Full text of "Psyche"

See other formats

IlfllllSNI ""'nVINOSHJLI 3 I H V d a !1 "* L I B R A R I ES^ SMITHSONIAN^INSTITUTli
r- , z r* 2 r-

- St I —

z \ m

CO « CO H CO

3 R ARIES SMITHSONIAN INSTITUTION NOlinilXSNi NVINOSHillNS S3IHVtfa
<2 ^ ^ z: co z 2

1 %% 1 I

v z t t \3X4AKsv = >

linillSNI ~NVIN0SHJLIWS^$3 I dVd a h2li b RAR I ES^SMITHSONIAN^INSTITUTI

■ 00 - CO —

5 W 2 1

z 'X' -I z — _| z

3 R ARIES SMITHSONIAN INSTITUTION NOlinilJLSNI NVIN0SH1IWS S3 I H Vd €
5 „ ^ z r- z

Oi
2/
of
o!

m “ Vt'

IXfUIXSNI NVINOSHXIWS S3 I U Vd 3 n~LI B RAR I ES^SMITHSONIAN^NSTITUTI

I ^SL

MR z >wv\' _i ^

> 5 S

O) z co * z — ^ *•

SRARIES SM ITHSON IAN _ INSTITUTION NOIXfUIXSNI_NVINOSHXIWS S3ldVdf

O ><1«XI XV>^ Q

linillSNI ^NVINQSHillNS^S 3 I d VH QIl^LIBRARI ES^SMITHSONIAN^INSTITUTI
r* * z r- z r-

m • Z p-j

3RAR I ES SMITHSON IAN ""INSTITUTION NOlifUIJLSNI ~ NVIN0SH1IWS S3 I HVHE

z co z co

^ — p -

> \!vass>- ^ > '' 2

linillSNI NVINOSHJUIAJS^SS I d VH a llZLI B RAR I ES^SMITHSONIAN ^INSTITUTI
~ 00 = co

O Q

lini!iSNI_lNVIN0SHllWS:ZS3 I avaa n^LIBRAR I ES^SMITHSONIANJNSTITUTIC

00
3D
>

3RAR I ES SMITHSONIAN INSTITUTION NOlinillSNI NVIN0SH1MS S31BVH9

CO 21 r C/> 2

T/7?V

X O

in
O

? *

> 2 ' )^v > ' 2 .

linillSNlf NVIN0SH1IWS^S3 I HVH 8 ll\l BRAR I ES^SMITHSONIAN JNSTITUTK

l | /#%. l S <jfkf l

J S ' K (tfLoSl) c (|§. .31) < c W

kE©?' 5 " x^ss^ 5 5

2 ' «l 2 " j Z

3RARIES SMITHSONIAN INSTITUTION NOlinillSNI NVIN0SH1IWS S3IBVaa

- - z r- _ # 2

mniliSNI NVINOSHJLIWS S3 I dVd 8 n-LI B RAR I ES SMITHSONIAN INSTITUTI

C/5 Z

2 /■»«* A -- • I 1 i A;-

O X

S I x\ >

BRAR I ES^SMITHSONIAN INSTITUTION NOlinillSNI NVINOSHIIKIS^SS I UM& E
</> =: co

eg
<
c/ a

m § - B ^ ■

ilinillSNI^NVINOSHHIAIS^SS I d VB 8 II^LI B RAR I ES^SMITHSONIAN^INSTITUTI

BRARIES SMITHSONIAN INSTITUTION NOlinillSNI NVIN0SH1MS S3IHVH8

PSYCHE

A Journal of Entomology

Volume 88
1981

Editorial Board

Frank M. Carpenter, Editor P. J. Darlington, Jr.

W. L. Brown, Jr. H. W. Levi

E. O. Wilson Alfred F. Newton, Jr.

B. K. Holldobler R. E. Silberglied

Ronald J. McGinley

Published Quarterly by the Cambridge Entomological Club
Editorial Office: Biological Laboratories
16 Divinity Avenue
Cambridge, Massachusetts, U.S.A.

The numbers of Psyche issued during the past year were mailed on the following
dates:

Vol. 87, no. 1-2, for 1980, April 27, 1981

Vol. 87, no. 3-4, for 1980, August 31, 1981

Vol. 88, no. 1-2, for 1981, December 28, 1981

Wj /

PSYCHE

A JOURNAL OF ENTOMOLOGY

founded in 1874 by the Cambridge Entomological Club

Vol. 88 1981 No. 1-2

CONTENTS

Anti-predator Strategies. II. Grasshoppers (Orthoptera, Acrididae) Attacked
by Prionyx parkeri and Some Tachysphex Wasps (Hymenoptera, Sphecinae

and Larrinae): A Descriptive Study. A. L. Steiner 1

A Comparison of the Nest Phenologies of Three Species of Pogonomyrmex

Harvester Ants (Hymenoptera: Formicidae). William P. MacKay 25

Laboratory Evaluation of Within-species, Between-Species, and Partheno-
genetic Reproduction in Reticulitermes flavipes and Reticulitermes vir-
ginicus. Ralph W. Howard, Eldon J. Mallette, Michael /. Haverty, and

Richard V. Smythe 75

Ecology and Life History of the Rhytidoponera impressa Group (Hymenop-
tera: Formicidae). I. Habitats, Nest Sites, and Foraging Behavior.

Philip S. Ward 89

Ecology and Life History of the Rhytidoponera impressa Group (Hymenoptera:
Formicidae). II. Colony Origin, Seasonal Cycles and Reproduction.

Philip S. Ward 109

The Ontogeny of Lyssomanes viridis (Walckenaer) (Araneae: Salticidae) on
Magnolia grandiflora L.

David B. Richman and Willard H. Whitcomb 127

The Emigration Behavior of Two Species of the Genus Pheidole (Hymenoptera:

Formicidae). Robert Droual and Howard Topoff 135

Statary Behavior in Nomadic Colonies of Army Ants: The Effect of Overfeed-
ing. Howard Topoff, Aaron Rothstein, Susan Pujdak, and

Tina Dahlstrom 151

Life History of Antaeotricha sp. (Lepidoptera: Oecophoridae: Stenomatinae) in

Panama. Annette Aiello 163

Polistes gallicus in Massachusetts (Hymenoptera:Vespidae).

Mary Hathaway 169

Notes on the Population Ecology of Cicadas (Homoptera: Cicadidae) in the
Cuesta Angel Forest Ravine of Northeastern Costa Rica.

Allen M. Young

175

CAMBRIDGE ENTOMOLOGICAL CLUB
Officers for 1980-1981

President

Vice-President

Secretary

Treasurer

Executive Committee

Jo Brewer Winter
Barbara L. Thorne
Heather Hermann
Frank M. Carpenter
William D. Winter, Jr.
Ernest Leblanc

EDITORIAL BOARD OF PSYCHE

F. M. Carpenter (Editor), Fisher Professor of Natural History,
Emeritus, Harvard University

W. L. Brown, Jr., Professor of Entomology, Cornell University and
Associate in Entomology, Museum of Comparative Zoology
P. J. DARLINGTON, Jr., Professor of Zoology, Emeritus, Harvard
University

B. K. HoLLDOBLER, Professor of Biology, Harvard University
H. W. Levi, Alexander Agassiz Professor of Zoology, Harvard University
R. J. McGlNLEY, Assistant Professor of Biology, Harvard University
Alfred F. Newton, Jr., Curatorial Associate in Entomology, Harvard
University

R. E. SlLBERGLlED, Smithsonian Tropical Research Institute, Panama
E. O. WILSON, Baird Professor of Science, Harvard University

PSYCHE is published quarterly by the Cambridge Entomological Club, the issues
appearing in March, June, September and December. Subscription price, per year,
payable in advance: $11.00, domestic and foreign. Single copies, $3.50.

Checks and remittances should be addressed to Treasurer, Cambridge
Entomological Club, 16 Divinity Avenue, Cambridge, Mass. 02138.

Orders for missing numbers, notices of change of address, etc., should be sent to
the Editorial Office of Psyche, 16 Divinity Avenue, Cambridge, Mass. 02138. For
previous volumes, see notice on inside back cover.

IMPORTANT NOTICE TO CONTRIBUTORS
Manuscripts intended for publication should be addressed to Professor F. M.
Carpenter, Biological Laboratories, Harvard University, Cambridge, Mass. 02138.

Authors are expected to bear part of the printing costs, at the rate of $27.50 per
printed page. The actual cost of preparing cuts for all illustrations must be borne by
contributors: the cost for full page plates from line drawings is ordinarily $18.00
each, and for full page half-tones, $30.00 each; smaller sizes in proportion.

Psyche, vol. 87, no. 3-4, for 1980, was mailed August 31, 1981

The Lexington Press, Inc., Lexington, Massachusetts

PSYCHE

Vol. 88

1981

No. 1-2

ANTI-PREDATOR STRATEGIES.

II.* GRASSHOPPERS (ORTHOPTERA, ACRIDIDAE)
ATTACKED BY PRIONYX PARKERI AND SOME

TACHYSPHEX WASPS (HYMENOPTERA, SPHECINAE
AND LARRINAE): A DESCRIPTIVE STUDY

By A. L. Steiner

Department of Zoology, University of Alberta
Edmonton, Alberta, Canada, T6G 2E9

Introduction

Predator and anti-predator adaptations, strategies, have been
studied extensively in recent years (see for instance Curio 1976 and
Edmunds 1974 for some recent reviews). Problems of predator-prey
coevolution, mimicry, protective coloration (e.g., Cott’s monu-
mental work, 1940), optimal strategies, etc., have received a great
deal of attention. Defense mechanisms are extremely diverse and
can even involve use of a commensal species (e.g. Ross 1971). A
variety of sensory channels can be used such as visual (e.g. Cott
1940; Robinson 1969), acoustical (e.g. Roeder 1965), chemical (e.g.
Eisner and Meinwald 1966; Eisner 1970), mechanical, vibratory (e.g.
Tautz and Markl 1975) to mention only a few examples. Predators
such as mammals, birds, reptiles (e.g. Curio 1970), fish, mollusks
have been extensively studied.

Among insects, solitary and social wasps have also been inten-
sively studied but on the whole surprisingly little is known about the
defensive mechanisms of their “helpless” prey. Prey capture is often
very difficult to observe and even more so to study extensively in
natural conditions. The few exceptions mostly deal with prey that
represent a potentially formidable opponent (e.g. spider, praying

*For part I see Steiner 1968 in the Literature Cited.
Manuscript received by the editor May 11, 1981

Psyche

[Vol. 88

mantis, etc.). Counter-attacks by such prey and occasional killing of
the predator have even been reported (e.g., Deleurance 1941, pp.
287-288, for a praying mantis attacked by the sphecid wasp Stizus
distinguendus; also 1945, p. 29 for Tachysphex costai Dest.). Dead
spider wasps have also been found in spider webs in natural
conditions (pers. obs.). Non-predaceous prey can also exhibit
defense reactions, however, as shown before for crickets attacked by
Liris nigra wasps (Steiner 1968).

The anti-predator system of acridid grasshoppers is now de-
scribed, analyzed, as observed both in nature and captivity (sum-
marized in Steiner 1976). The prey are: (1) mainly adult or subadult
Oedipodinae, but also a few Cyrtacanthacridinae, all attacked by
the sphecid wasp Prionyx parkeri Bohart and Menke, (2) to a much
lesser extent smaller, earlier, instars preyed upon by Tachysphex
wasps (details in next section). For the latter prey, defense reactions
were essentially the same, except for the ones involving the wings,
undeveloped at these stages. Prey hunting and stinging by Prionyx
parkeri are described in detail in Steiner 1981 (in press).

Materials and Methods
Field observations

Prionyx parkeri wasps were observed mainly in the grassland
desert and adjacent riparian habitat of S.E. Arizona, U.S.A., at the
foot of the Chiricahua Mountains, East of Willcox, during the
summer of 1972.

Observations in captivity

Individually marked Prionyx parkeri and Tachysphex [mostly
tarsatus (Say)] wasps were observed in controlled laboratory units
about 60 X 50 X 50 cm (general method described in Steiner 1965):
(1) at the Southwestern Research Station, Portal, Arizona, during
the sprng and part of -the summer 1973 (= Arizona study); (2) in
central Oregon, U.S.A., near Bend, using a field trailer, during the
summer of 1977 (= Oregon study). The following acridid grass-
hoppers taken from the wasps’ habitats were used in the Arizona
study; (1) for P. parkeri, adult or last instar nymphs of: Oedipodi-
nae, mostly Trimerotropis pallidipennis p. (Burm.), also Conozoa
carinata Rehn, a few Cibolacris parviceps (Walker) — Cyrtacan-
thacridinae, a few Psoloessa delicatula Scudder and an occasional

1981]

Steiner — Anti-predator Strategies

Eritettix variabilis Bruner; (2) for Tachysphex wasps, small acridid
nymphs of: Oedipodinae, mostly Conozoa carinata Rehn and also
a few Trimerotropis pallidipennis p. (Burm.); Cyrtacanthacridinae,
a few Psoloessa delicatula and an occasional Melanoplus sp.,
Derotmema sp., Rehnita sp. Rather similar but un-determined
grasshoppers were used in the Oregon study, in captivity. The grass-
hoppers were provided either ad libitum, or in staged encounters.

Observations were mostly continuous, with “all occurrences”
sampling of wasp-prey interactions. Precise quantifications were
difficult or impossible because initial stages of encounters were
often sudden and unpredictable. Generally speaking proof of effects
of escape-defense reactions is often very difficult to establish (e.g.
Edmunds 1974, p. 240). This study is basically descriptive.

Total observation times were; (1) for captive P. parkeri in the
Arizona study about 178 h over a period of 30 observation days
(X = about 6 h-day) and in the Oregon study about 142 h for 14
observation days (X= about 6V2 h-day); (2) for captive Tachysphex
wasps in the Arizona study about 224 h for 37 observation days (X =
about 6h-day) and in the Oregon study about 224^ h for 35
observation-day (X = 6!4 h-day).

Results: Description of Responses, Conditions

Common responses: escape by jumping (flying) away,
staying put = first line of defense.

a) Field observations

Visually hunting Prionyx (parkeri?) wasps were observed in the
short and sparse grassy vegetation, characteristic of the upper
Sonoran desert grassland. Acridid grasshoppers were abundant,
particularly Oedipodinae such as Mestobregma plattei rubripenne
(Bruner) adults, also found stored in the nests of these wasps. The
most common response to wasps approaching or pouncing was a
very sudden, even startling, escape by jumping (Fig. 5a) and flying
away (Fig. 5b). The bright flash of the colorful banded wings came
in sharp contrast with the sudden disappearance from sight, after
landing (crypticity: Fig. 5c). The wasps seldom followed the escap-
ing grasshoppers in flight, but occasionally did so (Fig. 5b) and even
managed to cling to them in mid air and to deliver stings before
landing. Most stung grasshoppers were apparently caught by sur-
prise or at the preparatory stages of escape. Close range and

4 Psyche [Vol. 88

Fig. 1: Attack of an adult acridid grasshopper (Oedipodinae) by a Prionyx parkeri
wasp. The wasp uses both the strong mandibles and long, powerful legs, to firmly
hold the prey and prevent escape. The grasshopper tries (in vain) to push away the
wasp with both powerful hind legs by applying strong pressure on the points where
the wasp is anchored (head and one fore leg). Several drops of regurgitated repelling
fluid are indicated by arrows. The wasp already assumes the appropriate posture for
the first paralyzing sting, delivered in the throat of the victim.

quantitative observations were almost impossible. At times the
grasshoppers stayed put instead of escaping, for no apparent reason.
Attack of the wasp does not necessarily follow detection of a
suitable prey, however, since hunting wasps go through periods of
temporary refractoriness (Steiner 1962, 1976, 1978, 1979). This
considerably complicates the study of possible effects of prey-
defenses on the wasps.

b) Observations in captivity

The same responses were also recorded in captivity. Flying away

1981]

Steiner — Anti-predator Strategies

Fig. 2: Regurgitation of a repelling fluid. An acridid grasshopper (Oedipodinae),
just paralyzed by a Prionyx parkeri wasp, lies on its back and a huge drop of fluid
covers a large surface of the ventral thoracic area where all four stinging sites are
located (indicated by white dots and arrows). Wasps often hesitate to dip their
abdomen tip into this viscous, probably offensive, fluid. Accidental contact triggers
vigorous body rubbing in an attempt to eliminate the unpleasant fluid from the body
surface.

and long-range escape were impossible, however, because of space
limitations.

There was no evidence of active avoidance of Prionyx or Tachy-
sphex wasps by grasshoppers (“predator recognition”), even after
repeated attacks. Escape was always in direct response to attack,
imminent attack, or at least sudden movements such as a wasp
running and/or pouncing. Thus predator and prey were often seen
basking together. Immediately following an attack, the escape
threshold was clearly lowered, however.

Mechanical defenses after contact: kicking, pushing and/or

brushing away the wasp: biting; wing fluttering and flying
= second line of defense (Fig. 1)

After contact, Prionyx wasps attempt to anchor themselves to the
struggling or escaping grasshopper. They try to gain a firm grip

Psyche

[Vol. 88

using their powerful spinose legs, terminal claws, and also mandi-
bles. These wasps tightly “embrace” the grasshopper, in an anti-
parallel posture and strongly cling to them (Fig. 1). In contrast,
many larrine wasps (e.g. Liris, Tachysphex ) are comparatively frail,
short-legged, and cannot physically overpower their prey as success-
fully as Prionyx wasps do. Their prey often struggles free, in
contrast to Prionyx prey which seldom succeed, after the “embrac-
ing” stage, in spite of frantic efforts to kick and / or brush, push away
the attacker with the powerful hind legs. Prionyx prey also try to
deny free access of the wasp to the dorsal side by raising their long,
folded, hind legs, often beyond the vertical, headwards (hind leg
raising: Fig. 5e). Powerful kicks (Fig. 5e) sometimes send the wasp a
few cm from the grasshopper, but this works mostly before the wasp
can secure a firm grip. Pushing action with the tarsi of the powerful
hind legs can also be recorded. They are very precisely directed at
the points seized by the wasp as shown in Fig. 1 . In the latter, drawn
from a photograph, the grasshopper tries, with its right hind leg, to
push away the left front leg of the wasp while it attempts, with the
left hind leg, to exercise strong pressure on the head, jaws, of the
attacker and presumably get the wasp to release its mandibular grip
(in Fig. 5f these “points of pressure” have been circled). Wing
fluttering and even flying attempts can also be observed in reponse
to the grasping action of the wasp. The orthopteran also performs
snapping motions with the jaws but is seldom able to bite the wasp.
The very globulous abdomen of Prionyx wasps appears to be
especially well adapted to prevent such biting. The abdomen is
particularly exposed since the wasp delivers the first sting in the
throat of the prey, dangerously close to the powerful jaws (Fig. 5g).

Chemical defenses: regurgitated fluid (Fig. 2)

In addition and often as a last ditch defense the grasshopper
regurgitates through the mouth a large drop of dark fluid (“tobacco
juice”) that usually spreads rapidly over the body areas closest to the
mouth, ventrally, namely the thoracic surface (Fig. 2). This surface
sometimes becomes completely covered with the substance. From
there it can spread to other body areas, if struggling is intense
enough. On Fig. 1 one drop can be seen on the right antenna of the
grasshopper and one on the tibia of the right hind leg (arrows).

1981]

Steiner — Anti-predator Strategies

Fig. 3: Postural defense replacing escape (startle and/or death feigning display?).
The attacked grasshopper froze into a hunched posture, with appendages tucked in,
thus protecting the vulnerable ventral surface. The colorful wings, showing striking
semi-circular dark markings, are fully extended and/or flutter convulsively. The
wasp, after many vain efforts, managed to slip under the grasshopper (one leg is still
visible on the right of the grasshopper head) and will attempt to reach the vulnerable
ventral surface of the thorax made less accessible by the posture and interposition of
appendages (obstruction behavior).

Uncommon and odd postural defenses replacing escape:
stationary wing flashing or extension; body arching; freezing
(Fig. 3) = first line of defense.

a) Field observations

These rare occurrences guarantee that such responses are not
reducible to captivity artifacts.

The first observation was made on Sept. 4, 1972, near the end of
the morning, in the Arizona grassland desert. One hunting Prionux
(parkeri?) suddenly pounced on a motionless grasshopper. Instead
of trying to escape, as usual, the latter was seen with the colorful
wings open, fluttering convulsively, with a startling suddenness,
thus producing a striking color flash. The hind legs were rigidly
extended behind like in the flying posture (Fig. 5b). However the

8 Psyche [Voi. 88

Fig. 4: A Prionyx parkeri wasp succeeded in overturning a “frozen” oedipodine
grasshopper. This makes the ventral surface of the thorax more accessible to the
stings of the wasp. One small drop of repellent fluid can be seen on the abdomen of
the wasp. After stinging is over, the wasp will vigorously rub its abdomen on the
substrate, in an effort to eliminate this unpleasant, perhaps noxious, fluid. Note (also
in Fig. 2) the dot of Testor paint on the dorsal surface of the wasp thorax, for
individual identification.

whole body was strongly arched downward as in Fig. 3. For the
observer, it looked as if the “frozen” grasshopper was disabled or
dying. The wasp left the grasshopper alone and pursued her hunting
trip. Under the impression that the prey had received a sting or two,
I picked it up only to see it instantly recover without the slightest
trace of paralysis. Obviously the grasshopper, later identified as an
adult Mestobregma plattei rubripenne (Bruner), had not been stung
and was not disabled at all. This species is an acceptable prey since it
was also found in two nests dug up the same day, nearby. In
another, 'Similar, instance the upper wings (tegmina) opened only
slightly, just enough to uncover the triangular base of the vivid red
wings that remained folded. Again the wasp failed to paralyze the
frozen grasshopper which later escaped just as suddenly as the first
one, unharmed. The latter case might be a less intense version of the
first case. Presumably all gradations could be observed.

1981]

Steiner — Anti-predator Strategies

The eliciting stimuli of such reactions could not be determined,
because of the suddenness and unpredictability of such encounters.
Sight of the rapidly approaching predator and/or mechanical
contact are likely candidates.

b) Observations in captivity (Figs. 3 and 4)

Similar or identical responses were also observed in captivity at
close range and in better conditions. Confinement seemed to even
somehow favor appearance of this behavior perhaps because of
restricted escape and/or greater concentration of attacks. Often the
extended wings and whole body were also strongly curved down-
wards, sometimes even tightly pressed against the substrate (Fig. 3).
The appendages and head were tucked in and more or less invisible
under the protective “umbrella” of the wings. The sudden flash of
the colorful wings and dark semi-circular markings, followed by the
appearance of convulsive movement and finally the illusion of a
disabled or dying grasshopper were, indeed, an arresting sight, at
least for a human observer.

Curiously such frozen grasshoppers mostly failed to suddenly
“resuscitate” and escape after it had become evident that their
postural defense had failed to stop the wasp attack. Such misfiring
might be a cost of this strategy because of the strong inhibitory
influences apparently involved. Sometimes wing fluttering resumed
as the wasp attempted to deliver the paralyzing stings. If left alone
by the wasp the grasshoppers would however invariably recover
without any sign of discomfort, like in the wild.

Such displays were never observed with Tachysphex wasps,
perhaps because the much smaller grasshopper nymphs they attack
have undeveloped wings . . . that cannot be used.

If the Prionyx wasps succeed in overcoming all these various
defense mechanisms or hurdles, as they often do, they then attempt
to deliver an average four successive stings, always on the same
stinging sites and in a predictable order (summarized in Steiner
1976; details in Steiner 1981). The paralyzed grasshopper can then
be safely and freely manipulated and stored in the nest, without any
resistance, obstruction.

Analysis, Discussion, Comparisons

Discussion is concerned mainly with possible or plausible inter-
pretations and evolutionary significance of these various defense

Psyche

[Vol. 88

reactions, their degree of predator-specificity. Comparisons are
made with other orthopterans, with similar and different anti-
predator strategies. Effectiveness, always difficult to prove, particu-
larly when attacks or lack thereof depend on the internal state of the
predator like in the present case, will be assessed rather than
analyzed mathematically.

All defenses described before (except crypticity) are secondary
rather than primary defenses since they are exhibited during
encounters (Edmunds 1974, pp. 1, 136). Defenses are often anti-
location, anti-capture or anti-consumption devices (i.e. Alcock
1975, p. 333). Furthermore, many species have several lines of
defense (integrated defense systems: Edmunds 1974, p. 243). Thus
the mantid Polyspilota aeruginosa may run, fly, give a startle
display, slash at the attacker. It can also feign death if persistently
handled in a rough way. It soon recovers, however. The brightly
colored abdomen might also represent flash behavior (in Edmunds
1974, p. 245). Each aspect of the defense system will now be
discussed separately.

Escape by jumping, flying away

This is a classical and common case of sudden startling (flash or
deimatic behavior Fig. 5b) followed by sudden disappearance into
crypsis (landing; Fig. 5c) (Edmunds 1974, pp. 146-148) by using
protective colors (e.g. Isely 1938). This is usually a very efficient
mechanism but Prionyx wasps occasionally dash at flying grass-
hoppers (Fig. 5b), even sting them in mid air, or take them by
surprise before they can escape. Pygmy mole crickets that escape by
flying away are also grasped and/or stung during flight by the
sphecid wasp Tachytes mergus (Yoshimoto, in Krombein and
Kurczewski 1963, p. 147) and also by Tachytes minutus (Kurczewski
1966). This defense is not especially aimed at digger wasp predators.

Detection of the predator is probably visual but could also be
based on hairs sensitive to airborne vibrations, as in some caterpil-
lars such as Barathra brassicae (Tautz and Markl 1978).

