techniques, it was generally only possible to theorize that a colony of
ants was composed of more than 1 nest. Since related problems with ants
have been solved using radioactive tracers, their use was attempted in this
situation. Kannowski (1959) introduced P mixed with honey into source
nests. He concluded that a colony of this .species was composed of 1 or
33

34
more nests, hence polydomous instead of monodcmous (single nest). Sanders
and Baldwin (1969) tagged carpenter ants, Camponctus spp., with Iridium-
193
192 to trace underground trails. Levieux (1971) used Au to study the
134
nest structure of 2 Camponotus spp. The biological half-lives of Cs in
3 castes of Solenopsis invicta Burn were determined by Chiu et al. (1974).
Materials and Methods
All experiments using radioisotope tracers on Conomyrma were conducted
iri forest-pasture areas of the Tall Timbers Research Station, Leon Co.,
134
Florida during August 1972. The selection of Cs for this experiment
was based on its relatively short half-life (2.6 yr compared to 30 yr for
137Cs). Six shell vials containing a mixture of 1.5 microcuries of 134Cs
and 0.5 g of sugar were prepared and dried with infrared heat. One min
counts with a single channel gamma spectrum analyzer, with a 3 in. Nal-Tl
137
crystal adjusted to optimum counting with a calibrated Cs source, ave
raged 45,000 cpm per 0.5 g of sugar for each of the vials containing the
prepared mixture.
Four areas of £. insana nest craters and 2 areas of C^. flavopecta nest
craters were selected. In each locale, 20 craters were marked with num
bered stakes; 1 crater was designated as the "source" at which the radio
isotope was incorporated into the nest. A diagrammatic map giving distance
and orientation of the nests to the source crater was constructed for each
area. One shell vial containing the sugar and 134Cs was placed adjacent
to the entrance hole of the source crater. The shell vial and crater en
trance were covered with an inverted quart glass jar to insure that only
ants from the source crater were able to feed on the radioactive sugar.
The jar and empty shell vial were removed after 24 hr. At 1, 2, 4, and 10

35
days after incorporation of the radioisotope into the source crater, it
and all marked peripheral nests were sampled. The ant samples were
taken with a standard aspirator equipped with a rubber squeeze bulb to
avoid inhalation of the isotope. Each sample of ants was labeled and
134
taken to the laboratory where total radioactivity of Cs was assayed
with a single channel analyzer.
Beginning in 1973, nest of both Conomyrma species were periodically
excavated and examined for structure and contents. The presence of
workers, brood, and reproductives was noted.
Results and Discussion
134
Conomyrma insana workers with Cs were recovered from 79 of the
80 nests sampled, including the 4 source nests. Workers taken from 1
nest in replicate A (Fig. 2.1) were not radioactive. The percentage of
134
Cs assayed in the workers taken from all sampled nests in replicates
A, B, C, and D (Fig. 2.1) was .46, .58, .63, and 1.27, respectively, of
134
ca. 45,000 cpm of Cs placed at each source nest. The recovery of
radioactive workers from apparently independent nest units indicated
food trophyllaxis between individuals other than nest mates, in addition
to exchange of individuals among the nests. Therefore, the data indicate
that a colony of insana is composed of multiple nests, hence polydo-
mous. The probability that only a single colony was present at each
locale was obtained by dividing the number of nests with radioactive
ants by the number of nests sampled. The results of the analysis are
presented in Table 2.1. The averages over the replications can be
viewed as tne level of confidence possessed in rejecting the hypothesis
that more than a single colony of C^. insana was present at each locale.
The greatest distance separating any 2 sampled nests of the same colony

Fig. 2.1. Location of Conomyrma insana craters in replication A, B, C,
and D, showing the relation of sampled and nonsampled craters
to the source crater.

37
C. insana craters: ^
J
I
o
0 Source
0
Non-radioactive
Radioactive
O
o Not sampled
o
1 meter



o e
e


o



0
o'
o
%




o

0
o
A o
B
o
0
o
c

o e
o
o
e
o



*
o
.
3
O
O


e
e


3
c

0
O C
o D

38
Table 2.1. Probability that only a single colony of Conomyrma insana
occurred within each replicate of 20 nests sampled for workers
containing Cesium-134.
Days following application
Replicate
Ia
2a
4a
10a
A
0.45
0.75
0.85
0.95
B
0.55
0.70
0.85
1.00
C
0.45
0.90
1.00
1.00
D
1.00
1.00
1.00
1.00
Average over Reps.
0.613
0.838
0.925
0.988
Probability (a single colony) =
of nests with
Total no.
radioactive
of nests
ants

39
was approximately 5.5 m. The finite number of nests comprising a colony
was not determined in the experiment.
Cesium-134 was recovered only from the source nest of £. flavo-
pecta. All samples of ants taken from nest exits other than the source
nest registered normal background radioactivity. Thus, £. flavopecta was
monodomous as there was no evidence of exchange of individuals or food
between the individual nests.
For the purposes of this research, a terrestrial ant "nest" means a
series of closely interconnected tunnels and chambers which have either
a single opening to the surface or several closely interconnected open
ings; the entire structure obviously spatially separated from other
similar nests. In the genus Conomyrma, each "nest" usually has a single
opening plus a vertical or slanted main tunnel with branching tunnels and
with lateral chambers located at various distances from the surface.
A total of 27 clustered groups of C. insana nests has been excavated
to date. Of these, dealate queens were found in 5 of the clustered
2
groups. On 25 May 1973, 18 nests were excavated in a 20 m area. Sixteen
dealate queens were located in 1 nest at depths ranging from 4-13 cm. On
2
15 June 1973, 32 nests were excavated in a 21 m area. One nest con
tained 13 dealate queens located within 5-15 cm of the soil surface.
This nest had large masses of new eggs in addition to larvae and pupae in
all stages of development. All other nest sites excavated in this series
contained larvae and pupae, but no eggs. On 5 April 1973, a series of 18
2
nests located in an area of 25 m was excavated. No queens were recov
ered from 17 of the nests excavated in the area. In 1 nest, 9 dealate
queens were recovered from a single chamber 20-22 cm below the surface.
There was no surface indication that this nest differed from others

40
excavated. This was the only nest of those excavated in which eggs were
found in addition to brood of all stages. On 18 July 1974, 26 nests were
2
excavated in an area of 29 m Fourteen deal ate queens were located in 1
nest at depths ranging from 2-17 cm. No queens were found in the other
excavated nests. On 11 February 1975, 13 nests were excavated in an area
2
covering 9m. A single C^. insana queen was found in a chamber 2 cm
below the soil surface in 1 nest. Four queens were located in chambers
5-10 cm below the surface in a second nest 1.3 m from the nest in which
the single queen was located. No additional queens were located in any
of the remaining nests excavated in this location.
All C^. insana queens recovered during the excavation of the clus
tered mounds were dealate and apparently reproducing, based on the en
larged appearance of the gaster. Eggs were not found in any of the
excavated nests unless a queen was present. If a queen was not present,
only larvae and pupae were noted. Prior to excavation, each crater was
examined for visual differences, such as a larger crater compared to
others, indicating the possible presence of queens. No differences were
detected.
Three clustered groups of £. flavopecta nests were excavated in
O
1973. On 26 May, 11 nests located in an area 37 m were excavated. In
each of 5 of the 11 nests, a single dealate queen was found. On 27
p
September, 9 nests were excavated in a 36 m area. A single queen was
found in each of 2 nests. On 9 October, 14 nests were excavated in a 43
2
m area. A single queen was found in each of 2 nests. On 9 October, 14
2
nests were excavated in a 43 m area. A single queen was located in each
of 3 nests.

41
Conclusions
Two basic assumptions were made in this research in relation to
insana: (1) the center of the colony could be selected, and (2) there
would be equal distribution of the isotope throughout the colony. The
first assumption could be fallacious as no criteria existed upon which
to base a decision; therefore, the decision was purely subjective.
Possibly, the isotope may have been placed at the edge of the colony.
The second assumption also had some invalidity because of possible
"storage" individuals in the colony which could accumulate excessive
amounts of isotope. In addition, when samples were taken and assayed,
the individuals removed were not replaced in the colony, resulting in a
net loss of radioactivity, not only by biological half-life (Chiu et al.
1974), but also by removal of radioactive individuals. Nevertheless,
the data appear to validate the conclusion that, despite the relatively
large numbers of "exit" holes, only 1 colony of C^. insana was present in
each of the tested populations. Conomyrma insana can thus be assumed to
have polydomous nesting habits and more than 1 queen per colony.
Conomyrma flavopecta, in contrast, appears to have monodomous
nesting habits with only 1 queen per colony. Although queens were not
found in 24 out of a total of 34 nests excavated, in those 10 nests
where a queen was found, there was only a single queen per nest. These
data, plus the lack of distribution of ^Cs among the nests, support
the hypothesis that there is only 1 queen per nest, and each nest is a
distinct colony. An alternate hypothesis that single females could be
serving as the single colony queen for a number of nests is not sup
ported by the ^^Cs data. This failure to find queens in 24 of the 34
nests can more reasonable be attributed to.escape responses of the

42
queens in the large quantity of soil which must be excavated and searched,
or failure to find the true bottom of the nest.
The radioisotope 134Cs appears to be an excellent tagging material
for this group of ants. The isotope was recovered in sufficient quanti
ties for assay, and the evidence indicated that it was rapidly and
evenly distributed throughout the £. insana colony.
Summary
The organization of the colony and number of queens per colony of
Conomyrma insana (Buckley) and C. flavopecta (M. R. Smith) were investi
gated with the aid of the radioisotope cesium-134 and by excavation of
nests. Conomyrma insana was determined to be polydomous (more than 1
nest), had a single colony occupying several square meters, and had
multiple queens located in a central nest. Conomyrma flavopecta was
monodomous (1 nest per colony) and had 1 colony queen.
/

CHAPTER III
MIXED NESTS OF Conomyrma insana and C. flavopecta--
EVIDENCE OF TEMPORARY SOCIAL PARASITISM
Introduction
Two species of the genus Conomyrma are common in Florida. For the
present, these are identified as Conomyrma insana (Buckley) and C^.
flavopecta (M. R. Smith) by the identification criteria given by
Creighton (1940) and the revision by Snelling (1973). The 2 species
differ in color, morphology, and biology. Conomyrma insana is nearly
uniformly dark, greyish-black, has a broad, straight or slightly concave
occipital border, and has a distinct declivity on the mesonotum when
seen in profile; while £. flavopecta is light yellowish-brown, with the
head and apex of gaster weakly to moderately infuscated, the occipital
border rounded or straight only medially, and the mesonotum in profile
evely convex without trace of a declivity. Colony structuring differs
markedly in the 2 species. Conomyrma insana exists in enclaves of
multiple nests with much traffic between the nests, transfer of brood,
food exchange, and many queens localized together in only 1 of many
nests; while in C. flavopecta, there is no worker traffic or food ex
change between workers from separate nests and 1 queen in each nest.
Reactions between the 2 species in the field are entirely inimical.
When v/orkers of the 2 species meet, the C. insana worker usually holds
its ground and may open its mandibles, while the C_. flavopecta worker
immediately breaks contact and runs rapidly and erratically away. This
43

44
reaction is termed "panic retreat" paralleling the terminology of Wilson
and Regnier (1971) who distinguished between "panic alarm" and "aggres-
- sive alarm" reactions in ants. In contrast, accidental or other contacts
between insana workers never lead to antagonistic or panic retreat
reactions. Chance encounters between 2 foraging C^. flavopecta workers
can either be non-antagonistic or can initiate panic retreat (both
workers sometime display this response) according to whether the meet
ing is between workers of the same colony or from 2 different colonies.
In all respects, the differences between the 2 taxa are striking,
and the evidence that 2 distinct species are involved is unequivocal.
Therefore, the discovery of mixed nests of the 2 species was totally
unanticipated. The specific records in Florida for these mixed nests are
as follows: Highlands Co., Archbold Biological Station, Lake Placid, 18
April 1973; Polk Co., Winter Haven, Camp's citrus grove, 19 April 1973;
Polk Co., Winter Haven, Camp's citrus grove, 13 April 1974; Alachua Co.,
Gainesville, Depot Road, 30 April 1974; Alachua Co., Gainesville, 15 May
1974 (Wm. F. Burn); Alachua Co., Gainesville, 10 April 1975 (J. F.
Carroll); Alachua Co., Gainesville, 23 May 1975 (VJm F. Burn); and
Wakulla Co., Medart, 20 June 1975 (D. P. Wojcik). In all mixed nests,
workers of both species were observed participating in normal activities
as nest-mates (foraging, casting out soil particles, and moving together
in the nest burrows) without antagonism. The mixed C^. insana-flavopecta
nests were never found within an established enclave of £. insana nests;
rather, they were found only in areas where C_. flavopecta occurred. On
one occasion, a mixed nest was found ca. 10 m from a insana enclave.
These observations and records indicate that temporary social
parasitism might be occurring in these ants. The Conomyrma, in 1 area
I

45
where a mixed nest was found, were investigated in order to determine if
this explanation for the observations could be fact. Temporary social
parasitism is well known in ants, especially in the subfamilies Formi-
cinae and Myrrnicinae (Brian 1965, Sudd 1967, and Wilson 1971). In the
subfamily Dolichoderinae, species of the genus Bothriomyrmex are temp
orary social parasites on species of Tapi noma (Santschi 1906, 1920).
Several species of the genus Azteca are also suspected of being tempor
ary parasites because they have microgynous queens (Wheeler 1910) and,
therefore, cannot be expected to employ the usual claustral method of
colony establishment.
Methods
The mixed nest of Conomyrma insana-flavopecta, found on 23 May 1975
on the University of Florida Experimental Farm, was excavated on 27 May.
The queen, workers, and brood present in this colony were collected and
counted. This queen, 25 C_. insana workers, 25 C^. flavopecta workers,
and the brood were maintained in the laboratory.
A 45 x 20 m area was marked off into 1 m squares. A diagrammatic
map shewing major vegetation and the relative location of all Conomyrma
nest craters was constructed on 1-10 June and 26 June. The species of
Conomyrma constructing each nest was recorded.
Results
Three hundred twenty-six Conomyrma flavopecta workers and 726 £.
insana workers were collected from the mixed nests excavated on 27 May.
A single deal ate queen of C^. insana was captured in this nest. All
pupae (ca. 75) mature enough to identify (by head shape) were insana

46
rather than C. flavopecta and all callow workers (n=36) were also C_.
insana. Although the ratio of C_. Insana to flavopecta workers in the
original mixed nests was about 7:3, the ratios of £. insana to £. flavo
pecta in other mixed nests observed at the time of mapping (1-10 June)
was about 20:1, judging from the workers appearing at the nest entrance.
Distributional patterns of the craters of the Conomyrma sp. and the
mixed nests are shown in Fig. 3.1 and 3.2. The C. insana craters are
clustered into 3 localized enclaves, whereas the C. flavopecta craters
have a more scattered, but probably not random, spatial arrangement. No
Conomyrma nests were found in the area of high, dense grass. While the
location and number of craters constructed by C^. flavopecta appeared
stable, the number of C_. insana nests increased from the time the area
was first mapped. A new C_. insana enclave also developed in the area
where mixed nests had previously been located. Only 1 mixed nest could
be found on 26 June. This nest had an estimated ratio of 100:1 £.
insana to C^. flavopecta workers.
The colony taken from the mixed nest was maintained in the labora
tory for 1 year. The last individual of the original 25 C. flavopecta
workers died 42 days after capture. During this time, the overall
mortality rate of the group of C^. flavopecta workers appeared due to
normal age-related factors. No antagonism between the C_. flavopecta
and the C^. insana workers was detected. Normal colony duties, such as
brood tending and food gathering, were equally performed by both. In
addition, trophyllactic exchange of food between the 2 species was
repeatedly observed. At 6 months, the colony reached a peak population

47
Fig. 3.1. Condition of study area 1 June to 10 June 1975, showing (1)
location of original excavated Conomyrma insana-flavopecta
nest, (2) additional mixed C_. insana-flavopecta nest found
several days after excavation, (3) small, localized enclaves
of £. insana nests, and (4) scattered occurrence of
flavopecta nests.
fe
e
e
O
o
rC
O
or

o
o
"O
o
Lu
o
o
>>
"O
c
03
oo
o
1
oo
o & o
O no
O O ^
I
CP O oo oQ Q
'
o o
flavopecta
Ij*
* %
flavopecta
^ Peach Tree
iV-V -V JV-\ w:
5 meters
Fig. 3.2. Condition of study area on.26 June 1975 showing growth of
the C^. insana enclaves and appearance of a new enclave of C^.
insana in the same area as the previously present mixed nests,
while the unmixed C^. flavopecta nests appear to have remained
unchanged.

48
of an estimated 800 workers. Altogether, based on counts of dead
workers, ca. 4,300 insana workers were reared in the colony during
the year.
Discussion
The mixed nests of £. insana-flavopecta found Florida and the
localized enclaves of C^. insana give presumptive evidence that temporary
social parasitism followed by hesmosis (budding to form new nests) was
occurring in these ants. In addition, the data indicate that £. insana
was parasitic on C_. flavopecta. The modus operandi for the adoption of
the parasitic queen, elimination of the host queen, and eventual matura
tion to a £. insana nest should be similar, in a general way, to numerous
other temporary parasitic ant species.
Further evidence of temporary social parasitism may be given by the
morphology of the queens. The head of C^. insana is large, wider than
the thorax, while the gaster is small and slender in alate queens. The
head of £. flavopecta is of moderate size, about equal in width to the
thorax, and the gaster is large and voluminous in both alate and deal ate
queens. The morphological form of the C^. insana queens is similar to
the modifications already known for many temporary parasitic ants.
According to Wilson (1971), queens of temporary parasitic species are
all atypical to a degree. In many species, the head is enlarged and the
gaster is reduced in relation to the thorax. In addition, the mandibles
may be strongly modified to form falcate, piercing organs; or the queen
may be small (microgynous) in relation to the worker and/or may have
various other characterisitics such as modified pilosity or integument.
In contrast, females of free living species, which practice the normal

49
claustral method of new colony formation, have voluminous gasters and
relatively small heads with mandibles, pilosity, and integument not
modified from the norm for the genus.
The existence of incipient nests is additional evidence that a
species has the normal claustral type of new colony establishment as
temporary parasitic queens have lost the ability to begin a new colony
on their own (Emery 1909). Incipient nests of £. flavopecta have been
found repeatedly in Florida; whereas, no incipient nests of £. insana
have been found.
Apparently, the £. flavopecta queen had already been eliminated
when the mixed nest was excavated on 27 May as only a C_. insana queen
was found, and only brood and callows of C. insana were present. In many
temporary parasitic ants, adoption of the parasitic queen by a host
colony is difficult and not often successful (Sudd 1967). Mixed colo
nies of these species and their hosts are rare. Nevertheless, the
parasitic species may become locally abundant by the process of colony
fission (hesmosis) to form an enclave of closely spaced, interconnected
nests (Emery 1909). Further adoption of newly mated parasitic queens,
additional to the original parasitic queen, is a necessary step in the
growth of such an enclave. Colony "towns" of such ants often have been
noted by observers. In the case of the allegheny mound building ant,
Formica exsectoides Forel, enclaves containing up to 1,700 individual
nests have been recorded (Wheeler 1901). Similarly, enclaves of C^.
insana that exceed 2,000 nests have been located in Leon Co., Fla.
The clustering of the mixed nests in the study area indicates that
the process of hesmosis begins to occur even while the nests are still

50
mixed. Obviously, 1 or more additional £. insana queens were present in
these mixed nests, as the new £. insana enclave continued to develop
even after the £. insana queen found in the original mixed nest was
removed.
This is the first report of temporary social parasitism in the
genus Conomyrma. It may be noted that mixed nests of Conomyrma have
been recorded (as Dorymyrmex) previously by Wheeler (1901) who gave the
remarks of Forel upon finding mixed nests of ants identified as Dory
myrmex pyramicus var. niger Pergande and £. pyramicus var. flavus
McCook near Faison, North Carolina. Forel attributed these mixed nests
to a queen of each taxon fortuitously associating together when first
founding a new colony. Such a chance association of 2 species has never
been confirmed for any ants, although reported in older literature
(Forel 1874, Wheeler 1901). In the Forel colonies, all alates found
were identified as niger except for 1 male which Forel identified as
flavus on the basis of color. In the absence of any positive statement,
I assume no deal ate queens were found to confirm Forel's hypothesis.
Summary
Conomyrma insana and C^ flavopecta were found in the same nest in
Florida. Conomyrma insana queen, brood, and workers were in one nest
with C. flavopecta workers. Later, clusters of £. insana nests were
formed at locations where the mixed nests were found. Conomyrma
insana is a temporary social parasite on £. flavopecta.

CHAPTER IV
PREDATION ON FOUNDING QUEENS OF Solenopsis invicta
BY WORKERS OF Conomyrma insana
Introduction
The founding queens of the red imported fire ant, Solenopsis in
victa Burn, are especially exposed to predation during the period
following the nuptial flight and before entering the soil (Whitcomb et
al. 1973). The major predators at this point appear to be its own
workers and those of other ant species. Hung (1974), Markin et al.
(1971), Whitcomb et al. (1973) have reported the dolichoderine ant
Conomyrma insana (Buckley) as one of the more effective ant predators of
the Su invicta founding queens. Initial observations indicate that
these queens have a great deal of difficulty establishing where C_.
insana occur in north Florida. Although £. insana is sporadic in dis
tribution in this region, it may be an important factor in limiting
local populations of the red imported fire ant.
Competition for preferred nesting sites and foraging territories
appears to be the major factor determining the distribution of many for-
micine species (Bhatkar et al. 1972, Leston 1973, Van Pelt 1966). No
where is this more true than in the aggression of the formicine species
toward founding queens. Predation of founding queens alighting in a
territory dominated by a given ant species would reduce or eliminate
future competition for available resources in favor of the dominant
species. Carroll and Janzen (1973) stated that founding queens were
51

52
usually killed by worker ants of the same or other species. This strong
statement is supported by research of several workers (Janzen 1967,
Levieux 1971, Marikovsky 1961, Pontin 1960). Janzen (1973) presented
research on Pseudomyrmex ferruginia F. Smith and described in detail how
workers destroyed new queens of their own species. Hung (1974) analyzed
the refuse pile of a Conomyrma insana colony and reported the recovery
of discarded heads of the queens of 5 species of myrmecines, including
those of Solenopsis invicta queens.
The predator-prey relationship of Conomyrma insana workers and the
founding queens of Solenopsis invicta is of special interest because of
the apparent complexity of the encounters leading to the death of the
founding queens or the rare successful escape from predation by the
latter. The first objective of the present research was to gain infor
mation of the general behavior pattern of a do!ichoderine ant worker
attacking a myrmecine founding queen. Furthermore, such data are needed
on the predator-prey relationship of these 2 species, in particular, to
evaluate the effectiveness of C^. insana as a predator and its potential
in limiting the local increase of SL invicta. Conversely, this informa
tion was collected to determine the ability of S_. invicta to survive
within an area under these high stress conditions.
Methods
This research consisted of field observations of the behavior(s)
of the predator (£. insana workers) and prey (S_. invicta founding queens)
under natural conditions. These events were observed in detail in Leon
county in northern Florida in areas where Conomyrma insana nests were
present. Of approximately 2,000 Solenopsis invicta founding queens
(

53
observed while under attack, 240 were watched from the time they alighted
until they were either destroyed by predators or escaped. When possible,
the action was verbally described and recorded on tape as it occurred.
Most observations were made on bare ground, either on dirt roadways or
in fallow fields. Additional observations were made in mowed fields with
sparse vegetation cover. This report is based on observations of queens
which alighted naturally; information on queens which had been captured
and handled was discarded.
In 1972, the effectiveness of predation by C^. insana was deter
mined. Nests of £. insana were examined and excavated to determine the
numbers of S_. invicta queens taken as prey. In addition, five 3 x 8 m
plots were randomly located in an open field. These plots were examined
at 1/2 hr intervals from 1400 to 1830 on the dates of heavy SL invicta
nuptial flights. Predation of S. invicta queens within these plots was
noted and recorded. For 3 consecutive days following nuptial flight,
this field was searched for additional :S. invicta females on the surface
and for signs of brood chamber construction.
Results and Discussion
A summary of the responses of Su invicta queens to attack(s) by £.
insana workers is shown in the flow diagram in Fig. 4.1. While the re
sponses of the queen to attack are varied, they can be separated into
artificial categories which recur frequently.
An abridged summary of a single attack sequence by <3. insana work
ers on 1 S. invicta queen is presented in Table 4.1. This was taken
from observations made 21 June 1972 of a queen that had alighted at
1626. The nuptial flight began after 2 days of intermittent storms

Fig. 4.1. Flow diagram of responses of 240 Solenopsis invicta founding
queens to initial contact, pursuit, and attack by Conomyrma
insana workers. The contrasting behavior leading to nest
founding is included.

55
TABLE 4.1. Abridged sunmary of observation of attack by Conomyrma insana
(Ci) on a founding queen of Solenopsis invicta (Si) beginning
at 1625 on 21 June 1972. Leon Co., Fla.
Elapsed
time (min)
Observations
0 Si_ alights, immediately moves under debris and into
soil crevices.
4 Met by 1st £i_; duration of contact momentary. Si
movement rapid and erratic.
9 Si reverted to normal rate of movement.
13 £i attacked (n=2) and pursued (n=l). Si hid under
leaf.
20 Si left leaf. attacked Si at base of wing (n=l),
Head (n=l), and thorax (n=lTT Si fled.
21 Si_ has traveled 8.5 m from alighting point, wings
intact.
30 Ci_ attacked (n=l) Si_ at head. Si_ fled.
33 Ci attacked (n=9), Si^ shook Ci loose by rapidly
flexing body.
34 Ci attacked (n=4), Ci clinging to wing, Si_ shook Ci
loose. Si climbed blade of grass, attempted to fly,
failed. *
35 pursuing (n=4), Ci_ attached (n=3) to Si_. All Ci_
dislodged from Si_.
36 Ci clinging to thorax and wing (n=2). Si_ dislodged
38 Ci attacked (n=4), Si motionless (6 sec), fled. Ci_
pursued and attacked (n=2).
39 Ci pursuing (n=7) and clinging to Si_ head (n=l).
40 Si moving rapidly, attempting to break off wings. Ci
Tn=3) pursuing and biting at legs and wings. Si
flexed abdomen rapidly, curled into "C" position and
bit at Ci. Si had traveled 16.8 m since alighting.

56
TABLE 4.1 Continued.
Elapsed
time (min) Observations
41 Si knocked left forewing off. Ci_ cling to Si_ legs.
Sibit and killed Ci_ (n=l). Si_ now moving rapidly,
Ci_ pursuing (n=9). Ci_ clinging (n=3), Si shook all
off, still moving rapidly. Ci clinging~[h=3) and
pursuing (n=ll). Sj_ curled into "C" position and bit
at Ci_, then moved away with Ci_ in pursuit (n=14).
48 §2 still moving with no pursuit.
50 Ci attack (n=2) Si. Si_ shook off attack.
51 Si_ under leaf momentarily, then emerged.
52 Si attacked by Ci (n=l). Si curled into "C" position
and bit CL Ci (n=17) pursuing SL Si shook off
pursuit and attack, moved under clod of soil.
53 Si emerged, immediately attacked by Ci (n=l). Ci at
tacked (n=3). Si dislodged attackers. Ci clinging
to wing. Si climbed blade of grass, immediately
descended with Ci_ (n=l) still clinging to wing.
54 (n=l) attacked Si at right metathoracic leg. Si
curled into "C" position and bit Ci_, forcing it to
release. Ci (n=2) attacked, forced away by Si.
55 Ci attacked (n=2), Si curled into "C" position and
bit at Ci. Si moved with Ci still clinging to leg.
Ci (n=7T~biting and pursuing, Si flexing abdomen and
thorax, intermixed with assuming "C" position and
biting.
58 Ci (n=3) attacking Si. Si shook loose CL Additional
Ci (n=8) attacked, dislodged all £i by shaking
and biting.
6C Si moved forward and went under a leaf.
61 Si emerged from under the leaf and was immediately
attacked by (n=l). Si curled into "C" position
and bit Ci until it left. Ci (n=3) in pursuit of
Si. Ci TFT=3) caught Si. Si bit at Ci on leg. Ad
ditional attacked Xn=9). Si assumed "C" position
and bit clinging to leg but did not dislodge.

57
TABLE 4.1 Continued.
Elapsed
Time (min) Observations
63 Si moving with Ci_ clinging. Additional Ci_ in pursuit
I=7) and attacking (n=15+) Si^ moving rapidly while
shaking and flexing body. Stopped, assumed "C" posi
tion bit and killed £i_ (n=l). Ci_ attacking (n=4).
66 Ci^ (n=3) clinging to Si. Si_ dislodged Ci amd moved
forward rapidly. Ci_ T=4) pursuing, Ci~Tn=l) cling
ing to leg. Si_ curled into "C" position and bit Ci_.
68 Ci^ (n=12) attacking. Sj^ halted, assumed "C" position,
bit Ci_ (n=l), now moving with Ci_ continuing attack.
Si biti ng at Ci_ ki 11 ed 1.
72 Si_ nearly subdued (= taken as prey by Ci_). Little
response by Ci_ and only feeble attempts to escape.
75 Ci_ have severed left prothoracic and both metathoracic
legs.
76 Ci_ have severed left mesothoracic leg.
77 All legs severed.
80 Ci_ moving appendageless Sj_ to nest. Si_ alive and trying
to bite Ci_. Distance from alighting point = 53 m.

58
resulting in 18 cm rainfall. The temperature at 1414 was 35 C and the
v/ind velocity varied from 8 to 24 kmh. The alates of SL invicta emerged
from the mounds at 1331. The first queen observed returning from the
nuptial flight alighted at 1450.
Consistencies and Variations in Behavior
Detection. When a S. invicta queen alighted in an area heavily
populated by £. insana, a period, varying from a few seconds to several
minutes, elapsed prior to initial contact by (1. insana. During this
period, if sufficient, the queen shed her wings and commenced searching
for a suitable site to construct the brood chamber. Detection of the S^.
invicta queen by C_. insana workers occurred any time after the queen
alighted, depending upon the proximity of the workers. The first con
tact with the queen was made by a single worker approaching from the
front or side with mandibles spread and maintaining only momentary con
tact with the queen. It is probable that the queen was marked with an
identifying compound by the £. insana worker. Following attacks were
not comparable to the initial attack in that subsequent attacks were
progressively more frequent and prolonged. In later attacks, the workers
remained attached to the queen until dislodged. In the first approach,
penetration by the mandibles was never observed and contact was made at
any portion of the body.
During a short span of time following the initial contact, the
queen was unmolested by C^. insana workers. Again, the queen may attempt
to remove the wings (if not previously shed), but the usual reaction of
the queen was to flee from the point of attack. In most observed cases,
this was followed by a halcyon period as evidenced by the less erratic

59
and hurried rate of movement. Subsequently, the queen would renew the
search for a suitable location in which to excavate the brood chamber.
Secondary contacts. The time between first contact and subsequent
attacks varied from 15 sec to 35 min depending on the local population
densities and the degree of surface activity of £. insana. These secon
dary contacts differed from the first encounter in several respects.
The £. insana attacks were progressively more frequent and prolonged;
.the workers tended to cling to the queen until dislodged. Pursuit of
the queen by workers also occurred, although individual workers did not
take part continuously. The intensity of pursuit was again dependent
upon the local density of £. insana. As the attack progressed, the
number of individuals of C^. insana workers taking part increased up to
the time the queen was taken as prey.
Queen response to attack. The responses of the Sk invicta founding
queens to the attack by the C^. insana workers can be divided into 5
basic reactions: (1) escape, (2) concealment, (3) cessation of all
movement, (4) mandibular defense, and (5) rapid flexing or shaking of
the body. All or several of these responses were utilized by the queens
in attempting to avoid or fight off the attack of C. insana worker(s).
Escape attempts from the attack of £. insana workers occurred in 2
forms: (1) flight and (2) surface movement. If a queen was attacked
prior to shedding of the wings, she could avoid predation by further
flight. This was accomplished by less than 2% of the queens under
observation. The queens, when under attack, crawled up vegetation or
any other available elevation in an attempt to gain the height necessary
to become airborne. Queens not under attack were seldom observed to

60
attempt flight after landing. The response most often elicited by the
attack of C^. insana was attempted escape on the ground surface. The
distance covered varied with the topography, presence of vegetation, or
other natural barriers. The escaping queen usually covered distances
of 20 to 80 cm per min; some covered distances of 3 m or more. This re
sponse was often temporarily successful in avoiding immediate predation
by C^. insana workers.
While under attack or when unmolested by C^. insana workers, the S^.
invicta queens were continuously moving under twigs, leaves, or into
soil cracks. This appeared to be a concealment response, even though
it may be difficult to distinguish from the normal instinct of the queen
to commence construction of the brood chamber as soon as possible after
landing. Hiding was usually effective in momentarily avoiding predation.
If the queen abandoned the protection of the leaf or twig, C^. insana
workers immediately resumed their attack.
In further attempts to avoid predation, the queen often remained
motionless briefly (6-10 sec) when attacked or investigated by a £.
insana worker; afterwards, attempted escape or defense methods were
used. In a number of observed cases, the cessation of motion was tem
porarily successful as the C_. insana worker either did not commence
further attack or broke off the attack and left the vicinity of the
female.
The queens also attempted to take a defensive stand against the
attack of the C_. insana workers by aggressively biting. This was most
often seen when the invicta queen had failed to dislodge attacking
workers and 1 or more had seized an appendage. To use the mandibles
effectively, the queen would lie on either side and assume a "C" position

61
with the tip of the abdomen curled ventrally toward the mandibles. In
this position, the SL invicta queen could use the mandibles effectively
for defense. On a number of occasions, a queen severed an abdomen from
a C_. insana worker's thorax at the pedicel or decapitated a worker
hanging onto 1 of her legs. The worker's decapitated head often re
mained attached to the seized appendage. This method of defense was
effective against an individual worker, although the queen could not
remain in the position long as she was vulnerable to further attack by
other £. insana workers.
The S. invicta queens frequently attempted to dislodge clinging £.
insana workers by rapid shaking or flexing of the abdomen. This was
used by all queens under attack by C_. insana workers. The queen was
always successful in dislodging attacking workers by this means until
she became exhausted or was overcome by the mass attack of the £.
insana.
Predation and disposal of queens. When the queens were preyed upon
successfully by £. insana workers, the appendages were removed by a
combination of tugging and biting at the joints. Appendageless queens
were dragged, while still alive, to nests by the workers.
On the day following predation, bits of discarded body parts and
heads of invicta queens were found on the periphery of the C_. insana
nests. After a major nuptial flight of 21 June 1972, a series of 24
nests were excavated on each of 2 successive days. On the 1st day, live
appendageless queens were found in all nests. The fewest found in a
single colony was 7 and the most was 18. A total of 329 S_. invicta
queens was recovered from the £. insana nests. An additional 83
t

62
queen heads were counted from the discarded debris around the colonies
that were excavated. On the 2nd day, only 1 £. insana nest had live
appendageless invicta queens (n=2).
Effectiveness of C. insana as a Predator
Nests of £. insana tend to be concentrated in localized areas. The
greater the number of C^. insana nests within the area, the more intense
the attack on fire ant queens. Mortality of S^. invicta queens did not
depend on the effectiveness of the bite or toxin of the individual work
ers, but rather on the ability of C_. insana to continue the attack until
the invicta queens were taken as prey. When the process was inter
rupted prior to removal of appendages, the queen survived and laid eggs.
Where the process was followed from beginning to end, however, only 6
queens out of 240 contacted by £. insana workers managed to escape de
struction by any means: 4 by flying away from the C_. insana area and
2 by escaping on the ground surface to outside of the foraging range
of C^. insana.
In 1972, an effort to quantify predation by £. insana was made in
five 3 x 8 m plots. On 9 May, 613 postnuptial S_. invicta queens were
counted in these plots; 60% of these were completely subdued by the C^.
insana workers, 36.4% were under continuous attack, and 3.6% were
momentarily unmolested.
A field at Tall Timbers Research Station was searched for S^.
invicta queens for 3 successive days following a nuptial flight. On 10
May, 19 S_. invicta queens were being dragged to the C_- insana colony by
a coordinated effort of workers. Only 4 S_. invicta queens were still
under attack by C^. insana. On 11 and 12 May, no live queens were found

63
even though the area was extensively searched. An identical search was
conducted following the nuptial flight on 21 June. On 22 June, 24 S^.
invicta queens were found that had been subdued by £. insana workers and
were being taken into the colony. No queens were found on 23 and 24
June within the £. insana enclave.
Summary
Solenopsis invicta Burn founding queens are especially vulnerable
to predation during and following the nuptial flight. Conomyrma insana
(Buckley) workers are effective predators of the queens in some areas of
northern Florida. The respective behaviors that occurred during the
confrontation of the 2 species were studied in the field in Leen county
of northern Florida. Detection of a founding queen alighting within a
C_. insana enclave by workers varied from a few seconds to several min
utes. The founding queen's reactions to attack by C_. insana consisted
of 5 basic responses: (1) escape, (2) concealment, (3) cessation of
movement, (4) mandibular defense, and (5) rapid flexing or shaking.
Less than 3% of the queens observed alighting within a C^. insana en
clave were successful in escaping predation. No SL invicta queens were
found that survived predation by £. insana workers long enough to con
struct a brood chamber and begin a new colony within an area occupied by
C. insana.