Use of hind legs other than for jumping: kicking or obstructive

behavior such as hind leg raising or interpositions, brushing
away, pushing away

Hind leg autotomy used by crickets (Steiner 1968) was never
observed in grasshoppers in the present study but Prionyx wasps

1981]

Steiner — Anti-predator Strategies

Fig. 5: Summary of oedipodine grasshopper anti-predator actions and Prionyx-
prey interactions, a: the wasp detected a grasshopper which is at the preparatory
stage of jumping (J); b: the prey flies away (F), suddenly opening very colorful wings
with conspicuous semi-circular dark markings (startle display) and in some cases the
wasp follows the grasshopper in flight and even stings it in midair; c: the escaping
prey suddenly lands and blends with the substrate (crypsis); d: instead of escaping the
grasshopper sometimes “freezes” into an odd posture somewhat remindful of an
inhibited flying action; the posture and convulsive wing fluttering give the impression
that the orthopteran is disabled, dying (disablement display? thanatosis?); at the same
time the posture and hunching appear to emphasize the semi-circular dark markings
on the wings (eyespot intimidation display, “bluff’?); furthermore in this posture,
access of the vulnerable ventral side of the thorax, where stings are delivered, is
reduced or impossible for the wasp (obstruction behavior); e: hind leg raising ( HLR)
is another obstructive behavior that makes initial posturing of the wasp difficult or
impossible; kicking can also send the wasp a few cm away; f: hind legs are also used
for brushing (B) and/or pushing away (P) the wasp; pressure is applied on the circled
areas so as to try to force the wasp to release her mandibular and leg grip; g: as a last
ditch defense the grasshopper can release a repellent fluid through the mouth, which
rapidly spreads over the ventral thoracic surface where all stinging sites are located;
the wasp often hesitates to dip into this pool her abdomen tip (circled); the latter is
also exposed to powerful bites from the grasshopper; therefore the throat of the prey
must be quickly stung to stop these mouth-based defenses. Solid and dashed arrows
indicate prey and wasp movements, respectively; open arrows show possible
sequences of events but these sequences can also be broken if the defenses are
effective and the wasp gives up.

Psyche

[Vol. 88

usually seize the wing base(s) or abdomen rather than one hind leg
(Fig. 1). Hind legs of crickets, grasshoppers, sometimes phasmids,
often covered with strong spines, are one of their major systems of
escape and/or defense. Some wingless phasmids can jab the spines
into an aggressor (Robinson 1968b). Overt defense by kicking has
also been described in some large aphids (in Edmunds 1974, p. 245)
and in a number of orthopterans such as crickets (Steiner 1968) and
Locusta migratoria for instance (Parker et al. 1974). In the latter
case it can be so violent that the attacker is knocked 20-30 cm away.
According to Parker et al. (1974) hind leg raising often precedes
kicking (threat?). It is also part of the defense postures of male L.
migratoria, the giant weta ( Deinacrida ) of New Zealand (in Sebeok
1977, Fig. 5a, p. 342) and mormon crickets when attacked by the
digger wasp Pa/modes laeviventris (Parker and Mabee 1928, p. 9).
In the latter case, as in Prionyx and Tachysphex, the wasps
succeeded in stinging only with considerable difficulty. In the
present study hind legs were often raised past the vertical line (Fig.
5e) and even as far forward as the level of the head, as in Fig. 1 for
instance, in addition to tail or body raising. This was also observed
once in response to an approaching Tachysphex tarsatus. Freezing
into such postures made access to the dorsal area and wasp
posturing very difficult, sometimes impossible (Fig. 5e) (obstructive
behavior) and the efficiency of this behavior appeared even to
increase as a result of repeated attacks. Interposition of legs
(obstruction behavior) was also observed in mole crickets attacked
by Larra wasps (Williams 1928).

Brushing and pushing away (Figs. 1 and 5f) are more difficult to
evaluate since they are more graded and variable responses which
are not easy to detect, let alone quantify, in the confusion of the
attack. Plausibly these responses work best (if at all) at early stages
of contact with the wasp, also if the prey is very large and vigorous
or if the wasp is more likely to easily give up, for instance at early
stages of hunting (Steiner 1976). It is doubtful that a firmly
anchored wasp can easily be dislodged in this way.

[Remark: some orthopterans extend or raise their fore legs,
vertically, as part of a threat-intimidation posture (e.g., Neobarettia:
Cohn, in Sebeok 1977, p. 342, Fig. 5b)].

Orthopteran hind legs are often given special attention and are
paralyzed first by some predatory waps such as Liris and Tachy-
sphex (Steiner 1962, 1976). Prionyx wasps can give priority to the

1981]

Steiner — Anti-predator Strategies

mouth-based defenses (biting, regurgitating), now discussed, since
they effectively neutralize hind leg defenses with their powerful
“embracing” legs. Correspondingly, these wasps deliver the first
sting in the throat, not around the hind legs (Steiner 1976).

Biting and retaliation ( aggressive defense:

Edmunds 1974, p. 182)

Orthopterans commonly use their powerful jaws for threat,
intimidation or even active defense, retaliation, if not for predation.
The predaceous North American katydid Neobarettia severely bites
and displays the open mandibles as part of the threat-intimidation
display (Cohn, in Sebeok 1977, p. 342, Fig. 5b).

In one observation in captivity (Arizona, June 24 1973, 1335 h) a
wrongly positioned Tacky sphex tarsatus (No + 1042) was clearly
and severely bitten by a nymph Trimerotropis pal/idipennis p.
(Burm.) (No + 1098) during a stinging attempt. This suggests that
the wasp is particularly vulnerable before proper positioning is
achieved and that strong selection pressures in the direction of
minimum risk must have shaped the usual stinging postures. The
penalty for wrong posturing can be very heavy. Thus the above
wasp was found dying in the cage the next day, June 25, most likely
as a result of this violent retaliation of the prey.

Importance of mouth-based defenses is confirmed by the fact that
many orthoptera-hunters deliver a special throat sting (Steiner 1962,
1976) sometimes even before any other sting (e.g. Prionyx parkeri).
This also eliminates opposition to prey-transport and storage in the
nest (and furthermore “de-activates” the prey that recovers in part
from paralysis, later: Steiner 1963a). In sharp contrast, Oxybelus
uniglumis wasps omit the throat sting when they paralyze their non-
recovering fly-prey devoid of subesophageal ganglion and of poten-
tially dangerous mouth parts (Steiner 1978, 1979). Orthoptera-
hunting wasps with missing legparts or damaged antennae are often
found, particularly late in the season. This might be a testimony to
the efficiency of bites of their prey but also result from intra-specific
fighting (see for instance Brockmann and Dawkins 1979, for Sphex
ichneumoneus ) and/or accidents during nesting. A female Pal-
modes carbo with two deep dents on the back of her abdomen was
found in southern British Columbia. It is probable that this
represented severe bites received from one of their large, often

Psyche

[Vol. 88

predaceous, decticine grasshopper-prey rather than beak marks of
some bird.

Chemical defenses: regurgitated fluid (R)

Chemical defenses are particularly widespread among insects (see
for instance Eisner and Meinwald 1966; Wallace and Blum 1971,
etc.) including Opthopterans. Some of them have specialized glands
and the substance can be ejected with considerable force (e.g.
Poekilocerus buforus, from an opening located on the first abdom-
inal tergite: Fishelson 1960). A froth can also be discharged through
a thoracic spiracle (e.g. Romalea microptera: in Eisner and Mein-
wald 1966). Such repellents make their owner distasteful or un-
palatable. The same apparently holds for fluids regurgitated from
the gut through the mouth (Edmunds 1974, p. 199) by grasshoppers
for instance = enteric discharges (Matthews and Matthews 1978, p.
335). Digger wasps, however, do not consume their prey usually but
avoid contact with this fluid which is apparently a contact repellent.
Functioning of the receptors located around the stinger could be
impaired (jamming effect?) chemically and/or mechanically (Steiner
1976). Stinging remains possible, however, even with stinging sites
covered with the fluid (Figs. 2 and 5b) but the wasp clearly hesitates
or even gives up half way through stinging. Contact triggers
vigorous, sometimes frantic, rubbing against the ground and/or
hyper-grooming as in ants (Matthews and Matthews 1978, p. 335) as
in hunters of regurgitating caterpillars like cutworms (e.g., Am-
mophila, Podalonia wasps). Body contact is clearly unpleasant if
not deleterious, particularly for some small Tachysphex wasps
(Steiner 1976).

One of the latter ( tarsatus No + 874) had her abdomen tip covered
with a thick coat of sand particles as a result of her attempts to rub
off the sticky substance. The wasp was found dying the next day,
June 19 (Arizona study) (the same probably happened to another
tarsatus (No + 887) which died on June 6).

The same wasp (No + 874) was also observed the day before (June
18, 1405 h) in the process of carefully removing with the mandibles,
bit by bit, a large crust of dried up fluid, from the ventral surface of
the thorax and throat of a grasshopper. This was done right after
“malaxation” of the fore leg bases which in some larrine wasps is a
preparatory stage of egg-laying (details in Steiner 1971). Since the

1981]

Steiner — Anti-predator Strategies

egg is invariably glued right behind the fore legs, where the crust was
also located, this would indicate that the regurgitated fluid could
also be a serious obstacle to egg-laying or egg development. Prionyx
wasps lay their egg at the base of one hind leg . . . where the risk of
such “flooding” is clearly much reduced or even nil! Furthermore,
paralyzed grasshoppers cannot remove the spilled fluid by groom-
ing, as they normally do. Consequently “cleaning” of the soiled prey
can be done only by the wasps, if at all.

This chemical defense is apparently even more effective in mole
crickets against another larrine wasp: Larra (Williams 1928). Thus
Larra sanguinea wasps were found with their mouthparts com-
pletely glued together by the very viscous fluid. Remarkably, some
of these wasps managed to catch their mole cricket in spite of such
crippling handicap! Ants are repelled by fecal material or chryso-
melid beetle larvae (in Matthews and Matthews 1978, p. 343), and
refuse to carry away pieces of grasshopper treated with their own
repelling fluid (Eisner 1970).

In conclusion, the importance of mouth-based regurgitative
defenses can be assessed by (1) the care with which these wasps try
to eliminate the fluid from the prey and from their own body, (2)
evolution of a specialized sting in the throat that abolishes mouth-
based defenses, (3) the priority given by Prionyx wasps to mouth-
based defenses (first sting in the throat), (4) dramatic effects,
including death, observed on some wasps like small Tachysphex, (5)
toxic effects reported in the literature, for mammals, such as topical
irritation of eyes, vomiting when swallowed and severe symptoms
caused by injection (Matthews and Matthews 1978, p. 335).

Such defenses are therefore particularly efficient against smaller
predators like arthropods, wasps included. More experimentation is
clearly needed, however.

Postural defenses, displays, replacing escape
(Figs. 3, 4 and 5d)

Such complex postures and displays will be analyzed in terms of
their various components or aspects.

a) Color flash, startle response

Sudden display of colored wings, of hidden and bright structures
(deimatic behavior) is common in insects, particularly in otherwise
cryptically colored moths such as Catocala scripta, Triphaena

Psyche

[Vol. 88

pronuba (in Edmunds 1974) and also many orthopterans. For the
latter, wing opening (lifting) is for instance part of the dramatic
threat-intimidation display of Neobarettia already mentioned (in
Sebeok 1977, p. 342) or the one of Phymateus. Since these latter
species are potentially dangerous and/or distasteful such displays
are usually interpreted as warning (in Edmunds 1974, pp. 148, 154;
see also for instance Frazer and Rothschild 1962). The first species
bites severely while the latter has strong hind leg spines and secretes
a repelling fluid if further molested. When exhibited by harmless
species such as the stick insect Metriotes diocles (e.g. Bedford and
Chinnick 1966; Robinson 1968a) or common grasshoppers it is
considered as mere “bluff’ based on a startle effect and/or an
apparent increase in size, height, volume, etc. (intimidation beha-
vior). Similar actions are reported from some cicadas and mantids
and are particularly dramatic in the African mantid Idolium
diabolicum (in Wickler 1968).

b) Display of dark markings or “eyespots”

Eyespots are commonly displayed by moths (see for instance Blest
1957, 1964). If even very imperfect imitations are considered
effective then perhaps this also applies to the semi-circular dark
markings displayed by grasshoppers (Figs. 3 and 5d). Rarity of the
display is essential (in Edmunds 1974, p. 168).

c) Appearance of disabled, dying or dead insect (thanatosis) with
freezing, hunching and appendages tucked in (Fig. 5d).

Inhibition of movement in itself or freezing is likely to lower the
probability of detection and / or attack by predators that hunt
moving live prey visually (e.g., Steiner 1962, 1976 for cricket-
hunting Liris wasps). This probably includes many digger wasps.
Thanatosis is known from a number of insects, also orthopterans
(Edmunds 1974, p. 172; Robinson 1968a). The prey might also be
considered unsuitable because of the unusual appearance as such
(oddity effects). The latter is illustrated by “protean defenses” an
unpredictable, erratic and highly diverse behavior (in Edmunds
1974, pp. 144-145; see also Chance and Russel 1959; Humphries
and Driver 1971, etc.).

Furthermore, grasshoppers with wings spread, appendages
tucked in and body strongly arched (Fig. 3) also seem less exposed
because of reduced access to the vulnerable stinging sites, all located
on the well protected ventral surface of the thorax (Steiner 1981).

1981]

Steiner — Anti-predator Strategies

Some Prionyx wasps experienced great difficulties in squeezing
themselves under such grasshoppers (Fig. 3, one leg of the wasp
visible). Sometimes also the wasps succeeded in turning over such
grasshoppers, venter up (Fig. 4). Even so, stinging was difficult.

Reduced accessibility might be an accidental by-product of the
“disablement” display or a more direct result of wasp-grasshopper
coevolution. The apparent immunity of Acrotylus grasshopper
nymphs to Tachysphex peetinipes was also attributed to restricted
accessibility linked with dense and long pilosity (Ferton 1910, p.
158). Body arching has also been observed on some other orthop-
terans and is sometimes associated with the release or violent
expulsion of repellent fluid, as in Poekilocerus buforus (Fishelson
1960).

d) “Intimidating” and aggressive defensive elements (Fig. 5d).

If the posture shown in Figs. 3 and 5d is also an eyespot display
then it has an intimidating as well as “bluff’ value.

Sideways rocking, known from some mantids (Crane 1952) and
also forward-backward rocking were often observed in crickets, just
before or after contact with Liris wasps (Steiner 1968), suggesting an
intimidating function. This was also observed in Empusa egena in
response to attacks by the sphecid wasp Stizus distinguendus Handl.
(Deleurance 1941, pp. 287-288), along with other aggressive re-
sponses such as wings open, striking with the raptorial fore legs.
Rocking was also observed in some phasmids (Crane 1952) and
roaches such as Periplaneta fuliginosa (Simon and Barth 1977, p.
307). Crickets also sometimes froze into odd or intimidating erect
postures difficult to interpret as “death feigning” (Steiner 1962,
1968). Absence of stinging in such cases, if related at all to the
display, might depend on: (1) the oddity of the posture, as Chauvin
and Chauvin (1977) suggest (the vertical posture is in sharp contrast
with the usual horizontal one), or (2) the possible intimidating
effects associated with increased height (bluff behavior), (3) preda-
tor mimicry, namely a mantis-like appearance (see Steiner 1968,
Fig. i, p. 267). [Remark: this latter possibility was considered far-
fetched by one reviewer of the paper cited and consequently
eliminated from the text. . .and yet Simon and Barth (1977, p. 307,
Fig. 2) describe a somewhat comparable rare posture from the roach
Periplaneta fuliginosa which they interpreted (probably rightly) as a
“Mantis-threat”!]

Psyche

[Vol. 88

Fig. 6: Proportions of cases where negative effects on the wasp were present
(hatched bars = wasp inhibited or stopped, stinging incomplete or no stinging at all)
or not present (bars in solid black = complete stinging without apparent negative
effects). For cases where prey defenses were recorded (left pair of bars) the
proportion of negative effects is greater than no effects, whereas it is the reverse for
cases where no defenses were recorded or this information was unavailable (right pair
of bars). This indicates that prey defenses (all cases pooled) do have some negative
effects on the wasps. It is only a trend, however, since the differences do not reach
significance.

It has been suggested that some protective and intimidating
displays (e.g. in saturniid and sphingid moths) could have evolved
from flight movements (Blest 1957) and can be classified as (1)
rhythmic, (2) static, (3) mixed and (4) cryptic. Category (1), that
appears to best fit the data (Figs. 3 and 5d) would be closest to the
original flight movements. Extension of hind legs, wing beats, even
if convulsive, are clearly part of flying which is strongly inhibited.

1981]

Steiner — Anti-predator Strategies

Similar explanations would seem to apply to the odd cricket
postures (Steiner 1968) but in the form of “frozen jumping and/ or
kicking” rather than “frozen flight” and reduced access to the
vulnerable ventral stinging sites is also indicated. Startle displays
have also been interpreted in terms of conflict between flying and
freezing for some mantids (Crane 1952).

Efficiency of such defenses has been clearly demonstrated in only
a few cases. Parker et al. (1974), for instance, showed that defense
postures exhibited by Locusta migratoria had a significant negative
effect on bout continuance between conspecifics. With wasp studies
the problem is further complicated by wide moment-to-moment
fluctuations in responsiveness of the hunting wasps (Steiner 1962,
1976, 1979). Such variables must be controlled, manipulated or
eliminated to get clear answers and this was not done in the present
study.

Quantitative Data

Quantifications were too limited and inappropriate to make a
statistical analysis of the effectiveness of such defenses very mean-
ingful. Only 128 cases were known in sufficient detail to be included
in the analysis. In 41 .27% (n = 26) of the cases the defenses (lumped
together) had no apparent effect and complete stinging followed and
in 58.73% (n = 37) at least some possible effects were recorded, such
as temporary, permanent, interruption or even deletion of stinging.
When no defenses were observed (or unknown status) the percent-
ages of complete vs incomplete stinging were approximately re-
versed as predicted: 55.38% (n = 36) and 44.62% (n = 29). These
differences in proportions (Fig. 6) were not significant, however,
since the calculated x2 was only 3.689 for a critical value of 5.991
(p ^ 0.05; df = 2; G-test of independence of rows and columns:
Sokal and Rohlf 1969, p. 599). A slight advantage can have a
decisive selective value in the long run, however.

Conclusion

Prey as harmless as herbivorous crickets and grasshoppers pos-
sess a rather complex, well integrated, system of anti-predator
devices they can use against their wasp enemies. Even if some of
these responses are merely obstructive, they do in fact increase the

Psyche

[Vol. 88

cost of predation to the wasps by making capture more difficult,
more costly, and/or less probable. Natural selection should there-
fore promote evolution of such anti-predator strategies which in the
long run increase the fitness of the prey.

Some components of the system such as flying away and cryptic-
ity, perhaps regurgitation, are of a very generalized nature whereas
other devices are more predator-specific. Thus startle displays with
exposure of dark semi-circular markings are probably most efficient
against small avian predators, whereas biting, mouth regurgitation,
hind leg raising and obstruction behaviors are presumably more
useful against smaller, more vulnerable predators such as other
insects, including digger wasps. Matthews and Matthews (1978, p.
352) state that “protective adaptations in insects are intimately
related to the behavior and physiology of their predators.” This also
applies well to wasp predators.

Acknowledgements

The Arizona study was part of a sabbatical project, while on an
exchange program with the American Museum of Natural History,
New York, in 1972-73. Research was conducted at the Southwest-
ern Research Station, Portal, Arizona, and in the surrounding
areas, including the Chiricahua National Monument and the Erick-
son Ranch. Help, advice and hospitality of many persons and
friends, I cannot mention individually, are gratefully acknowledged.
Wasp specimens were kindly identified by A. S. Menke, U.S.
National Museum (Entomology), Washington; R. M. Bohart, Uni-
versity of California, Davis, and W. J. Pulawski, Wroclaw Univer-
sity, Poland, and grasshopper specimens by D. C. Rentz, the
Academy of Natural Sciences, Philadelphia, Pennsylvania. The
study was supported in part by an operating grant (A3499) from the
National Research Council of Canada and funds from the Univer-
sity of Alberta, Edmonton, Canada. I would like to thank J.
Scheinas for typing the manuscript.

Summary

Harmless herbivores such as acridid grasshoppers exhibit a
complex anti-predator behavior when attacked by Prionyx and
Tachysphex sphecid wasps. Besides jumping and flying away with

1981]

Steiner — Anti-predator Strategies

exposure of colorful wings (flash behavior) and sudden return to
crypticity upon landing, these insects show freezing, often in odd
postures, with the colorful wings and dark markings (“eyespots”?)
prominently exposed. Such postures also reduce access to the
vulnerable ventral surface usually stung by these wasps (obstruction
behavior). After contact with the wasp a second line of defense
comes into effect such as kicking, brushing and pushing actions. In
addition to these hind-leg based defenses, the attacked prey can also
use mouth-based defenses: biting and / or regurgitating a repelling,
perhaps even noxious, fluid (“tobacco juice”). Such defenses pre-
sumably lower the probability of capture or at least increase the cost
to the predator and have therefore a selective value.

Literature Cited

Alcock, J.

1975. Animal Behavior, an evolutionary approach. Sunderland, Mass.: Si-
nauer, 547 p.

Bedford, G. O. and L. J. Chinnick

1966. Conspicuous displays in two species of Australian stick insects. Anim.
Behav. 14: 518-21.

Blest, A. D.

1957. The evolution of protective displays in the Saturnioidea and Sphingidae
(Lepidoptera). Behaviour 11: 257 309.

1964. Protective display and sound production in some New World arctiid and
ctenuchid moths. Zoologica, New York 49: 161-81.

Brockmann, H. J. and R. Dawkins

1979. Joint nesting in a digger wasp, an evolutionary stable preadaptation to
social life. Behaviour 71: 203-45.

Chance, M. R. A. and W. M. S. Russell

1959. Protean displays: a form of allaesthetic behaviour. Proc. Zool. Soc.
Lond. 132: 65-70.

Chauvin, R. and B. Chauvin

1977. Le monde animal et ses comportements complexes. Plon, Paris.

Cott, H. B.

1940. Adaptive coloration in animals. London, Methuen.

Crane, J.

1952. A comparative study of the innate defensive behaviour in Trinidad
mantids (Orthoptera, Mantoidea). Zoologica, New York 37: 259-93.

Curio, E.

1970. Die Selektion dreier Raupenformen eines Schwarmers (Lepidopt.,
Sphingidae) durch einen Anolis (Rept. Iguanidae). Z. Tierpsychol. 27:
899 914.

1976. The ethology of predation. Berlin, Springer Verlag.

Psyche

[Vol. 88

Deleurance, E. P.

1941. Contributions a l’etude biologique de la Camargue - Observations
entomologiques. Bull. Mus. Hist. Natur. Marseille 1(4): 275-89.

1945. Sur Fethologie d’un Tachytes chasseur de Mantes Tachysphex costai
Dest. (Hym. Sphegidae). Bull. Mus. Hist. Natur. Marseille 1(1-2):
25-29.

Edmunds, M.

1974. Defence in Animals. A survey of anti-predator defences. New York,
Longman.

Eisner, T.

1970. Chemical defense against predation in arthropods. In Chemical Ecology,
E. Sondheimer and J. B. Simeone (eds.). New York, Academic Press,
(pp. 157-215).

Eisner, T. and J. Meinwald

1966. Defensive secretions of arthropods. Science 153: 1341-50.

Ferton, C.

1910. Notes detachees sur l’lnstinct des Hymenopteres melliferes et ravisseurs
(6e Serie). Ann. Soc. Entom. Fr. 79: 145-78.

Fishelson, L.

1960. The biology and behaviour of Poekilocerus bufonius Klug, with special
reference to the repellent gland (Orth. Acrididae). Eos, Madrid 36:
41-62.

Frazer, J. F. D. and M. Rothschild

1962. Defence mechanisms in warningly-coloured moths and other insects.
Internat. Congr. Entom. 11,3: 249-56.

Humphries, D. A. and P. M. Driver

1971. Protean defence by prey animals. Oecologia 5: 285-302.

Isely, F. B.

1938. Survival value of acridian protective coloration. Ecology 19: 370-89.

Krombein, K. V. and F. E. Kurczewski

1963. Biological notes on three Floridian wasps. Proc. Biol. Soc. Wash. 76:
139-52.

Kurczewski, F. E.

1966. Behavioral notes on two species of Tachytes that hunt pygmy mole-
crickets (Hym.: Sphecidae, Larrinae). J. Kans. Ent. Soc. 39: 147-55.

Matthews, R. W. and J. R. Matthews

1978. Insect behavior. New York, Wiley, 507 p.

Parker, G. A., G. R. G. Hayhurst and J. S. Bradley

1974. Attack and defence strategies in reproductive interactions of Locusta
migratoria, and their adaptive significance. Z. Tierpsychol. 34: 1-24.

Parker, J. R. and W. B. Mabee

1928. Montana insect pests for 1927 and 1928. Univ. Montana Agric. Exper.
Station, Bozeman, Mont., Bulletin 216, p. 9.

Robinson, M. H.

1968a. The defensive behaviour of the Javanese stick insect, Orxines macklotti
De Haan, with a note on the startle display of Metriotes diocles Westw.
(Phasmatodea, Phasmidae). Ent. month. Mag. 104: 46-54.

1968b. The defensive behavior of the stick insect Oncotophasma martini

1981]

Steiner — Anti-predator Strategies

(Griffini) (Orthoptera: Phasmatidae). Proc. Royal. Entom. Soc. Lond.
43: 183-7.

1969. Defenses against visually hunting predators. In: Evolutionary Biology 3,
T. Dobzhansky, M. K. Hecht and W. C. Steere (eds.). New York,
Meredith, (pp. 225-59).

Roeder, K. D.

1965. Moths and ultrasound. Sci. Amer. 212: 94 102.

Ross, D. M.

1971. Protection of hermit crabs (Dardanus spp.) from Octopus by commensal
sea anemones ( Calliactis spp.). Nature, London 230: 401-2.

Sebeok, T. A.

1977. How animals communicate. Bloomington, Indiana University Press.

Simon, D. and R. H. Barth

1977. Sexual behavior in the cockroach genera Periplaneta and Blatta. III.
Aggression and sexual behavior. Z. Tierpsychol. 44: 305-22.

SOKAL, R. R. AND F. J. ROHLF

1969. Biometry. San Francisco, Freeman, 776 p.

Steiner, A. L.

1962. Etude du comportement predateur d‘un Hymenoptere Sphegien: Liris
nigra V.d.L. (= Notogonia pompiliformis Panz.) Ann. Sci. Natur. (Zool.
Biol. Anim.) Ser. 12,4: 1-126.

1963. Interpretation neuro- et psycho-physiologique de l’etat des victimes de
certaines Guepes paralysantes ( Liris nigra V.d.L. = Notogonia pompili-
formis Panz.). (Contribution a letude des facteurs motivationnels et
activateurs chez l’lnsecte). C.R. Acad. Sci. Ser. D (Sci. Nat.) Paris 257:
3480-82.

1965. Mise au point d’une technique d’elevage d — Hymenopteres fouisseurs en
laboratoire (Note preliminaire). Bull. Soc. Entom. Fr. 70: 12-18.

1968. Behavioral interactions between Liris nigra V.d.L. (Hym. Sphecidae)
and Gryllus domesticus L. (Orthoptera: Gryllidae) Psyche 75: 256-73.