CHAPTER V
SOURCES OF CARBOHYDRATES UTILIZED BY Conomyrma insana
Introduction
The colonies of the ant Conomyrma insana (Buckley) consist of many
clustered nests that form enclaves. Conomyrma insana is omnivorous, how
ever, it is a voracious predator (Van Pelt 1966, Hung 1974, Markin et al.
1971, Whitcomb et al. 1972). Way (1968) suggested that, in terms of food
supply, any species of omnivorous ant, in order to maintain continuously
large colonies, had to change foraging grounds periodically as doryline
ant do, or to tend Homoptera, which provide honeydew to carry the ants
over periods when animal prey is scarce. The workers of £. insana appar
ently derive a large portion of the energy required to maintain their
colonies from honeydew excreted by homopterans and nectar from extrafloral
and floral structures of plants.
Two species of Conomyrma, insana and C_. flavopecta (M. R. Smith),
are currently recognized in the southeastern United States (Snelling 1973).
Prior to Snelling's revision, the North American species of Conomyrma
were lumped as Dorymyrmex pyramicus (Roger) or Conomyrma pyramicus. Thus,
the information contained in reports of aphid tending by £. or £. pyrami-
cus is difficult to assign correctly to the presently accepted species
of Conomyrma. Nielsson et al. (1971) reported the association of several
aphid species with (1. pyramicus, but did not attempt to establish if a
mutualistic relationship occurred with the species of aphids listed. Van
64

65
Pelt (1966) noted that Dorymyrmex pyramicus (= Conomyrma insana based
on Van Pelt's description of the nesting habits) tended 2 aphid species.
Research by Odum and Kuenzler (1963), Wiegert et al. (1967), and de la
Cruz and Wiegert (1967), using ^2P to trace the plant-aphid-ant food
chain, demonstrated that £. pyramicus obtained honeydew from aphids of
the genera Aphis and Macrosiphum. Jones (1929) listed 3 aphids (Aphis
sp., Chaitophorous negundinus Thos, and £. populella Gill. and Pal.)
tended by [). pyramicus.
The objective of this research was to determine the sources of
carbohydrates available to £. insana within their foraging area.
Methods
Investigations were conducted at Tall Timbers Research Station,
Leon Co., Florida from July 1972 through August 1975. Eight locations
with high populations of Conomyrma insana were selected. The flora
within and adjacent to the C^. insana enclaves were examined for Homop-
tera tended by ants and ants feeding from floral and extrafloral nectar
ies. With Homoptera, determination of tending was based on observing
the workers antennuate an individual and subsequently receive a droplet
of exudate. The presence of the ant on the same plant as the Homoptera
was not accepted as prima facie evidence of ant-aphid association. All
new Homoptera and their host plants, and plants with nectaries were
collected and identified.
Beginning in 1973, a distribution map was constructed annually of
the Z. insana nests around a solitary short leaf pine located in an open
field. To construct the map, the distance of each nest from the base
of the pine was measured. Using the pine tree as a base point, the
compass angle of each nest was also determined.

66
Results and Discussion
Homoptera Tended by Conomyrma insana
Sixteen species of aphis (Table 5.1) and 10 other species of Homop
tera (Table 5.2) viere tended by Conomyrma insana. These homopterans
were colonizing plants within the foraging area of the ant. The Homop
tera were collected from 13 plants families; these were Pinaceae, Gra-
mineae, Fagaceae, Chenopodiaceae, Amaranthaceae, Rosaceae, Caesalpinia-
ceae, Fabaceae, Onagraceae, Labiatae, Scrophulariaceae, Rubiaceae, and
Compositae.
The aphids Cinara watsoni Tissot, Sipha flava (Forbes), Aphis gossy-
pii Glover, A. craccivora Koch, A. spriaecola Patch, Cerosipha rubifoli
Thomas, and Hysteroneura setariae (Thomas) were common in all areas where
£. insana occurred. Rhopalosiphum maidis (Fitch) were numerous on Paspa-
lum boscianum Flugge and corn, Zea mays L., in only one area. Another
aphid, Myzocallis bella (Walsh), was found only on running oak, Quercus
pumila Walter, in one field. The remaining aphids presented in Table
5.1 were rarely collected.
The aphids Dactynotus tissoti (Boudreaux), collected on Solidaqo sp.
(Goldenrod); ]3. ambrosine (Thomas), collected on Solidago sp. and Pyrrho-
pappus carolinianus (Walt.) DC.; ID- helianthicola Olive, collected on He-
lianthus augustifolius L.; and CL tutaiae Olive, collected on Ambrosia
artemissiifolia L. (Ragweed) were within the foraging areas of £. insana.
The ant was often found on the same plant as the aphids, but there v/as
no evidence that these aphid species were tended by £. insana.
Adelges sp. and Toumeyella parvicornis (Ckll.) were both tended by
£. insana on short leaf pine. Both were attractive to the ant, and large
numbers of foraging workers were constantly present on the pine.

Table 5.1. Aphids tended by Conomyrma insana and the plant host(s) that the aphids were found on. Tall
Timbers Research Station, Leon Co., Fla. 1972-75.
Plant Host
Aphid Species
Family
Species (Common Name)
Cinara watsoni Tissot
Pinaceae
Pinus echinata Miller (Short leaf pine)
Cinara melaina Boudreaux
Pinaceae
Pinus echinata Miller (Short leaf pine)
Lachnochaitophorus obscurus (Tissot)
Fagaceae
Quercus nigra L. (Water oak)
Fagaceae
Quercus pumila Walter (Running oak)
Sipha flava (Forbes)
Gramineae
Eragrostis spectabilis (Pursh.) Steud (Love grass)
Gramineae
Digitaria villosa (Walt.) Pers. (Crabgrass)
Gramineae
Paspa!urn setaceum Michx.
Aphis coreopsidis (Thomas)
Compositae
Bidens bipinnata L. (Beggar ticks)
Aphis goss.ypii Glover
Caesalpiniaceae
Cassia obtusifolia L. (Sickle-pod)
Rubiaceae
Richardia scabra L. (Mexican clover)

Compositae
Gnaphalium obtusifolium L. (Everlasting)
Aphis oestlundi Gill
Onagraceae
Oenothera sp. (Evening primrose)
Aphis spiraecola Patch
Chenopodiaceae
Chenopodium ambrosioides L. (Mexican tea)
Compositae
Eupatorium capillifolium (Lam.) Small (Dog-fennel)

Table 5.1 .Continued.
Plant Host
Aphid Species
Family
Species (Common Name)
Aphis spiraecola Patch Compositae
Aphis fabae Scop.
Compositae
Compositae
Aphis craccivora Koch
Caesalpiniaceae
Caesalpiniaceae
Aphis sp.
Fagaceae
Amaranthaceae
Rubiaceae
Scrophulariaceae
Compositae
Compositae
Schizaphis sp.
Labiatae
Cerosipha rubifoli Thomas
Rosaceae
Pyrrhopappus carol ini anus (Walt.) DC
Conyza canadensis (L.) Crong. (Horseweed)
Ambrosia artemisiifolia L. (Ragweed)
Cassia fasciculata Michx. (Partridge pea)
Cassia nictitans L. (Wild sensitive plant)
Quercus nigra L. (Water oak)
Amaranthus hybridus L.
Richardia scabra L. (Mexican clover)
Agalinis fasiculata (Ell.) Rat.
Ambrosia artemisiifolia L. (Ragweed)
Eupatorium capillifolium (Lam.) Small (Dog-fennel)
Trichostema dichotomum L. (Blue curls)
Rubus cunneifolius Pursh. (Sand blackberry)

Table 5.1 .Continued.
Aphid Species
Plant Host
Family
Species (Common Name)
Hysteroneura setariae (Thomas)
Gramineae
Eragrostis spectabilis (Pursh.) Steud. (Love grass)
Gramineae
Digitaria violascens Link (Crabgrass)
Gramineae
Paspalum boscianum Flugge
Gramineae
Paspalum plicatulum Michx.
Gramineae
Paspalum setaceum Michx.
Gramineae
Cenchrus echinatus L. (Sandspurs)
Gramineae
Andropogon virginicus L. (Broom sedge)
Rhopalosiphum maidis (Fitch)
Gramineae
Paspalum boscianum Flugge
Gramineae
Zea mays L. (Corn)
Myzocallis bella (Walsh)
Fagaceae
Quercus pumila Walter (Running oak)
Neosymdobius lutens Tissot
Fagaceae
Quercus nigra L. (Water oak)

Table 5.2. Homoptera, other than aphids, tended by Conomyrma insana and the host plant(s) they were found
on. Tall Timbers Research Station, Leon Co., Fla. 1972-75.
Plant Host
Species
Family
Species (Common Name)
Phylloxieridae
Adelqes sp.
Pinaceae
Coccidae
Saisettia sp.
Compositae
Pulviaria urbicola Ckll.
Compositae
Chenopodiaceae
Toumeyella parvicornia (Ckll.)
Pinaceae
Coccus hesperidum Linn.
Fabaceae
Fabaceae
Coccus sp.
Compositae
Pseudoccidae
Oracella sp.
Fagaceae
Pinus echinata Miller (Short leaf pine)
Eupatorium capillifolium (Lam.) (Don-Fennel) ^
Pyrropappus carolineaus (Walt.)
Chenopodium ambrosiodes L.
Pinus echinata Miller (Short leaf pine)
; Glycine max Merr. (Soybean)
Lespedeza angust'folia (Pursh.) Ell.
Ambrosia artemisiifolia L. (Ragweed)
Quercus virginiana Miller (Live oak)

Table 5.2.Continued.
Plant Host
Species
Family
Species (Common Name)
Eriococcidae
Eriococcus quercus
Eriococcus smithii
Kermesidae
(Comst.)
Lobdell
Fagaceae
Gramineae
Quercus virginiana Miller (Live oak)
Andropogon virginicus L. (Broom sedge)
Kermes sp.
Membracidae
Fagaceae
Quercus nigra L. (Water oak)
Unidentified sp.
Compositae
Ambrosia artimisiifolia L. (Ragweed)

72
Coccus hesperidium linn annually built up heavy infestations in a
soybean field. The scale was also tended by Solenopsis invicta Burn on
the interior of the field.
The mealybug Oracella sp. was found only on a single live oak tree,
Quercus virginiana Miller. This population persisted in large numbers
on the oak.
The first collected specimens of Eriococcus smithii Lobdell were
taken from 7 clumps of broom sedge, Andropogon virginicus L. in a shal
low ditch between a forest road and a soybean field. By September 1974,
colonized broom sedge clumps had increased to 43, spaced out over
approximately 30 m within the shallow ditch.
Kermes sp. were common on water oak, Quercus nigra !_., and were
heavily tended by £. insana. A species determination was impossible as
they were so heavily parasitized that no mature specimens were found.
Nectaries of Plants
Conomyrma insana constantly fed from the extrafloral nectaries of
Cassia nictitans L. and £. fasciculata Michx. and to a lesser degree on
C_. obtusifolia L. Cassia nictitans was the most prevalent plant at most
C^. insana locales. In cultivated fields, mexican clover, Richardia sca-
bra L., was a common weed. The floral nectaries of this plant were
frequently visited by £. insana.
Influence of Carbohydrate Food Source
The influence of a carbohydrate food source on selection of nesting
sites by C^. insana is illustrated in Figs. 5.1-5.4. In this field, vege
tation was composed of scattered clumps of broom sedge, Andropogon vir-
o
ginictis L. (ca. 2.1 clumps per nr), with sparse occurrence of Cassia

73
o
c

e
o
o
Fig. 5.1. Distribution of Conomyrma insana nests in 1273 around a
short leaf pine, Pinus echinata, located in an open field
at Tall Timbers Research Station. The short leaf pine was
heavily infested with Cinara watsoni.

74

75

76
Fig. 5.4. Distribution of Conomyrma insana nests in 1976 around a
short leaf pine, Pinus echinata, located in an open field
at Tall Timbers Research Station. The short leaf pine was
heavily infested with Cinara watsoni.

77
nictitans L. and some composites. The distribution of the plants were
uniform with most of the soil surface exposed to sunlight. An occasional
short leaf pine, Pinus echinata Miller, was present in the field; an en
clave of C.. insana surrounded each of these trees. The only attractant
in each of the pine trees was the pine aphid, Cinara watsoni. The work
ers of C^. insana, from the nests surrounding the trees, continuously
tended these aphids.
Throughout the 4 years of monitoring this population of ants, nests
were maintained around the trees as diagrammed in Figs. 5.1-5.4. No
other major carbohydrate food sources were near this location, although
an occasional group of the aphid Hysterouneura setariae was on broom
sedge. Apparently, while the presence of the aphid on the pine tree was
instrumental in determining the occurrence of £. insana, the quantity of
carbohydrates available from this and other sources was insufficient to
maintain a larger colony. This hypothesis is supported, in part, by the
recorded increase of the number of nests from 31 in 1973 to 50 in 1976,
while there was corresponding increase in the pine from ca. 2 m to 4.5
m in height at the same time. Theoretically, the pine would be capable
of supporting a larger population of aphids which, in turn, would
support a small increase in the ant colony.
Summary
Twenty-six species of Homoptera were tended by Conomyrma insana
(Buckley). The Homoptera were in the families Aphididae (10), Phyllox-
ieridae (1), Coccidae (5), Pseudoccidae (1), Eriococcidae (3), Kermes-
idae (1), and Membracidae (1). Conomyrma insana also fed from the
nectaries of Cassia nictitans L., C. fasciculata Michx., C. obtusifolia,
and Richardia scabra L.

78
Conomyrma insana construct their nests adjacent to carbohydrate
food sources. The population size in an area may be proportional to
the quantity of carbohydrates available.

CHAPTER VI
PRESENCE OF Spissistilus festinus TENDED BY ANTS:
A FACTOR AFFECTING EGG PREDATION BY ANTS IN SOYBEANS
Introduction
Ants are important predators on arthropod pests in agroecosystems
(Cook 1904, Pierce et al. 1912, Whitcomb and Bell 1964, Whitcomb et al.
1972). Many predaceous ants supplement their diet by collecting honey-
dew from Homoptera (Carroll and Janzen 1973). This stationary and re
newable food source can be an attractant resulting in locally high ant
population densities that may be maintained throughout the year (Finne
gan 1974). The presence of Homoptera could consequently influence the
effectiveness of ants as predators.
In 1975, ants were found tending nymphs of the threecornered alfalfa
hopper, Spissistilus festinus (Say), on soybeans in northern Florida.
Since the ants were observed more often on plants with the nymphs than
on plants without the nymphs, I hypothesized that predation on soybean
pests increased in the presence on nymphs. My objective was to test
this hypothesis by examining predation on eggs of the soybean looper,
Pseudoplusia includens Walker, under field conditions.
Methods
Experiments were conducted between 15 September and 15 October,
1975 in a 15-ha field of Bragg soybeans in Alachua County, Florida.
Initially, the field was surveyed for ant species tending nymphs of
79

80
Spissistilus festinus. Two areas were then selected for experimentation,
2
an 1100-m area in the interior of the field in which the nymphs were
being tended primarily by Solenopsis geminata (F.) (designated as area
O
1) and a 45-m area at the perimeter of the field in which the nymphs
were being tended by Conomyrma insana (Buckley) (designated as area 2).
On 3 and 15 October, between 1200 and 1700 hr, nymphs were counted in
10 randomly selected 3.05-m sections of soybean row in the S.. geminata
area and three 3.05-m sections of soybean row in the C_. insana area.
On both dates, the ant species present on each plant and the number of
soybean plants in each 3.05-m section of soybean row were recorded. An
analysis of variance (Steel and Torrie 1960) was performed on the
squareroot-transformed counts of nymphs. Data were arranged in a 2 x 2
factorial design (= 2 areas of the field x 2 dates).
In each of the 2 areas of the field, 14 plants were selected, 7
with 1 or more nymphs tended by ants and 7 without nymphs. The absence
o
of nymphs was confirmed by shaking each plant over a 0.58-m white cloth.
Previous tests had established that nymphs were easily shaken from the
soybean plants. A trifoliate leaf on the upper, middle, and lower 1/3
of each plant was marked with a small spot of water-soluble paint. Eggs
of the soybean looper, Pseudoplusia includens, were obtained from a
laboratory colony maintained by N. C. Leppla, Insect Attractants, Be
havior and Basic Biology Research Laboratory, ARS, USDA, Gainesville,
FI. Three eggs were attached with casein glue to each marked leaf (1
on the upper surface of each leaflet). Therefore, 9 eggs (3 at each
height on the plants) were attached to each of the 14 plants in
an area for a total of 126 eggs/area. Twenty-one of the eggs were at
I