1971. Behavior of the hunting wasp Liris nigra V.d.L. (Hym., Larrinae) in
particular or in unusual situations. Can. J. Zool. 49: 1401-15.

1976. Digger wasp predatory behavior (Hym., Sphecidae). II. Comparative
study of closely related wasps (Larrinae: Liris nigra, Palearctic; L.
argentata and L. aequalis, Nearctic) that all paralyze crickets (Orthop-
tera, Gryllidae). Z. Tierpsychol. 42: 343-80.

1978. Evolution of prey-carrying mechanisms in digger wasps: possible role of
a functional link between prey-paralyzing and carrying studied in
Oxybelus uniglumis (Hym., Sphecidae, Crabroninae). Quaest. ent. 14:
393-409.

1979. Digger wasp predatory behavior (Hym., Sphecidae): fly hunting and
capture by Oxybelus uniglumis (Crabroninae: Oxybelini); a case of
extremely concentrated stinging pattern and prey nervous system. Can.
J. Zool. 57: 953-62.

1981. Digger wasp predatory behavior (Hym., Sphecidae). IV. Comparative
study of some distantly related Orthoptera- hunting wasps (Sphecinae
vs. Larrinae), with emphasis on Prionyx parkeri (Sphecini). Z. Tier-
psychol. (in press).

Psyche

[Vol. 88

Tautz, J. and J. Markl

1978. Caterpillars detect flying wasps by hairs sensitive to airborne vibration.
Behav. Ecol. Sociobiol. 4: 101-10.

Wallace, J. B. and M. S. Blum

1971. Reflex bleeding: a highly refined defence mechanism in Diabrotica
larvae (Coleoptera: Chrysomelidae). Ann. Ent. Soc. Amer. 64: 1021-4.
Wickler, W.

1968. Mimicry in plants and animals. London, Weidenfeld and Nicholson.
Williams, F. X.

1928. Studies in tropical wasps — their hosts and associates (with descriptions
of new species). Bull. Exper. Station Hawaii. Sugar Planter’s Assoc.
(Entom.) (19): 1-179.

A COMPARISON OF THE NEST PHENOLOGIES OF
THREE SPECIES OF POGONOMYRMEX HARVESTER
ANTS (HYMENOPTERA: FORMICIDAE)*

By William P. MacKay
Departamento de Entomologia
Colegio de Graduados
Escuela Superior de Agricultura
Ciudad Juarez, Chih. Mexico

Introduction

Ants are among the most abundant animals in most habitats
(Petal 1967) and may even be the dominant insects in many
ecosystems (Nielsen 1972; Nielsen and Jensen 1975). Harvester ants
of the genus Pogonomyrme x are a major component of the energy
flux through ecosystems (Golley and Gentry 1964). Ants of this
genus have become increasingly important in ecological studies,
including mutualism (O’Dowd and Hay 1980), competition (Mares
and Rosenzweig 1978; Reichman 1979; Davidson 1980), predation
(Whitford and Bryant 1979), foraging (Whitford and Ettershank
1975; Holldobler 1976a; Whitford 1976, 1978a; Davidson 1977a, b;
Taylor 1977), community structure (Davidson 1977a, b; Whitford
1978b), and impact on ecosystems (Clark and Comanor 1975;
Reichman 1979). It is difficult to investigate harvester ants as
seasonal processes occurring inside the nest are generally unknown
and the nest populations are usually underestimated.

This investigation compares the nest phenologies of three species
of Pogonomyrmex harvester ants: P. montanus MacKay, P.
subnitidus Emery, and P. rugosits* Emery, which occur at high, mid,
and low altitudes respectively. These data form the basis for a
comparison of the ecological energetics of the three species
(MacKay 1981).

Materials and Methods
The species investigated.

The altitudinal comparison is based on three species of harvester

*This research constitutes Chapter 3 of a dissertaion submitted to the faculty of the
University of California, Riverside, in partial fulfillment of the requirements for the
Degree of Ph.D. in Population Biology.

Manuscript received by the editor May 28, 1981.

Psyche

[Vol. 88

ants: Pogonomyrmex montanus MacKay, P. subnitidus Emery, and
P. rugosus Emery. All three belong to the subgenus Pogonomyrmex.
Pogonomyrmex subnitidus and P. montanus are very closely
related, both belong to the occidentalis complex (MacKay 1980).
Pogonomyrmex rugosus belongs to the barbatus complex (Cole
1968). Pogonomyrmex montanus is unusual for the genus in being a
high mountain species occurring in pine forests in the mountains of
southern California. Pogonomyrmex subnitidus is a mid-altitude
species in the San Jacinto Mountains. Pogonomyrmex subnitidus is
distributed throughout southern and south central California and
Baja California, occurring at lower elevations throughout much of
its range. Pogonomyrmex subnitidus is sympatric with P. rugosus in
parts of Riverside County, but is uncommon in such areas.
Pogonomyrmex rugosus is a low altitude species near Riverside and
occurs at lower elevations throughout much of southwestern United
States. It rarely occurs at higher elevations. For example, in the
Joshua Tree National Monument it is present up to 1350 meters, in
New Mexico it occurs at over 2100 meters.

Study areas.

Populations of all three species were studied in southern Cali-
fornia: P. montanus — in a yellow pine forest community between
Fawnskin and Big Pine Flat at 2100 meters elevation in the San
Bernardino Mountains of San Bernardino Co., P. subnitidus — in
the chaparral near the Vista Grande Ranger Station at 1500 meters
in the San Jacinto Mountains of Riverside Co., P. rugosus — in the
coastal sage scrub community at Box Springs at 300 meters near
Riverside, Riverside Co. The three species occur in clearings within
these different plant communities.

Estimation of nest populations.

Two primary methods are used in the estimation of ant nest
populations: mark-recapture methods and nest excavation. Mark-
recapture methods are used to compare a population before and
after seasonal production. This method has been criticized as one of
the assumptions is that workers mix randomly in the nest. The
workers of all three species are stratified within the nests and there is
strong evidence that other species are stratified as well (MacKay
1981). Also I could find no reliable way to mark the individuals such
that the marks were permanent, could not be passed on to other
individuals, and would not disrupt normal activities. In any case,

1981] Mac Kay— Nest Phenologies of Pogonomyrmex

such a method would only estimate the numbers of foragers in a
Pogonomyrmex nest, not the actual nest population. In addition,
mark-recapture methods do not provide an estimate of the repro-
ductives produced in a nest.

Excavation of nests destroys them for further study and requires a
large expenditure of time and effort. I chose periodic nest excava-
tion as the method of estimating production as counts of the
sexuals, brood, and workers can be made.

Our experience indicates that most of the nest population is
collected. Pogonomyrmex spp. colonies may live 15 to 20 years
(Barnes and Nearney 1953), and will live at least two years after the
removal of the queen (pers. obs.). Nest longevity is unknown in the
three species investigated, but based on data from other species, I
expect at least 5%rl0% of the nests should not have queens. The
high proportion of nest queens collected (84% in P. montanus, 77%
in P. subnitidus, and 80% in P. rugosus ) supports the hypothesis
that most of the nest population is collected. The queens do not
reside in any special “queen chamber” and are of a similar size as a
worker. Therefore, it is not any easier to find the queen than it is to
find any individual worker in the nest. In all cases excavation was
continued at least 50 cm deeper than the position of the last ant
found or the end of a burrow.

Nest excavation procedure.

The procedure was as follows: The surface dimensions of the nest
were determined by removal of the top 10 cm of the nest. The hole
was then extended at least 50 cm on all sides. A square ditch was
dug around the perimeter of the nest to a depth of one meter in the
case of P. montanus nests and over 1.5 meters around the nests of P.
rugosus and P. subnitidus. We were able to sit in the ditches while
carefully excavating the nests in 10 cm levels. As the hole became
deeper, the ditches were proportionally deepened. All of the
contents of the burrows, including ants, brood, guests, stored seeds,
and dirt were placed in labeled half or one liter plastic containers.
Later the animals were separated from the dirt, and counted. Nest
excavation usually began between 06:00 and 07:00, before the ants
became active. If foragers were needed for other investigations,
excavation began later in the morning or early in the afternoon.
Excavation and counting of a P. montanus nest requires 6-10 hours,
of a P. subnitidus nest 20-30 hours and of a P. rugosus nest 60-90

Psyche

[Vol. 88

hours. Whenever excavation was stopped to be continued on the
following day, the nest was covered with a heavy vinyl cloth and 10
cm deep layer of dirt. This was necessary to keep the inhabitants,
especially the males, in the nest. A total of 80 P. montanus, 26 P.
subnitidus, and 20 P. rugosus nests were completely excavated
between 1977 and 1980.

It appeared that the excavation procedure disrupted stratification
of individuals within the nest only slightly. When nest chambers
were exposed, many individuals emerged, but most of the popula-
tion remained in the chambers, and assumed a defensive position
involving opening of the mandibles and forward extension of the
antennae.

The numbers of workers at each level and the position of the
queen were recorded. When the nests were in production, the
presence or absence of eggs was noted, but the eggs were not
counted, as they were extremely small and are easily lost in the dirt.
The larvae, pupae, females, males, and callows (immature, under-
pigmented workers) were counted when they were present in the
nests. The contents of each level were summed to obtain an estimate
of the entire nest population.

Seed storage in nests.

The seeds were separated from the soil by filling a 1000 ml beaker
about full of soil and seeds. The contents were washed into a sieve
with 0.5 mm mesh. The washing and swirling were continued until
all of the seeds were removed from the soil. The material caught in
the sieve was washed again until only seeds remained in the sieve.
The seeds were then dried (60° C) to constant weight.

Nest structure.

In the process of nest excavation it was noted that the general
form and shape of the nests were comparable in all three species.
The P. montanus nest structure was studied by pouring a thin
solution of plaster of Paris (3 tablespoons /liter of water) into one
nest. The solution was dilute enough that the walls of most of the
tunnel system were coated with plaster. The nest was excavated in
1-2 cm layers and the tunnel structure at each layer was measured
and sketched. The resulting series of “cross sections” of the nest
resulted in a composite drawing of the nest.

Nest temperature and humidity.

Temperature data were recorded from approximately weekly

1981]

Mac Kay — Nest Phenologies of Pogonomyrmex

readings of thermisters permanently implanted in nests of the three
species. The data were supplemented with readings taken during
nest excavation, following the procedure of Rogers et al. (1972).
Soil temperatures taken within the excavation hole (at least 20 cm
distant from ant burrows) and within the adjacent undisturbed soil
at the same level were not significantly different in two cases
involving P. montanus nests (F = 0.00001ns, F = 0.13ns). Similar
comparisons were not made in the cases of P. rugosus and P.
subnitidus as the soils were too compacted to allow the insertion of
a thermometer in undisturbed soil to a depth of 30 or 40 cm.

Soil samples (160 grams) were collected at various depths and
oven dried (60° C) to constant weight to determine water content. At
least three replicates of soil temperature and soil moisture content
were collected at each level. It was anticipated that these parameters
would determine the position of the brood within the nest. I
assumed a correlation existed between the humidity within the
burrows and water content of the soil as well as a uniformity of the
soil structure in the first 100 cm of the nest where most of the
seasonal changes in the positions of the inhabitants occurred. Sandy
soils would release more water vapor to burrows than would clay
soils, if both had the same level of soil moisture (Marshall and
Holmes 1979). The amount of water present within the soil changes
continuously under field conditions (Marshall 1959), which would
also modify the relative humidity.

Food input into nest.

Food input was estimated by channeling the flow of foragers and
sampling a fraction of foragers at regular intervals to determine the
numbers of trips made and the amount of food brought back to the
nest.

Twenty-eight nests of the three harvester ant species (13 P.
montanus, 10 P. subnitidus, and 5 P. rugosus ), were surrounded by
strips of 25 gauge sheet metal. The diameters of the enclosures were
approximately one meter for P. montanus, 1.5 meters for P.
subnitidus, and 2 meters for P. rugosus. The sheet metal strips were
buried to a depth such that 10 cm of the metal were exposed. Sheet
metal with a total width of 20 cm was sufficient. The ants could not
normally climb over the enclosure as the sheet metal was very
smooth. The ants would occasionally begin to climb the enclosure at
the junction of the two ends. In such cases the area was covered with
Tanglefoot(R).

Psyche

[Vol. 88

In some cases, especially with P. montanus, the ants would
attempt to tunnel under the enclosure. When this occurred, the ants
were removed from the site of the tunneling and placed near the nest
entrance inside the enclosure. In such cases the tunneling was
completely controlled by destroying the tunnel system and replacing
it with soil.

The ants were allowed to enter and exit the colony through two 2
cm diameter vinyl tubes, 6 cm in length. Entrance of the ants to the
colony through the “exit” tube was prevented by having a 0.5-1 cm
distance between the end of the tube and the soil. In a similar
manner exit via the “entrance” tube was prevented. The ants were-
apparently not affected by this short distance, they either simply
dropped with no hesitation or rapidly climbed down from the tube
to the soil. The tubes were within 15 cm of each other and were
placed on the side of the nest where most of the foraging occurred.
A 0.448 liter glass jar could be placed under the tube by which the
ants entered the nest, thus collecting the foragers with the food items
they carried. The foragers were counted and the food items
collected. The foragers were released into the nest enclosure with a
quantity of food (native seeds) which approximated the amount of
food removed. The nests were sampled at approximately weekly
intervals throughout the foraging seasons, during 1978 to 1980. All
of the foragers entering P. montanus nests were collected, 1/5 to 1/6
of those entering P. subnitidus nests, and 1 / 60 of those entering the
P. rugosus nests. With these proportions, one person could handle
the activity of 5 nests during a single day. The forager populations
were estimated by capturing all of the foragers throughout the day,
as they returned to the nests.

Statistical analysis.

Unless otherwise indicated, the 5% level of significance was used
in all comparisons. A single asterisk indicates statistical significance
at the 5% level, double asterisks at the 1% level, triple asterisks
indicate significance at the 0.1% level. Means are listed plus or
minus one standard error. The percentages of the nest populations
were used to make comparisons between the species possible. The
data obtained were fit to least squares polynomial regressions
(Snedecor and Cochran 1967). The curves were constructed from
the equations.

1981] Mac Kay — Nest Phenologies of Pogonomyrmex

Figure 1. The structure of a typical Pogonomyrmex montanus nest.

CENTIMETERS

Psyche

[Vol. 88

Results

Nest structure.

The nest of P. montanus has numerous burrows in the upper
levels (Figure 1). Below this, there is often only a single main tunnel
to the bottom of the nest. Most of the ants are found in the burrows
which branch from the main tunnel. The main tunnel contains few
ants and is apparently used only for movement between the side
burrows. In many cases there are two separate “major tunnels”, as is
shown in Figure 1. In P. subnitidus the two major tunnels may be
separated by more than 100 cm and may appear as two separate
nests. One major tunnel may contain no brood and the other may
contain all of the brood in the nest. The queen and brood are usually
found in the major tunnel which goes to the deeper level.

The structure of the nests of P. subnitidus and P. rugosus are not
shown, but are similar except that they are larger and deeper, often
extending to 300 or 400 cm deep. There was no relationship between
the worker populations and the nest depth (for P. montanus r =
0.16ns (65), for P. subnitidus r = 0.03ns (26), and for P. rugosus r =
0.32ns (20)).

Nest microclimatology: temperature.

The seasonal changes in nest temperatures are similar for all three
species (Figure 2). The nest warms rapidly in the spring and
temperatures reach a maximum at the end of June or July. The soil
temperature begins to drop in August and levels out during the
winter months. As the species occur at different altitudes, the
temperature ranges are different. The range of P. montanus extends
from slightly below zero to 20° C, that of P. subnitidus from slightly
above zero to 25° C, and that of P. rugosus from slightly below 10
to 30° C.

Only the changes at the 20 and 50 cm depths are shown in Figure
2 as the other levels are similar. The differences between the levels
deeper than 40 cm were generally not significant. The only
important difference between the curves of the 20 cm level and 50
cm level is that the shallow level warmed sooner in the spring and
cooled sooner in the fall.

Nest microclimatology: humidities.

The seasonal changes in soil moisture are similar in the nests of all
three species (Figure 3). Soil moistures are high in the winter and

1981] Mac Kay — Nest Phenologies of Pogonomyrmex

MONTHS

Figure 2. Seasonal changes in the mean daily nest temperatures of three species
of Pogonomyrmex harvester ants.

Free water in soil (%) Free water in soij_(%)

Psyche

[Vol. 88

Figure 3. Seasonal changes in the nest humidities of three species of
Pogonomyrmex harvester ants.

spring and low in the summer and fall. Throughout the winter, the
soils receive relatively large amounts of rain or snow which raise the
soil moistures to high levels. After this time, the surface and upper
levels lose water rapidly by evaporation. The lower levels of the nest
retain water throughout the entire season, although the percentage
decreases. Soil moistures at levels below 30 cm are essentially the
same for all three species. Summer showers rapidly increase soil
moistures of the upper levels (note the peaks in the Figure 3), but
have little effect on the levels below 30 cm. This water input into the
soil is rapidly lost by evaporation.

The soil moisture of the lower levels is generally higher than that
of the upper levels, possibly forming a relative humidity gradient.
There are more fluctuations in the higher levels, both in soil

1981] Mac Kay — Nest Phenologies of Pogonomyrmex

moisture and temperature. This probably accounts for much of the
brood being kept in the lower nests levels.

The harvester ants apparently obtain water from several sources.
Some metabolic water may be available to the ants, as it has been
shown that harvester ants increase their metabolism when they are
water stressed, without increasing their activity (Ettershank and
Whitford 1973; Kay and Whitford 1975). Morning dew would not
normally be available as foraging begins after dew has evaporated. I
have seen harvester ants actively drink rain drops on the soil
surface, demonstrating a curious pumping action of the gaster, but
precipitation is not common in the three habitats during the summer
(U.S. Weather Bureau Climatological Data). Capillary condensa-
tion occurs in the soil at relative humidities above eighty percent
(Rode 1955) and may allow the ants free water. Arthropods,
especially insects, are able to actively absorb water vapor from
unsaturated air, although the mechanism is not understood (Edney
1974; Cloudsley-Thompson 1975). It is not known if harvester ants
have the ability to actively absorb water vapor.

Seasonal changes in nest populations.

The data on nest populations obtained from the nest excavations
are summarized in Appendix 1. Absolute counts could not be easily
compared because the numbers of individuals present in the nests of
the three species are very different. To reduce this variation between
nest populations of the three species, the data are compared in the
form of percentages. The seasonal changes in the brood and sexual
populations are similar for all three species, when the percentage
composition of each of the classes are compared (Figs. 4 & 5). In the
three species, egg laying begins in late April to late May, similar to
P. owyheei (Willard and Crowell 1965) and P. occidentalis (Lavigne
1969). Development from egg to callow in the species requires five
to six weeks compared to 25 days for P. badius (Gentry 1974) and 30
days in P. occidentalis (Cole 1934). It is very difficult to determine
the number of larval instars in the development of ants (Wheeler
and Wheeler 1976), although Marcus (1953) suggests that there are
four instars in P. marcusi. As a consequence, all of the instars were
combined into a single group. The first larvae appear about a week
after the eggs are laid, first pupae about two weeks later. Callows
are found in the nest about 5 or 6 weeks after the eggs were laid and

Psyche

[Vol. 88

MONTHS

Figure 4. Seasonal changes in the brood populations of three species of
Pogonomyrmex harvester ants. The arrows indicate the dates when eggs were first
found in the nests. Nests excavated which contained only adult workers are not
represented in the figure.

1981] Mac Kay — Nest Phenologies of Pogonomyrmex

remain pale for about three weeks. Thus, development from the egg
through the larval instars requires about three weeks, the pupal
stage 2 3 weeks, and the callow stage three weeks.

Most of the eggs are laid in the spring as large amounts are found
early in the season. The amounts found in later excavations decrease
and eggs are rarely found after the pupae begin to appear in the nest.

The larval population reaches a maximum in late July in P.
montanus, and mid August in P. subnitidus and P. rugosus. The
pupal population reaches a maximum in mid August in P.
montanus and late August in P. subnitidus and P. rugosus. The
callow population reaches a maximum in early to mid September in
all three species. The callows are easy to distinguish from adult
workers in P. montanus as they remain pale for at least three weeks
(based on laboratory observations). The callows of P. rugosus and
P. subnitidus are much more difficult to distinguish from the adult
workers. Pogonomyrmex rugosus callows darken to a color indis-
tinguishable from mature workers within five days. Pogonomyrmex
subnitidus mature workers are pale making it difficult to distinguish
them from the callows, even if the callows remain pale for many
days.

As the majority of the first individuals produced are sexuals, most
of the larvae and pupae formed in the first part of the season
become reproductives. Workers are also produced early in the
season, especially in P. rugosus. All of the later brood become
workers as was also found in P. owyheei (Willard and Crowell
1965). The reproductives remain in the nest only until late August or
early September. In P. owyheei they remain in the nests until mid
December (Willard and Crowell 1965).

The first winged reproductives appear in the nests in late June (P.
rugosus ) or late July ( P . montanus and P. subnitidus ). The mating
flights are completed by the first part of September. The highest
sexual populations occur in mid August. Therefore the colony
begins production of reproductives early in the year and allows
them to remain in the nest for extensive periods of time, even
though they are consuming food. This is true to a lesser extent in P.
subnitidus, where the reproductives appear in the nest in late July
and most have left the nest by mid August (Figure 5).

There are several interesting points in Figs. 4 & 5. Although P.
rugosus begins production earlier in the year than do the other two
species, the populations of brood in the nest reach peaks later in the

Psyche

[Vol. 88

year. Pogonomyrmex rugosus spreads reproduction out over the
year to a greater extent than does P. montanus. Pogonomyrmex
montanus produces relatively more sexuals than does P. rugosus or
P. subnitidus and in general the production is much higher.

Mating flights.

The mating flights occur either in the morning (P. subnitidus ) or
the afternoon (P. montanus and P. rugosus ). Reproductives of P.
montanus first appeared on the nest surface on 10 August 1978. The
reproductives emerged from the nest entrance, scurried over the
mound for a few seconds and then returned to the nest. They may
have been evaluating environmental conditions to determine when it
was optimal for the mating flight. This behavior was found in all
three species. A small flight occurred on 29 August 1978 between
15:30 and 16:20, a second larger flight occurred on 9 September

1978 between 13:20 and 14:10. The nests of P. montanus normally
have a single entrance-exit hole. During the large flight on 9
September 1978 the nests had 2.7 ± 0.3SE (12) exit holes per nest
(range = 2 to 4). These supplemental exit holes allowed the
reproductives to exit the nest more rapidly. I did not observe this
behavior in the other two species. Reproductives of P. subnitidus
were seen on the nest surface as early as 23 July 1980. The flights
occurred on 6, 7, and 8 August 1980 between 8:00 and 9:30. In P.
rugosus , reproductives first appeared on the nest surfaces on 1
August 1979. A large mating swarm was observed on 24 October

1979 between 14:00 and 15:00.

During the time the reproductives left the nest, the surfaces of the
nests swarmed with workers. Apparently most or all of these
workers were foragers as they were lighter in weight than the other
ants in the nest (MacKay, unpubl.). The reproductives often had
considerable difficulty becoming airborne, especially the females,
which usually climbed up plant stems before flying.

Large mating swarms were observed in P. rugosus and were
similar to those described by Holldobler (1976b). The males waited
on the tops of hills (over 100 m altitude above surrounding terrain)
for the females. The males displayed considerable competition for
females as was shown by Markl et al. (1977). As a result mating was
a frenzied activity in which numerous males competed for single
females by biting, pushing, and in general attempting to exclude

1981]

Mac Kay — Nest Phenologies of Pogonomyrmex

Figure 5. Seasonal changes in the populations of reproductives of three species of
Pogonomyrmex harvester ants. Note that the percentage scale for reproductives in P.
montanus has twice the range of the scales for reproductives in the other two species.

Psyche

[Vol. 88

other males (See Figure 2 of Holldobler 1976b and Figure 4 of
Markl et al. 1977). Prior to the mating flight, male respiratory rates
doubled or tripled (MacKay 1981). The individuals with higher
activity levels may be able to increase their fitness by excluding
other males from a female or by capturing a female quickly and
moving into the copulatory position before other males arrive.

After the female has copulated for a short time, she bites the
gaster of the male which is copulating with her. He usually
relinquishes his position to another male. There is considerable
fighting and tumbling so it is difficult to determine the numbers of
times a female mates. Observations suggest that a single female
mates at least 3 or 4 times. She may have mated previously with one
or more of her brothers in the nest. I observed one mating within the
nest of a laboratory colony of P. montanus. In all three species, the
males attempt to mate with their sisters during emergence from the
nest, although a complete copulation was never observed.

After several copulations the females leave the mating swarm
either by flying or walking away. The males no longer show interest
in such females, as the females apparently stop releasing a phero-
mone (Holldobler 1976b). Most females then fly away from the
area. A few remain and within a few minutes begin excavating nests
near the mating site. As the density of such nests is very high (more
than 4 per square meter) the success rate is undoubtedly low.
Several times I saw females near the mating area attempt to “steal”
the excavation hole of another female, but were chased away by the
resident female. Such attempts are common and are occasionally
successful (Markl et al. 1977).

Seasonal changes in the positions of inhabitants within the nests.

The seasonal movements in the positions of the inhabitants of the
nests depicted in Figures 6, 7 and 8 are similar to those described in
P. owyheei (Willard and Crowell 1965) and P. occidentalis (Lavigne
1969). The depths are not comparable between the three species as
the nests of P. rugosus are deeper than those of P. subnitidus which
are in turn deeper than those of P. montanus (Appendix 1). In most
cases the time axis is expressed in months of the year with the
exception of the sexuals in which only four months are shown. In all
cases, the proportions represent means of all nests excavated.

Most of the nest population of P. montanus, including the

1981] MacKay—Nest Phenologies of Pogonomyrmex 41

Figure 6. Seasonal movements of the populations of the various member groups in the nests of P. man, anus. The grid has a
value of zero. The value of the proportion of each element in the array is represented both by the height of the box above the
grid and the linear dimensions of the box.

Psyche

[Vol. 88

workers and the nest queen, overwinter near the 40 cm level of the
nest (Figures 6 and 9). In the early spring the soil temperatures are
low (Figure 2) and the ants are very sluggish. When the snow begins
to melt, the lowest chambers of the nest fill with water. If the ants
were at the lowest levels, they would probably be killed. In April
and May the P. montanus worker population begins to spread
throughout the nest. In June, July, and August, nearly 80% of the
worker population moves into the upper 10 cm of the nest (Figure
6). During this time the nest temperatures are high and much of the
worker population is involved in foraging, brood care, and nest
construction. In September as the soil temperature begins to cool,
foraging decreases and the workers begin to spread throughout the
levels of the nest. In December the workers are again at the 40 or 50
cm level of the nest. The worker population in the 20 and 30 cm
levels remains low and relatively constant throughout the year.
There is apparently no temporal movement in the larvae or pupae,
but they are present within the nest for only part of the year. In
general, they are located at the 30 or 40 cm level where temperature
and humidity are relatively constant throughout the season. The
callows tend to occur in the deeper levels of the nest together with
the brood. As most of the worker population is in the upper levels of
the nest, the responsibilities of brood care are left to the callows.

It is difficult to make inferences concerning the sexuals as
individuals begin to leave the nest in the middle of August. Thus,
what appears to be a downward movement may simply be the result
of the individuals in the upper levels leaving the nest. The females do
tend to occur deeper in the nest than do the males. They may be in
lower levels in the nest in order to assist in caring for the brood, as
has been observed in the laboratory. It has been shown in Formica
polyctena that workers must learn brood care during an early period
of their lives or they will never care for brood (Jaisson 1975). This
could occur in Pogonomyrmex where the female reproductives may
“learn” brood care so they can later rear their own brood.

The seasonal movement in P. subnitidus nests is similar to that
found in P. montanus nests (Figure 7). A high proportion of the
workers remains in the upper 30 cm of the nest. In October there is a
dispersion throughout the nest. By December, much of the popula-
tion is at the 120 to 180 cm level, with little of the population in the
lowest parts of the nest. The study area receives less snow than the

1981] Mac Kay — Nest Phenologies of Pogonomyrmex

Figure 7. Seasonal movements of the populations of the various member groups in the nests of P. subnitidus.

Psyche

[Vol. 88

area containing P. montanus, but the lower levels of the nest may
also become flooded when the snow melts. Many of the larvae and
pupae are found in the upper levels of the nest, but there is
apparently a downward movement of the brood and callows in
October and November. By December there is no brood in the nest.
Most of the reproductives are found in the upper 30 cm of the nest
(Fig. 7).

The seasonal movements in P. rugosus nests are similar to the
other two species (Figure 8). Most of the worker population is in the
upper levels of the nest throughout the spring and summer. In
September and October until December, the ants become distri-
buted throughout the nest. The larvae are dispersed throughout the
nest during most of the year, but appear to be moved into the deeper
levels of the nest at the beginning of the winter. The pupae are
located in the upper levels of the nest but also appear to be moved
into the deeper regions of the nest in the fall. The callows also
demonstrate a movement into the deeper nest levels in the fall.
Again, it is difficult to make inferences concerning the sexuals as
they are in the nest for a short period of time, but both sexes appear
to be in the upper levels.

In the winter the ants seem to be dispersed throughout the nest
and do not avoid the lowest levels of the nest. There is no winter
snow at Riverside and the temperatures are higher than those in the
mountains (Figure 1), therefore the ants remain somewhat active
throughout the year.

The seasonal patterns of distribution within the nests are similar
in all three species. The reproductives (when present) and workers
are most abundant in the upper levels of the nest, except in the
winter. The brood are in the deeper levels where the microclimate
undergoes little change. The callows are in the lower levels of the
nests in all three species and apparently care for the brood. This is
common in ants in general (Wilson 1971) and in P. badius (Gentry
1974). No callows were ever seen foraging. They do not quickly
darken on exposure to sunlight.

It is commonly stated that ants keep the larvae and pupae
separate within the nest to take advantage of the optimal conditions
for the development of each (Wheeler 1910; Protomastro 1973). In
Pogonomyrmex, at least P. marcusi is reported to practice such
behavior (Marcus and Marcus 1951). I have no evidence that the

200
250 /

1981]

MacKay — Nest Phenologies of Pogonomyrmex

Figure 8. Seasonal movements of the populations of the various member groups in the nests of P. rugosus

Psyche

[Vol. 88

Table 1. Three-way analysis of variance comparisons of the positions of larvae and
pupae in 26 nests of P. montanus collected in 1978 and 1979, 3 nests of P. subnitidus
collected in 1979, and 9 nests of P. rugosus collected in 1979. (As the data were
expressed as percentages of the total nest population, they were subjected to an arcsin
transformation before analysis.)

Source

P. montanus

Different nests

0.005

0.556 ns

Positions of larvae and pupae

0.007

0.778 ns

Levels in nests

0.315

35.000***

Nests X brood

0.004

0.444 ns

Nests X levels

175

0.110

12.222***

Brood X levels

0.028

3.111**

error

174

0.009

P. subnitidus

Different nests

0.000

0.000 ns

Positions of larvae and pupae

0.000

0.000 ns

Levels in nests

0.017

5.667***

Nests X brood

0.000

0.000 ns

Nests X levels

0.017

5.667***

Brood X levels

0.003

1.000 ns

error

0.003

P. rugosus

Different nests

0.000

0.000 ns

Positions of larvae and pupae

0.000

0.000 ns

Levels in nests

0.019

9.500***

Nests X brood

0.000

0.000 ns

Nests X levels

312

0.010

5.000***

Brood X levels

0.003

1.500 ns

error

311

0.002

larvae and pupae are placed in separate levels of the nests in any of
the three species (Table 1). There is a significant difference between
the levels of the nests, which is evident in Figures 6, 7, and 8. The
brood tend to be in the lower levels of the nest. Although it is
commonly assumed there is segregation of the larvae and pupae,
statistical analysis has not been performed in the past to support the
assumption.

In one instance, a P. montanus nest placed a large number of
brood on the soil surface near the nest entrance after a late-summer

1981] Mac Kay — Nest Phenologies of Pogonomyrmex

Table 2. Analysis of variance comparisons of the positions of males and females in
17 nests of P. montanus collected in 1978 and 1979, 4 nests of P. subnitidus collected
in 1980, and one P. rugosus nest collected in 1979. (The data were subjected to an
arcsin transformation before analysis.)

Source

P. montanus

Different nests

0.007

0.538ns

Males and females

0.000

0.000ns

Levels in nests

0.232

17.846***

Nests X sexuals

0.005

0.385ns

Nests X levels

112

0.072

5.538***

Sexuals X levels

0.062

4.769***

error

111

0.013

P. subnitidus

Different nests

0.000

0.044ns

Males and females

0.002

0.231ns

Levels in nests

0.455

45.083***

Nests X sexuals

0.001

0.065ns

Nests X levels

0.006

0.630ns

Sexuals X levels

0.040

3.924**

error

0.010

P. rugosus

Males and females

0.002

1.000ns

Levels in nest

0.006

3.000*

error

0.002

rain, possibly because the upper levels of the nest had become
waterlogged. A considerable number of workers guarded the brood
during this time and when disturbed, the workers immediately
moved the brood back into the nest. This behavior has not been
observed in the other two species.

The positions of the males and females were compared with an
analysis of variance (Table 2). Although it appears from Figures 6,
7, and 8 and our impressions in the field, that females are in deeper
levels of the nest than the males, there is no statistical support
(Table 2). There were significant differences between the levels.
Figures 6, 7, and 8 illustrate that the reproductives tend to be in the
upper levels of the nests.

In Pogonomyrmex spp. there is evidence that little mixing of
adult workers occurs within the nests (Chew 1960; Golley and

Psyche

[Vol. 88

Gentry 1964; Gentry 1974). MacKay (1981) presents data on the
respiratory rates and fat contents of workers taken from the
different levels of the nests of the three species. In winter, spring,
and fall, there are significant differences between the levels with
regard to both of these parameters. If mixing of the workers did
occur between the different levels of the nest, we would not have
found these consistent differences between workers taken from
different levels.

There is little evidence of seasonal movements of the nest queens
(Figure 9). In the spring P. occidental is queens ascend into the
upper levels from the lower levels (Lavigne 1969). The queens may
be moved into the deeper regions during the winter for greater
protection. In the spring, the soil begins to warm sooner in the
superficial levels. The queen may be moved to the higher warmer
levels in order to increase her metabolism for initiation of egg
production.

Guests.

Many species of insects and spiders were collected within the ant
nests. The occurrence of most of these species is probably accidental
and individuals of most species were found only in small numbers
(one or two individuals per nest). Those species most commonly
found include: Orthoptera — Myrmecophila manni Schimmer, in the
nests of all three species; Coleoptera — Echinocoleus setiger Horn, in
P. montanus and P. subnitidus nests, Hetarius hirsutus Martin and
H. sp.#l with P. montanus, H. morsus Leconte and H. sp.#2 with P.
subnitidus, Cremastocheilus westwoodi Horn in the nests of P.
subnitidus. There are at least two species of unidentified staphylin-
ids that are common in P. subnitidus nests (more than 10 per nest).
Hymenoptera — Solenopsis molesta (Say) is common in P. mon-
tanus and P. subnitidus nests, Pheidole spp. in P. rugosus nests. Of
the three harvester ant species, P. subnitidus has the greatest
number of guests and diversity of species.

Food input into nests.

All three species demonstrate similar seasonal changes in their
foraging patterns, with much activity in mid-summer and no activity
in the winter and early spring (Figures 10 and 11). There are
important differences between the three species. Foraging in P.
rugosus begins earlier in the spring and extends later into the fall
than in the other two species. Pogonomyrmex subnitidus has an

1981] Mac Kay — Nest Phenologies of Pogonomyrmex

MONTHS

Figure 9. Seasonal changes in the positions of the nest queen in three species of
Pogonomyrmex harvester ants.

especially short foraging period. Pogonomyrmex montanus begins
the spring with an abrupt increase in foraging (Figure 10). The lower
altitude species, P. rugosus, is exposed to many sunny days during
the winter. During most of this time the nests of the high altitude
species, P. montanus, are covered with snow. The nests of the mid
altitude species, P. subnitidus, are covered by snow part of the time.
In May or June foraging begins, increases throughout the summer
and decreases again in the fall. This foraging pattern corresponds
well with the production of workers and reproductives within the
nest.

Only a small portion of the population is involved in foraging.
The mean number of foragers per day (recorded during July and
August, the months of peak foraging) were 378 ± 73.2 (6) for P.
montanus, 648 + 177.3 (4) forP. subnitidus, and 1427 ± 187.3 (5) for
P. rugosus. Later excavation of the nests indicated that the
population of foragers comprised 22.9%, 19.4%, and 18.4% of the
total nest populations of P. montanus, P. subnitidus, and P.
rugosus, respectively. Others have estimated that 10% of the
population is involved in foraging in such species as P. badius
(Golley and Gentry 1964), P. calif ornicus (Erickson 1972) and P.

Psyche

[Vol. 88

MONTHS

Figure 10. A comparison of the number of daily foraging trips in three species of
Pogonomyrmex harvester ants. The horizontal lines indicate the means, the black
rectangles the standard errors on each side of the mean, and the vertical lines indicate
the ranges.

Table 3. Nest densities, populations and biomasses of several ant species of the genus Pogonomyrmex. The values are
±1 standard error, n is presented in parenthesis.

1981] Mac Kay — Nest Phenologies of Pogonomyrmex 51

\o ~o

On C

— cd

C C

<u aj

a o

c/3

C cd

o Q

X "O

£ g

-- (N
X r-

£ ^

1 *

O *-! O

£■1
a> u.
X ,0

3 X
O 3
C/3 U

-3 <u

1 ^

cd 0

C On

2 £

C4 3
O a -3

Z. Cl "3
^ S' -Cl

desertorum 115 400-600 New Mexico Whitford and

Bryant 1979

Table 3 continued

Psyche

[Vol. 88

6 o
£

■g m ^

3 t, c

_ ° c
£

3 3

T3 *n

3 o S

tt £ e

C/3

>> .2

5 |

1 "

2 3

3 33
O 3
C/3 U

ON
NO
O ON
NO —
ON

£ 60
Si ’>

33 efl

U J

3 .s

§ I

■si

< £

OO II <N

“ y>3 £5

m 60 I O ’'3’
NO o c ON O (N

vo fC sj vo r~ o
— ' w u- m oo cn

.03 .«n

2 2

s* c

8 8

o o

NT)

■o 2
c ~

rsi

03 1-^

« ON
60 —
O

04 3

do o
o o
o o
— m

ON _
60

O ” —

^ W ^ u

2 D

o *3
C/3 U

II CN

S 15

^ © oo

^ M ^

r» w <n

>3 -2

5 l

I &

S 3

O as

£ U

•5 <2

3 33
O 3
C/3 U

- ^

£ £

> §
<D 2

II NO

m 3 22

+l 2 7

^ o
^ m
ON fO oo

mo —

1981] Mac Kay — Nest Phenologies of Pogonomyrmex 53

in in
Os Os

> >

o o

N N

CL> <D

3 C

C/3 C/3

3 3

< <

< CQ

'JS

* Estimations based on data from literature.
** From Brian (1965)

Specie

(II)

0.8-8.3*

1. 2-3.9* 8.6-27.9*

Psyche

[Vol. 88

occidentalis (Rogers et al. 1972). Chew (1960) estimated that no
more than !/2 of P. occidentalis workers were out of the nest at any
one time. In a mark recapture analysis, Whitford et al. (1976)
estimated the forager population at 2786 in P. rugosus. This
estimate is higher than the one I determined which may indicate that
the nest populations of P. rugosus in New Mexico are larger than
those in southern California. My estimates are minimal: there may
have been foragers which remained within the nest. Also the
experimental channeling of the forager population may have
affected the natural foraging activity. The whole work force may not
have been activated because of a reduction of recruitment (Hbll-'
dobler, Pers. Comm.).

A comparison of the number of foragers given above and the
number of foraging trips per day (Figure 10) indicates that
individual P. montanus foragers make two or three trips per day, P.
subnitidus foragers about nine, and P. rugosus foragers make more
than ten trips per day. There are considerable differences between
the three species in the numbers of foraging trips made (Figure 10),
which compares with the differences in the sizes of the nest
populations (Table 3).

The seasonal changes in the daily amount of food brought to the
nest are similar to those found in the numbers of foraging trips
(Figure 11). As with the forager number, P. rugosus brings in food
earlier in the spring and extends foraging later into the fall,
compared to the other two species. Pogonomyrmex montanus
abruptly increases the food input once foraging begins and de-
creases it slowly until fall. Pogonomyrmex montanus is the only
species of the three which does not store seeds in the nests. It may
have to bring in large amounts of food once the larvae begin to
appear in the nest. The other two species have seed reserves and may
thus avoid such an abrupt increase in foraging in the spring.

Comparisons of the food sources of the three species (Figure 12)
indicate that the harvester ants utilize a wide variety of food items,
although most materials are either seeds or plant parts. Pogono-
myrmex rugosus relies almost exclusively on seeds. Pogonomyrmex
subnitidus and especially P. montanus bring a much greater
diversity of food items to the nest. Pogonomyrmex montanus relies
more heavily on plant parts and insects than does P. subnitidus.
Pogonomyrmex subnitidus brings in a greater proportion of feces
than does P. montanus, although the ratio of bird to mammal feces

DAILY FOOD INPUT (g)

1981]

Mac Kay — Nest Phenologies of Pogonomyrmex

1.4

0.6 -

0.4 -

0.2 -

101-

P montonus

P. subnitidus

^ ..

P. rugosus

JFMAMJ JASOND
MONTHS

Figure 11. A comparison of the daily food input (grams) in the three species of
Pogonomyrmex harvester ants.

Psyche

[Vol. 88

SOURCES OF FOOD

Plant parts
Seeds
Soft insects
Hard insects
Bird feces
Mammal feces

Figure 12. A comparison of the food sources in the three species of Pogono-
myrmex harvester ants.

1981] MacKay — Nest Phenologies of Pogonomyrmex

is similar in both species. Pogonomyrmex rugosus brings in more
bird feces than mammal feces, P. montanus and P. subnitidus bring
in more mammal feces than bird feces. A distinction was made
between “hard” insects and “soft” insects. Hard insects included
those heavily chitinized forms, especially the Coleoptera and certain
Formicidae. Soft insects included Homoptera, most Hemiptera,
most Diptera, larvae and pupae of most orders and a few non-
insects such as spiders. It appears that the degree of chitinization
may not be important as the proportions of hard and soft insects
were similar. All three species have chitinase activity in their gasters
(MacKay, unpub. data).

Plant parts consist of pieces of leaves and flowers and in the case
of P. montanus, pine resin. Flowers of Penstemon spp. and
Arctostaphylos spp. are transported to the nest and placed around
the brood, possibly to increase the humidity. Later the intact flowers
are discarded at the nest surface. This indicates the flowers are not
placed around the brood to protect them from predators. In the case
of pieces of leaves, apparently they are eaten by the ants as they do
not later appear on the nest surface. There is considerable seasonal
change in the food composition of P. montanus and P. subnitidus
(Figure 13). The percentages of insects brought into P. montanus
nests changes little seasonally. There is a seasonal reduction in the
percentage of utilization of insects in P. subnitidus. There is little
seasonal change in the proportion of the food sources composed of
feces in the two species, although a slight reduction may occur. In
both species, especially P. montanus, there is a seasonal decrease in
the proportion of plant parts brought to the nest. In both species,
there is a dramatic increase in the utilization of seeds after July. This
increase is probably related to a greater availability of seeds after the
flowering period of annual plants. A similar comparison was not
made in the case of P. rugosus as non-seed materials are a very small
portion of their diet (Fig. 12). In P. rugosus, there was a seasonal
drop in the proportion of the diet composed of Erodium eicutarium
(L.) L’Her. seeds (May 90.3%, June 91.0%, July 88.9%, August
89.7%, September 84.1%, and October 80.9%). Other seeds, espe-
cially those of Pectocarya linearis DC and Festuca octoflora Walt.,
made up most of the difference.

Caloric analysis of the food entering the nests of the three species
indicates that a P. montanus colony receives an average of 166.6

PERCENT OF TOTAL FOOD INPUT

Psyche

[Vol. 88

Figure 13. The seasonal changes in the food sources of P. montanus and P.
subnitidus.

1981] Mac Kay — Nest Phenologies of Pogonomyrmex

kcals, a P. subnitidus nest 1267.0 kcals, and a P. rugosus nest 7613.6
kcals of food during a year (MacKay 1981). Of these amounts, a P.
rugosus colony discards seed husks and other such materials, a
quantity consisting of 5004.5 kcals or 65.7% of the intake. This is
indicated in the field by large discard piles of seed husks being
deposited around the nests. A few seeds are discarded and germinate
from the piles in the spring. Another harvester ant, Veromessor
pergandei (Mayr) forages in the piles and removes many of the
discarded seeds. Pogonomyrmex montanus and P. subnitidus
discard few materials, the amounts are too small to be estimated.

Seed storage.

Seasonal changes in seed storage in P. subnitidus and P. rugosus
are shown on Figure 14. Pogonomyrmex rugosus began both 1979
and 1980 (data for January) with 0.04^0.06 grams of seed storage
per ant. The correlation of ant number vs. seed weight was very high
(r = 0.997, p < 0.01). This amount dropped until May, possibly the

Figure 14. A comparison of the seasonal changes of seed storage in P. rugosus
and P. subnitidus.

Psyche

[Vol. 88

result of seed consumption by the developing larvae. I have no
explanation for the other two peaks which appear. There is some
evidence of a drop in seed storage in the spring in P. subnitidus, but
it is not as great as that found in P. rugosus. Pogonomyrmex
subnitidus also appears to begin the season with a constant amount
of seeds, about 0.002-0.004 g/ant, much smaller quantities than P.
rugosus. There are also many unexplained peaks in P. subnitidus
seed storage, especially the high peak in September. Pogono-
myrmex montanus does not store seeds in the nest. In the
population at Big Pine Flats in the San Bernardino Mountains, we
occasionally encountered very small caches of seeds (less than
0.0001 g/ant) which were apparently only small daily accumulations
of seeds that had not been eaten at that time.

Production.

Production in the three species is summarized in Table 4. The
proportion of energy invested in production varies considerably
between the three species, but in all cases it is relatively low. Total
production constitutes 12.2, 8.3, and 7.9 per cent of the total energy
flow in P. montanus, P. subnitidus, and P. rugosus respectively
(MacKay 1981). In all three species, a higher percentage of the total
production is invested in workers than reproductives (Table 4).
Pogonomyrmex subnitidus and P. rugosus both invest heavily in
workers, P. montanus invests heavily in reproductives. The data on
Table 4 suggest that the three species invest more in the production
of females than in males. The costs of respiration of males are higher
than of females (MacKay 1981). When respiration costs are taken
into account, the colonies of each species invest about equally in the
production of males and females (MacKay 1981). More numbers of
males than females are produced in all three species (Table 4).
Individual females are more expensive to produce than are indi-
vidual males (MacKay 1981).

Most of the workers are replaced each year. Pogonomyrmex
montanus colonies produce 1516 workers per year (Table 4), which
is similar to the mean worker population of 1665 (Table 3).
Pogonomyrmex subnitidus colonies produce 3988 workers as
compared to a worker nest population of 5934; P. rugosus colonies
produce 5298 workers per year compared to a worker nest
population of 7740.

1981]

Mac Kay — Nest Phenologies of Pogonomyrmex

Table 4. A comparison of the investments in production in three species of
Pogonomyrmex harvester ants.

Species

Group

Number of
Individuals

Dry wt
(g)

kcals

Percent Total
Production

montanus

Workers

1516+ 95

2.4

12.7

51.8

Females

187+ 30

1.2

7.8

31.8

Males

239+ 41

0.8

4.0

16.3

24.5

subnitidus

Workers

3988+438

11.6

87.4

91.5

Females

111+ 65

0.9

5.4

5.7

Males

251+ 87

0.6

2.7

2.8

95.5

rugosus

Workers

5298+763

30.3

208.2

86.6

Females

118+100

2.7

18.9

7.9

Males

312+ 73

2.5

13.4

5.6

240.5

Discussion

Comparison with other species in the genus Pogonomyrmex.

The genus Pogonomyrmex belongs to the tribe Myrmicini, one of
the most primitive tribes in the subfamily Myrmicinae. The genus
has existed at least since the Oligocene (Burnham 1978), and is
distributed throughout North and South America from Canada to
Patagonia, from sea level to at least 4500 meters in altitude. At the
present time there are 24 valid species in North America and about
33 in Central and South America. The genus may have originated in
South America and migrated northward (Kusnezov 1951) or
originated in North America and migrated southward (Wheeler
1914; Creighton 1952).

Considerable work has been done on nest densities, populations,
and biomasses of ants of various species of the genus Pogono-
myrmex (Table 3). Examples of biomasses from other genera would
include the following (expressed as mg dry weight/ m2), Tetramori-
um caespitum at 200 (Brian et al. 1967) and 1480 (Nielsen 1974),
Lasius niger at 60 (Odum and Pontin 1961) and 1060 (Nielsen 1974),
L. alienus at 2090 (Nielsen 1974), L. flavus at 1400 (Odum and
Pontin 1961) and 15,000 (Waloff and Blackith 1962), Leptothorax
acervorum at 3000 (Brian 1956), and Formica rufa at 12,000

Psyche

[Vol. 88

(Marikovsky 1962). In general, the biomasses of Pogonomyrmex
are much lower than those found in other genera.

The species investigated, especially P. subnitidus and P. rugosus,
are comparable to most of the North American representatives of
the genus (Table 3). The South American species apparently have
much smaller populations, but few nests have been excavated and
most were partial excavations in which the queen was not found or
after the excavation was finished, additional ants were found later.
Species from arid regions tend to have larger colonies than those
from mesic environments, with the exception of P. laticeps. The
colonies of North American species live longer than South Ameri-*
can species (Kusnezov 1951). Pogonomyrmex montanus is some-
what atypical for the genus in occurring at higher altitudes, but is
similar to other species in several aspects. The number of nests per
hectare is comparable to several other species including P. badius,
P. barbatus, P. occidentals, P. owyheei, P. rugosus, and P.
subnitidus. The nest populations of P. montanus are smaller than
those of most of the other species, but the number of workers/ m.sq.
and/or the dry wt/m.sq. are comparable to P. badius, P. calif orni-
cus, P. occidentals, P. owyheei, P. rugosus, and P. subnitidus.

With regards to the populations, the three species investigated
appear to be “typical” North American Pogonomyrmex harvester
ants. It would be very interesting to do a comparable study of
“typical” South American Pogonomyrmex harvester ants.

Effect of altitude.

It was anticipated that altitude would have three primary
effects: 1) The higher altitude species, P. montanus, would be

subjected to lower average temperatures. 2) The higher altitude
species would be subjected to shorter foraging seasons, thus
reducing the yearly food input into the nest, resulting in lower
production. 3. The higher altitude’s shorter growing season would
result in fewer available seeds from annual plants.

Although P. montanus is subjected to the lowest seasonal
temperatures of the specific populations of the three species
investigated (Figure 2), it metabolically compensates for this by
having higher respiratory rates than the other species (MacKay
1981). Apparently altitude has an effect on foraging, although it was
not as large as expected. The foraging season was somewhat
reduced in P. montanus and P. subnitidus, when they are compared

1981]

Mac Kay — Nest Phenologies of Pogonomyrmex

with P. rugosus (Figure 10). Pogonomyrmex montanus, and to
some extent P. subnitidus, are in habitats with winter snow cover. In
such habitats foraging during the winter is not possible. Pogono-
myrmex rugosus occupies a low altitude habitat where there are
many warm sunny days during the winter. During these days, it does
not forage, although a few workers are on the nest surface either
sunning themselves or working on nest reconstruction.

The higher altitudes had shorter growing seasons, resulting in
fewer annual seed producing plants. As a result P. montanus and P.
subnitidus foraged on various materials but began to rely heavily on
seeds later in the year (Figure 13). This was especially the case in P.
montanus, which relied heavily on plant parts early in the year.
Later when seeds became more available, they almost completely
replaced plant parts in the diet (Figure 13).

Allocation of resources between worker and reproductive produc-
tion.

As was expected, the highest altitude species was exposed to a
shorter foraging season, but this did not result in lower production.
The highest altitude species, P. montanus, invests a larger propor-
tion of energy into production than do the other two species. The
amount invested in reproductives is especially high (Table 4).
Pogonomyrmex subnitidus and P. rugosus invested about equally in
production, with investment in reproductives very low compared to
P. montanus (Table 4).