81
each height on plants with nymphs, and 21 eggs were at each height on
- plants without nymphs. The fraction of eggs missing out of each group
of 21 eggs was recorded after the eggs had been in the field for 24 hr.
Experiments were repeated during five 24-hr periods. An analysis of
variance was performed on the arcsine-transformed data arranged in a 2
x 2 x 3 factorial design (= nymphs present or absent x 2 areas of the
field x 3 heights on the plants) replicated on 5 dates.
On 14 October, an additional experiment was conducted in three 45-
2
m areas of the field in which the nymphs were being tended by £.
insana. Experimental procedures were the same as those used in the
earlier experiments except that 1 set of 14 plants (7 with 1 or more
nymphs tended by insana and 7 without nymphs) was selected in each of
the 3 areas of the field. In addition, the eggs were examined after
they had been in the field 4, 8, and 24 hr. An analysis of variance was
performed on the arcsine-transformed data (numbers of eggs missing/21
eggs) arranged in a 2 x 3 x 3 factorial design (= nymphs present or
absent x 3 elapsed periods of exposure x 3 heights on the plants)
replicated in 3 areas of the field.
Results and Discussion
Workers of Solenopsis geminata, Conomyrma insana, C. flavopecta (M.
R. Smith), Iridomyrmex pruinosus (Roger), and Pheidole morrisi Forel
tended threecornered alfalfa hopper nymphs in the soybean field in which
the investigations were conducted. Although tending of S.. festinus
nymphs by ants in soybeans has not previously been reported, Jordan
(1952) observed an unspecified ant species tending nymphs in alfalfa.
In the experimental field, nests of Pheidole morrisi and Iridomyrmex
f

82
pruinosus were uncommon, and workers of each species were observed
tending nymphs on only 2 occasions. Conomyrma flavopecta and Solenopsis
geminata occurred over most of the interior of the field. Workers of S_.
geminata foraged on soybean plants from underground trails radiating
from their mounds and opening at the bases of plants on which nymphs
were being tended. Three clusters of Conomyrma insana nests were pre-
2
sent along 1 edge of the field. Each enclave covered ca. 45 m and
extended ca. 3 m into the field. No other nests of £. insana were found
in the field.
In the interior of the field (area 1), workers of S_. geminata or,
occasionally, workers of flavopecta were observed tending three-
cornered alfalfa hopper nymphs. On 3 and 15 October, workers of
geminata were found on 94.6% of the plants with nymphs and on 15.0% of
the plants without nymphs. Workers of flavopecta were on the remain
ing 5.4% of the plants with nymphs and on 0.8% of the plants without
nymphs. Within a C^. insana enclave (area 2), C^. insana was the only ant
species observed tending threecornered alfalfa hopper nymphs. On 3 and
15 October, at least 1 insana worker was found on every plant with
nymphs. Conomyrma insana workers were present on 83.0% of the plants
without nymphs. Conomyrma insana workers and, to a lesser extent, S_.
geminata workers were present on plants without nymphs but at lower
frequency than on plants with nymphs.
Estimates of the densities of the threecornered alfalfa hopper
nymphs are given in Table 6.1. Although nymphs were found on 50.4% of
all plants examined on 3 and 15 October, only a single nymph was present
on 77.7% of the plants with nymph(s). Densities of nymphs were not

Table 6.1. Numbers of Splsslstilus festinus nymphs in 2 areas of a Florida soybean field: one area in
which nymphs were tended by Solenopsis geminata and another in which nymphs were tended by
Conomyrma insana.
% of plants examined with:
Ant
Species
Survey
date
0
nymphs
i
nymphs
2
nymphs
3-5
nymphs
Nymphs/plant
x + 95% Cl (a)
Nymphs/ha
x + 95% Cl la)
S. geminata
Oct. 3
36.0
49.0
11.0
4.0
0.84 + 0.22
111.3 + 32.7
(area 1)
Oct. 15
67.0
28.0
5.0
0.0
0.38 + 0.20
63.8 + 38.1
C. insana
Oct. 3
30.0
43.3
20.0
6.7
1.10 + 0.90
140.7 + 140.6
(area 2)
Oct. 15
56.7
36.7
6.7
0.0
0.50 + 0.25
63.5 + 42.0
(a)mean + 95% confidence interval.

84
significantly different between area 1 and area 2 (p>0.1), but signifi
cantly more nymphs were present in both areas on 3 October than on 15 Oc
tober (p<0.005). By 15 October the soybean plants were senescent, and
many plants were apparently unacceptable as hosts for the nymphs.
The percentage of soybean looper eggs missing after 24 hr from
plants in area 1 and 2 are presented in Fig. 6.1. A significantly
greater percentage of eggs was missing from plants in area 2 (88.6%)
than from plants in area 1 (57.0%) (p<0.005). Within area 1, a signi
ficantly greater percentage of eggs was missing from plants with nymphs
(77.1%) than from plants without nymphs (36.8%) (p<0.005). Within area
2, a greater percentage of eggs also was missing from plants with nymphs
(90.8%) than from plants without nymphs (86.3%), but the difference
between the 2 sets of plants was not significant (p>0.1).
The percentages of eggs missing after 4, 8, and 24 hr from plants
within the boundaries of 3 £. insana enclaves are presented in Fig. 6.2.
Significantly greater percentages of eggs were missing from plants with
nymphs than from plants without nymphs after 4 hr (36.0% vs. 17.5%) and
after 8 hr (65.6% vs. 42.9%) (p>0.005). After 24 hr the percentage of
eggs missing from plants with nymphs (96.8%) was not significantly
different from the percentage of eggs missing from plants without nymphs
(92.8%) (p>0.1).
In the interior of the field, rates of predation were greater on
plants with SL festinus nymphs tended by ants than on plants without
nymphs. I hypothesize that workers of geminata were attracted to
specific plants by the presence of the nymphs, and that, at any point
in time, more ants were foraging on plnats with nymphs than on plants

Fig. 6.1. Percentages of Pseudoplusia includens eggs missing after 24 hr exposure at 3 heights on
soybean plants with or without Spissistilus festinus nymphs. Plants were in an area where
Solenopsis geminata was the dominant ant (area 1) or in an area where Conomyrma insana was
the dominant ant (area 2).

PERCENT of EGGS MISSING
Nymphs Nymphs
| present absent
SolenoBsis gemlnata
Position
on Plant
1 1 upper 1/3
middle l/3
L,,.* J lower 1/3
Nymphs Nymphs
present absent
insana

Percent of Eggs Missing
87
Fig. 6.2. Percentages of Pseudoplusia includens eggs missing after 4, 8,
and 24 hr exposure at 3 heights on soybean plants with or with
out Spissistilus festinus nymphs. Plants were in 3 areas where
Conomyrma insana was the dominant ant.

88
without nymphs. Thus the probability of a prey item being encountered
by ants would be higher on plants with nymphs than on plants without
nymphs. Within the C^. insana enclaves, rates of predation were also
greater on plants with nymphs than on plnats without nymphs as long as
a supply of eggs remained available to the ants. Supplies of eggs were
almost exhausted after 24 hr, however, and predation approached 100% on
all plants. The data indicate that, at least on cool autumn days,
workers of C^. insana patrol most plants within an enclave. Thus, chances
are high that, given a 24-hr period, a worker would encounter an egg or
other stationary prey item on plants within the boundaries of an enclave
regardless of the presence of an attractant such as the nymphs. Through
out the rest of the field, where egg predation was low on plants without
nymphs, the presence of nymphs could be an important factor affecting
survival of pest insects.
In both predation experiments, the percentages of eggs missing were
greatest from the lowest 1/3 of the plant and least from the upper 1/3
of the plant. Differences were statistically significant (p<0.005).
This suggests that the ants spent more time foraging on the lower parts
of the plants than on the upper parts of the plants or that few ants
foraged at the top of the plants. Eggs oviposited on the upper parts of
the plants would, therefore, be encountered less frequently by ants than
eggs oviposited on the lower parts of the plants and may have a greater
probability of survival.
Conclusions
The threecornered alfalfa hopper is common in soybean fields and
can kill young soybean plants (Tugwell et al. 1972, Caviness and Miner

89
1562, Mueller and Dumas 1975). At the present time, however, the econo
mic impact of the insect on soybean production is unclear. Mueller and
Dumas (1975) found that no yield losses were incurred under Arkansas
conditions even when 30-50% of the soybean plants in a field were de
stroyed by S_. festinus. Threecornered alfalfa hopper nymphs attract ants
to soybean plants. The ants, by preying on pest insects, may protect the
soybean plants, thus offsetting damage caused by the feeding of the
nymphs. Any beneficial effects of the association of ants with Homoptera
should be considered when assessing the economic status of homopteran
pests. Further studies of ant-Homoptera relationships and their effects
of predation should provide a basis for developing strategies that would
increase the effectiveness of ants as predators.
Summary
Five ant species, Solehopsis geminata (F.), Conomyrma insana (Buck-
ley), C. flavopecta (M. R. Smith), Iridomyrmex pruinosus (Roger), and
Pheidole morrisi Fore!, tended nymphs of the threecornered alfalfa hopper,
Spissistilus festinus (Say), on soybeans in a north Florida field. The
presence of nymphs tended by ants was examined as a factor affecting
predation on eggs of the soybean looper, Pseudoplusia includens Walker.
Where S^. geminata tended the nymphs, the percentage of eggs missing after
24 hr from plants with nymphs (77.1%) was significantly greater than the
percentage missing from plants without nymphs (36.8%). Where insana
tended the nymphs, the percentage of eggs missing from plants with nymphs
was significantly greater than the percentage missing from plants without
nymphs after 4 hr (36.8% vs. 17.5%) and 8 hr (65.6% vs. 42.9%) but not

90
after 24 hr (90.7% vs. 86.3%). In all experiments, the proportion of
eggs missing was greatest from the lowest 1/3 of the plant and least from
the upper 1/3.

PROGNOSIS
From the foregoing studies, it is obvious that the distribution of
Conomyrma insana (Buckley) is severely limited by ecological requisites.
The most obvious limiting factors are the availability of suitable nest
ing sites, food sources, and its temporary host, £. flavopecta (M. R.
Smith).
In general, C_. insana nests must be located in open areas fully ex
posed to sunlight. The majority of the sites where C^. insana nested
were acceptable owing to recent disturbances such as mowing, cultivation,
or burning. The ant ceased to nest in sites where normal vegetational
succession occurred. This can be accepted as proof of their dependence
upon open areas for nesting. Additionally, the ant is excluded from
excessively hydric locations.
A further restriction on nesting sites is that a location must in
clude or be adjacent to an adequate food supply. For £. insana this
means that areas of heavy vegetation must be nearby, as they are depen
dent on carbohydrate food sources for the energy required to retain a
site. The carbohydrate food can be from plant nectaries or honeydew
excreted by Homoptera.
The successful colonization of an area is further limited by the
nature of C. insana's habits. Conomyrma insana is temporarily parasitic
on C. flavopecta. Although no analysis is available, it is reasonable
to assume that, based on prior studies with other temporary parasitic
ants, the percentage of successful parasitism is extremely low (Wilson
91

92
1971). Parasitic species of ants must successfully parasitize the host
and produce adequate numbers of offspring to dominate an area. This
requires that the ant increase by a process of hesmosis (budding) until
large areas are their domain with other ants of the same trophic level
excluded. They must then accept additional queens of the same species,
as is true with C. insana. Thus, by chance, they must initially parasi
tize the host species within an area that has the prerequisites of suita
ble nesting conditions with readily available carbohydrate food. This,
in effect, decreases the chances of successful new colonizations. That
success is low is demonstrated by the relatively low percentage of via
ble £. insana colonies as compared to the much greater number of sites
that appear to have the requisites of suitable nesting conditions with
adequate food sources.
As noted by Van Pelt (1966) and Whitcomb et al. (1972), £. insana
is a general predator. The ant is omnivorous and utilizes all avail
able food sources for energy to maintain and expand its colonies.
Conomyrma insana is, without any question, an effective predator
of imported fire ant queens, Solenopsis invicta Burn. A fire ant queen
alighting in a £. insana enclave has little chance of surviving. Yet,
£. insana appears to be a poor candidate for deliberate use as a control
ling agent of fire ants via destroying the queens before they succeed in
beginning a new colony. The relatively sparse distribution and the com
pact enclaves must be considered when assessing the predatory effective
ness of the ant in relation to the total number of fire ant queens
attempting to begin new colonies. At best, even though £. insana workers
do destroy nearly all fire ant queens landing in their enclaves, only a
small fraction of all fire ant queens out of a nuptial flight land where

93
insana occur. It would be a fallacy, though, to dismiss C^. insana as
being ineffective. As Whitcomb et al. (1973) pointed out, all mortality
factors must be considered when determining the percentage of survival
of fire ant queens. Therefore, as £. insana has stable and persistent
populations, it is an important predator of fire ant queens.

LITERATURE CITED
Bhatkar, A., W. H. Whitcomb, W. F. Burn, P. Callahan, and T. Carlysle.
1972. Confrontation behavior between Lasius neoniger (Hymenoptera:
Formicidae) and the imported fire ant. Environ. Entomol. 1:274-9.
Blum, M. S. 1969. Alarm pheremones. Annu. Rev. Entomol. 19:57-80.
Blum, M. S. 1973. Comparative exocrinology of the formicidae. Proc.
VII Congr. IUSSI, London. 23-40.
Blum, M. S. and S. L. Warter. 1966. Chemical releasers of social beha
vior. VII. The isolation of 2-heptanone from Conomyrma pyramica
and its modus operandi as a releaser of alarm and digging behavior.
Ann. Entomol. Soc. Amer. 59:774-9.
Brian, M. V. 1965. Social insect populations. Academic Press, Inc.
London. 135 p.
Buckley, S. B. 1866. Descriptions of new species of North American
Formicidae. Proc. Entomol. Soc. Philadelphia 6:152-72, 335-49.
Carroll, C. R. and D. H. Janzen. 1973. Ecology of foraging by ants.
Ann. Rev. Ecol. Syst. 4:231-57.
Caviness, C. F. and F. D. Miner. 1962. Effects of stand reduction of
soybeans simulating threecornered alfalfa hopper injury. Agron. J.
54:300-2.
Chiu, Y. J., R. Levy, and H. L. Cromroy. 1974. Biological half-life of
cesium-134 in the red imported fire ant, Solenopsis invicta. Fla.
Entomol. 57:141-4.
Cole, A. C., Jr. 1936. Descriptions of seven new western ants (Hymenop.
Formicidae). Entomol. News 47:118-21.
Cole, A. C., Jr. 1940. A guide to the ants of the great Smoky Mountains
National Park, Tennessee. Amer. Midi. Nat. 24:1-88.
Cole, A. C., Jr. 1942. The ants of Utah. Amer. Midi. Nat. 28:358-88.
Cole, A. C. 1957. Notes on western ants (Hymenoptera:Formicidae). N.
Y. Entomol. Soc. 65:129-31.
94

Comstock, J. H. 1879. Report on cototn insects. U. S. Dep. Agr. Ento
mologist Report. 511 p.
Cook, 0. F. 1904. Habits of the Kelep, or Guatemalan cotton-boll-weevil
ant. U. S. Dep. Agr. Entomol. Bull. 49. 15 p.
Creighton, W. S. 1950. The ants of North America. Bull. Mus. Comp.
Zool. 104:1- 585.
de la Cruz, A. A. and R. G. Wiegert. 1967. 32-Phosphorus tracer studies
of a horseweed-aphid-ant food chain. Amer. Midi. Nat. 77:501-9.
DeMoisy, R. G. 1949. Forest surveying. Part 1. The use of steel
tape, compass, Abney level, and aneroid barometer in forest survey
ing and mapping. 0. S. C. Cooperative Assoc., Corvallis, Oregon
123 p.
Dennis, C. A. 1938. The distribution of ant species in Tennessee with
reference to ecological factors. Ann. Entomol. Soc. Amer. 31:267-
308,
Donisthorpe, H. 1943. A list of the type-species of the genera and
subgenera of the Formicidae. Annu. Mag. Nat. Hist. 10:617-737.
Eisner, T. 1957. A comparative morphological study of the proventriculus
of ants. Bull. Mus. Comp. Zool. 116:439-90.
Emery, C. 1909. Uber den ursprung der dulotischen, parasitischen und
myrmekophilen ameisen. 3iologisches Centralblatt 29:352-62.
Fincher, G. T. and H. 0. Lund. 1967. Notes of the biology of the im
ported fire ant, Solenopsis saevissima richteri Forel (Hymenoptera:
Formicidae) in Georgia. J. Ga. Entomol. Soc. 2:91-4.
Finnegan, R. J. 1974. Ants as predators of forest pests. Entomophaga
Memoires Hors-Serie 7:53-9.
Forel, A. 1874. Les fourmis de la Suisse. Nouveaux Memoires de la Soc.
Helvetique de Sci. Nat. 24:1-447.
Forel, A. 1899. Biologia Centrali-Americana. Hymenoptera. Vol. III.
169 p.
Forel, A. 1913. Formicides du Congo beige recoltes par M. M. Bequarert,
Luja, etc. Rev. Zool. Africaine 2:306-51.
Gallardo, A. 1916. Las hormigas de la Repblica Argentina. Subfamilia
Dolichcderinas. Anal. Mus. Nac., Buenos Aires 28:1-130.
Girault, A. A. 1907. The lesser peach borer, hi Papers on decidous
fruit insects and insecticides. U. S. Dep. Agr. Bur. Entomol. 68,
Part IV.

96
Hung, A. C. F. 1974. Ants recovered from the refuse pile of the pyramid
ant Conomyrma insana (Buckley) (Hymenoptera:Formicidae). Ann.
Entoniol. Soc. Amer. 67:522-3.
Hunter, W. D., and W. D. Pierce. 1912. The Mexican cotton boll-weevil:
a summary of the investigation of this insect up to December 21,
1911. U. S. Dep. Agr. Bur. Entomol. Bull. 114. 118 p.
Janzen, D. H. 1967. Interaction of the bull's-horn acacia (Acacia cor-
nigra L.) with an ant inhabitant (Pseudomyrmex ferruginea F. Smith)
in eastern Mexico. Univ. Kans. Sci. Bull. 47:96-110.
Janzen, K. H. 1973. Evolution of polygynous obligate acacia ants in
western Mexico. J. Anim. Ecology 42:727-50.
Jones, C. R. 1929. Ants and their relation to aphids. Col. Exp. Sta.
Bull. 341. 96 p.
Jordan, Jr., C. R. 1952. The biology and control of the three-cornered
alfalfa hopper, Spissistilus festinus (Say). Ph.D. Dissertation.
Texas A & M University, College' Station 129 p.
Kannowski, P. B. 1959. The use of radioactive phosphorus in the study
of colony distribution of the ant Lasius minutus. Ecology 40:162-5.
Knight, P. 1944. Insects associated with the Palay rubber vine in
Haiti. J. Econ. Entomol. 37:100-2.
Kusnezov, N. 1952. El estado real del grupo Dorymyrmex Mayr. Acta
Zoll Lilloana 10:427-48.
Kusnezov, N. 1959. Die Do!ichoderinen-Gattungen von Sud-Amerika. Zool.
Anz. 162:38-51.
Leston, D. 1973. The ant mosaicTropical tree crops and the limiting
of pests and diseases. Pans 19:311-41.
Letendre, M. and J. Piln. 1973. Nids et micromilieux de nidification
utilises par les fourmis dans les peuplements forestiers des basses
Laurentides, Quebec (Hymenoptera:Formicidae). LeNaturaliste
Canadien 100:237-46.
Levieux, P. J. 1971. Mise en evidence de la structure des nids et di
11 implantation des zones de classe de deux especies de Camponotus
a l'aide de radioisotopes. Insectes Soc. 18:29-48.
McCook, H. C. 1879. Formicariae. p. 182-9. Jr^J. H. Comstock.
Report on cotton insects. U. S. Dep. Agr. Entomol. Report 511 p.
McGurk, D. J,, J. Frost, G. R. Waller, E. J. Eisenbraun, K. Vick, W. A.
Drew, and J. A. Young. 1968. Iridodial isomer variation in
dolichoderine ants. J. Insect Physio. 14:841-5.

97
Marikovsky, P. I. 1961. Material on sexual biology of the ant Formica
rufa L. Insectes Soc. 8:23-30.
Markin, G. P., J. H. Dillier, S. 0. Hill, M. S. Blum, and H. R. Hermann.
1971. Nuptial flight and flight ranges of the imported fire ant,
Solenopsis saevissima richteri (Hymenoptera-.Formicidae). J. Ga.
Entorno!. Soc. 6:145-56.
Markin, G. P., J. O'Neal, and H. L. Collins. 1974. Effects of mirex on
the general ant fauna of a treated area in Louisiana. Environ.
Entomol. 3:895-8.
Mayr, G. 1866. Myrmecologische Beitrage. Sitz. Adak. Wiss. Wien. 63:
484-517.
Metcalf, R. A. and R. L. Metcalf. 1970. Effects of isosteres of 2-
heptanone on the alarm behavior of the ant Conomyrma pyramica.
Ann. Entomol. Soc. Amer. 63:34-5.
Mueller, A. J. and B. A. Dumas. 1975. Effects of stem girdling by the
threecornered alfalfa hopper on soybean yields. J. Econ. Entomol.
63:511-2.
Naves, M. A. 1974. Ecological manipulation affecting ant populations.
(Hymenoptera:Formicidae). M. S. thesis. Univ. Fla., Gainesville
77 p.
Nielsson, R. J., A. P. Bhatkar, and H. A. Denmark. 1971 A preliminary
list of ants associated with aphids in Florida. Fla. Entomol. 54.
245-8.
Norton, E. 1875. Wheeler's Report of Geological Exploration. Zool.
p. 734.
Odum, E. P. and E. J. Kuenzler. 1963. Experimental isolation of food
chains in old-field ecosystem with the use of phosphorus-32. Li
Schultz, Vincent and A. W. Klenent, Jr. (eds.). Radioecology.
Reinhold Pub. Corp., N.Y.
Pierce, W. C., R. A. Cushman, C. E. Hood, and W. D. Hunter. 1912. The
insect enemies of the cotton boll weevil. U. S. Dep. Agr. Bur.
Entomol. Bull. 100. 99 p.
Pontin, A. J. 1960. Field experiments on colony foundation by Lasius
niger (L.) and l. flavus (F.). Insectes Soc. 7:227-30.
Quaintance, A. L., and E. L. Jenne. 1912. The plum curculio. U. S.
Dep. Agr. Bur. Entomol. Bull. 103. 205 p.
Roe, R. A. 1973. A biological study of Solenopsis invicta Burn, the
red imported fire ant, in Arkansas with notes on related species.
M. S. Thesis. Univ. Ark., Fayetteville. 135 p.

98
Roger, J. 1863. Die ner aufgefuhrten Gattungen und arten meines
Formiciden-Verzeichnisses. Berliner Entomologische Zeitschrift.
7:131-214.
Sanders, C. J., and W. F. Baldwin. 1969. Iridium-192 as a tag for car
penter arits of the genus Camponotus (Hymenoptera:Formicidae). Can.
Entomol. 101:416-8.
Santschi, F. 1906. A propos des moeurs parasitiques temporaires des
fourmis de genre Bothriomyrmex. Annales de la Societie Entomolo-
gioue de France. 75:363-92.
Santschi, F. 1920. Fourmis du genre Bothriomyrmex Emery. Revue
Zoologizue Africaine. 7:201-24.
Santschi, F. 1922. Myrmecines, Do!ichoderines et autres formicides
neotropiques. Bull. Soc. Sci. Nat. Laussane 54:345-78.
Schwarz, E. A. 1907. Note on Dorymyrmex pyramicus Roger. Proc. Entomol
Soc. Wash. 7:4.
Smith, M. R. 1924. An annotated list of the ants of Mississippi (Hym.).
Entomol. News 35:45-54, 77-85, 121-7.
Smith, M. R. 1936. The ants of Puerto Rico. J. Agr. Univ. Puerto Rico
20:819-75.
Smith, M. R. 1944. Additional ants recorded from Florida, with
descriptions of two new subspecies. Fla. Entomol. 27:14-7.
Snellinq, R. R. 1973. The ant genus Conomyrma in the United States.
(Hymenoptera:Formicidae). Contributions in Science. Los Angeles
Co. Natr. Hist. Mus. 238. 6 p.
Steel, R. G. D. and J. H. Torrie. 1960. Principles and Procedures of
Statistics. McGraw-Hill Book Co., Inc., N.Y. 481 p.
Sudd, J. H. 1967. An introduction to the behaviour of ants. Edward
Arnold Ltd., London 200 p.
Tugwell, P., F. D. Miner, and D. E. Davis. 1972. Threecornered alfalfa
hopper infestations and soybean yields. J. Econ. Entomol. 65:1731-
3.
Van Pelt, A. F. 1966. Activity and density of oldfield ants of the
Savannah River Plant, South Carolina. J. Elisha Mitchell Sci.
Soc. 82:35-43.
Van Pelt, Jr., A. F. 1958. The ecology of the ants of the Welaka Re
serve, Florida (Hymenoptera:Formicidae). Part II. Annotated List.
Amer. Midi. Nat. 59:1-57.
Vick, D., W. A. Drew, J. Young, D. J. McGurk, and E. J. Eisenbraun. 1969
Chemotaxonomic studies of ants: Volatile compounds. Can. Entomol.
101:879-83.

99
Vick, K., W. A. Drew, J. Young, and E. J. Eisenbraun. 1969. Chemotaxo-
nomic studies of ants: Free amino acids. Can. Entomol. 101:1207-
13.
Way, M. J. 1968. The influence of other ant species on biological con
trol of Oecophylla longinoda (Latr.). Proc X Int. Congr. Entomol.
4:595-3.
Wheeler, G. C., and J. Wheeler. 1951. The ant larvae of the subfamily
Dolichoderinae. Proc. Entomol. Soc. Wash. 53:169-210.
Wheeler, W. M. 1901. The compound and mixed nests of American ants.
Amer. Nat. 35:431-8, 513-39, 701-24.
Wheeler, W. M. 1902. A consideration of S. B. Buckley's "North
American Formicidae". Trans. Tex. Acad. Sci. 4:1-15.
Wheeler, W. M. 1905. The ants of the Bahamas, with a list of the
known West Indian species. Bull. Amer. Mus. Natur. Hist. 21:79-137.
Wheeler, W. M. 1906. The ants of the Grand Canyon. Bull. Amer. Mus.
Natur. Hist. 22:329-45.
Wheeler, W. M. 1910. Ants: Their structure, development and behavior.
Columbia Univ. Press., N.Y. 663 p.
Wheeler, W. M. 1932. A list of the ants of Florida with descriptions
of new forms. N. Y. Entomol. Soc. 40:1-17.
Whitcomb, W. H. 1974. Natural populations of entomophagous arthropods
and their effect on the agroecosystem, p. 150-169. _In_ F. G. Max
well and F. A. Harris (eds). Proc. Summer Institute Biol. Control.
Plant Insects and Diseases. Univ. Press. Miss., Jackson, Miss.
647 p.
Whitcomb, W. H., H. A. Denmark, P. A. Bhatkar, and G. L. Greene. 1972.
Preliminary studies on the ants of Florida soybean fields. Fla.
Entomol. 55:129-142.
Whitcomb, W. H., A. P. Bhatkar, and J. C. Nickerson.- 1973. Predators
of Solenopsis invicta prior to successful colony establishment.
Environ. Entomol. 2:1101-3.
Whitcomb, W. H. and K. Bell. 1964. Predaceous insects, spiders, and
mites of Arkansas cotton fields. Ark. Agr. Exp. Sta. Bull. 690.
84 p.
Wiegert, R. G., E. P. Odum and J. Schnell. 1967. A radionuclide tracer
study of forb-arthropod food chains in a one-year experimental field
Ecology 43:75-83.
Wilson, E. 0. 1957. Sympatry of the ants Conomyrma bicolor (Wheeler)
and pyramica (Roger). Psyche 64:76.

TOO
Wilson, E. 0. 1964. The ants of the Florida Keys. Breviora Mus. Comp.
Zool. 14 p.
Wilson, E. 0. 1971. The insect societies. The Belknap Press of
Harvard University Press, Cambridge, Mass. 548 p.
Wilson, E. 0. and F. E. Regnier, Jr. 1971. The evolution of the alarm-
defense system in the formicine ants. Amer. Naturalist. 105:279-89.

APPENDICES

APPENDIX A
GLOSSARY OF SPECIFIC TERMS
Alate. Winged, having wings.
Brood chamber. Chamber constructed for purpose of rearing brood.
Callow. Immature; green. Freshly pupated adult ant prior to darkening.
Claustral. Form of colony establishment in which a freshly inseminated
queen ant digs and seals brood chamber, lays eggs, and feeds first
developing brood from stored energy.
Contiguous. Adjoining; near, though not in contact.
Deal ate. Wingless; a term applied to formerly winged females or queen
ants that have shed their wings.
Declivity. Moderately steep. Declivity of mesonotum; the inclined
surface of the mesonotum.
Enclave. An enclosed tract or territory with definite boundaries.
Epinotum. That part of the thorax behind the mesoepinotal suture; in
worker ants the term is loosely applied, since the region morpho
logically is composed of two segments, an anterior metathorax and
a posterior epinotum.
Extrafloral nectaries. Plant structures not associated with flowers that
produce a sweet secretion or nectar relished by ants as food.
Floral nectaries. Certain parts of flowers that secrete a sweet liquid
or nectar relished by ants for food.
Gaster. That portion of the abdomen behind the petiole in ants with a
single segmented petiole, and behind the postpetiole in ants with
a two-segmented petiole.
Gyne. Woman; female. In ants, refers to the reproductive form, or
queen ant.
Hesmosis. Process of new colonies forming by "budding" or "colony
fission" from parent colony.
Honeydew. A sweetish alimentary excretion produced by certain insects
such as plant lice, mealybugs, and scales, and greatly relished as
food by ants.
102

103
Incipient. Beginning to be; commencing; initial. Colony of ants com
posed of queen and underdeveloped first brood workers, in addition
to brood.
Infuscation. A smoky gray-brown or blackish tinge.
Microgynous. See gyne.
Nuptial flights. The mating flight of female and male ants. It should
be understood that all ants do not need to take flight and mate in
the air; some mate inside of the nest, others on the soil, plants,
or other objects outside of the nest.
Occipital border. The hind margin of the head.
Pedicel. The one or two basal segments of the abdomen between the epino-
tum and gaster.
Queen (female). The individual that normally carries on reproduction;
she is usually larger than the worker or male, and before mating
normally possesses two pairs of wings; in most colonies the queen,
after establishing her nest and rearing her first brood alone, func
tions only as an egg producer, with the care and feeding of the
brood relegated to the workers.
Temporary host. The species of ant whose colony is invaded temporarily
by a queen of an alien species. See temporary parasite.
Temporary parasite. The female of certain ants who seeks adoption in a
colony of an alien species. To accept such a queen the alien colony
must be queenless, the queen is killed by her own workers, or by the
invading queen; after the death of the host workers, the colony
eventually becomes purely one of the same species as the invading
queen.
Tend. Applied to the visits of ants on insects for the purpose of
obtaining honeydew for food.
Worker. Usually an undeveloped female, differing from the queen in its
smaller size, lack of wings, and other morphological features; the
normal functions of the worker are foraging for food, care of the
brood and nest, and defense of the colony.
I

APPENDIX B
PUBLICATIONS
The following publications are based, entirely or in part, on infor
mation and data contained in this dissertation:
Nickerson, J. C. 1973. Predators of Solenopsis invicta prior to suc
cessful colony establishment. Environ. Entomol. 2:1101-3. With
W. H. Whitcomb and A. P. Bhatkar.
Nickerson, J. C. 1974. Eriococcus smithii (Eriococcidae:Homoptera):
A new species record for Florida. Fla. Entomol. 57(3):285. With
W. H. Whitcomb, and G. W. Dekle.
Nickerson, J. C. 1975. Predation on founding queens of Solenopsis in
victa by workers of Conomyrma insana. Fla. Entomol. 58:75-82.
With W. H. Whitcomb, A. P. Bhatkar, and M. A. Naves.
Nickerson, J. C. 1975. Colony organization and queen numbers in two
species of Conomyrma. Ann. Entomol. Soc. Amer. 68:1083-5. With
H. L. Crornroy, W. H. Whitcomb and J. A. Cornell.
Nickerson, J. C. 1975. Mixed nests of Conomyrma insana and £. flavo-
pecta--evidence of parasitism (Hymenoptera:Formicidae). Psyche
83:306-14. With W. F. Burn and C. R. Thompson.
Mickerson, J. C. Presence of Spissistilus festinus (Horn.:Membracidae)
tended by ants (Hym.:Formicidae): A-factor affecting egg preda
tion by ants in soybeans. (Submitted for publication.) With
C. A. Rolph Kay, L. L. Buschman, and W. H. Whitcomb.
104

BIOGRAPHICAL SKETCH
James Charles Everett Nickerson was born on July 5, 1942, in De-
Queen, Arkansas. His parents are James Lawrence Nickerson and Katie
Aline (Miller) Nickerson of Horatio, Arkansas. In 1966, he married
Norma Jean Doss. They have 3 children: Genie Lene born in 1967, James
Everett born in 1970, and Jon Carlos born in 1972.
He attended the Horatio Public Schools grades 1-12, graduating from
high school in 1961. In 1963, he enrolled in the University of Arkansas
at Fayetteville and received his B. S. A. degree in 1969. Immediately,
he commenced graduate study at the same institution in the Department of
Entomology. The M. S. degree was awarded in 1972.
Study towards a Ph. D. degree was initiated in September 1971 at
the University of Florida in the Department of Entomology and Nematology.
105

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
Jr. w. H. Whitcomb, Chairman
Professor of Entomology
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
Dr. Wm. F. Burn
Professor of Entomology
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
Dr. P. S. Callahan
Professor of Entomology

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
A.^A ?vc
Dr. S. H. Kerr
Professor of Entomology
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
Professor of Zoology
This dissertation was submitted to the Graduate Faculty of the College
of Agriculture and to the Graduate Council, and was accepted as partial
fulfillment of the requirements for the degree of Doctor of Philosophy.
December, 1976
cudl <^/wy
College of Agrjic
ulture
Dean, Graduate School



48
of an estimated 800 workers. Altogether, based on counts of dead
workers, ca. 4,300 insana workers were reared in the colony during
the year.
Discussion
The mixed nests of £. insana-flavopecta found Florida and the
localized enclaves of C^. insana give presumptive evidence that temporary
social parasitism followed by hesmosis (budding to form new nests) was
occurring in these ants. In addition, the data indicate that £. insana
was parasitic on C_. flavopecta. The modus operandi for the adoption of
the parasitic queen, elimination of the host queen, and eventual matura
tion to a £. insana nest should be similar, in a general way, to numerous
other temporary parasitic ant species.
Further evidence of temporary social parasitism may be given by the
morphology of the queens. The head of C^. insana is large, wider than
the thorax, while the gaster is small and slender in alate queens. The
head of £. flavopecta is of moderate size, about equal in width to the
thorax, and the gaster is large and voluminous in both alate and deal ate
queens. The morphological form of the C^. insana queens is similar to
the modifications already known for many temporary parasitic ants.
According to Wilson (1971), queens of temporary parasitic species are
all atypical to a degree. In many species, the head is enlarged and the
gaster is reduced in relation to the thorax. In addition, the mandibles
may be strongly modified to form falcate, piercing organs; or the queen
may be small (microgynous) in relation to the worker and/or may have
various other characterisitics such as modified pilosity or integument.
In contrast, females of free living species, which practice the normal


Fig. i.5. Conomyrma insana nests (stippled areas) in Persimmon Field
during July, 1974.