Most Pogonomyrmex spp. are low altitude desert species (Cole
1968). Pogonomyrmex montanus appears to be in a marginal
habitat for Pogonomyrmex spp. in that it occurs in a high altitude
pine forest. The nest populations are among the smallest for the
genus (Table 3) and the nests are also very shallow (Appendix 1).
Both P. montanus and P. subnitidus have shorter foraging seasons
and apparently are not able to exploit their optimal food source
(seeds) until late in the season (Figure 13). Simulations of the effects
of bad years on the nests indicate that P. rugosus and P. subnitidus
are able to withstand moderately large reductions in food input
whereas P. montanus is not (MacKay in prep.). As a result, nests
may be short-lived as compared to the other two species and nest-
extinction may be a common phenomenon. Apparently, as a
response to such conditions, P. montanus invests a larger propor-
tion of energy in the production of reproductives than do the other

Psyche

[Vol. 88

two species. It might be expected that the South American species
would be ecologically similar to P. montanus as they share many
characteristics (Table 3).

Production as well as foraging and food input were spread over
more of the season in P. rugosus than in the other two species (Figs.
4, 5, 10 & 1 1). This is easily explained as P. rugosus lives in a more
moderate climate than the other two species. Actually it was
expected that these processes would occur over the entire year as
there are many warm sunny days at lower elevations during the
winter. Yet, activities almost stop. Perhaps these processes do not
continue as the nest temperatures are lower during the winter than
they are in the summer (Figure 2).

The sex ratio was not constant between years (see data in
Appendix 1). In P. montanus the female:male ratio was 0.88:1 in
1978, 1.41:1 in 1979, and 0.42:1 in 1980. In 1980 the number of
males produced was three times those of the other years. An excess
of females in 1 979 was not found in P. rugosus (0.38: 1 ) as was found
in P. montanus. An excess of males was found in P. subnitidus
(0.42:1) in 1980 as was found in P. montanus.

Nests are extremely heterogeneous in regards to sex ratio
(Appendix 1). Correlations were investigated between the female:
male ratio and the apparent age of the nest. Twelve P. montanus
nests at the peak levels of production were used in the analysis. The
age of a nest should be related to the numbers of adult workers
present in the nest and the depth of the nest: older nests should be
deeper and have a larger worker population. The product-moment
correlation coefficients (Sokal and Rohlf 1969) of the sex ratio with
worker population size and nest depths were both 0.17. Although
the coefficients were not statistically significant, both were positive,
suggesting that older nests produced greater proportions of females.
The product-moment correlation coefficient comparing the sex
ratio with the numbers of workers produced by the nest during the
year was negative (r = —0.38). Although the relationship was not
statistically significant, it suggested that nests involved in an
increase in the worker population (i.e., younger nests) produced a
smaller proportion of females. Data were presented (MacKay 1981)
which indicated that food stressed nests produced a greater
proportion of females; nests given extra food produced a greater
proportion of males.

1981]

Mac Kay — Nest Phenologies of Pogonomvrmex

The factors influencing the determination of sex ratios in the
Hymenoptera are currently of much interest (Herbers 1979).
Experimental manipulation of food input and excavation of
colonies of known age may provide information on the factors
which determine the sex ratio in a harvester ant nest.

Summary

This investigation compares the phenologies of foraging and
reproduction in three species of Pogonomvrmex harvester ants
along an altitudinal transect in southern California, USA. Periodic
excavations of 126 nests of the three species, P. montanus, P.
subnitidus, and P. rugosus, reveal that seasonal changes occur
within the nests. The three species have similarities in the physical
environment of the nest although P. montanus, the highest altitude
species, has lower nest temperatures. Both P. montanus and P.
subnitidus are snowbound during part of the season. Egg laying
begins in late April or May; development to adult requires five to six
weeks. The brood reach maximum numbers in late July to late
August. Most of the larvae and pupae formed in the first part of the
season become reproductives. Mating flights begin in late July and
are completed by the first part of September. The highest reproduc-
tive populations occur in mid August.

Much of the nest population is in the upper levels of the nest
during the summer and in the lower levels during the winter. During
the summer, temperature and humidity gradients exist in the nests
with deeper levels being cooler and moister. These gradients may
account for the placement of the brood in the lower levels. There is
no evidence of segregation of the larvae and pupae within the nest,
which has been reported by other investigators.

All three species demonstrate similar seasonal changes in foraging
patterns, with much activity in the mid summer and no activity
during the winter. Only about 20% of the nest population is
involved in foraging. Individual foragers make up to 9 or more
foraging trips per day. The ants utilize a wide variety of food items,
although most materials are either seeds or plant parts. There is a
considerable seasonal change in the food composition of P.
montanus and P. subnitidus.

The highest altitude species, P. montanus, allocates more energy
to reproduction than do the mid or low altitude species. The nests

Psyche

[Vol. 88

invest about equally in the production of males and females.
Evidence presented suggests that the sex ratio may be ecologically
determined and that there may be a yearly change in the sex ratio.

Acknowledgements

I would like to thank Clay Sassaman, Rodolfo Ruibal, Robert
Luck and Bert Holldobler for the critical review of the manuscript.
Walter Whitford provided unpublished information on several
Pogonomvrmex spp., Charles Kugler provided unpublished data on
Pogonomyrmex mayri. Jessie Halverson, Cecil Hoff and the U.S.
Forest Service generously granted permission to conduct the
investigation on property under their jurisdiction. I am especially
grateful to Emma MacKay, who assisted in all aspects of the field
and laboratory work, prepared the figures and made important
contributions to the manuscript. Kenneth Cooper, Fred Andrews,
Gary Alpert, David Kistner, and Stewart Peck kindly identified the
beetles found in the ant nests, and provided much stimulating
information on the ecologies of the beetles. Oscar Clarke identified
the plant seeds.

The research was supported by the Theodore Roosevelt Memori-
al Fund of the American Museum of Natural History, three Grants-
in-Aid of Research from Sigma Xi, The Scientific Research Society
of North America, the Chancellor’s Patent Fund of the University
of California, and the Irwin Newell Award of the Department of
Biology of the University of California at Riverside. The Depart-
ment of Entomology of the Colegio de Graduados of Ciudad
Juarez, Mexico, paid the costs of publication.

Literature Cited
Barnes, O. L. and N. J. Nearney.

1953. The red harvester ant and how to subdue it. USDA Farmers Bull,
number 1668, 11 pages.

Brian, M. V.

1956. The natural density of Mvrmica rubra and associated ants in west
Scotland. Ins. Soc. 3: 474-487.

1965. Social insect pupulations. Academic Press, vii + 135 pp.

Brian, M. V., G. Elmes, and A. F. Kelly.

1967. Populations of the ant Tetramorium caespitum L. J. Anim. Ecol. 36:
337-342.

Bruch, C.

1916. Contribucion al estudio de las hormigas de la provincia de San Luis.
Revista del Museo de La Plata, Argentina. 23: 291-357.

Burnham, L.

1978. Survey of social insects in the fossil record. Psyche 85: 85-134.

1981]

Mac Kay — Nest Phenologies of Pogonomyrmex

Chew, R. M.

I960. Note on colony size and activity in Pogonomyrmex oecidentalis
(Cresson). J. New York Entomol. Soc. 68: 81 82.

Clark, W. H. and P. L. Comanor.

1975. Removal of annual plants from the desert ecosystem by western
harvester ants, Pogonomyrmex oecidentalis. Environ. Entomol. 4:
52-56.

Cloudsley-Thompson, J. L.

1975. Adaptations of arthropods to arid environments. Ann. Rev. Entomol.
20: 261-283.

Cole, A. C.

1934. A brief account of aestivation and overwintering of the Occident ant,
Pogonomyrmex oecidentalis Cresson, in Idaho. Can. Entomol. 66:
193-198.

1954. Studies of New Mexico ants. VII. The genus Pogonomyrmex with
synonymy and a description of a new species ( Hymenoptera:For-
micidae). J. Tenn. Acad. Sci. 29: 115-121.

1968. Pogonomyrmex harvester ants, a study of the genus in North America.
Knoxville Univ. Press, x + 222 pp.

Creighton, W. S.

1952. Studies on Arizona ants (3). The habits of Pogonomyrmex huachucanus
Wheeler and a description of the sexual castes. Psyche 59: 71-81.

Davidson, D. W.

1977a. Foraging ecology and community organization in desert seed-eating
ants. Ecology 58: 725-737.

1977b. Species diversity and community organization in desert seed-eating ants.
Ecology 58: 7 1 1-724.

1980. Some consequences of diffuse competition in a desert ant community.
Amer. Nat. 116: 92-105.

Edney, E. B.

1974. Desert arthropods. In: Desert Biology Vol. II, ed. by G. W. Brown Jr.
pp. 311 384. Academic Press, New York.

Erickson, J. M.

1972. Mark-recapture techniques for population estimates of Pogonomyrmex
ant colonies: An evaluation of the 32 P technique. Ann. Entomol. Soc.
Am. 65: 57-61.

Ettershank, G. and W. G. Whitford.

1973. Oxygen consumption of two species of Pogonomyrmex harvester ants
(Hymenoptera: Formicidae). Comp. Biochem. Physiol. 46A: 605 611.

Gentry, J. B.

1974. Response to predation by colonies of the Florida harvester ant,
Pogonomyrmex hadius. Ecology 55: 1328-1338.

Gentry, J. B. and K. T. Stiritz.

1972. The role of the Florida harvester ant, Pogonomyrmex hadius, in old field
mineral nutrient relationships. Environ. Entomol. 1: 39-41.

Golley, F. B. and J. B. Gentry.

1964. Bioenergetics of the southern harvester ant, Pogonomyrmex hadius.
Ecology 45: 217-225.

Psyche

[Vol. 88

Herbers, J. M.

1979. The evolution of sex-ratio strategies in hymenopteran societies. Amer.
Nat. 114: 818-834.

Holldobler, B.

1976a. Recruitment behavior, home range orientation and territoriality in
harvester ants, Pogonomyrmex. Behav. Ecol. Sociobiol. 1: 3 44.

1976b. The behavioral ecology of mating in harvester ants. ( Hymenoptera:
Formicidae: Pogonomyrmex). Behav. Ecol. Sociobiol. 1: 405 423.

Jaisson, P.

1975. L’impregnation dans l’ontogenese des comportements de soins aux
cocons chez la jeune fourmi rousse ( Formica polvctena Eorst.). Beha-
viour 52: 1-37.

Kay, C. A. and W. G. Whitford.

1975. Influences of temperature and humidity on oxygen consumption of five
Chihuahuan desert ants. Comp. Biochem. Physiol. 52A: 281 286.

Kusnezov, N.

1951. El genero Pogonomyrmex Mayr. Acta Zoologica Lilloana 11: 227-333.

Lavigne, R.

1969. Bionomics and nest structure of Pogonomyrmex occidentalis (Hymen-
optera: Formicidae). Ann. Entomol. Soc. Amer. 62: 1166-1175.

Mack ay, W. P.

1980. A new harvester ant from the mountains of southern California
(Hymenoptera: Formicidae). Southwest. Nat. 25: 151-156.

1981. A comparison of the ecological energetics of three species of Pogono-
myrmex harvester ants (Hymenoptera: Formicidae). Ph.D. dissertation.
University of California at Riverside. 240 pp.

Marcus, H.

1953. Estudios mirmecologicos 11. Observaciones en Pogonomyrmex marcusi
(Kusn.). Folia Universitaria de la Universidad de Cochabamba 6: 45 55.

Marcus, H. and E. E. Marcus.

1951. Los nidos y los organos de estridulacion y de equilibrio de Pogono-
myrmex marcusi y de Dorymyrmex emmaericaellus. (Kusn.). Folia
Universitaria, Universidad Mayor de San Simon, Cochabamba, Bolivia
5: 117-143.

Mares, M. A. and M. L. Rosenzweig.

1978. Granivory in North and South American deserts: rodents, birds, and
ants. Ecology 59: 235-241.

Marikovsky, P. 1.

1962. On intraspecific relations of Formica rufa L. Ent. Rev. 1: 47-51.

Markl, H., B. Holldobler and T. Holldobler.

1977. Mating behavior and sound production in harvester ants {Pogono-
myrmex, Formicidae). Ins. Soc. 24: 191 212.

Marshall, T. J.

1959. Relations between water and soil. Technical Communication Number
50. Commonwealth Bureau of Soils, Harpenden.

Marshall, T. J. and J. W. Holmes.

1979. Soil physics. Cambridge University Press. * + 345 pp.

1981]

Mac Kay — Nest Phenologies of Pogonomyrmex

Nielsen, M. G.

1972. Production of workers in an ant nest. Ekologia Polska 20: 65-71.
Nielsen, M. G.

Number and biomass of worker ants in a sandy area in Denmark.
Natura Jutlandica 17: 93-95.

Nielsen, M. G. and T. F. Jensen.

1975. Ecological studies on Lasius alienus ( Forst) ( Hymenoptera: Formicidae).
Entomologiske Meddelelser 43: 5-16. (In Danish).

O’Dowd, D. J. and M. E. Hay.

1980. Mutualism between harvester ants and a desert ephemeral: seed escape
from rodents. Ecology 61: 531-540.

Odum, E. P. and A. J. Pontin.

1961. Population density of the underground ant, Lasius flavus as determined
by tagging with P22. Ecology 42: 186-188.

Peck, S. B.

1976. The myrmecophilous beetle genus Echinoeoleus in the southwestern
United States (Leiodidae, Catopinae). Psyche 83: 51-62.

Petal, J.

1967. Productivity and the consumption of food in the Myrmica laevinodis
Nyl. population. Proceedings of the International Biological Program
on secondary productivity of terrestrial ecosystems. K.. Petrusewicz (ed).
pp. 841 857.

Protomastro, J. J.

1973. Relacion entre el comportamiento de crianza y un ritmo diario de
preferencia termica en la hormiga Camponotus mus Roger. Physis,
Seccion C. 32: 123-128.

Reichmann, O. J.

1979. Desert granivore foraging and its impact on seed densities and distribu-
tions. Ecology 60: 1085 1092.

Rode, A. A.

1955. The moisture properties of soils and underground strata. Academy of
Sciences of the USSR, Moscow. 117 pp.

Rogers, L., R. Lavigne and J. L. Miller.

1972. Bioenergetics of the western harvester ant in the shortgrass plains
ecosystem. Environ. EntomoU 1: 763-768.

SOKAL, R. R. AND F. J. ROHLF.

1969. Biometry. W. H. Freeman and Co. xxi + 776 pp.

Snedecor, G. W. and W. G. Cochran.

1967. Statistical Methods. 6th edition. Iowa State University Press, xiv + 593

pp.

Taylor, F. W.

1977. Foraging behavior of ants: Experiments with two species of Myrmicine
ants. Behav. Ecol. Sociobiol. 2: 147 167.

Waloff, N. and R. W. Blackith.

1962. The growth and distribution of the mounds of Lasius flavus (Fabricius)
(Hym: Formicidae) in Silwood Park, Berkshire. J. Anim. Ecol. 31:
421 437.

Psyche

[Vol. 88

Wheeler, G. L. and J. Wheeler.

1976. Ant larvae: review and synthesis. Mem. Entomol. Soc. Wash. #7, vi +

108 pp.

Wheeler, W. M.

1910. Ants, their structure, development and behavior. Columbia University
Press, xxv + 663 pp.

1914. New and little known harvesting ants of the genus Pogonomyrmex.
Psyche 21: 149-157.

Whitford, W. G.

1 972. Demography and bioenergetics of herbivorous ants in a desert ecosystem
as functions of vegetation, soil type and weather variables. Unpublished
manuscript.

1976. Foraging behavior of Chihuahuan desert harvester ants. Amer. Midi.
Nat. 95: 455-458.

1978a. Foraging in seed-harvester ants Pogonomyrmex spp. Ecology 59:
185-189.

1978b. Structure and seasonal activity of Chihuahua desert ant communities.
Ins. Soc. 25: 79-88.

Whitford, W. G. and M. Bryant.

1979. Behavior of a predator and its prey: the horned lizard ( Phrynosoma
cornutum) and harvester ants ( Pogonomyrmex spp.) Ecology 60:
686-694.

Whitford, W. G. and G. Ettershank.

1975. Factors affecting foraging activity in Chihuahuan desert harvester ants.
Environ. Entomol. : 689-696

Whitford, W. G., P. Johnson and J. Ramirez.

1976. Comparative ecology of the harvester ants Pogonomyrmex barbatus (F.
Smith) and Pogonomyrmex rugosus (Emery). Ins. Soc. 23: 117-132.

Wildermuth, V. L. and E. G. Davis.

1931. The red harvester ant and how to subdue it. U.S.D.A. Farmers’ Bull.

#1668, 21 pp.

Willard, J. R. and H. H. Crowell.

1965. Biological activities of the harvester ant, Pogonomyrmex owyheei in
central Oregon. J. Econ. Entomol. 58: 484-489.

Wilson, E. O.

1971. The insect societies. Belknap Press, x + 548 pp.

Appendix 1. List of the populations of the excavated Pogonomyrmex spp. nests,
including workers (W), larvae (L), pupae (P), callows (C), males (M), and females
(F). The position of the queen and maximum depth of the nests are expressed in
centimeters. The dates indicated are when the excavation was begun.

Position

Date W L P C F M of queen Depth

Pogonomyrmex montanus

21 Sept 77 793 7 514 0 0 0 50 50

23 Sept 77 1918 123 578 0 0 0 40 40

1981] MacKay — Nest Phenologies of Pogonomyrmex

Appendix 1. cont.

Position

Date W L P C F M of queen Depth

1 1 Apr 78
13 Apr 78

16 Apr 78

3 May 78

8 May 78

9 May 78
18 May 78

25 May 78
31 May 78

6 June 78

8 June 78
13 June 78

13 June 78

23 June 78
27 June 78

7 July 78
12 July 78
20 July 78

24 July 78

4 Aug 78
1 1 Aug 78

18 Aug 78

22 Aug 78
30 Aug 78

9 Sept 78

19 Sept 78

23 Sept 78

24 Sept 78

26 Sept 78

26 Sept 78
30 Sept 78
30 Sept 78
30 Sept 78

7 Oct 78

14 Oct 78
16 Oct 78
16 Oct 78

27 Oct 78
7 May 79
7 May 79

25 May 79
25 May 79

3142

1654

1874

2372

2240

3641

2277

1752

1951

491

1598

499

1841

695

1087

141

1488

359

1652

367

1604

426

1033

228

996

41 1

1013

290

1158

277

1057

244

1552

306

1113

1443

631

2271

2573

634

1734

1785

1024

652

1245

2105

1343

538

2812

1304

3585

2194

3141

0 0

38 0

160 0

177 0

477 0

131 0

678 438

323 496

358 529

359 784

330 766

163 254

73 281

0 80

2 128

6 96

0 40

3 117

12 52

0 13

1 1 1

0 1 1

3 63

0 0

0 4

0 0

188 54

52 138

64 87

262 0

147 148

48 327

111 405

162 83

81 25

0 0

1 0

0 1

0 I

0 0

1 0

0 0

105

1 10

Psyche

[Vol. 88

Appendix 1. cont.

Position

Date

of queen

Depth

3 June 79

1605

15 June 79

1979

20 June 79

537

20 June 79

1768

8 July 79

599

193

16 July 79

563

22 July 79

2614

1318

750

290

29 July 79

947

813

710

305

198

29 July 79

1777

383

708

286

146

30 July 79

369

137

30 July 79

1 1 1 1

835

693

149

3 Aug 79

1853

325

490

609

601

120

4 Aug 79

1261

483

1049

595

305

543

100

4 Aug 79

1625

555

1013

225

221

277

12 Aug 79 (a)

976

345

534

752

167

12 Aug 79 (b)

640

124

231

701

144

13 Aug 79 (a)

1063

619

529

495

390

13 Aug 79 (b)

697

244

466

660

387

18 Aug 79

755

234

719

729

113

19 Aug 79 (c)

370

107

221

19 Aug 79 (c)

520

108

641

322

262

19 Aug 79 (d)

836

168

632

1068

142

20 Aug 79 (d)

792

401

655

141

20 Aug 79 (e)

596

169

318

20 Aug 79 (e)

1079

574

839

289

436

7 Sept 79

1796

468

931

110

8 Sept 79

711

8 Sept 79

808

126

112

559

14 Oct 79

1308

5 Dec 79

1504

19 Aug 80

1039

780

216

309

20 Aug 80

1388

501

255

343

21 Aug 80

1346

144

274

21 Aug 80

1687

351

634

22 Aug 80

1800

185

310

516

Pogonomyrmex subnitidus

5 Nov 78

5033

150

210

29 Aug 79

2507

894

1212

2819

140

170

15 Sept 79

3612

867

892

4147

120

150

7 Oct 79

5442

109

321

999

220

230

3 Nov 79

3452

315

160

230

30 Nov 79

5182

180

240

1981]

Mac Kay — Nest Phenologies of Pogonomyrmex

Appendix 1. cont.

Position

Date

of queen

Depth

25 Jan 80

6864

130

210

12 Mar 80

13056

100

200

16 May 80

5009

210

22 May 80

7687

150

6 June 80

4962

300

12 June 80

8679

270

30 June 80

10160

531

160

180

8 July 80

3515

470

153

260

13 July 80

2440

784

541

260

17 July 80

1784

609

579

140

19 July 80

9385

1238

767

368

270

280

22 July 80

3619

1977

1 139

143

270

280

29 July 80

5215

1817

1301

570

260

4 Aug 80

4060

850

1 127

314

554

300

7 Aug 80

4734

1816

1646

794

271

190

250

12 Aug 80

2420

559

953

1207

1 10

150

13 Aug 80

2362

750

484

1355

230

14 Aug 80

3877

2145

1300

1305

270

23 Aug 80

4168

2301

2062

2524

250

280

4 Sept 80

6901

1359

1567

1591

270

Pogonomyrmex

rugosus

19 Nov 78

4569

1 10

200

25 Mar 79

3707

130

3 Apr 79

3778

120

240

8 Apr 79

14742

100

360

20 Apr 79

3115

270

27 Apr 79

11802

125

1 1 May 79

7275

160

28 May 79

10588

160

6 June 79

8033

532

300

20 June 79

9214

141 1

594

170

190

2 July 79

2485

1 136

1057

101

505

240

260

10 July 79

9374

1522

1368

270

280

24 July 79

3086

1753

1693

838

418

219

160

180

6 Aug 79

5648

2440

3072

1501

181

300

31 Aug 79

7219

1839

1565

365

342

160

270

28 Sept 79

1 1640

2204

2483

1440

350

360

27 Oct 79

4655

465

633

305

280

7 Dec 79

10538

370

400

1 Feb 80

7503

210

23 Feb 80

11239

300

Psyche

[Vol. 88

Appendix 1. cont.

* Nest queen not found.

** Nest queen found but level not recorded.

(a) Nest received extra food in June 1979.

(b) Nest received extra food in July 1979.

(d) Control nest.

(e) Nest received extra food throughout 1979 season.
See MacKay (1981) for further details.

nr= not recorded.

LABORATORY EVALUATION OF WITHIN-SPECIES,
BETWEEN-SPECIES, AND PARTHENOGENETIC
REPRODUCTION IN RETICULITERMES FLAVIPES
AND RETICU LITE RATES VIRGINICUS 1

By Ralph W. Howard,2,3 Eldon J. Mallette,2 3
Michael I. Haverty,4 and Richard V. Smythe5

Introduction

Considerable interest currently exists regarding the reproductive
strategies of social insects (Blum and Blum, 1979; Crozier, 1979).
Among termites (Order Isoptera) colony foundation by alate pairs,
fusion of existing colonies, splitting of existing colonies, and
parthenogenesis have all been reported (Nutting, 1969). Little
information is available regarding the relative importance of each of
these strategies.

The genus Reticulitermes (Rhinotermitidae) contains six Nearctic
and twelve Palearctic species, three of which have been critically
examined for reproductive modes. Pickens (1932) and Weesner
(1956) studied colony foundation of R. hesperus Banks by male +
female dealate pairs, as well as by parthenogenesis. Buchli (1950)
studied similar strategies for R. lucifugus Rossi. Clement (1979)
studied interspecific hybridization of R. santonensis Feytaud and R.
lucifugus. More limited studies on colony foundation by male +
female dealate pairs of R.flavipes (Kollar) were conducted by Beard
(1974).

Field studies with R. flavipes (Howard and Haverty, 1980)
suggest that an important reproductive strategy for this species is
colony splitting with subsequent production of numerous (several
hundred) neotenic reproductives. However, sizeable alate flights are
also a prominent feature of the biology of Reticulitermes spp. and

1 Manuscript received by the editor July 6, 1981.

2Forestry Sciences Laboratory, Southern Forest Experiment Station, P. O. Box 2008
GMF, Gulfport, MS 39503.

3Author to whom correspondence should be addressed.

4Pacific Southwest Forest and Range Experiment Station, P. O. Box 245, Berkeley,
CA 94702.

5North Central Forest and Range Experiment Station, 1992 Folwell Avenue, St.
Paul, MN 55108.

Psvche

[Vol. 88

suggest that alate-based reproductive strategies may also be impor-
tant. Field studies of colony foundation by alate pairs are extremely
difficult. We have accordingly chosen to investigate by laboratory
studies the potential of alate based reproductive modes. We have
examined incipient colony formation with males and females of the
same species, males and females of different species, and pairs of
conspecific females. Our results are reported here.

Methods and Materials

Termites

Unflown alates were collected from fallen logs in the De Soto
National Forest in southern Mississippi. R. flavipes alates were
collected from mid-March to early April of 1968, 1969, and 1970
(one source colony each year) and from mid-September to early
October of 1967, 1968, and 1969 (one source colony each year). R.
virginicus alates were collected in mid- April to mid-May of 1968,
1969, and 1970 (one source colony each year). At least 500
alates/ source colony were anesthetized with CO2 (200 ml/ min),
sexed, placed in separate petri dishes lined with moistened filter
paper, and transferred to a dark incubator at 25 +1°C for less than
one week. R. flavipes alates to be paired with R. virginicus alates
were held in their source wood for up to 30 days in an incubator at
15 +1°C until R. virginicus alates were available. All alates were
allowed to lose their wings naturally before pairing.

Pairing Procedures

Each experimental unit consisted of an 8.3- X 12.7-cm piece of
single strength window glass to which 2.5 mm X 1 cm strips of
plexiglass had been glued to form a 6.3- X 8.9-cm rectangular cell
(Howard, 1980). Washed and ovendried sand was placed in the
upper third of the cell and moistened with deionized water. Two 2-
X 4-cm X 1- to 2-mm weathered strips of southern pine were gently
inserted about 1 cm apart into the border of the moistened sand.
Termites were placed in the cell, the cell was closed by covering the
opening with four 2.5- X 7.5-cm microscopic slides, and then the
cell was sealed along the edges with hot paraffin. A small opening
(about 1 mm) was left to allow for air exchange.

Dealated termites were randomly selected, paired, and placed in
an experimental unit. Each unit was examined daily during the first
week and any termites caught in condensed moisture were freed.