59
and hurried rate of movement. Subsequently, the queen would renew the
search for a suitable location in which to excavate the brood chamber.
Secondary contacts. The time between first contact and subsequent
attacks varied from 15 sec to 35 min depending on the local population
densities and the degree of surface activity of £. insana. These secon
dary contacts differed from the first encounter in several respects.
The £. insana attacks were progressively more frequent and prolonged;
.the workers tended to cling to the queen until dislodged. Pursuit of
the queen by workers also occurred, although individual workers did not
take part continuously. The intensity of pursuit was again dependent
upon the local density of £. insana. As the attack progressed, the
number of individuals of C^. insana workers taking part increased up to
the time the queen was taken as prey.
Queen response to attack. The responses of the Sk invicta founding
queens to the attack by the C^. insana workers can be divided into 5
basic reactions: (1) escape, (2) concealment, (3) cessation of all
movement, (4) mandibular defense, and (5) rapid flexing or shaking of
the body. All or several of these responses were utilized by the queens
in attempting to avoid or fight off the attack of C. insana worker(s).
Escape attempts from the attack of £. insana workers occurred in 2
forms: (1) flight and (2) surface movement. If a queen was attacked
prior to shedding of the wings, she could avoid predation by further
flight. This was accomplished by less than 2% of the queens under
observation. The queens, when under attack, crawled up vegetation or
any other available elevation in an attempt to gain the height necessary
to become airborne. Queens not under attack were seldom observed to


CHAPTER VI
PRESENCE OF Spissistilus festinus TENDED BY ANTS:
A FACTOR AFFECTING EGG PREDATION BY ANTS IN SOYBEANS
Introduction
Ants are important predators on arthropod pests in agroecosystems
(Cook 1904, Pierce et al. 1912, Whitcomb and Bell 1964, Whitcomb et al.
1972). Many predaceous ants supplement their diet by collecting honey-
dew from Homoptera (Carroll and Janzen 1973). This stationary and re
newable food source can be an attractant resulting in locally high ant
population densities that may be maintained throughout the year (Finne
gan 1974). The presence of Homoptera could consequently influence the
effectiveness of ants as predators.
In 1975, ants were found tending nymphs of the threecornered alfalfa
hopper, Spissistilus festinus (Say), on soybeans in northern Florida.
Since the ants were observed more often on plants with the nymphs than
on plants without the nymphs, I hypothesized that predation on soybean
pests increased in the presence on nymphs. My objective was to test
this hypothesis by examining predation on eggs of the soybean looper,
Pseudoplusia includens Walker, under field conditions.
Methods
Experiments were conducted between 15 September and 15 October,
1975 in a 15-ha field of Bragg soybeans in Alachua County, Florida.
Initially, the field was surveyed for ant species tending nymphs of
79


21
N


10
Distribution
At present, there is no real means of delineating the range of
insana. The range of Dorymyrmex/Conomyrma pyramicus is recorded to be
as far north as North Dakota and, according to Creighton (1950), occurs
in all states except the extreme northeastern United States.
The confirmed records of insana are from southern states.
These include Texas (Buckley 1866, Wheeler 1901), Arkansas (Roe 1973),
Mississippi (Markin et al. 1971), Florida (Whitcomb et al. 1973), Georgia
(Van Pelt 1966), and Louisiana (Markin et al. 1974).


10 meters


75


Table 5.1. Aphids tended by Conomyrma insana and the plant host(s) that the aphids were found on. Tall
Timbers Research Station, Leon Co., Fla. 1972-75.
Plant Host
Aphid Species
Family
Species (Common Name)
Cinara watsoni Tissot
Pinaceae
Pinus echinata Miller (Short leaf pine)
Cinara melaina Boudreaux
Pinaceae
Pinus echinata Miller (Short leaf pine)
Lachnochaitophorus obscurus (Tissot)
Fagaceae
Quercus nigra L. (Water oak)
Fagaceae
Quercus pumila Walter (Running oak)
Sipha flava (Forbes)
Gramineae
Eragrostis spectabilis (Pursh.) Steud (Love grass)
Gramineae
Digitaria villosa (Walt.) Pers. (Crabgrass)
Gramineae
Paspa!urn setaceum Michx.
Aphis coreopsidis (Thomas)
Compositae
Bidens bipinnata L. (Beggar ticks)
Aphis goss.ypii Glover
Caesalpiniaceae
Cassia obtusifolia L. (Sickle-pod)
Rubiaceae
Richardia scabra L. (Mexican clover)

Compositae
Gnaphalium obtusifolium L. (Everlasting)
Aphis oestlundi Gill
Onagraceae
Oenothera sp. (Evening primrose)
Aphis spiraecola Patch
Chenopodiaceae
Chenopodium ambrosioides L. (Mexican tea)
Compositae
Eupatorium capillifolium (Lam.) Small (Dog-fennel)


Fig. 2.1. Location of Conomyrma insana craters in replication A, B, C,
and D, showing the relation of sampled and nonsampled craters
to the source crater.


92
1971). Parasitic species of ants must successfully parasitize the host
and produce adequate numbers of offspring to dominate an area. This
requires that the ant increase by a process of hesmosis (budding) until
large areas are their domain with other ants of the same trophic level
excluded. They must then accept additional queens of the same species,
as is true with C. insana. Thus, by chance, they must initially parasi
tize the host species within an area that has the prerequisites of suita
ble nesting conditions with readily available carbohydrate food. This,
in effect, decreases the chances of successful new colonizations. That
success is low is demonstrated by the relatively low percentage of via
ble £. insana colonies as compared to the much greater number of sites
that appear to have the requisites of suitable nesting conditions with
adequate food sources.
As noted by Van Pelt (1966) and Whitcomb et al. (1972), £. insana
is a general predator. The ant is omnivorous and utilizes all avail
able food sources for energy to maintain and expand its colonies.
Conomyrma insana is, without any question, an effective predator
of imported fire ant queens, Solenopsis invicta Burn. A fire ant queen
alighting in a £. insana enclave has little chance of surviving. Yet,
£. insana appears to be a poor candidate for deliberate use as a control
ling agent of fire ants via destroying the queens before they succeed in
beginning a new colony. The relatively sparse distribution and the com
pact enclaves must be considered when assessing the predatory effective
ness of the ant in relation to the total number of fire ant queens
attempting to begin new colonies. At best, even though £. insana workers
do destroy nearly all fire ant queens landing in their enclaves, only a
small fraction of all fire ant queens out of a nuptial flight land where


Comstock, J. H. 1879. Report on cototn insects. U. S. Dep. Agr. Ento
mologist Report. 511 p.
Cook, 0. F. 1904. Habits of the Kelep, or Guatemalan cotton-boll-weevil
ant. U. S. Dep. Agr. Entomol. Bull. 49. 15 p.
Creighton, W. S. 1950. The ants of North America. Bull. Mus. Comp.
Zool. 104:1- 585.
de la Cruz, A. A. and R. G. Wiegert. 1967. 32-Phosphorus tracer studies
of a horseweed-aphid-ant food chain. Amer. Midi. Nat. 77:501-9.
DeMoisy, R. G. 1949. Forest surveying. Part 1. The use of steel
tape, compass, Abney level, and aneroid barometer in forest survey
ing and mapping. 0. S. C. Cooperative Assoc., Corvallis, Oregon
123 p.
Dennis, C. A. 1938. The distribution of ant species in Tennessee with
reference to ecological factors. Ann. Entomol. Soc. Amer. 31:267-
308,
Donisthorpe, H. 1943. A list of the type-species of the genera and
subgenera of the Formicidae. Annu. Mag. Nat. Hist. 10:617-737.
Eisner, T. 1957. A comparative morphological study of the proventriculus
of ants. Bull. Mus. Comp. Zool. 116:439-90.
Emery, C. 1909. Uber den ursprung der dulotischen, parasitischen und
myrmekophilen ameisen. 3iologisches Centralblatt 29:352-62.
Fincher, G. T. and H. 0. Lund. 1967. Notes of the biology of the im
ported fire ant, Solenopsis saevissima richteri Forel (Hymenoptera:
Formicidae) in Georgia. J. Ga. Entomol. Soc. 2:91-4.
Finnegan, R. J. 1974. Ants as predators of forest pests. Entomophaga
Memoires Hors-Serie 7:53-9.
Forel, A. 1874. Les fourmis de la Suisse. Nouveaux Memoires de la Soc.
Helvetique de Sci. Nat. 24:1-447.
Forel, A. 1899. Biologia Centrali-Americana. Hymenoptera. Vol. III.
169 p.
Forel, A. 1913. Formicides du Congo beige recoltes par M. M. Bequarert,
Luja, etc. Rev. Zool. Africaine 2:306-51.
Gallardo, A. 1916. Las hormigas de la Repblica Argentina. Subfamilia
Dolichcderinas. Anal. Mus. Nac., Buenos Aires 28:1-130.
Girault, A. A. 1907. The lesser peach borer, hi Papers on decidous
fruit insects and insecticides. U. S. Dep. Agr. Bur. Entomol. 68,
Part IV.


74


90
after 24 hr (90.7% vs. 86.3%). In all experiments, the proportion of
eggs missing was greatest from the lowest 1/3 of the plant and least from
the upper 1/3.


9
Homoptera Tending
Nielsson et al. (1971) reported the association of several aphid
species with £. pyramicus, but did not attempt to establish if a mutu
al istic relationship occurred between the ant and species of aphids
listed. In fact, they reported associations with Dactynotus sp., which
are not thought to be tended by any ant species. Van Pelt (1966) noted
that £. pyramicus (= £. insana based on Van Felt's description of the
nesting habits) tended 2 aphid species, Hysteroneura setariae (Thomas)
and Pteroconuna smithiae (Monel!). Research by Odum and Kuenzler (1963),
Wiegert et al. (1967), and de la Cruz and Wiegert (1967), using ^P to
trace the plant-aphid-ant food chain, demonstrated that £. pyramicus
was obtaining honeydew from aphids of the genera Aphis and Macrosiphum.
Jones (1929) listed 3 aphids, Aphis sp., Chaitophorous negundinus Thos.,
and C. populella Gill, and Pal., tended by £. pyramicus. Knight (1944)
reported that the hemispherical scale, Saisetta hemispherica Targioni,
was tended by £. pyramicus var. niger Perg. on Palay rubber vines.
Comstock (1879) reported that ID. insanus (= £. insana) feeds upon the
sweet exudations of plants, galls, and sweet excretions of aphids.
Habitat
While no habitat studies have been made on ID. pyramicus, numerous
researchers, including Cole (1940, 1942), Dennis (1938), Smith (1924,
1936), Van Pelt (1958), Wheeler (1905, 1932), and Wilson (1964), have
noted that this ant prefers to nest in open sandy areas. Markin et al.
(1974) stated that £. insana was common in a part of an open field in
Louisiana. According to Naves (1974), £. insana are more populous on
roadways or other open areas with sparse vegetation.


Fig. 1.7. Conomyrma insana nests (stippled areas) in Persimmon Field
during July, 1975.


41
Conclusions
Two basic assumptions were made in this research in relation to
insana: (1) the center of the colony could be selected, and (2) there
would be equal distribution of the isotope throughout the colony. The
first assumption could be fallacious as no criteria existed upon which
to base a decision; therefore, the decision was purely subjective.
Possibly, the isotope may have been placed at the edge of the colony.
The second assumption also had some invalidity because of possible
"storage" individuals in the colony which could accumulate excessive
amounts of isotope. In addition, when samples were taken and assayed,
the individuals removed were not replaced in the colony, resulting in a
net loss of radioactivity, not only by biological half-life (Chiu et al.
1974), but also by removal of radioactive individuals. Nevertheless,
the data appear to validate the conclusion that, despite the relatively
large numbers of "exit" holes, only 1 colony of C^. insana was present in
each of the tested populations. Conomyrma insana can thus be assumed to
have polydomous nesting habits and more than 1 queen per colony.
Conomyrma flavopecta, in contrast, appears to have monodomous
nesting habits with only 1 queen per colony. Although queens were not
found in 24 out of a total of 34 nests excavated, in those 10 nests
where a queen was found, there was only a single queen per nest. These
data, plus the lack of distribution of ^Cs among the nests, support
the hypothesis that there is only 1 queen per nest, and each nest is a
distinct colony. An alternate hypothesis that single females could be
serving as the single colony queen for a number of nests is not sup
ported by the ^^Cs data. This failure to find queens in 24 of the 34
nests can more reasonable be attributed to.escape responses of the


99
Vick, K., W. A. Drew, J. Young, and E. J. Eisenbraun. 1969. Chemotaxo-
nomic studies of ants: Free amino acids. Can. Entomol. 101:1207-
13.
Way, M. J. 1968. The influence of other ant species on biological con
trol of Oecophylla longinoda (Latr.). Proc X Int. Congr. Entomol.
4:595-3.
Wheeler, G. C., and J. Wheeler. 1951. The ant larvae of the subfamily
Dolichoderinae. Proc. Entomol. Soc. Wash. 53:169-210.
Wheeler, W. M. 1901. The compound and mixed nests of American ants.
Amer. Nat. 35:431-8, 513-39, 701-24.
Wheeler, W. M. 1902. A consideration of S. B. Buckley's "North
American Formicidae". Trans. Tex. Acad. Sci. 4:1-15.
Wheeler, W. M. 1905. The ants of the Bahamas, with a list of the
known West Indian species. Bull. Amer. Mus. Natur. Hist. 21:79-137.
Wheeler, W. M. 1906. The ants of the Grand Canyon. Bull. Amer. Mus.
Natur. Hist. 22:329-45.
Wheeler, W. M. 1910. Ants: Their structure, development and behavior.
Columbia Univ. Press., N.Y. 663 p.
Wheeler, W. M. 1932. A list of the ants of Florida with descriptions
of new forms. N. Y. Entomol. Soc. 40:1-17.
Whitcomb, W. H. 1974. Natural populations of entomophagous arthropods
and their effect on the agroecosystem, p. 150-169. _In_ F. G. Max
well and F. A. Harris (eds). Proc. Summer Institute Biol. Control.
Plant Insects and Diseases. Univ. Press. Miss., Jackson, Miss.
647 p.
Whitcomb, W. H., H. A. Denmark, P. A. Bhatkar, and G. L. Greene. 1972.
Preliminary studies on the ants of Florida soybean fields. Fla.
Entomol. 55:129-142.
Whitcomb, W. H., A. P. Bhatkar, and J. C. Nickerson.- 1973. Predators
of Solenopsis invicta prior to successful colony establishment.
Environ. Entomol. 2:1101-3.
Whitcomb, W. H. and K. Bell. 1964. Predaceous insects, spiders, and
mites of Arkansas cotton fields. Ark. Agr. Exp. Sta. Bull. 690.
84 p.
Wiegert, R. G., E. P. Odum and J. Schnell. 1967. A radionuclide tracer
study of forb-arthropod food chains in a one-year experimental field
Ecology 43:75-83.
Wilson, E. 0. 1957. Sympatry of the ants Conomyrma bicolor (Wheeler)
and pyramica (Roger). Psyche 64:76.


49
claustral method of new colony formation, have voluminous gasters and
relatively small heads with mandibles, pilosity, and integument not
modified from the norm for the genus.
The existence of incipient nests is additional evidence that a
species has the normal claustral type of new colony establishment as
temporary parasitic queens have lost the ability to begin a new colony
on their own (Emery 1909). Incipient nests of £. flavopecta have been
found repeatedly in Florida; whereas, no incipient nests of £. insana
have been found.
Apparently, the £. flavopecta queen had already been eliminated
when the mixed nest was excavated on 27 May as only a C_. insana queen
was found, and only brood and callows of C. insana were present. In many
temporary parasitic ants, adoption of the parasitic queen by a host
colony is difficult and not often successful (Sudd 1967). Mixed colo
nies of these species and their hosts are rare. Nevertheless, the
parasitic species may become locally abundant by the process of colony
fission (hesmosis) to form an enclave of closely spaced, interconnected
nests (Emery 1909). Further adoption of newly mated parasitic queens,
additional to the original parasitic queen, is a necessary step in the
growth of such an enclave. Colony "towns" of such ants often have been
noted by observers. In the case of the allegheny mound building ant,
Formica exsectoides Forel, enclaves containing up to 1,700 individual
nests have been recorded (Wheeler 1901). Similarly, enclaves of C^.
insana that exceed 2,000 nests have been located in Leon Co., Fla.
The clustering of the mixed nests in the study area indicates that
the process of hesmosis begins to occur even while the nests are still


56
TABLE 4.1 Continued.
Elapsed
time (min) Observations
41 Si knocked left forewing off. Ci_ cling to Si_ legs.
Sibit and killed Ci_ (n=l). Si_ now moving rapidly,
Ci_ pursuing (n=9). Ci_ clinging (n=3), Si shook all
off, still moving rapidly. Ci clinging~[h=3) and
pursuing (n=ll). Sj_ curled into "C" position and bit
at Ci_, then moved away with Ci_ in pursuit (n=14).
48 §2 still moving with no pursuit.
50 Ci attack (n=2) Si. Si_ shook off attack.
51 Si_ under leaf momentarily, then emerged.
52 Si attacked by Ci (n=l). Si curled into "C" position
and bit CL Ci (n=17) pursuing SL Si shook off
pursuit and attack, moved under clod of soil.
53 Si emerged, immediately attacked by Ci (n=l). Ci at
tacked (n=3). Si dislodged attackers. Ci clinging
to wing. Si climbed blade of grass, immediately
descended with Ci_ (n=l) still clinging to wing.
54 (n=l) attacked Si at right metathoracic leg. Si
curled into "C" position and bit Ci_, forcing it to
release. Ci (n=2) attacked, forced away by Si.
55 Ci attacked (n=2), Si curled into "C" position and
bit at Ci. Si moved with Ci still clinging to leg.
Ci (n=7T~biting and pursuing, Si flexing abdomen and
thorax, intermixed with assuming "C" position and
biting.
58 Ci (n=3) attacking Si. Si shook loose CL Additional
Ci (n=8) attacked, dislodged all £i by shaking
and biting.
6C Si moved forward and went under a leaf.
61 Si emerged from under the leaf and was immediately
attacked by (n=l). Si curled into "C" position
and bit Ci until it left. Ci (n=3) in pursuit of
Si. Ci TFT=3) caught Si. Si bit at Ci on leg. Ad
ditional attacked Xn=9). Si assumed "C" position
and bit clinging to leg but did not dislodge.


BIOGRAPHICAL SKETCH
James Charles Everett Nickerson was born on July 5, 1942, in De-
Queen, Arkansas. His parents are James Lawrence Nickerson and Katie
Aline (Miller) Nickerson of Horatio, Arkansas. In 1966, he married
Norma Jean Doss. They have 3 children: Genie Lene born in 1967, James
Everett born in 1970, and Jon Carlos born in 1972.
He attended the Horatio Public Schools grades 1-12, graduating from
high school in 1961. In 1963, he enrolled in the University of Arkansas
at Fayetteville and received his B. S. A. degree in 1969. Immediately,
he commenced graduate study at the same institution in the Department of
Entomology. The M. S. degree was awarded in 1972.
Study towards a Ph. D. degree was initiated in September 1971 at
the University of Florida in the Department of Entomology and Nematology.
105


32
Research Station. The ants were found more commonly in open areas adja
cent to heavy vegetation. Nests were absent in heavily cultivated areas,
in areas of heavy vegetation, and where drainage was poor.