1981] Howard, Mallet te, Haver tv, & S my the — Reticulitermes 77

Also during this period dead dealates were replaced with live ones.
After the first week dead dealates were removed but not replaced.
Subsequent inspections were made approximately three times a
week for the following month, then two times a week for 3 months,
once a week for 2 months, and once every 2 weeks thereafter until
the dealates were dead or the experiment was terminated. At each
inspection the number of live dealates, eggs, larvae (by stage),
presoldiers, and soldiers were recorded.

Seven combinations involving the following pairings were exam-
ined: (1) R. flavipes male + R. flavipies female (spring); (2) R.
flavipes male + R. flavipes female (fall); (3) R. virginicus male +
R. virginicus female; (4) R. flavipes male + R. virginicus female; (5)
R. virginicus male + R. flavipes female; (6) R. flavipes female + R.
flavipes female; and (7) R. virginicus female + R. virginicus female.
Each combination was replicated a minimum of ten times per year.

Determination of Larval Stage

Larvae from six of the seven combinations were randomly
selected from at least five of the experimental units in each
combination and placed in Bouin’s solution. Measurements were
then made of the number of antennal segments and head width.
These data were used for determining the instars of live larvae
through the fourth stage.

Summarization of Data

For each of the seven combinations the following two kinds of
data were gathered: (1) mean number of days to the first appearance
of an egg, 1st-, 2nd-, 3rd-, and 4th-stage larva, presoldier and
soldier; and (2) mean caste composition of incipient colonies at
selected intervals. Not all colonies were established nor observed on
the same days. To make the various pairings comparable, dates of
establishment and observation were converted to Julian dates.
Observations were then grouped by the number of days past the
data of pairing. Results of pairings replicated over 2 or 3 years were
combined. To simplify data summarization, observations were
grouped in 3-day intervals. Because of the different times of
observation, not all of the colonies were included in each 3-day
interval, giving rise to fluctuations in the number of colonies
included in the summarized data.

When both dealates had died, colonies were removed from the
experiment. Some of the colonies remained viable for more than 2

Psyche

[Vol. 88

years. However, when the number of colonies included in any 3-day
interval for any of the combinations dropped below five, we
discontinued summarization of the data for that combination, since
average caste compositions based on less than five colonies would
have little meaning.

Examination of Symbiotic Protozoa

Fourth stage or older larvae from five of the R. virginicus male +
R. flavipes female pairing and from four of the R. flavipes male +
R. viginicus female pairing were examined for the presence of
species-characteristic protozoa. R. flavipes contains mainly Dine-
nympha fimbriata Kirby, D. gracilis Leidy, Pyrsonympha major
Powell, P. vertens Leidy, Spirotrichonympha flagellata (Grassi) and
Trichonympha agilis Leidy. R. virginicus contains fewer species of
protozoa, with the primary species being D. fimbriata, P. minor, S.
flagellata, and T. agilis (Yamin, 1979).

Results

Egg and Larval Development

The time to the first appearance of eggs and larval instars was
rather similar for four of the seven combinations examined ( R .
flavipes male + R. flavipes female, spring; R. virginicus male + R.
virginicus female; R. virginicus male + R. flavipes female; and R.
virginicus female + R. virginicus female; see Table 1). Egg
production began in these combinations about 8 to 15 days after
pairing, and the first larvae were produced in about 38 to 45 days.
Fourth stage larvae were present by about 75 days after pairing.

In contrast, the fall R. flavipes male + R. flavipes female
combination required almost 25 days for egg production to begin
and almost 60 days for the first larva to be produced. Fourth stage
larvae were present by 83 days after pairing. The combination of R.
flavipes male + R. virginicus female began egg production within 15
days of pairing, but only one larva was ever produced from this
cross. This larva appeared normal and successfully molted twice
(Table 1). Although the R. flavipes female + R. flavipes female
combination was prolific in egg production, none of these eggs ever
hatched.

Soldier Production

The proportion of incipient colonies producing soldiers in the

1981] Howard, Mallette, Haverty, & Smythe—Reticulitermes 79

Table 1. Mean number of days until the appearance of the first egg and first
individual of an instar in incipient colonies of different parentage

Instar

Parentage1

Egg

Rf^ + Rf$ (spring)

8.1

38.2

46.1

52.9

74.7

Rf$ + Rf$ (fall)

24.8

59.7

65.5

73.1

83.3

RvS +

11.5

40.2

50.5

58.6

65.1

Rf(5 + Rv9

15.0

48.5

60. 42

70. 62

R vS + Rf?

11.7

42.2

50.5

57.0

Rf? + Rf?

7.0

Rv? + Rv?

14.6

45.2

57.8

64.6

1 Reticulitermes flavipes; Reticulitermes virginicus.
2Values from one experimental unit only.

3Not measured.

4No larvae ever produced.

various combinations varied from 8 to 58.8 percent (Table 2). The
R. flavipes male + R. flavipes female (spring) combination produced
only about one-third as many colonies with soldiers as did the R.
flavipes male + R. flavipes female (fall) combination. The R.
virginicus male T R. virginicus female combination, however,
produced approximately the same proportion of colonies with
soldiers as did the R. virginicus female + R. virginicus female
combination.

Temporal Colony Composition and Numbers
The relative proportions of eggs and larvae in the R. flavipes male
+ R. flavipes female (spring) combination (Fig. 1 A), R. virginicus

Table 2. Soldier production in incipient colonies of different parentage

Parentage1

Number of colonies
producing soldiers

Total number
of colonies

Percent of colonies
producing soldiers

Rf<3 + Rf? (spring)

20.3

Rf S + Rf? (fall)

58.8

Rv<3 + Rv$

8.0

RiS + Rv2$

1.2

R v$+ Rf$

25.9

Rf S + Rf3$

Rv$ + Rv$

11.1

1 Reticulitermes flavipes; Reticulitermes virginicus.

2Only one pair produced a larva.

3No eggs hatched in this crossing.

Psyche

[Vol. 88

• * *

• •••

••••

•>/A*

* * v* •

CD'

z
<
LU CM

• .

o o

COLONY AGE. DAYS

Figure 1 . Mean number of eggs and larvae produced by incipient colonies headed
by male and female dealates of the same species. A. R. flavipes (spring); B. R.
flavipes (fall); C. R. virginicus. Legend: * eggs, + larvae. Each point is a mean of
at least five replicates.

MEAN NUMBER

1981] Howard, Mallette, Haverty, & Smythe — Reticulitermes 81

Figure 2. Mean number of eggs and larvae produced by incipient colonies headed
by male and female dealates of different species. A. R. virginicus male + R.
flavipes female; B. R. flavipes male + R. virginicus female. Legend: * eggs,
+ larvae. Each point is a mean of at least five replicates.

rOOL T00Z

MEAN NUMBER

Psyche

[Vol. 88

• •

o o

CD CD

CD 00 O CM

COLONY AGE, DAYS~

Figure 3. Mean number of eggs and larvae produced by incipient colonies headed
by two female dealates of the same species. A. R. virginicus; B. R. flavipes.
Legend: * eggs, + larvae. Each point is a mean of at least five replicates.

100

1981] Howard, Mallette, Haverty, & Smythe — Reticulitermes 83

male + R. virginicus female combination (Fig. 1C), R. virginicus
male + R. flavipes female combination (Fig. 2A), and R. virginicus
female + R. virginicus female combination (Fig. 3 A) were similar at
all time intervals examined. A maximum mean number of six to
eight eggs were present by day 40 and a maximum mean number of
six to eight larvae were present by 60 to 100 days. The fall R.
flavipes male + R. flavipes female combination produced similar
numbers of eggs and larvae (Fig. 1 B) as the above combinations, but
at a slower rate.

In contrast, the R. flavipes male + R. virginicus female combina-
tion (Fig. 2B) and the R. flavipes female + R. flavipes female
combination (Fig. 3B) each produced a mean of up to 15 eggs within
the first 40 to 60 days, but produced essentially no larvae.

Symbiotic Protozoa in Progeny of Between-Species Pairings

R. flavipes male + R. virginicus female: Ten larvae beyond the
third instar from four experimental units were examined for
protozoa. Seven larvae contained protozoa, and of these, two
contained protozoa typical of R. flavipes, two contained protozoa
typical of R. virginicus, and the remaining three contained mixtures
of protozoa characteristics of both termite species.

R. virginicus male + R. flavipes female: Seventeen larvae from
five experimental units were examined for protozoa. All larvae
contained protozoa. Three contained protozoa typical of R. fla-
vipes, five contained protozoa typical of R. virginicus, and the
remaining nine contained mixtures of protozoa typical of both
termite species.

Discussion

Successful incipient colony foundation by male and female
dealates of R. flavipes and R. virginicus occurred readily in the
laboratory. The young colonies were provided with abundant food,
plentiful water, an absence of predators, and near optimum
temperatures. Despite this, the growth rate of all colonies was slow,
with no more than 20 to 30 larvae being produced within the first
year. These results agree closely with published laboratory data on
several other rhinotermitids. Buchli (1950) obtained ca. 30 individ-
uals from R. lucifugus dealate pairs after 8 months, Weesner ( 1956)
and Pickens (1932) obtained 15 to 20 individuals from R. hesperus

Psyche

[Vol. 88

dealate pairs after one year, and Beard (1974) obtained ca. 53
individuals from R. flavipes dealate pairs after one year.

King and Spink (1974) and Akhtar (1978) working with Copto-
termes formosanus Shiraki and C. heimi (Wasmann), respectively,
both obtained ca. 30 individuals from dealate pairs the first year.
Since each of these workers utilized different temperature and
rearing methods, and still obtained similar results, the observed
growth rates are probably a fair approximation of that to be
expected from field colonies. Such a growth rate implies that R.
flavipes and R. virginicus dealate pairs (and probably other
rhinotermitids as well) are K strategists (Matthew, 1976). Incipient
colonies will be successful only if the dealate pairs establish nests in
sites that are sparsely occupied by other members of the same
species, and which possess adequate food and defense requirements
necessary for slow, long term colony growth (Oster and Wilson,
1978).

Our laboratory data suggest that at least for R. virginicus, it
might be possible for female dealates alone to parthenogenetically
establish a colony with a reproductive potential equal to that of the
normal male + female dealate combination.6 This finding raises the
question of whether all progeny resulting from the R. virginicus
male + R. virginicus female combinations are sexual offspring, or
whether some fraction might have been of parthenogenetic origin.

We have also found that R. virginicus males readily mate with R.
flavipes females in the laboratory, producing apparently viable
progeny at rates comparable to those from same-species pairings. In
contrast, the pairing of R. flavipes males with R. virginicus females
results in nuptial cell construction, but only a very low rate of
progeny production. We infer from our data that the progeny of the
R. virginicus male + R. flavipes female combinations are true
interspecific hybrids rather than parthenogenetic progeny, since
paired R. flavipes females laid many eggs, but only one of them ever
hatched. Since the larvae resulting from these mixed-species mat-

6None of the females were dissected to verify the absence of sperm. All alates however
were taken from the logs before their normal flight period, and had fully developed
wings which presumably rendered them incapable of copulation within the confines
of the galleries of the logs. Furthermore, no instances are known of any termite
species that copulate until they have flown, shed their wings, and constructed a
nuptial cell. We consider it extremely unlikely that the females used in our
experiments had been inseminated.

1981] Howard, Mallette, Haverty, & Smythe — Reticulitermes 85

ings contained protozoa typical of both parents, we also infer that
the larvae engage in proctodeal feeding with both parents. We do
not know whether such mixed-species pairing occurs in the field.
The main flight periods of R. flavipes and R. virginicus are
separated by about one month (late February to early April for R.
flavipes and mid-April to mid-May for R. virginicus). But unpub-
lished records from the Forestry Sciences Laboratory in Gulfport,
Mississippi, indicate that R. flavipes, at least, may have flights every
month of the year, rendering it at least theoretically possible for
interspecific pairing to occur.

Despite the success of incipient colonies in the laboratory, their
intrinsically slow growth rates raise serious questions regarding the
importance of such pairs as a major means of population expansion.
As noted in the introduction, our field studies (Howard and
Haverty, 1980) sugget that R. flavipes frequently undergoes popula-
tion expansion by colony fission with subsequent production of
multiple neotenic reproductives. Since such new colonies pre-
sumably consist of several thousand individuals, their ability to
survive should be markedly greater than that of dealate headed
incipient colonies. It is, of course, possible that dealate individuals
or pairs could be adopted by an established colony, but we know of
no data to support such a position.

Clearly, considerably more work should be done to verify the
findings of these studies. The success of intraspecific matings is not
in question. The confounding results of the interspecific matings
demand further cytological and experimental evaluation. The
mechanisms of reproductive isolation should be clarified as well as
the integrity of these two sympatric species.

Summary

Incipient colony foundation in the laboratory by dealates of
Reticulitermes flavipes (Kollar) and R. virginicus (Banks) was used
to examine several possible reproductive strategies available to these
sympatric subterranean termite species. Successful colony forma-
tion and progeny production occurred with pairings of R. flavipes
males + R. flavipes females (from either spring or fall flights), R.
virginicus males + R. virginicus females, R. virginicus males + R.
flavipes females, and R. virginicus females + R. virginicus females.
Few progeny resulted from pairing R. flavipes males + R. virginicus

Psyche

[Vol. 88

females, or from pairing R. flavipes females + R. flavipes females.
All colony growth rates were slow, producing no more than 20 to 30
individuals within the first year.

Literature Cited

Akhtar, M. S.

1978. Some observations on swarming and development of incipient colonies
of termites of Pakistan. Pakistan J. Zool. 10(2): 283-290.

Beard, R. L.

1974. Termite biology and bait-block method of control. Conn. Agric. Exp.
Stn. Bull. 748. 19 p.

Blum, M. S., and N. A. Blum.

1979. Sexual Selection and Reproductive Competition in Insects. Academic
Press, New York. 463 p.

Buchli, H.

1950. Recherche sur la fondation et le developpement des nouvelles colonies
chez le termite Lucifuge ( Reticulitermes lucifugus Ressi [sic]). Physiol.
Comp. Oecologia 2: 145-160.

Clement, J. L.

1979. Hybridation experimentale entre Reticulitermes santonensis Feytaud et
Reticulitermes lucifugus Rossi. Ann. Sci. Nat., Zool. 1: 251-260.

Crozier, R. H.

1979. Genetics of sociality. P. 223-286. In Social Insects, Vol. 1 . H. R. Herman
(ed.). Academic Press, New York, San Francisco, and London. 437 p.

Howard, R. W.

1980. Effects of methoprene on colony foundation by alates of Reticulitermes
flavipes (Kollar). J. Ga. Entomol. Soc. 15(3): 281-285.

Howard, R. W., and M. I Haverty.

1980. Reproductives in mature colonies of Reticulitermes flavipes: abundance,
sex-ratio, and association with soldiers. Environ. Entomol. 9(4): 458-460.

King, E. G., Jr., and W. T. Spink.

1974. Laboratory studies on the biology of the Formosan subterranean termite
with primary emphasis on young colony development. Ann. Entomol.
Soc. Am. 67(6): 953-958.

Matthews, E. G.

1976. Insect Ecology. Univ. of Queensland Press, St. Lucia, Queensland. 226
P-

Nutting, W. L.

1969. Flight and colony foundation. P. 233-282. In Biology of Termites, Vol.
1. K. Krishna and F. M. Weesner (eds.). Academic Press, New York and
London. 598 p.

Oster, G. F., and E. O. Wilson.

1978. Caste and Ecology in the Social Insects. Princeton Univ. Press,
Princeton, N. J. 352 p.

1981] Howard, Mallet te, Haverty, & Smythe — Reticulitermes 87

Pickens, A. L.

1932. Distribution and life histories of the species of Reticulitermes Holmgren
in California; a study of the subterranean termites with reference to(l)
Zoogeography, and (2) life histories. Ph. D. Thesis. Univ. Calif.

Weesner, F. M.

1956. The biology of colony foundation in Reticulitermes Hesperus Banks.
Univ. Calif. Pub. Zool. 61(5): 253-314.

Yamin, M. A.

1979. Flagellates of the orders Trichomonadida Kirby, Oxymonadida Grasse,
and Hypermastigida Grassi and Foa reported from lower termites
(Isoptera families Mastotermitidae, Kalotermitidae, Hodotermitidae,
Termopsidae, Rhinotermitidae, and Serritermitidae) and from the
wood-feeding roach Cryptocercus (Dictyoptera: Cryptocercidae). Socio-
biology 4(1): 1-119.

ECOLOGY AND LIFE HISTORY OF THE
RHYTIDOPONERA IMPRESS A GROUP
(HYMENOPTERA:FORMICIDAE)

I. HABITATS, NEST SITES, AND FORAGING BEHAVIOR

By Philip S. Ward1

Department of Zoology, University of Sydney,

N.S.W. 2006, Australia

Introduction

The ponerine ants of the genus Rhytidoponera constitute a rich
assemblage of species, widespread throughout Australia, with lesser
representation in Melanesia and adjacent regions (Brown, 1958;
Wilson, 1958). On the Australian mainland they have collectively
occupied a broad range of habitats, and often rank among the more
abundant members of an ant community. Considerable interest
centers on the unusual habit, apparently widespread in the genus, of
reproduction by mated “workers” in lieu of a morphologically
differentiated dealate queen (Brown, 1953, 1954; Whelden, 1957,
1960; Haskins & Whelden, 1965).

The Rhytidoponera impressa group consists of a small, distinctive
cluster of species occurring in mesic habitats (mostly rainforest and
wet sclerophyll) along the east coast of Australia and in New
Guinea. Until recently, the impressa group was thought to comprise
no more than three species, all reproducing by means of distinct
winged queens (Brown, 1953, 1954; Haskins & Whelden, 1965).
However, recent studies of systematic relationships and colony
structure in the impressa group have revealed the presence of at least
5 close^ related species and the occurrence of reproduction by both
queens and mated workers (Ward, 1978, 1980).

There is a notable paucity of detailed ecological studies on
rainforest ponerines in general, and there have been no extensive
field studies on Rhytidoponera. This paper summarizes information
on habitat and nest site preferences, colony densities, and various
aspects of foraging, in the impressa group. A second paper describes
life cycle and reproductive patterns (Ward, 1981).

'Present address: Department of Entomology, University of California, Davis,
California 95616

Manuscript received by the editor April 15, 1981.

Psyche

[Vol. 88

Methods

Data were gathered during a survey of the Rhytidoponera
impressa group from approximately 100 mesic forest sites in eastern
Australia and New Guinea. A detailed tabulation of these collection
sites is given in Ward (1978). Field work was carried out from
October, 1974 to October, 1978, with a few additional collections in
May-July, 1980. Voucher specimens from these collections have
been deposited in the Australian National Insect Collection (ANIC),
CSIRO, Canberra.

In rainforest and wet sclerophyll forest the collection procedure
was as follows: colonies of the impressa group were sought by
examining all rotting logs, loose stones and other potential nest sites
which were encountered during a more or less random (i.e.
undirected) walk through a tract of suitable forest. In most localities
a tally was kept of the number of “potential nest sites” (logs and
stones) sampled. The “rotting log” count was confined to moist
rotten logs in middle to late stages of decay, with numerous
preformed cavities (corresponding roughly to the “zorapteran” and
“passalid” stages of Wilson, 1959), since field observations showed
that recently fallen or dessicated logs were rarely inhabited. If a
single large log was dissected in two places more than 1 meter apart
it was counted as two potential nest sites. Records from rotting logs
include a few instances where ants also nested in soil below the log.
Stones ranging in areal size from about 100 to 1500 cm2 were
recorded as potential nest sites if they rested completely on the
ground and could be easily overturned. Fallen epiphytic fern masses
on the rainforest floor were also considered potential nest sites and
were examined and counted in areas where they occurred. Almost
invariably, a single colony occupied only one nest site, so the terms
“colony” and “nest” are used in equivalently in this paper.

When an impressa group colony was located, an attempt was
usually made to collect the entire colony contents, i.e. all workers,
reproductives, and brood. This entailed considerable excavation of
rotting wood and/or soil. Where only colony fragments were
believed to be collected, this was noted.

Collected colonies were returned to the lab and their contents
enumerated. A few were maintained in modified Janet or Lubbock
nests. The majority were frozen for electrophoresis.

Field observations of foraging behavior, colony movement, alate

1981]

Ward — Rhytidoponera impressa. I

dispersal, and mating behavior were also made. In addition, field
observations and collections of related Rhytidoponera species from
Australia, New Guinea, and New Caledonia provided some com-
parative data.

Results

Habitat Preferences

The known members of the Rhytidoponera impressa group and
their respective distributions are as follows (Ward, 1980): chalybaea
Emery (= cyrus Forel), New South Wales, southern Queensland,
New Zealand (introduced); confusa Ward, Victoria, New South
Wales, southern Queensland; enigmatica Ward, New South Wales;
impressa Mayr, Queensland; and purpurea Emery (= splendida
Forel), northern Queensland, New Guinea.

Most species in the impressa group occupy a considerable range
of latitude, altitude and forest types; and all species show partial
sympatry with at least one other species (Table 1). In this context, a
sympatric association is defined as the occurrence of two (or more)
species within the dispersal range of their alates. In all cases of
sympatry, non-conspecific nests were located within several hun-
dred meters of one another, and in most instances within 50 meters.
Despite the overlap between species, differences in habitat prefer-
ences are apparent.

R. confusa is essentially a species of wet sclerophyll forest and
temperate rainforest. In Victoria and southern New South Wales it
is principally confined to lowland wet sclerophyll, and does not
occupy cool temperate rainforest of the type dominated by such
trees as Nothofagus, Quintinia, and/or Atherosperma. At the
northern limit of its range, confusa is restricted to temperate and
subtropical rainforest at moderate to high elevations. Thus, there is
an inverse relationship between elevation and latitude (Figure 1),
and the regression of altitude on latitude indicates an average shift
of about 70m per degree latitude.

In contrast to confusa, chalybaea is common in subtropical
rainforest of northern New South Wales and southern Queensland
(where confusa is rare or absent). At the southern limit of its
distribution, chalybaea is confined to disturbed lowland habitats.
Thus, in the Sydney region, it occurs commonly in well-watered
parks and gardens, and only penetrates wet sclerophyll and

Psyche

[Vol. 88

C O T3

£.2

c >

<L> • =

C/3 2> 3

<D O'

b ^ 0

til D cn

2 o r

* o V
*"2 D. ^ ' —

cuo c/3 <L> oo

.2 <- ‘ — ■

I- 4i

3 CS c

^ o "3

o £

u H „

D. ° O

e *-> 'g.

3-0

•2 <»
3 T3

! I*

c3 'S -

O >> "*t
a- — -a
a 8 e

o i ^ w

M 2 (N ^

^ X) C3

^ c/3 '3 «-h

S, “ H

5 ■=

I »
§ °

* 2
M r-~ 3

3 ON J-;

° 3 &

<3 0

3 o -3
-3 <U •§

g 2* *

o ^ -

" ° r-

n On

£ 2 s c

“ S 2 «
>> ^

!■§

O <U ^ o

° -C 3 °

— *7 E "S

*i O ■" u

3 +-* X i-

•3 i_ O O
-o OJ <t3

3 ^ D-’S

•SS^
Jl §.§ ..

3 ~ p. <*).*■*

g c - « —

«- <D

cd o
D. o

3 ^ - Z

•- .2 E jd

1) o
b oo
3 u

-g 5b _
h£ o

pu^i^j^d
imqjnqns
pus imqjft

JS9J0JUIBJ

Ajq

JS3JOJUIBJ

iBJo«n

1S9J0JUIBJ

psoidojx

JS9JOJUIBJ

jBDidojjqns

JS9JOJUIBJ

9JBJ9dui9X

JS9J0J

pAqdojgps

9SUB>1

lEuipmiqv

ggiie-y

jeuipmimi

pgipnjs

suoqBindoj

•OM

I ^

I I

E E B S £

o o o o o

o o oo >n <n

° O — o ■ ■

in in i i o

— cn
m

C/3 C/2 C/3 c/3 C/3

On On
n m

>3 3

5 -Cl

^ -b
sr a
o c:

■~ a

S ^

.£p ci, £

5 S ~

115(25) 7°-38°S 5- 1250m 24(11) 35(6) 30(3) 10(2)

1981] Ward — Rhytidoponera impressa. I 93

DEGREES LATITUDE (South)

Figure 1 . Altitude and latitude of 57 populations of confusa (open circles) and 34
populations of chalybaea (closed circles). Regressions of altitude on latitude for
confusa (upper line) and chalybaea (lower line) are highly significant (p < .001).

rainforest gullies which are ecologically very disturbed, i.e. heavily
encroached with introduced weeds such as Lantana, Ligustrum and
Tradescantia.

Sympatric associations between chalybaea and its sibling species,
confusa, occur in some of these disturbed gully sites, with confusa
preferentially occupying the vegetationally less disturbed portions
of the gully. These two species also occur sympatrically in stands of
undisturbed temperate and subtropical rainforest in northern New
South Wales and southern Queensland. In this region chalybaea
tends to occupy more xeric microhabitats than confusa, but in one
locality (an isolated patch of rainforest at Boonoo Boonoo Falls,
N.S.W.) no obvious nest site or microhabitat differences were found
between the two species, which nested within a few meters of one
another.

R. chalybaea also shows an altitudinal shift with increasing
latitude (Figure 1) and tends to occur at lower elevations than
confusa. The general picture is one of partial ecological differentia-
tion between these two species despite their very close morpho-
logical resemblance (cf. Ward, 1980).

Psyche

[Vol. 88

Table 2. Nest site records for the Rhytidoponera impressa group, excluding small,
incipient colonies (< 20 workers). Figures in parentheses represent the percentages
(for each species) of colonies occupying a given type of nest site.

Species

Rotten

Logs

Stones

Fallen

Epiphytes

Total

confusa

258

143

412

(62.6)

(34.7)

(2.7)

chalybaea

145

165

(87.9)

(11.5)

(0.6)

impressa

(92.9)

(7.1)

(0.0)

purpurea

(100.0)

(0.0)

enigmatica

(0.0)

(100.0)

(0.0)

all species

450

(69.7)

184

(28.5)

(1.9)

646

R. enigmatica is a localized species, known only from wet
sclerophyll vegetation in sandstone gullies (6 sites, including two
ANIC records) and urban parkland (1 site), the latter record coming
from an area where the original habitat would have been sandstone
gully vegetation. The range of elevation from which it has been
recorded is 10 to 180 meters. Thus, with regard to habitat preference
enigmatica is the most stenotopic species. Most of the known
populations are in sympatry with, or in close proximity to,
populations of confusa and/or chalybaea.

The 7 impressa populations studied come from tropical rainforest
(1), subtropical rainforest (5), and dry rainforest (1). These data,
along with 30 other collection records in the ANIC, indicate that
impressa is confined to Queensland rainforest at altitudes ranging
from 30m to 1050m.

Based on the 12 populations studied here plus additional records
from the ANIC and from Wilson (1958), purpurea is recorded from
subtropical and tropical rainforest (and one population from dry
microphyll rainforest on the Mt. Windsor Tableland) in northern
Queensland (30m to 1200m), and from tropical montane rainforest
(600m to 1300m) in Papua New Guinea. In north Queensland it
occurs in both primary-growth and partially disturbed rainforest,

1981]

Ward — Rhytidoponera impressa. I

while New Guinea records indicate a predilection for second-growth
montane rainforest.

Nest Site Preferences and Densities

Members of the impressa group are found nesting mostly in
rotten logs and under stones. Nests are multi-chambered, but not
highly fragmented, seldom penetrating deeper than 15- 20cm into
soil, or occupying more than lm length of rotting log. Nest
entrances are cryptic, without conspicuous mounds of excavated
material.

Fallen epiphytes on the rainforest floor are occasionally utilized
as nest sites by confusa and chalybaea. Duringthe present study no
colonies were found in living epiphytes on trees, although there are
single records of a colony-founding purpurea queen (Brown, 1954)
and a mature purpurea colony (Wilson, 1958) from fern epiphytes
on rainforest trees.

Nest site records from the present study are summarized in Table
2 which lists, for each species, the number of colonies collected from
rotten logs, under stones, and in fallen epiphytes. Excluded from
this table are a small number of single records from other nest sites.
Thus confusa was also found nesting in a Banksia lignotuber, in a
rotting bracket fungus, directly in the soil, and (twice) in an
abandoned termite mound in rainforest. A chalybaea colony was
located under the bark sheath of an Archontophoenix palm, and in
urban areas this species occupied less orthodox nest sites (e.g. in and
under rusting metal, under concrete slabs, and in crevices along a
stone wall). Three purpurea colonies (two in north Queensland, one
in Papua New Guinea) were observed nesting in cavities in the
trunks of living rainforest trees, and in New Guinea this species may
be primarily an arboreal nester (Wilson, 1958; records in ANIC).

Table 2 shows that there is a clear trend towards greater
specialization in the rotten log nest site in species of more tropical
latitudes. The difference between confusa and chalybaea with
respect to numbers of logs and stones utilized is highly significant
(x? = 33.0, p < .001) and the difference between chalybaea and
purpurea is also significant (x? = 4.4, p < .05). In contrast to all
others, enigmatica (the localized species of wet sclerophyll gullies)
appears to nest exclusively under stones.

In 70 populations (from 63 localities, due to some sympatry) a
tally was kept of the number of “potential” nest sites (rotten logs,

Psyche

[Vol. 88

00 I .0

LlJ

m -b h

A -

.2-

.2 .4 .6 .8 1.0

FREQUENCY OF POTENTIAL NEST SITES

Figure 2. Within-species frequencies of utilized nest sites as a function of potential
nest site frequencies, for 5 impressa group species. Closed circles refer to log nest sites,
open circles to stones.

stones, fallen epiphytes) encountered as well as the number of actual
nest site occupancies (Table 3). It seems clear that nest site
availability varies from species to species. For both rotten log and
stone nest sites there are positive correlations (r = 0.94, p < .02, in
both instances, arcsine transformed data) between the proportion of
a species’ colonies found in a particular nest site and the relative
frequency of that nest site for the species (Figure 2). This suggests
that species-specific preferences are partly a function of nest site
availability. (No such correlation is found for fallen epiphytes —
confusa showns the highest preference for this nest site despite its
relative rarity in the southern rainforests; however, the numbers are
in all instances rather low.)

The relative abundances of species can be crudely compared by

1981]

Ward — Rhytidoponera impressa. I

Table 3. Numbers of potential nest sites (pns) sampled and actual nests encoun-
tered, for 70 impressa group populations.

Species

No.

populations

Logs

Stones Epiphytes

Total

confusa

no. pns

1838

2984

4914

no. nests

227

333

nests/ pns

.124

.033

.087

.068

chalybaea

no. pns

1 164

1136

2370

no. nests

141

159

nests/ pns

.121

.015

.014

.067

impressa

no. pns

260

126

393

no. nests

nests/ pns

.031

.008

.000

.023

purpurea

no. pns

404

109

537

no. nests

nests/ pns

.052

.000

.039

enigmatica

no. pns

105

561

666

no. nests

nests/ pns

.000

.027

all species

no. pns

3771

4916

193

8880

no. nests

397

131

537

nests/ pns

.105

.270

.047

.060

examining the proportion of potential nest sites which are occupied.
(The desirable complementary data on absolute densities of poten-
tial nest sites for different geographical regions and habitats are not
available). Comparing the density figures (Table 3) for confusa and
chalybaea, the former occupies a significantly greater proportion of
stone nest sites than chalybaea (x? = 9.7, p < .01), but no differences
exist in the proportion of suitable rotten logs occupied, and the
overall nest densities (considering all potential nest sites) are the
same for the two species. Nest densities are considerably lower for
impressa, purpurea, and enigmatica. Rhytidoponera confusa and
chalybaea utilize a significantly greater proportion of rotten logs
than impressa and purpurea (contingency x\ p <.001, for all four
comparisons), despite the greater importance of rotting logs as nest
sites in the more northerly (tropical) species. This may be partly the
result of greater competition for nest sites in the species-rich tropical
rainforests. R. confusa and chalybaea are often common and
dominant ants in temperate and subtropical rainforests, respec-
tively, of New South Wales and southern Queensland where the

Psyche

[Vol. 88

numbers of sympatric rainforest ant species are probably about one-
quarter to one-half that experienced by purpurea in north Queens-
land rainforest.

It is unclear why there is a disproportionate decline in the
utilization of stones as nest sites in the more tropical members of the
impressa group (Table 3) and perhaps for tropical rainforest ants in
general (cf. Wilson, 1959, p. 440). One possibility is that in
subtropical and tropical rainforests on well-drained soils, stones
frequently lie on subsoil below the thin organic horizon and offer an
environment poorer in immediate food resources and more de-
manding for nest excavation than rotting logs. In temperate and
some subtropical rainforests of New South Wales, soil horizons
tend to be less sharply stratified and/or litter decomposition is
slower, so that humic material extends below the level of loose
stones.

Effects of Sympatry

Nest site densities for sympatric and allopatric populations of
confusa and chalybaea are given in Table 4. Both species occupy a
significantly greater proportion of log nest sites in allopatric
populations (contingency x2, p < .01 and p < .001, for confusa and
chalybaea respectively) and confusa inhabits a greater proportion of
stone nests sites allopatrically (x2 = 5.4, p < .05). The lower
sympatric densities of confusa and chalybaea could be a result of
sympatric associations occurring in more marginal environments.
However, the combined sympatric nest densities are very similar to
the allopatric densities of both species. There are no significant
differences between the total proportion of rotting logs occupied
sympatrically and the proportion utilized allopatrically by either
confusa (x2 = 0.7) or chalybaea (x2 = 1.8). The combined sympatric
nest density under stones is the same as that for allopatric confusa
populations. While these results could be coincindental, it seems
more reasonable to conclude that sympatry has a depressant effect
on relative abundance, and that competition for nest sites, food, or
foraging space is important.

Other Sympatric Congeners

Other, more distantly related Rhytidoponera species also co-
occur with members of the impressa group. R. victoriae (s.l.) is a
common species (or complex of species) present in rainforest and
other mesic habitats along the entire east coast of Australia. R.

1981]

Ward — Rhytidoponera impressa. I

O (3
<1

3 ^

Z. ’2

XI >

b s.

oo r- cn Tt

o >r

& *
o 3

£ °

£5

No. potential
nest sites

650

2334

650

486

.0 ^3

Proport

occupit

.071

.133

.043

.146

oc,

No. potential
nest sites

280

1558

280

884

< 3

53 Js

t >

1 8

3 -G

5 1

ic wi
c)

itric

trie)

o3 ~ & a,

si * =

« o y\ os

■Ss ,»

a <3 33

5 S5 <) <>

e 3 3
O O -3 C

3 3 3 3

100

Psyche

[Vol. 88

victoriae is considerably smaller than the impressa group species,
and nests preferentially under stones.

In some north Queensland localities, purpurea or impressa
coexist with one of several small Rhytidoponera species (e.g.
chnoopyx and kurandensis nesting in logs and under stones) and
with one of several larger species (scaberimma and related species,
nesting in logs and directly in the soil). There are no rainforest
Rhytidoponera of comparable size to the impressa group species
that regularly coexist with the latter with the exception of croesus
(s.l.), which nests in rotten logs and in tree trunks in rainforest and
wet sclerophyll of New South Wales and southern Queensland. R.
croesus appears to be generally uncommon, and in fact averages
slightly smaller than chalybaea, confusa and impressa to a degree
which may significantly reduce prey size overlap (see below).

Colonies of other Rhytidoponera species are virtually never
found occupying the same nest site as an impressa group colony
even though other medium to large ponerines such as Amblyopone
australis, Leptogenys hackeri and Prionogenys podenzanai are
occasionally found nesting in close proximity to an impressa group
colony (e.g. under the same stone, or in adjacent cavities in a log).

Colony Movement

It appears that species in the impressa group are prone to move
colonies from one nest site to another rather frequently. For
example, in one rainforest population of confusa (Royal National
Park, N.S.W.) eight stones under which colonies had been briefly
located and otherwise left undisturbed were examined one week
later: half were unoccupied. Three weeks later, only two colonies
remained under the stones. While the censussing no doubt consti-
tuted a disturbance conducive to nest-movement, it demonstrates
nevertheless the readiness with which colony movement is carried
out.

During the course of field collections, vacated nest chambers were
occasionally encountered (under stones or in rotten logs) whose
previous occupants could be traced to an impressa group species on
the basis of cocoon remains in the middens. Moreover, colony
movement involving transport of brood and other workers was
observed several times in chalybaea (and in other Rhytidoponera
species outside the impressa group) (Ward, 1981).

1981]

Ward — Rhytidoponera impressa. I

101

Foraging and Food- Retrieval

Members of the Rhytidoponera impressa group are partly
predacious on other arthropods, but also scavenge for dead insects,
seeds, animal feces, etc. Capture of live prey is achieved by a short
lunge forward, coincindent with rapid closure of the outstretched
mandibles. Prey thus captured are subdued by stinging.

In most species, foraging occurs principally on the ground,
among leaf litter and rotting logs. However, purpurea workers were
frequently observed foraging on low foliage of understorey plants,
as well as on the rainforest floor, in north Queensland. In Papua
New Guinea this species nests (at least partly) arboreally, but limited
observations (Wau; September, 1975) suggests that it tends to
forage downward from the nest entrance. Urban and suburban
populations of chalybaea, noted for their unusual nest sites (above),
usually forage on the ground and on low vegetation, in damp tree-
shaded situations. On one occasion chalybaea workers were ob-
served foraging in a house in an urban residential area of Sydney.

Foraging is not restricted to any particular time of the day or
season, although activity decreases noticeably towards the middle of
the day (and in the winter). Periods of clear warm weather after rain
seem particularly conducive to high levels of foraging activity.

Field observations indicate that workers are usually lone foragers,
although occasionally several individuals co-operatively transport a
large food item back to the nest. Sometimes this occurs close to the
nest entrance, seemingly as a result of fortuitous encounters of a
heavily-laden forager with other workers. In lab colonies of
chalybaea, single workers struggling with a large prey item in a food
arena were observed to make movements of the gaster suggesting
stridulation. On the other hand, chemical recruitment to food
sources does occur, although this behavior is rudimentary in
comparison to the mass-recruitment patterns of some higher ants. It
is readily demonstrated by placing large food baits (e.g. chunks of
tuna fish or large insects) close to a nest. Workers which discover
the food and return to the nest with a portion of the bait can be
observed dragging the tips of their gasters along the ground, and
subsequent outward-bound foragers follow the same path to the
food (field observations on chalybaea and purpurea). Large pieces
of the bait are retrieved co-operatively by several workers; smaller
portions are carried by single foragers.

102

Psyche

[Vol. 88

When such baiting experiments are carried out, there appears to
be little active defense of the food by Rhytidoponera workers. When
baits are partially occupied by other smaller but mass-recruiting ant
species, such as Pheidole, Rhytidoponera workers adopt a “grab-
and-run” strategy. This is illustrated by the following observations
on purpurea in rainforest near Cape Tribulation, north Queensland
(5 June 1980).

A purpurea colony was located in the trunk of a living palm tree,
in a cavity 60cm above ground. Workers were foraging down the
palm trunk and on the adjacent rainforest floor. A small chunk of
tuna fish was placed on a stone, 1.5m from the palm tree, and close
to a purpurea forager which soon located the bait. It grasped a small
piece of the tuna and returned to the nest, dragging the end of its
gaster along the ground. A few minutes later, a worker (possibly the
same individual) emerged from the nest entrance and returned to the
bait by exactly the same trail. By this time, the remaining tuna bait
was in two pieces, each attended by 2-3 workers of a Meranoplus
sp. The purpurea worker carefully circled around one piece of tuna
to an unoccupied corner and grabbed it, inadvertently getting a
Meranoplus worker at the same time. The two briefly grappled, and
the purpurea worker dropped the food and retreated several
centimeters. It then approached the second piece of tuna, edged in
towards another exposed corner, swiftly grabbed it (this time
without a Meranoplus worker), and hurriedly departed for the nest
by a different route.

Unrecruited workers of the impressa group apparently forage
randomly, without laying a continuous odour trail, but upon
locating food they return directly to the nest. It is unclear what
method(s) of orientation are utilized. Any explanation must take
into account the observation that foraging occurs nocturnally as
well as diurnally (at least in confusa and chalybaea).

Food Diversity and Size

The great majority of food items collected by impressa group
workers are small, individual objects brought in by single foragers.
Eighty-one food items were returned to a single chalybaea nest
observed over a total of 8 hours (Table 5). Of these, one item (an
earthworm) was transported by four workers; the remaining food
items (encompassing 56 arthropods, 17 Ficus seeds or pieces of fruit,
and 7 pieces of miscellaneous organic material) were carried by

1981]

Ward — Rhytidoponera impressa. I

103

-C

■5

.© 00

O O
o <U

t 6

oo '■£

<— > c

o .2

.s 5

j e:

?•§

H 2

2^
cd u
D.^

OO— hOOOO — <N — <N — — < <N © m m <N <N — — H CN — * I I I v©

— I — . >/-> <N — coo— • — « O O <N © © © ~ O — © ro <N CN © © I

i-, u o

<D

Xj M >

Uh U- L_

O O o

£ £ c

dd loo

£ £ «
i- i- cd
2 .2, *

Sij

(N <N

-Si -Si

I :§

CL, ft,

C i
%

<u <u

Cd 03

t &

<u
cd
-a

•e

.2 £ ~

5 0712 2 2
. . . . <D <U

cd cd

3
^ T3
*S cd
3 w

■a 2
cd ,2
' — ' JO

flj

cd «

*2 8
:c o

S £3

X U .3

§•3
8 T3
>> cd

2 e

4>
cd
T3 2
■ — cd

^.*2
x> £
o 2

6 I

O g
e **

W O

a «2 _

0 aa
8 ° °
g'2’2 «
8 'E/o, «

D U

1 J J

a&a

o o o

4) o 4>

B'.&’o o o
QQUUU

4) <U
0,0..
o 2

6 x

o 6

X w .

c« jrt cd
t; •§ T3

S3 © ©

o.-g 2-

w C j-
O O D,
<o =3 e
2 0 0
£ U <

Isopoda: Oniscidae . . .

Diplopoda

Annelida: Lumbricidae

Ficus seeds

Ficus fruit (pulp)

104

Psyche

[Vol. 88

Table 6. List of 1 9 food items returned to a single nest of Rhytidoponera croesus
s.l. (Royal National Park, N.S.W., 26 January, 1976) over a three-hour observa-

tion period.

Hymenoptera: Formicidae: Paratrechina sp. (worker) 1

" " Solenopsis sp. (worker) 1

" " Chelaner sp. (worker) 1

" " Myrmicinae (male alate) 1

Hymenoptera: Pteromalidae (adult) 1

Lepidoptera: adult microlepidopteran 1

Lepidoptera: larvae 1

Diptera: Nematocera (adults) 2

Diptera: Brachycera (adults) 2

Coleoptera: Chrysomelidae (adult) 1

Homoptera: Coccoidea (nymph) 1

Homoptera: Cicadellidae (nymph) 1

Insect larva, unidentified 1

Unidentified insect legs 2

Acarina (small mite) 1

Mammalian (?) excrement, with veg. matter and insect parts 1

single workers. Thirty-one (55%) of the 56 arthropod items were
alive when retrieved from their captors (near the nest entrance).
Some of the remaining items may have been killed or paralyzed
during capture; others were clearly scavenged as dead material.

It is of some interest to note that 19 (34%) of the 56 arthropod
items consisted of other ant species (including alates). Some of these
ants, particularly alates, may have been injured or dying when
collected. On the other hand, predation on healthy, active worker
ants was observed first-hand in the field: chalybaea workers from
the Sydney University population were seen preying at the soil
entrances of Pheidole nests, grabbing workers as they emerged.

For comparison with another similar-sized, rainforest species of
Rhytidoponera outside the impressa group, Table 6 lists the food
items returned to a Rhytidoponera croesus nest over a three-hour
observation period. The mean head widths for workers of croesus
and chalybaea are 1.25 ± 0.03 s.d. (n=8) and 1.36 mm ± 0.08 s.d.
(n=80), respectively. Although there is considerable similarity in
food items taken by the two species as measured by ordinal
taxonomic categories, an analysis of food size (Figure 3) reveals that
the mean food item length of croesus (2.5 mm) is significantly less
than that of chalybaea (3.5 mm) (t-test, p< .02). However, the food

NO. OF ITEMS

1981]

Ward — Rhytidoponera impressa. I

105

Figure 3. Frequency distributions of the lengths of food items taken from 80
chalybaea foragers and 19 croesus foragers (see Tables 5 and 6). Each distribution is
based on workers from one colony only.

size distributions are based on limited single-nest samples, and there
is likely to be significant temporal and spatial heterogeneity within,
as well as between, species.

Additional studies on food item diversity and overlap in Rhy-
tidoponera are desirable. Such studies are feasible for ants which are
primarily lone-foraging predators and scavengers, because of the
discrete, visible nature of most foraged items. However, difficulties
remain in assessing the importance of honeydew and other liquid
foods, which may be carried in the crop as well as between the
mandibles.

Two species in the impressa group were recorded collecting
honeydew from homopterans. Workers of chalybaea were seen
tending coccids on a fresh shoot emerging from the trunk of a
camphor laurel tree ( Cinnamomum eamphora), in the Sydney
region. R. purpurea workers were observed tending aphids on
ginger plants ( Alpinia caerulaea ) in several places at Lake Eacham,
Qld.

106

Psyche

[Vol. 88

In one of the latter instances, observations were made inter-
mittently over a period of 2 days, during which time a force of about
15 workers was regularly maintained on the plant. These workers
gave outward-facing aggressive displays (mandibles barred) when
the plant was disturbed. A small contingency of workers was also
clustered among leaf litter at the base of the plant, apparently
controlling access to the plant and aphids. Detense of “tending
rights” may be important since other aggressive, aphid-tending ants
such as Pheidole were present in the same locality. The colony of the
purpurea workers was located in a rotten log 5m distant. Workers
returning to the colony from the ginger plant showed high fidelity to
a particular route which involved following the ground for half the
distance and then proceeding along a decumbent liana (one of
many) which led back to the log.

Thus, despite the “lone forager” status of most impressa group
workers, short-term recruitment, co-operative food retrieval, and
(in at least one species) persistent, long-range trails, may be used.
Excepting persistent trails, species in the impressa group appear to
show a level of individual foraging and recruitment similar to that
described for the myrmicine ant, Novomessor (Holldobler, et al.,
1978).

The species in the impressa group with the most sophisticated
foraging and recruitment behavior {purpurea ) is the only member
whose colonies are entirely monogynous and queenright. It is
tempting to speculate that widespread polygyny and worker repro-
duction in other Rhytidoponera species may have constrained
ergonomic improvements because of a reduction in the efficacy of
colony-level selection (cf. Oster & Wilson, 1978).

Summary

The five known species of the Rhytidoponera impressa group
collectively inhabit a variety of mesic forest habitats (from wet
sclerophyll to tropical rainforest) along the east coast of Australia,
with one species {purpurea ) also occurring in montane rainforest of
New Guinea. R. chalybaea has invaded mesic anthropogenic
habitats (parks and gardens) in the Sydney region. All species show
partial sympatry with at least one other species.

Most colonies are located in rotten logs or under stones. There
are significant differences between species in the frequencies with

1981]

Ward — Rhytidoponera impressa. I

107

which different nest sites are utilized, and these preferences are
correlated with the availability of potential nest sites. The more
tropical species ( impressa and purpurea) show a stronger preference
for rotten logs, but occur at lower nest densities, than inhabitants of
temperate and subtropical rainforest (< confusa and chalybaea).
Where confusa and chalybaea occur sympatrically, they have
significantly lower nest densities than allopatrically.

Workers of the impressa group are generally lone-foraging
predators and scavengers, but co-operative food retrieval and
recruitment to food sources occur to a limited degree. The majority
of food items are small arthropods: other ant species may be a
significant component of the diet. Foraging usually occurs among
leaf litter and logs on the ground but at least two species ( chalybaea
and purpurea ) also forage on low foliage and tend homopterans.

References

Brown, W. L.

1953. Characters and synonymies among the genera of ants. Part I. Breviora,
11, 1-13.

1954. Systematic and other notes on some of the smaller species of the ant
genus Rhytidoponera Mayr. Breviora, 33, 1 11.

1958. Contributions toward a reclassification of the Formicidae. II. Tribe
Ectatommini (Hymenoptera). Bull. Mus. Comp. Zool. Harvard, 118,
175-362.

Haskins, C. P. and W. M. Whelden.

1965. “Queenlessness”, worker sibship, and colony versus population structure
in the formicid genus Rhytidoponera. Psyche, 72, 87-112.
Holldobler, B., R. C. Stanton and H. Markl.

1978. Recruitment and food-retrieving behavior in Novomessor (Formicidae,
Hymenoptera). I. Chemical signals. Behav. Ecol. Sociobiol., 4, 163-181.
Oster, G. F. and E. O. Wilson

1978. Caste and ecology in the social insects. Princeton University Press,
Princeton, N.J.

Specht, R. L., E. M. Roe and V. H. Boughton

1974. Conservation of major plant communities in Australia and Papua New
Guinea. Aust. J. Bot. Suppl. No. 7.

Ward, P. S.

1978. Genetic variation, colony structure, and social behaviour in the Rhy-
tidoponera impressa group, a species complex of ponerine ants. Ph.D.
thesis. University of Sydney.

1980. A systematic revision of the Rhytidoponera impressa group (Hymenop-
tera: Formicidae) in Australia and New Guinea. Aust. .1. Zool. 28,
475-498.

108

Psyche

[Vol. 88

1981. Ecology and life history of the Rhytidoponera impressa group (Hymen-
optera: Formicidae). II. Colony origin, seasonal cycles, and reproduc-
tion. Psyche, 88: 109-126.

Webb, L. S.

1978. A structural comparison of New Zealand and south-east Australian rain
forests and their tropical affinities. Aust. J. Ecol. , 7-21.

Whelden, W. M.

1957. Anatomy of Rhytidoponera convexa. Ann. Ent. Soc. Am., 50, 271-282.
1960. Anatomy of Rhytidoponera metalliea. Ann. Ent. Soc. Am., 53, 793-808.

Wilson, E. O.

1958. Studies on the ant fauna of Melanesia. III. Rhytidoponera in western
Melanesia and the Moluccas. IV. The tribe Ponerini. Bull. Mus. Comp.
Zool. Harvard, 119, 303-371.

1959. Some ecological characteristics of ants in New Guinea rain forests.
Ecology, 40, 437-447.

ECOLOGY AND LIFE HISTORY OF THE
RHYTIDOPONERA IMPRESS A GROUP
(HYMENOPTERA:FORMICIDAE)

II. COLONY ORIGIN, SEASONAL CYCLES,

AND REPRODUCTION

By Philip S. Ward1

Department of Zoology, University of Sydney,

N.S.W. 2006, Australia

Introduction

This paper is concerned with colony foundation and with
seasonal cycles in brood composition and alate production in the
Rhytidoponera impressa group, a species complex of ponerine ants
restricted to rainforest and other mesic habitats in eastern Australia
and New Guinea.

Life cycle information is most complete for confusa and chaly-
baea, and most of what follows refers to those species. Relevant
data on the other three members of the impressa group ( enigmatica ,
impressa, and purpurea ) are given where available. When pertinent
to the discussion, some observations on related Rhytidoponera
species outside the impressa group are also included.

Methods

Collection methods are described in Ward (1981). Most of the
data are based on field observations and collections. Where ap-
propriate, suspected reproductive females were dissected to ascer-
tain the condition of the ovaries and spermatheca.

Results

Colony origin

In the Rhytidoponera impressa group there are two methods by
which colonies can originate:

(i) from lone, colony-founding winged females (queens), in the
manner characteristic of many ants; or

‘Present address: Department of Entomology, University of California, Davis,
California 95616

Manuscript received by the editor April 15, 1981.

109

110

Psyche

[Vol. 88

(ii) as a result of colony fission or budding (hesmosis), in which
one or more mated “workers”, accompanied by uninsemi-
nated nest-mates, leave the parent colony to found a new
daughter nest.

As the foregoing remarks imply, there are two kinds of reproduc-
tive females: queens and ergatoid (worker-like) gynes, the latter
indistinguishable morphologically from unmated workers. This is
the first record of reproductive workers in the impressa group (they
are common and well-documented in some other Rhytidoponera )
where previous reports suggested that the only functional reproduc-
tives were winged queens (cf. Brown, 1953, 1954; Haskins &
Whelden, 1965). Mated queens and ergatoid gynes never coexist in
the same nest, but they often occur in different nests in the same
population (in confusa, chalybaea and impressa ). This rather
remarkable dimorphism of female reproductives in the impressa
group and the resulting differences in colony structure and genetic
relatedness will be examined in more detail elsewhere (Ward, in
prep.).