53
observed while under attack, 240 were watched from the time they alighted
until they were either destroyed by predators or escaped. When possible,
the action was verbally described and recorded on tape as it occurred.
Most observations were made on bare ground, either on dirt roadways or
in fallow fields. Additional observations were made in mowed fields with
sparse vegetation cover. This report is based on observations of queens
which alighted naturally; information on queens which had been captured
and handled was discarded.
In 1972, the effectiveness of predation by C^. insana was deter
mined. Nests of £. insana were examined and excavated to determine the
numbers of S_. invicta queens taken as prey. In addition, five 3 x 8 m
plots were randomly located in an open field. These plots were examined
at 1/2 hr intervals from 1400 to 1830 on the dates of heavy SL invicta
nuptial flights. Predation of S. invicta queens within these plots was
noted and recorded. For 3 consecutive days following nuptial flight,
this field was searched for additional :S. invicta females on the surface
and for signs of brood chamber construction.
Results and Discussion
A summary of the responses of Su invicta queens to attack(s) by £.
insana workers is shown in the flow diagram in Fig. 4.1. While the re
sponses of the queen to attack are varied, they can be separated into
artificial categories which recur frequently.
An abridged summary of a single attack sequence by <3. insana work
ers on 1 S. invicta queen is presented in Table 4.1. This was taken
from observations made 21 June 1972 of a queen that had alighted at
1626. The nuptial flight began after 2 days of intermittent storms


7
Physiology
The isolation of 2-heptanone (methylpentyl ketone) from the anal
glands of Conomyrma pyramica and its role as the alarm pheromone were
described by Blum (1969) and Blum and Warter (1966). The substance
releases a complex behavioral pattern in the worker ant which includes
alarm, attraction, repellancy, and digging. Metcalf and Metcalf (1970)
reported that typical alarm behavior by £. pyramica was initiated in
the laboratory by exposing the worker ants to isosteres of 2-heptanone
(1,1,1-trifluro-methylheptanone, butyl acetate, and butylthiol acetate).
They also reported that a remarkable behavioral change was induced by
the isostere hexanoyl chloride, which invariably produced a violent
combat between worker ants which touched one another.
Blum (1973) reported that the anal glands of do!ichoderines are not
restricted to biosynthesis of terpenoid natural products. His state
ment is corrcbrated by research by McGurk et al. (1968), which showed
that C. pyramicus also produce the monoterpene iridodial.
Vick et al. (1969, 1969) investigated the analysis of volatile com
pounds and free amino acids in several ant species, including CL pyrami-
ca. to determine the potential of the chemicals as taxonomic characters
in the study of ant systematics. They concluded that valuable
information on ant classification and phylogeny may be obtained.
Predation
Comstock (1879) reported both D.. insanos (= CL insana) and D_. flavus
killing the "cotton-worm." Schwarz (1907) noted that CL pyramicus pro
tected solitary tree cotton from the boll weevil in Cuba. Pierce et al.
(1912) and Hunter and Pierce (1912) listed CL pyramicus var. flavus Mc-
Coo'< as an enemy of Alabama argil lacea (Hulsner) and Heliothis obsoleta



Fig. 6.1. Percentages of Pseudoplusia includens eggs missing after 24 hr exposure at 3 heights on
soybean plants with or without Spissistilus festinus nymphs. Plants were in an area where
Solenopsis geminata was the dominant ant (area 1) or in an area where Conomyrma insana was
the dominant ant (area 2).


47
Fig. 3.1. Condition of study area 1 June to 10 June 1975, showing (1)
location of original excavated Conomyrma insana-flavopecta
nest, (2) additional mixed C_. insana-flavopecta nest found
several days after excavation, (3) small, localized enclaves
of £. insana nests, and (4) scattered occurrence of
flavopecta nests.
fe
e
e
O
o
rC
O
or

o
o
"O
o
Lu
o
o
>>
"O
c
03
oo
o
1
oo
o & o
O no
O O ^
I
CP O oo oQ Q
'
o o
flavopecta
Ij*
* %
flavopecta
^ Peach Tree
iV-V -V JV-\ w:
5 meters
Fig. 3.2. Condition of study area on.26 June 1975 showing growth of
the C^. insana enclaves and appearance of a new enclave of C^.
insana in the same area as the previously present mixed nests,
while the unmixed C^. flavopecta nests appear to have remained
unchanged.


I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
A.^A ?vc
Dr. S. H. Kerr
Professor of Entomology
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
Professor of Zoology
This dissertation was submitted to the Graduate Faculty of the College
of Agriculture and to the Graduate Council, and was accepted as partial
fulfillment of the requirements for the degree of Doctor of Philosophy.
December, 1976
cudl <^/wy
College of Agrjic
ulture
Dean, Graduate School


Percent of Eggs Missing
87
Fig. 6.2. Percentages of Pseudoplusia includens eggs missing after 4, 8,
and 24 hr exposure at 3 heights on soybean plants with or with
out Spissistilus festinus nymphs. Plants were in 3 areas where
Conomyrma insana was the dominant ant.


APPENDIX B
PUBLICATIONS
The following publications are based, entirely or in part, on infor
mation and data contained in this dissertation:
Nickerson, J. C. 1973. Predators of Solenopsis invicta prior to suc
cessful colony establishment. Environ. Entomol. 2:1101-3. With
W. H. Whitcomb and A. P. Bhatkar.
Nickerson, J. C. 1974. Eriococcus smithii (Eriococcidae:Homoptera):
A new species record for Florida. Fla. Entomol. 57(3):285. With
W. H. Whitcomb, and G. W. Dekle.
Nickerson, J. C. 1975. Predation on founding queens of Solenopsis in
victa by workers of Conomyrma insana. Fla. Entomol. 58:75-82.
With W. H. Whitcomb, A. P. Bhatkar, and M. A. Naves.
Nickerson, J. C. 1975. Colony organization and queen numbers in two
species of Conomyrma. Ann. Entomol. Soc. Amer. 68:1083-5. With
H. L. Crornroy, W. H. Whitcomb and J. A. Cornell.
Nickerson, J. C. 1975. Mixed nests of Conomyrma insana and £. flavo-
pecta--evidence of parasitism (Hymenoptera:Formicidae). Psyche
83:306-14. With W. F. Burn and C. R. Thompson.
Mickerson, J. C. Presence of Spissistilus festinus (Horn.:Membracidae)
tended by ants (Hym.:Formicidae): A-factor affecting egg preda
tion by ants in soybeans. (Submitted for publication.) With
C. A. Rolph Kay, L. L. Buschman, and W. H. Whitcomb.
104


TOO
Wilson, E. 0. 1964. The ants of the Florida Keys. Breviora Mus. Comp.
Zool. 14 p.
Wilson, E. 0. 1971. The insect societies. The Belknap Press of
Harvard University Press, Cambridge, Mass. 548 p.
Wilson, E. 0. and F. E. Regnier, Jr. 1971. The evolution of the alarm-
defense system in the formicine ants. Amer. Naturalist. 105:279-89.


88
without nymphs. Thus the probability of a prey item being encountered
by ants would be higher on plants with nymphs than on plants without
nymphs. Within the C^. insana enclaves, rates of predation were also
greater on plants with nymphs than on plnats without nymphs as long as
a supply of eggs remained available to the ants. Supplies of eggs were
almost exhausted after 24 hr, however, and predation approached 100% on
all plants. The data indicate that, at least on cool autumn days,
workers of C^. insana patrol most plants within an enclave. Thus, chances
are high that, given a 24-hr period, a worker would encounter an egg or
other stationary prey item on plants within the boundaries of an enclave
regardless of the presence of an attractant such as the nymphs. Through
out the rest of the field, where egg predation was low on plants without
nymphs, the presence of nymphs could be an important factor affecting
survival of pest insects.
In both predation experiments, the percentages of eggs missing were
greatest from the lowest 1/3 of the plant and least from the upper 1/3
of the plant. Differences were statistically significant (p<0.005).
This suggests that the ants spent more time foraging on the lower parts
of the plants than on the upper parts of the plants or that few ants
foraged at the top of the plants. Eggs oviposited on the upper parts of
the plants would, therefore, be encountered less frequently by ants than
eggs oviposited on the lower parts of the plants and may have a greater
probability of survival.
Conclusions
The threecornered alfalfa hopper is common in soybean fields and
can kill young soybean plants (Tugwell et al. 1972, Caviness and Miner


Table 5.1 .Continued.
Aphid Species
Plant Host
Family
Species (Common Name)
Hysteroneura setariae (Thomas)
Gramineae
Eragrostis spectabilis (Pursh.) Steud. (Love grass)
Gramineae
Digitaria violascens Link (Crabgrass)
Gramineae
Paspalum boscianum Flugge
Gramineae
Paspalum plicatulum Michx.
Gramineae
Paspalum setaceum Michx.
Gramineae
Cenchrus echinatus L. (Sandspurs)
Gramineae
Andropogon virginicus L. (Broom sedge)
Rhopalosiphum maidis (Fitch)
Gramineae
Paspalum boscianum Flugge
Gramineae
Zea mays L. (Corn)
Myzocallis bella (Walsh)
Fagaceae
Quercus pumila Walter (Running oak)
Neosymdobius lutens Tissot
Fagaceae
Quercus nigra L. (Water oak)


Table 5.1 .Continued.
Plant Host
Aphid Species
Family
Species (Common Name)
Aphis spiraecola Patch Compositae
Aphis fabae Scop.
Compositae
Compositae
Aphis craccivora Koch
Caesalpiniaceae
Caesalpiniaceae
Aphis sp.
Fagaceae
Amaranthaceae
Rubiaceae
Scrophulariaceae
Compositae
Compositae
Schizaphis sp.
Labiatae
Cerosipha rubifoli Thomas
Rosaceae
Pyrrhopappus carol ini anus (Walt.) DC
Conyza canadensis (L.) Crong. (Horseweed)
Ambrosia artemisiifolia L. (Ragweed)
Cassia fasciculata Michx. (Partridge pea)
Cassia nictitans L. (Wild sensitive plant)
Quercus nigra L. (Water oak)
Amaranthus hybridus L.
Richardia scabra L. (Mexican clover)
Agalinis fasiculata (Ell.) Rat.
Ambrosia artemisiifolia L. (Ragweed)
Eupatorium capillifolium (Lam.) Small (Dog-fennel)
Trichostema dichotomum L. (Blue curls)
Rubus cunneifolius Pursh. (Sand blackberry)


TABLE OF CONTENTS
ACKNOWLEDGEMENTS
Page
i i i
ABSTRACT Vi
INTRODUCTION 1
REVIEW OF LITERATURE 3
Introduction 3
Taxonomy 4
Physiology 7
Predation 7
Homoptera Tending 9
Habitat 9
Distribution 10
CHAPTER
I. DISTRIBUTION AND HABITAT OF Conomyrrr.a insana II
Introduction 11
Methods and Materials 11
Results and Discussion 12
Summary 24
II. COLONY ORGANIZATION AND NUMBERS OF QUEENS 33
Introduction 33
Materials and Methods 34
Results and Discussion 35
Conclusions 41
Summary 42
III. MIXED NESTS OF Conomyrma insana and C. flavopectaEVIDENCE
OF TEMPORARY SOCIAL PARASITISM 43
Introduction 43
Methods 45
Results 45
Discussion 43
Summary 50
iv


84
significantly different between area 1 and area 2 (p>0.1), but signifi
cantly more nymphs were present in both areas on 3 October than on 15 Oc
tober (p<0.005). By 15 October the soybean plants were senescent, and
many plants were apparently unacceptable as hosts for the nymphs.
The percentage of soybean looper eggs missing after 24 hr from
plants in area 1 and 2 are presented in Fig. 6.1. A significantly
greater percentage of eggs was missing from plants in area 2 (88.6%)
than from plants in area 1 (57.0%) (p<0.005). Within area 1, a signi
ficantly greater percentage of eggs was missing from plants with nymphs
(77.1%) than from plants without nymphs (36.8%) (p<0.005). Within area
2, a greater percentage of eggs also was missing from plants with nymphs
(90.8%) than from plants without nymphs (86.3%), but the difference
between the 2 sets of plants was not significant (p>0.1).
The percentages of eggs missing after 4, 8, and 24 hr from plants
within the boundaries of 3 £. insana enclaves are presented in Fig. 6.2.
Significantly greater percentages of eggs were missing from plants with
nymphs than from plants without nymphs after 4 hr (36.0% vs. 17.5%) and
after 8 hr (65.6% vs. 42.9%) (p>0.005). After 24 hr the percentage of
eggs missing from plants with nymphs (96.8%) was not significantly
different from the percentage of eggs missing from plants without nymphs
(92.8%) (p>0.1).
In the interior of the field, rates of predation were greater on
plants with SL festinus nymphs tended by ants than on plants without
nymphs. I hypothesize that workers of geminata were attracted to
specific plants by the presence of the nymphs, and that, at any point
in time, more ants were foraging on plnats with nymphs than on plants


42
queens in the large quantity of soil which must be excavated and searched,
or failure to find the true bottom of the nest.
The radioisotope 134Cs appears to be an excellent tagging material
for this group of ants. The isotope was recovered in sufficient quanti
ties for assay, and the evidence indicated that it was rapidly and
evenly distributed throughout the £. insana colony.
Summary
The organization of the colony and number of queens per colony of
Conomyrma insana (Buckley) and C. flavopecta (M. R. Smith) were investi
gated with the aid of the radioisotope cesium-134 and by excavation of
nests. Conomyrma insana was determined to be polydomous (more than 1
nest), had a single colony occupying several square meters, and had
multiple queens located in a central nest. Conomyrma flavopecta was
monodomous (1 nest per colony) and had 1 colony queen.
/


CHAPTER V
SOURCES OF CARBOHYDRATES UTILIZED BY Conomyrma insana
Introduction
The colonies of the ant Conomyrma insana (Buckley) consist of many
clustered nests that form enclaves. Conomyrma insana is omnivorous, how
ever, it is a voracious predator (Van Pelt 1966, Hung 1974, Markin et al.
1971, Whitcomb et al. 1972). Way (1968) suggested that, in terms of food
supply, any species of omnivorous ant, in order to maintain continuously
large colonies, had to change foraging grounds periodically as doryline
ant do, or to tend Homoptera, which provide honeydew to carry the ants
over periods when animal prey is scarce. The workers of £. insana appar
ently derive a large portion of the energy required to maintain their
colonies from honeydew excreted by homopterans and nectar from extrafloral
and floral structures of plants.
Two species of Conomyrma, insana and C_. flavopecta (M. R. Smith),
are currently recognized in the southeastern United States (Snelling 1973).
Prior to Snelling's revision, the North American species of Conomyrma
were lumped as Dorymyrmex pyramicus (Roger) or Conomyrma pyramicus. Thus,
the information contained in reports of aphid tending by £. or £. pyrami-
cus is difficult to assign correctly to the presently accepted species
of Conomyrma. Nielsson et al. (1971) reported the association of several
aphid species with (1. pyramicus, but did not attempt to establish if a
mutualistic relationship occurred with the species of aphids listed. Van
64


Table 5.2. Homoptera, other than aphids, tended by Conomyrma insana and the host plant(s) they were found
on. Tall Timbers Research Station, Leon Co., Fla. 1972-75.
Plant Host
Species
Family
Species (Common Name)
Phylloxieridae
Adelqes sp.
Pinaceae
Coccidae
Saisettia sp.
Compositae
Pulviaria urbicola Ckll.
Compositae
Chenopodiaceae
Toumeyella parvicornia (Ckll.)
Pinaceae
Coccus hesperidum Linn.
Fabaceae
Fabaceae
Coccus sp.
Compositae
Pseudoccidae
Oracella sp.
Fagaceae
Pinus echinata Miller (Short leaf pine)
Eupatorium capillifolium (Lam.) (Don-Fennel) ^
Pyrropappus carolineaus (Walt.)
Chenopodium ambrosiodes L.
Pinus echinata Miller (Short leaf pine)
; Glycine max Merr. (Soybean)
Lespedeza angust'folia (Pursh.) Ell.
Ambrosia artemisiifolia L. (Ragweed)
Quercus virginiana Miller (Live oak)


96
Hung, A. C. F. 1974. Ants recovered from the refuse pile of the pyramid
ant Conomyrma insana (Buckley) (Hymenoptera:Formicidae). Ann.
Entoniol. Soc. Amer. 67:522-3.
Hunter, W. D., and W. D. Pierce. 1912. The Mexican cotton boll-weevil:
a summary of the investigation of this insect up to December 21,
1911. U. S. Dep. Agr. Bur. Entomol. Bull. 114. 118 p.
Janzen, D. H. 1967. Interaction of the bull's-horn acacia (Acacia cor-
nigra L.) with an ant inhabitant (Pseudomyrmex ferruginea F. Smith)
in eastern Mexico. Univ. Kans. Sci. Bull. 47:96-110.
Janzen, K. H. 1973. Evolution of polygynous obligate acacia ants in
western Mexico. J. Anim. Ecology 42:727-50.
Jones, C. R. 1929. Ants and their relation to aphids. Col. Exp. Sta.
Bull. 341. 96 p.
Jordan, Jr., C. R. 1952. The biology and control of the three-cornered
alfalfa hopper, Spissistilus festinus (Say). Ph.D. Dissertation.
Texas A & M University, College' Station 129 p.
Kannowski, P. B. 1959. The use of radioactive phosphorus in the study
of colony distribution of the ant Lasius minutus. Ecology 40:162-5.
Knight, P. 1944. Insects associated with the Palay rubber vine in
Haiti. J. Econ. Entomol. 37:100-2.
Kusnezov, N. 1952. El estado real del grupo Dorymyrmex Mayr. Acta
Zoll Lilloana 10:427-48.
Kusnezov, N. 1959. Die Do!ichoderinen-Gattungen von Sud-Amerika. Zool.
Anz. 162:38-51.
Leston, D. 1973. The ant mosaicTropical tree crops and the limiting
of pests and diseases. Pans 19:311-41.
Letendre, M. and J. Piln. 1973. Nids et micromilieux de nidification
utilises par les fourmis dans les peuplements forestiers des basses
Laurentides, Quebec (Hymenoptera:Formicidae). LeNaturaliste
Canadien 100:237-46.
Levieux, P. J. 1971. Mise en evidence de la structure des nids et di
11 implantation des zones de classe de deux especies de Camponotus
a l'aide de radioisotopes. Insectes Soc. 18:29-48.
McCook, H. C. 1879. Formicariae. p. 182-9. Jr^J. H. Comstock.
Report on cotton insects. U. S. Dep. Agr. Entomol. Report 511 p.
McGurk, D. J,, J. Frost, G. R. Waller, E. J. Eisenbraun, K. Vick, W. A.
Drew, and J. A. Young. 1968. Iridodial isomer variation in
dolichoderine ants. J. Insect Physio. 14:841-5.


Fig. 1.1 Tall Timbers Research Station with locations of Conomyrma in
sana populations indicated. Where populations were found in
open fields, the entire field is marked rather than the bound
aries of the limited distribution of C. insana within the field.
The locations are: (1) Jones Field, JZ) Field N-W5, (3) Gate
Field, (4) Middle Gate Road, (5) Sutty Field, (6) Butler Field,
(7) Gum Pond Course Road, (8) Field M-8, (9) Field M-9, (10) Gay
Field C, (11) Hall Island Drive Road, (12) Gay Field B, (13)
Field M-12, (14) Hall Island Course Road, (15) Field L-8 and
L-9, (16) Field L-3, (17) Whitcomb Plots Rep 3, (18) Whitcomb
Plots Rep 2, (19) Fennel Field, (20) Tripsacum Field, (21)
Field 1.-12, (22) Tripsacum Course Road, (23) Richard Field,
(24) Introduction Field, (25) Persimmon Field, (26) Field L-14,
(27) Pea Field, (28) Field M-15, (29) Stoddard Lab Field.


50
mixed. Obviously, 1 or more additional £. insana queens were present in
these mixed nests, as the new £. insana enclave continued to develop
even after the £. insana queen found in the original mixed nest was
removed.
This is the first report of temporary social parasitism in the
genus Conomyrma. It may be noted that mixed nests of Conomyrma have
been recorded (as Dorymyrmex) previously by Wheeler (1901) who gave the
remarks of Forel upon finding mixed nests of ants identified as Dory
myrmex pyramicus var. niger Pergande and £. pyramicus var. flavus
McCook near Faison, North Carolina. Forel attributed these mixed nests
to a queen of each taxon fortuitously associating together when first
founding a new colony. Such a chance association of 2 species has never
been confirmed for any ants, although reported in older literature
(Forel 1874, Wheeler 1901). In the Forel colonies, all alates found
were identified as niger except for 1 male which Forel identified as
flavus on the basis of color. In the absence of any positive statement,
I assume no deal ate queens were found to confirm Forel's hypothesis.
Summary
Conomyrma insana and C^ flavopecta were found in the same nest in
Florida. Conomyrma insana queen, brood, and workers were in one nest
with C. flavopecta workers. Later, clusters of £. insana nests were
formed at locations where the mixed nests were found. Conomyrma
insana is a temporary social parasite on £. flavopecta.


PERCENT of EGGS MISSING
Nymphs Nymphs
| present absent
SolenoBsis gemlnata
Position
on Plant
1 1 upper 1/3
middle l/3
L,,.* J lower 1/3
Nymphs Nymphs
present absent
insana


76
Fig. 5.4. Distribution of Conomyrma insana nests in 1976 around a
short leaf pine, Pinus echinata, located in an open field
at Tall Timbers Research Station. The short leaf pine was
heavily infested with Cinara watsoni.


61
with the tip of the abdomen curled ventrally toward the mandibles. In
this position, the SL invicta queen could use the mandibles effectively
for defense. On a number of occasions, a queen severed an abdomen from
a C_. insana worker's thorax at the pedicel or decapitated a worker
hanging onto 1 of her legs. The worker's decapitated head often re
mained attached to the seized appendage. This method of defense was
effective against an individual worker, although the queen could not
remain in the position long as she was vulnerable to further attack by
other £. insana workers.
The S. invicta queens frequently attempted to dislodge clinging £.
insana workers by rapid shaking or flexing of the abdomen. This was
used by all queens under attack by C_. insana workers. The queen was
always successful in dislodging attacking workers by this means until
she became exhausted or was overcome by the mass attack of the £.
insana.
Predation and disposal of queens. When the queens were preyed upon
successfully by £. insana workers, the appendages were removed by a
combination of tugging and biting at the joints. Appendageless queens
were dragged, while still alive, to nests by the workers.
On the day following predation, bits of discarded body parts and
heads of invicta queens were found on the periphery of the C_. insana
nests. After a major nuptial flight of 21 June 1972, a series of 24
nests were excavated on each of 2 successive days. On the 1st day, live
appendageless queens were found in all nests. The fewest found in a
single colony was 7 and the most was 18. A total of 329 S_. invicta
queens was recovered from the £. insana nests. An additional 83
t


55
TABLE 4.1. Abridged sunmary of observation of attack by Conomyrma insana
(Ci) on a founding queen of Solenopsis invicta (Si) beginning
at 1625 on 21 June 1972. Leon Co., Fla.
Elapsed
time (min)
Observations
0 Si_ alights, immediately moves under debris and into
soil crevices.
4 Met by 1st £i_; duration of contact momentary. Si
movement rapid and erratic.
9 Si reverted to normal rate of movement.
13 £i attacked (n=2) and pursued (n=l). Si hid under
leaf.
20 Si left leaf. attacked Si at base of wing (n=l),
Head (n=l), and thorax (n=lTT Si fled.
21 Si_ has traveled 8.5 m from alighting point, wings
intact.
30 Ci_ attacked (n=l) Si_ at head. Si_ fled.
33 Ci attacked (n=9), Si^ shook Ci loose by rapidly
flexing body.
34 Ci attacked (n=4), Ci clinging to wing, Si_ shook Ci
loose. Si climbed blade of grass, attempted to fly,
failed. *
35 pursuing (n=4), Ci_ attached (n=3) to Si_. All Ci_
dislodged from Si_.
36 Ci clinging to thorax and wing (n=2). Si_ dislodged
38 Ci attacked (n=4), Si motionless (6 sec), fled. Ci_
pursued and attacked (n=2).
39 Ci pursuing (n=7) and clinging to Si_ head (n=l).
40 Si moving rapidly, attempting to break off wings. Ci
Tn=3) pursuing and biting at legs and wings. Si
flexed abdomen rapidly, curled into "C" position and
bit at Ci. Si had traveled 16.8 m since alighting.


17
N


6
beer, correct in his suggestion, yet he persisted in calling the ant £.
insanus. Forel (1899) placed insana in synonomy to pyramicus. Wheeler
(1902) stated that insana was undobutedly a synonym of pyramicus.
Buckley's insana was uniformly black or brownish-black; Roger (1363)
stated that pyramicus possessed a yellowish-red head and thorax and
brownish gaster. Roger's pyramicus, as described, agreed closely with
the description of a new variety, bicolor, by Wheeler (1906).
Creighton (1950) retained insana as a synonym of pyramicus, yet he
indicated doubt as to the validity of his decision. He stated that
since neither Mayr nor Forel had seen Roger's (1863) types, there was
little to indicate that the "typical pyramicus" was the same insect that
Roger described or the same form to which that name is applied to in
North America. According to Snelling (1973), the ant long referred to
as pyramicus in North American literature has a mesonotum which is
sharply declivitous behind and is uniformly brownish to blackish with
the lower part of the head lighter. This ant clearly is not the same
entity as the South American pyramicus described by Roger (1863). The
earliest available name for this ant is insana. Based on the evidence,
Snelling (1973) recognized insana (Buckley), and gave the synonomy for
this name as follows:
insana (Buckley 1866)
= pyramicus, sensu Wheeler (1902), Creighton (1950),
not of Roger (1863)
= flavus McCook (1879)
= nigra Pergande
= antillana Forel
= smithi Cole (1936)
= brunea, Kusnezov (1952)
= wheeleri Kusnezov (1952)


103
Incipient. Beginning to be; commencing; initial. Colony of ants com
posed of queen and underdeveloped first brood workers, in addition
to brood.
Infuscation. A smoky gray-brown or blackish tinge.
Microgynous. See gyne.
Nuptial flights. The mating flight of female and male ants. It should
be understood that all ants do not need to take flight and mate in
the air; some mate inside of the nest, others on the soil, plants,
or other objects outside of the nest.
Occipital border. The hind margin of the head.
Pedicel. The one or two basal segments of the abdomen between the epino-
tum and gaster.
Queen (female). The individual that normally carries on reproduction;
she is usually larger than the worker or male, and before mating
normally possesses two pairs of wings; in most colonies the queen,
after establishing her nest and rearing her first brood alone, func
tions only as an egg producer, with the care and feeding of the
brood relegated to the workers.
Temporary host. The species of ant whose colony is invaded temporarily
by a queen of an alien species. See temporary parasite.
Temporary parasite. The female of certain ants who seeks adoption in a
colony of an alien species. To accept such a queen the alien colony
must be queenless, the queen is killed by her own workers, or by the
invading queen; after the death of the host workers, the colony
eventually becomes purely one of the same species as the invading
queen.
Tend. Applied to the visits of ants on insects for the purpose of
obtaining honeydew for food.
Worker. Usually an undeveloped female, differing from the queen in its
smaller size, lack of wings, and other morphological features; the
normal functions of the worker are foraging for food, care of the
brood and nest, and defense of the colony.
I


93
insana occur. It would be a fallacy, though, to dismiss C^. insana as
being ineffective. As Whitcomb et al. (1973) pointed out, all mortality
factors must be considered when determining the percentage of survival
of fire ant queens. Therefore, as £. insana has stable and persistent
populations, it is an important predator of fire ant queens.


89
1562, Mueller and Dumas 1975). At the present time, however, the econo
mic impact of the insect on soybean production is unclear. Mueller and
Dumas (1975) found that no yield losses were incurred under Arkansas
conditions even when 30-50% of the soybean plants in a field were de
stroyed by S_. festinus. Threecornered alfalfa hopper nymphs attract ants
to soybean plants. The ants, by preying on pest insects, may protect the
soybean plants, thus offsetting damage caused by the feeding of the
nymphs. Any beneficial effects of the association of ants with Homoptera
should be considered when assessing the economic status of homopteran
pests. Further studies of ant-Homoptera relationships and their effects
of predation should provide a basis for developing strategies that would
increase the effectiveness of ants as predators.
Summary
Five ant species, Solehopsis geminata (F.), Conomyrma insana (Buck-
ley), C. flavopecta (M. R. Smith), Iridomyrmex pruinosus (Roger), and
Pheidole morrisi Fore!, tended nymphs of the threecornered alfalfa hopper,
Spissistilus festinus (Say), on soybeans in a north Florida field. The
presence of nymphs tended by ants was examined as a factor affecting
predation on eggs of the soybean looper, Pseudoplusia includens Walker.
Where S^. geminata tended the nymphs, the percentage of eggs missing after
24 hr from plants with nymphs (77.1%) was significantly greater than the
percentage missing from plants without nymphs (36.8%). Where insana
tended the nymphs, the percentage of eggs missing from plants with nymphs
was significantly greater than the percentage missing from plants without
nymphs after 4 hr (36.8% vs. 17.5%) and 8 hr (65.6% vs. 42.9%) but not


80
Spissistilus festinus. Two areas were then selected for experimentation,
2
an 1100-m area in the interior of the field in which the nymphs were
being tended primarily by Solenopsis geminata (F.) (designated as area
O
1) and a 45-m area at the perimeter of the field in which the nymphs
were being tended by Conomyrma insana (Buckley) (designated as area 2).
On 3 and 15 October, between 1200 and 1700 hr, nymphs were counted in
10 randomly selected 3.05-m sections of soybean row in the S.. geminata
area and three 3.05-m sections of soybean row in the C_. insana area.
On both dates, the ant species present on each plant and the number of
soybean plants in each 3.05-m section of soybean row were recorded. An
analysis of variance (Steel and Torrie 1960) was performed on the
squareroot-transformed counts of nymphs. Data were arranged in a 2 x 2
factorial design (= 2 areas of the field x 2 dates).
In each of the 2 areas of the field, 14 plants were selected, 7
with 1 or more nymphs tended by ants and 7 without nymphs. The absence
o
of nymphs was confirmed by shaking each plant over a 0.58-m white cloth.
Previous tests had established that nymphs were easily shaken from the
soybean plants. A trifoliate leaf on the upper, middle, and lower 1/3
of each plant was marked with a small spot of water-soluble paint. Eggs
of the soybean looper, Pseudoplusia includens, were obtained from a
laboratory colony maintained by N. C. Leppla, Insect Attractants, Be
havior and Basic Biology Research Laboratory, ARS, USDA, Gainesville,
FI. Three eggs were attached with casein glue to each marked leaf (1
on the upper surface of each leaflet). Therefore, 9 eggs (3 at each
height on the plants) were attached to each of the 14 plants in
an area for a total of 126 eggs/area. Twenty-one of the eggs were at
I


45
where a mixed nest was found, were investigated in order to determine if
this explanation for the observations could be fact. Temporary social
parasitism is well known in ants, especially in the subfamilies Formi-
cinae and Myrrnicinae (Brian 1965, Sudd 1967, and Wilson 1971). In the
subfamily Dolichoderinae, species of the genus Bothriomyrmex are temp
orary social parasites on species of Tapi noma (Santschi 1906, 1920).
Several species of the genus Azteca are also suspected of being tempor
ary parasites because they have microgynous queens (Wheeler 1910) and,
therefore, cannot be expected to employ the usual claustral method of
colony establishment.
Methods
The mixed nest of Conomyrma insana-flavopecta, found on 23 May 1975
on the University of Florida Experimental Farm, was excavated on 27 May.
The queen, workers, and brood present in this colony were collected and
counted. This queen, 25 C_. insana workers, 25 C^. flavopecta workers,
and the brood were maintained in the laboratory.
A 45 x 20 m area was marked off into 1 m squares. A diagrammatic
map shewing major vegetation and the relative location of all Conomyrma
nest craters was constructed on 1-10 June and 26 June. The species of
Conomyrma constructing each nest was recorded.
Results
Three hundred twenty-six Conomyrma flavopecta workers and 726 £.
insana workers were collected from the mixed nests excavated on 27 May.
A single deal ate queen of C^. insana was captured in this nest. All
pupae (ca. 75) mature enough to identify (by head shape) were insana


62
queen heads were counted from the discarded debris around the colonies
that were excavated. On the 2nd day, only 1 £. insana nest had live
appendageless invicta queens (n=2).
Effectiveness of C. insana as a Predator
Nests of £. insana tend to be concentrated in localized areas. The
greater the number of C^. insana nests within the area, the more intense
the attack on fire ant queens. Mortality of S^. invicta queens did not
depend on the effectiveness of the bite or toxin of the individual work
ers, but rather on the ability of C_. insana to continue the attack until
the invicta queens were taken as prey. When the process was inter
rupted prior to removal of appendages, the queen survived and laid eggs.
Where the process was followed from beginning to end, however, only 6
queens out of 240 contacted by £. insana workers managed to escape de
struction by any means: 4 by flying away from the C_. insana area and
2 by escaping on the ground surface to outside of the foraging range
of C^. insana.
In 1972, an effort to quantify predation by £. insana was made in
five 3 x 8 m plots. On 9 May, 613 postnuptial S_. invicta queens were
counted in these plots; 60% of these were completely subdued by the C^.
insana workers, 36.4% were under continuous attack, and 3.6% were
momentarily unmolested.
A field at Tall Timbers Research Station was searched for S^.
invicta queens for 3 successive days following a nuptial flight. On 10
May, 19 S_. invicta queens were being dragged to the C_- insana colony by
a coordinated effort of workers. Only 4 S_. invicta queens were still
under attack by C^. insana. On 11 and 12 May, no live queens were found


15
The soil at all £. insana nesting sites v/as sandy with little clay
or loam content. The nest craters were found only in areas fully ex
posed to sunlight. No nests were found in heavy vegetation. The vege
tation at all sites v/as sparse to absent because of cultivation, well-
maintained fire lanes, annual burning (spring), sterile soil, roadways,
or closely mowed areas. At one location, vegetation was absent owing
to a rapidly expanding gulley. The most common plants occurring within
the nesting areas of C_. insana were Andropogon vlrqinicus L., Di gl tari a
sp., Pas pal urn sp., Richardia scabra L., Cassia fasciculata Michx., C_.
nictltans L., Eupatorium capiHi folium (Lam.) Small, and Ambrosia
artemisiifalla L.
Apparently, slope had little effect on the distribution of C^. insana
nests. Although the majority of the areas where the ant was found were
gently sloped (1-6"), at 2 locations the slope was greater than 30*.
All sites were good to well drained. Poorly drained areas adjacent to
established enclaves of C^. insana were not utilized for nesting.
The C_. insana enclaves in Introduction Field from 1973 to 1976 are
shown in Figs. 1.2-1.5. The boundaries of the enclaves are somewhat
stable, but some change is evident from year to year. The major annual
variation occurred on the v/est side of the field. In 1973, this portion
of the field had an extremely patchy vegetative cover (ca. 30%) consist
ing primarily of Paspalum bosianum Flugge, and £. insana nests were
numerous (Fig. 1.2). In 1974 and 1975, the vegetation was apparently
too rank for C. insana as they no longer nested in this portion of the
field (Fig. 1.3 and 1.4). In 1976, C^. insana were nesting in a fire lane
along the edge of the west side of the field (Fig. 1.5). This fire lane
had been maintained annually, but in 1976 the vegetation was less profuse
than in previous years.


31
Table 1.1. Number of nests per square meter within £. insana enclaves
in Introduction Field and Persimmon Field, Tall Timbers
Research Station, Leon Co., Fla.
Year
Introduction Field
x + 95% C.I.(a)
Persimmon Field
x + 95% C.I.'a'
1973
2.4 + .41
-(b)
1974
2.7 + .32
3.6 + .32
1975
2.5 + .25
3.5 + .22
1976
3.1 + .76
3.8 + .27
[a(Mean + 95% Confidence Interval.
^b'No Samples taken.