There is little information on the frequency of colony fission in
worker-reproductive colonies or on the size of newly-budded
daughter colonies. Occasionally small isolated clusters of workers
and brood are seen in the field, under stones or in rotten log cavities.
Table 1 summarizes the composition of four such clusters in the
impressa group, and two from other Rhytidoponera species ( tas -
maniensis and fulgens). Similar observations were made by Haskins
& Whelden (1965) on R. metallica. Note that in the two cases where
workers were dissected (Table 1), only one individual in each cluster
was found to be inseminated. In no instance in the impressa group
(or in any other Rhytidoponera species) was a single isolated
worker, with brood, located in the field, in contrast to the frequent
occurrence of single colony-founding queens (see below).

The process of colony fission is observationally difficult to
distinguish from the movement of a colony from one nest site to
another, and the two events may be inter-related. Table 2 summar-
izes observations made on colony movement in chalybaea and in
three other Rhytidoponera species (outside the impressa group). In
only one instance ( maniae ) was a single colony observed splitting
into two nests, but the same event may have been occurring during
the other observations, if some workers remained at the original
nest site.

1981]

Ward — Rhytidoponera impressa. II

Table 1. Composition of small, isolated clusters of workers and brood (incipient
worker-reproductive colonies?) in Rhytidoponera confusa, chalybaea, tasmaniensis,
and fulgens.

Species

Locality

Date

No.

workers

Brood

confusa

Royal Natl. Park,

5.xi. 1974

eggs, larvae

N.S.W.

Pearl Beach, N.S.W.

9.iv. 1977

25*

9 eggs, 8 larvae

Seal Rocks, N.S.W.

14. vi. 1977

several larvae

chalybaea

Whian Whian State

14. v. 1977

several larvae

Forest, N.S.W.

tasmaniensis

nr. Wonboyn Lake,

25.x. 1975

none seen

N.S.W.

fulgens

Mt. Koghis,

18. ii. 1977

several larvae.

New Caledonia

one worker cocoon

♦workers dissected, one inseminated.

In view of the apparent scarcity of very small isolated clusters of
workers and brood (of the size documented in Table 1, i.e. 5-25
workers), it seems likely that colony fission in the impressa group
often produces daughter colonies larger in size. (Mature worker-
reproductive colonies, i.e. those with alates, contain, on average,
about 150 workers.) More field observations on budding are
needed; the small amount of information accumulated thus far
suggests that nocturnal observations might be rewarding. It is also
possible that some worker-reproductive colonies develop from
former queen-right colonies in which the queen has died.

The origin and development of queen-founded colonies in the
impressa group has been more extensively documented. Incipient
queen-right colonies have been observed repeatedly in the field
(Table 3). Mated queens apparently disperse for some distance,
undergo dealation, and search for a suitable nest site (under stones,
rotting logs, etc.). Having located shelter, the queen excavates a
small cavity, lays several eggs, and rears a small brood of workers,
the first of these appearing within about 6 months (3-4 months in
lab colonies). Unlike the claustral colony foundation typical of
higher ants, queens forage outside the nest for food, and feed their
larvae partly on insect prey.

The available field information on incipient, queenright colonies
suggests that they are usually founded in the spring and early

Table 2. Field observations on colony movement and worker transport in Rhytidoponera.

Time

Species Date Locality (EDT) Weather Observations

112

Psyche

[Vol. 88

8 .2

5 ^

cz <&

t— I

c '-5

t/5 g
a ea

C 1/3

O ID

c/3 'l“'

D 2

-3 o
O <2
CO D

g 2

V> 3

o ^

&1 73

3 . ,

£ I
I ."2

o &
£ c

« 3

> t-

d o

>3 M

"O .2
3

_o
73

c o

0 +J

c «

1 -g

n. 2

c/3

cZ cd

£ (-HH

^ O

<D Qh
U 0

% 2

S' 3

2 a
2

D 3
■£

o O

T3 u

3 _o
2 d

o 5

o ^

8 -

o d

D g
cd f3
> £2

0 £

1 d
cd g

3* —

£ c
d 2

° <u

! ?

cd c

,3 '-3

^ 2

§ s

3 D

2 Id cd

Si £ ~

£ is S

H . . ”3

53 c ~

rv '3 i+h
2 O

§ £ 3

3 'td 3

c/3 3 3

O O
£ £

§ 2
3

S 3

2 ^
c/3 DO
<U 60

3 ^

O DO

I 'I

C c/5 O

> <5 ^

S oji

£ * O

O 3 7D
8^3

S3 2

o S

<u — -

2 ©
O c/3

2 .2
is

3 C
3

D. 2

G J-h O

(U iu Z,

c/3 ^

1 §

DO £

c/3 3

3 P

3 C

C/3 D
3 U
D 3
-* 3

O C/3

£ .2

DO 'C
-1 *
■§ ! i

£ O 3

8 e

o d
m 3

>T3 m c

o E.

VO O-,

vd ci.

Ov — 3

GO ‘g

0 5?

• o

1 3 U
. . DO

c; d

-GJ flJ
^ >

_ap

^ «»

^ C ^

C/5 3

CO 3

co 3 t:

Z D

£ D

z ^ u

976

1977

O O d

m r*d G

ri 3 d

-Ci

-C

.xi.1975 NSW: 15 km c.9.00 cloudy workers carrying brood and other workers to one of

N Coombah a.m. two new nest sites, 0.6 and 2.5 meters, respectively,

from old nest site, the directions at right angles to one
another; all nests directly in soil

Table 3. Field data on 43 incipient, queenright colonies (with < 20 workers). All Rhytidoponera confusa except the following
accessions: 2006 ( chalybaea ), 2620 ( chalybaea ) and 2580 ( impressa ).

Brood* Probable

Accession Population Dealate year of

no. code no. Date Female(s) Workers Eggs Larvae Cocoons origin

1981]

Ward — Rhytidoponera impressa. II

113

m r*~) m
r- r-
® O' O'

't t 't

r- r-~ r-

Ov O' O'

Tf Tt Tt Tt Tf

r — t — r — i~-~

O O' O' O' O'

r--

r-~

O' O' O' O' O' O' O' O' O'

O-j 1- OOOOOO IOOO I l©H h° 1+0 +

+ + + OOOOOO I+ + + I ,+ + + + I + O +

oo + oo©H — | — |- i-| — (- + i ioH — |-o i o + o

m<NTf OOOOOOOOOOOOOOOO(0-OOrn

* * ■*

—I 'O
ON (N

_a. . t M xj xf xl

+ + + + + + +

*. * '* * * 'y '* 'y

Tj- rf Tj-

t" f- ^

'y 'y 'y +

O' O' oi O'

— — ON

m «/-> JC?

r — r~~

r-~ r-~
> >
o o

r-~ r-' r-~

(N OO DO 00
- (N (N (N

'O rn OO OO
— — rr) r<->

m on O'
r- o r-

m Tt Tt

or^ooo— ■Tj-r'— 'ONmcn— — oo»^>
r-r-mr-fN>roiAovnoO'Ooo>o«ri«ro«r)t^-ocoo
fNfNfnm^t^tTfTtTtTtTtrtiniy-iiomiromoooo

Table 3 (continued).

114

Psyche

[Vol. 88

2 ° .s

o S .2P

<U t-

CU O

^3 ed

O >

5 s

to J

cd «

3 g
« 6
Q <U

a o

S c

■d d1 d- Tf

r- r- r- r-
O' O' O' O'

r- r- r-
r- r-"

NO NO *0 NO NO NO NO NO NO 'O NO
r-r-r-r-r-r-r-r-r-r-r-
O' O' O' O' O' O' O' O' O' O' O'

«n

r-~

t —

r»

I —

’>

’x

r-‘

[T; £; r- r- r'-

r': r- i" r-

.> .► > > -•

^ ® £ ~ o;

X X
oo on

oo — < o m
© oo t~- (N
oo oo O' m

no oo o

no Tf rsi
m >n ©
— — <N

— 00'0'0(NmTTt~-<NOO
hXO'OMf'l'J'thaOM
O'O'O'OOTtTj-i^Tj-m'o

— — - MM(N(N(N(N(N(N

O' o
Tf «n
00 oo

<N <N

brood: +, present; 0, absent; -, no information

1981]

Ward — Rhytidoponera impressa. II

115

Table 4. Composition of 322 confusa colonies, with respect to numbers of cocoons
and alates, and month of collection. Large standard deviations are due to variation in
colony size, and to the fact that not all nests produce alates in a given season.

Month

Sample Size
<#colonies)

# worker cocoons
mean ± S.D.

# alate cocoons
mean ± S.D.

# alates
mean ± S.D.

Sept.

0.0± 0.0

0.0+ 0.0

10.8+13.9

Oct.

0.2± 0.5

0.0+ 0.0

5.3+12.7

Nov.

3.6+ 6.5

0.0+ 0.0

0.2+ 1.5

Dec.

16.3+27.7

5.7+14.6

0.1+ 0.3

Jan.

43.0+55.3

16.9+38.0

0.0+ 0.0

Feb.

62.8+40.1

15.9+15.7

1.0+ 2.8

Mar.

20.5+25.2

3.9+ 5.2

6.9+ 8.3

Apr.

27.5+32.5

1.8+ 4.4

23.1+29.9

May

0.4+ 1.2

0.0+ 0.0

12.7+21.4

June

0.1+ 0.3

0.0+ 0.0

25.5+39.0

July

0.0+ 0.0

25.0+29.7

Aug.

0.0+ 0.0

21.9+34.1

summer, and that development proceeds rather slowly. Of the 17
dealate females collected with eggs or no brood at all, 13 came from
spring and early summer months (October-December) and only 4
from the fall (April-May) (Table 3). These findings are consistent
with the observation that virgin alates usually remain in the nests
throughout the winter, and fly in the spring. Nevertheless the
occurrence of a few incipient colonies in apparently early stages of
development in April and May requires some explanation: it seems
likely that either development was hindered in these colonies or that
occasional fall mating flights occur.

Colony foundation in the spring and early summer appears to be
the pattern followed \n purpurea: Brown (1954) noted many colony-
founding dealate females of this species in October and November
on the Atherton Tableland, north Queensland.

Forty-one of the 43 incipient colonies listed in Table 3 contained
only a single queen. The two instances of primary pleometrosis
(colony foundation by more than one queen) both involved colonies
in a very early stage of development, without brood. One of these
pairs (acc. no. 1996) was brought into the lab, and colony
development was monitored. The two queens cohabited peacefully
from 3 October, 1976 until the end of December, at which time the
colony contained 12 eggs, 8 larvae and 8 worker cocoons. The first
worker emerged 2 January, 1977; four days later (after a second

116

Psyche

[Vol. 88

ALATES

ALATE

COCOONS

WORKER

COCOONS

LARVAE

EGGS

i i i l l i i l l I i i i

SEP OCT NOV DEC JAN FEB MAR APR MAY JUN JUL AUG

(13) (47) (42) (30) (25) (13) (28) (30) (72) (107) (32) (39)

Figure 1. Seasonal changes in brood and alate composition in colonies of confusa
and chalybaea, as measured by the proportion of colonies with various life stages.
Maximum width (shown at either side of figure) indicates that 100% of colonies
contain the particular stage. Because there were no obvious differences between species
or between years, data covering both species over 3‘/2 seasons have been combined.
Figures in parentheses refer to the number of colonies sampled in each month. Total
sample size: 479 colonies.

worker had eclosed) one queen was found ousted from the nest and
almost dead. The colony (with one remaining queen) continued to
develop until artificially terminated 15 months later. Subsequent
spermathecal dissections and electrophoretic analysis using allo-
zyme markers confirmed that both females were inseminated, and
that both had contributed worker offspring to the incipient colony.

Seasonal cycle in mature colonies

There are consistent seasonal patterns in the occurrence of brood
and alates in mature colonies of the impressa group. These seasonal
patterns are essentially the same for both queenright and worker-
reproductive colonies, except that alates in the latter are pre-

1981]

Ward — Rhytidoponera impressa. II

117

MAR APR MAY JUN JUL AUG SEP OCT NOV DEC JAN

(7) (19) (20) (24) (27) (6) (13) (39) (20) (4) (4)

Figure 2. Maximum larval length per nest (measured in millimeters, with larva in
natural resting position), in relation to time of year, for 1 83 nests of confusa. One larva
(the largest) was measured in each nest. Figures in parentheses indicate the number of
nests sampled. No data available for February.

dominantly male only. Figure 1 summarizes the seasonal changes
for confusa and cha/ybaea, in terms of the proportion of colonies
containing various stages of brood or alates, for each month. Data
on absolute numbers of worker cocoons, alate cocoons, and alates
are given in Table 4 for 322 confusa colonies (similar patterns are
shown by chalybaea, but numbers average higher).

In confusa and chalybaea there appear to be two peaks of egg
production — one in the spring (September-October) and another in
the late fall (February- April). Larvae are continually present; those
overwintering are small to medium-sized and show little growth
until the spring, when development proceeds rapidly (Figure 2).
Worker cocoons first appear in October-November, and not until

118

Psyche

[Vol. 88

approximately two months later do the first alate cocoons appear
(December-January). Adult workers emerge from January until
June, while alates eclose over a shorter time period (February-
April). Most, or all, of the alates overwinter in the nest, and are
released in the spring, possibly in several bursts, since some nests
have been found with alates as late as November (and one nest with
a single male in early December).

Several points of interest emerge from the foregoing:

(1) There is only one crop of adults produced each year, and
alate production is restricted to a limited period of the total time
that new offspring are produced. At any given time, the standing
crop of new cocoons consists on average of no more than about 30%
alates (Table 5). These facts may be relevant to a consideration of
control of the sex ratio of investment.

(2) No cocoons are overwintered, and there is a period of 6
months (July-December) when no new individuals are added to the
workforce. At first glance, this would seem detrimental to the
increased foraging requirements during rapid larval growth in the
spring and early summer. However, because of a time lag between
worker eclosion and foraging (callow workers remain in the nest) it
may in fact produce an effective increase in the foraging force when
it is most needed.

(3) In the absence of data on sex- and caste-specific growth
rates, it is difficult to know whether alates arise from the overwin-
tering larvae (hence, from eggs laid the previous season) or from
eggs laid in the spring. However one piece of evidence suggests the
latter: the discrepancy between the appearances of the first worker
and first alate cocoons (two summer months) seems to be too great
to be explained by assuming that equivalent-sized overwintering
larvae require that extra period of time (and quantity of food) to
develop into reproductives. Rather, it would seem more likely that
the reproductives develop from spring-laid eggs or alternatively
from smaller overwintering larvae.

Not all nests of confusa and chalybaea contain alates in a given
season, alate production being associated with larger colony sizes
(Table 5). Nevertheless, there is considerable overlap in colony size
between nests with and without alates, partly due to the fact that
worker-reproductive colonies produce alates at a smaller size (and
probably younger age) than queenright colonies (Ward, 1978). It
seems likely that a variety of genetic, environmental, and develop-

1981]

Ward — Rhytidoponera impressa. II

119

Table 5. Mean colony size (number of workers) for nests with alates and for those
without alates at the time of year (February- September) when winged reproductives
are normally present.

Mean no. workers (± S.D.)

Species Alates present Alates absent

confusa 203. 1 + 179.9 (n= 132) 83.9+ 65.7 (n=41)

chalybaea 270.7+206.2 (n=68) 146.7+122.4 (n=41)

mental (ergonomic) factors influence the production of alates.

The available information on impressa and purpurea indicates a
seasonal brood cycle similar to that of confusa and chalybaea. Nests
of the two former species collected in the winter in Queensland
generally had small larvae (sometimes eggs), alates, and few or no
cocoons (sample of 8 impressa colonies, 16 purpurea colonies). A
lowland population of purpurea from near Cape Tribulation, north
Queensland, was exceptional in overwintering with mature larvae,
as well as worker cocoons and adult alates. Nothing is known of the
brood cycle in New Guinea populations of purpurea which inhabit
much less seasonal environments.

Thus, in Australia at least, four species in the impressa group
produce one brood of sexuals a year, most or all of which are
overwintered in the nest and released in the spring. This occurs
despite contrasting climatic regimes at the north-south extremes of
range (summer rainy season in the north, and winter rains in the
south) (cf. Brown, 1954).

Collections of enigmatica suggest a similar brood cycle (i.e. small,
overwintering larvae; cocoons present only in summer), with one
important distinction: alates are usually absent from nests in the
winter. Of 13 nests collected in the early winter (April 30- July 1)
only one contained alates (all males); on the other hand, four out of
five nests collected in the summer (January 12- March 7) contained
alate pupae (also all males). The differences are significant (p < .02,
two-tailed Fisher’s exact test), and suggest that alates fly principally
in the fall. If this is so, there would appear to be considerable
temporal isolation between enigmatica and its two sympatric
congeners (< chalybaea and confusa).

Mating Flights

Two pieces of indirect evidence suggest that reproductives of
confusa and chalybaea normally mate in the spring:

120

Psyche

[Vol. 88

(i) the proportion of nests containing alates is more or less
constant throughout the late summer, fall and winter,
dropping rapidly in the spring; and

(ii) there is a flush of colony-founding females in the early to
mid-summer. To the extent that impressa and purpurea
share the same seasonal cycle, it may be supposed that their
nuptial flights also occur in the spring.

Alates of confusa and chalybaea were observed actively dispers-
ing or swarming on several occasions in rainforest and urban
parkland in the Sydney region. All observations but one (out of 15)
were made in the spring (September 15-November 10), and the only
large-scale mating swarms were seen at this time. Most observations
involved congregations of males around nest entrances. On six
occasions, isolated male or female alates were observed away from
the nest, apparently in a dispersing phase. Spring mating flights
were observed for 3 consecutive years (1976-78) in the chalybaea
population occurring on the University of Sydney campus. Because
of the scarcity of information on this important stage of the life
cycle, the 1976 mating swarm is described in detail.

This flight took place on 4 October 1976, a mild overcast day with
brief periods of sunshine and light rain. At the time observations
were begun (10:15 a.m. EST) large numbers of chalybaea alates,
mostly males, were observed flying in parts of the University
campus. Alates were distinctly concentrated into clusters in tree-
shaded areas. Three of these concentrations were examined in detail
(Sites A, B and C in Figure 3).

Site A. This cluster was centered about a chalybaea nest entrance
between two slabs of sandstone which formed part of a stone wall.
Between 10:45 and 11:45 a.m. there were several hundred males
within 2 meters of the nest entrance. No alate females were seen.
Although males spent most of the time on the ground chasing other
individuals, the congregation appeared to be formed by males flying
into the site. There were large numbers of workers milling around
the nest entrance and most behaved aggressively towards the males,
but this did not deter the latter from making repeated attempts to
mate with workers (and with other males). Three apparently
successful male-worker matings were observed; in each instance the
pair was already in copulation when discovered, in a position
similar to that described by Holldobler & Haskins (1977) for R.
metallica. The worker dragged the male on the ground for about 30

1981]

Ward — Rhytidoponera impressa. II

121

Figure 3. Section of University of Sydney campus where chalybaea nuptial flights
were observed. A, B, and C represent sites of large clusters of alates, described in text.
Lesser numbers of alates were observed at locations U3, and elsewhere.

seconds, after which separation occurred. Wings were vibrated
rapidly during attempts by males to mount workers. A mating
attempt by one male often attracted others, resulting in a buzzing,
frolicking ball of males. Males were also observed to enter (and
leave) the nest, and may perhaps have mated with workers within
the nest. Several instances were noted of workers forcibly evicting
males from the nest, dragging them to a distance of 1 meter from the
nest entrance. Workers were still foraging during these events: two
which were observed returning with a dead honey-bee, and another
with a seed, were unmolested by males.

Sites B and C. Similar observations were made at these sites
(Figure 3), with large numbers (> 100) of alates and workers
clustered in the vicinity of nest entrances, along sandstone walls. A
few alate females were also seen among these swarms. Despite
persistent attempts by males, no successful worker-male or queen-
male matings were recorded. There was a noticeable decline in
swarming activity by early afternoon.

Male alates were observed in smaller numbers at several other
places on campus, particularly at Sites 1, 2, and 3 (Figure 3). A
single, inseminated dealate female was encountered at Site 3 in mid-
afternoon, apparently searching for a nest site. During the day,
samples of alates were collected from each observation site. Out of a
total of 293 alates, 279 (95.2%) were males, and 14 (4.8%) were
females.

122

Psyche

[Vol. 88

Later in the evening (10:45 p.m.), lesser numbers of alate males
were present, but inactive, on the ground at various locations. Over
the next 3 weeks small congregations of males were seen around nest
entrances, but never in such numbers or frenzied activity as during
the large-scale swarm of 4 October.

The Sydney University population of chalybaea consists princi-
pally of worker-reproductive colonies, so the preponderance of
males among alates is not surprising. The 1976 mating swarms
apparently involved insemination of both workers and queens. Only
one mated queen was found, however, and it remains unclear if
queens mate predominantly in the vicinity of nest entrances or in
separate rendevous sites.

Given the limited number of successful matings observed, it is
conceivable that the mating swarm had already passed a peak of
activity at the time that observations began (10:15 a.m.). This is also
suggested by the absence of workers in a sex pheromone-releasing
posture (as described by Holldobler and Haskins (1977) for R.
metallica). Such “calling” workers were observed in lab colonies of
chalybaea , where the behavior occurred both inside and outside the
artificial nest. The posture adopted was similar to that described for
metallica (i.e., head and mesosoma lowered, gaster raised and
arched, with tergites exposed). In addition, workers repeatedly
rubbed the sides of the gaster with their hind tibiae, presumably
facilitating release of pygidial (=tergal) gland pheromone. Such
rubbing movements have been reported in Amblyopone pallipes
queens (Haskins, 1979) but not previously in Rhytidoponera.

A mated worker from one of the copulating pairs observed at Site
A was isolated in a modified Janet (plaster-of-Paris) nest in the lab
and fed on honey and Drosophila. On 1 1 November the first egg
was seen, and by 21 December there were 2 eggs, 1 larva and 1
worker cocoon. Just before the colony was terminated, in March,
1977, this mated worker had produced three worker offspring (the
first had appeared on 19 January 1977). This is perhaps the first
record among the Formicidae of colony-foundation by a lone
worker. However, as mentioned previously, there is no evidence that
single workers found colonies in the field and it appears that they
are always accompanied by an entourage of uninseminated workers.

The inseminated dealate female, also collected on 4 October 1976,
was kept under similar lab conditions for five months. The first
worker appeared on 7 January 1977. At time of termination

1981]

Ward— Rhytidoponera impressa. II

123

(March, 1977) the colony consisted of 1 queen, 1 worker, 1 worker
^cocoon, 1 larva and several eggs.

The following spring, in the morning and early afternoon of 1
October 1977 another large mating swarm of chalybaea occurred on
the University of Sydney campus. As before, this consisted mostly
of male alates, concentrated into more or less discrete clusters
around several nest entrances. Large clusters were situated at Sites
A and C (Figure 3), at exactly the same places observed in 1976. No
matings were directly observed, but a timid worker which was being
mobbed by males was later found to be inseminated. Workers were
generally very aggressive towards males, but the latter persisted in
attempts to mate. Once again, samples of alates were collected from
various sites, of which 97.0% (195) were males and 3.0% (6) were
females. These figures are not significantly different from those of
1976.

On October 12, 1978 small swarms (20-30 individuals) of
Rhytidoponera chalybaea males were observed at Site C and at
several other locations on campus (but not Site A). At 10:15 a.m.
males were mostly at nest entrances, apparently in the process of
emerging. One alate female was observed; this individual emerged
from a nest entrance, and flew off into open sky, ascending rapidly.
Similar behavior was observed in males. By 1 1:30 a.m. many males
appeared to be flying into the area, congregations had formed
outside nest entrances, and males made repeated attempts to mate
with workers.

On the afternoon of the same day two chalybaea queens (one
alate, one partially dealate) were seen floundering on the sidewalk in
a heavily built-up section of downtown Sydney. Both were unin-
seminated. This suggests that alate females may disperse a con-
siderable distance before mating.

Colonly Structure and Life Cycle

In most populations of confusa, chalybaea, and impressa, queen-
right and worker-reproductive colonies coexist, in intermediate
proportions. Despite the likely disparity between mating sites of
winged queens and workers, genetic data from electrophoretic
studies (Ward, 1978, 1980) reveal no indication of extensive
inbreeding or assortative mating with respect to colony type. This is
consistent with the observation that brood development and alate
production proceed at similar rates in the two colony types, and that

124

Psyche

[Vol. 88

release of alates in worker-reproductive colonies occurs syncronous-
ly with (or at least in the same season as) queenright colonies.

As for the remaining species in the impressa group, only
queenright nests are known in purpurea and this species shows a
brood development pattern similar to the three others. By contrast,
distinct winged queens are unknown in enigmatica (all recorded
colonies worker-reproductive), and this species diverges from its
closely related congeners by releasing most alates in the fall,
although males were found overwintering in one nest. The limited
information indicates a possible relaxation of synchrony in the
release of ergatoid-seeking male alates, a pattern which would be
predicted with the loss of the winged queen caste, especially if the
sexual calling behaviour of ergatoid gynes is temporally dispersed.
This trend is continued in some other Rhytidoponera species
outside the impressa group, in which functional queens are rare or
absent, and flights of alates (males) are reported to be highly non-
specific with respect to season (Brown, 1958; Haskins & Whelden,
1965; Haskins, 1979). However, since most of the data come from
lab colonies of one species ( metalliea ) additional field observations
are desirable.

Scattered collections of colonies from different times of the year
may give a misleading impression of patterns of alate production. In
at least two species of the impressa group, alates can be found in
some nests from February to November. Although this superficially
suggests aseasonal production of alates, a detailed examination of
brood development demonstrated that only one crop of alates is
produced each year and that alates are released over a limited time
period. Additional field studies are necessary to determine whether
brood development in Rhytidoponera species without queenright
colonies is less constrained by the need for synchronous alate
release. For comparison with the impressa group, such studies
would be most appropriately directed towards other species of east
Australian mesic forests, in order to minimize climatic and other
environmental differences.

Summary

In the Rhytidoponera impressa group there are two kinds of
colonies, which are distinguished by the type of reproductive female
present: queenright colonies with a single dealate queen, and

1981]

Ward — Rhytidoponera impressa. II

125

worker-reproductive colonies in which one or more mated “work-
ers” occur in lieu of a queen. It appears that worker-reproductive
colonies normally reproduce by colony fission or budding, although
information on this process is fragmentary. Queenright colonies are
founded by lone queens. Colony-founding queens are most fre-
quently encountered in the spring and early summer; such queens
leave the brood chamber to forage for food.

In mature colonies of confusa and chalybaea, the development of
brood and production of alates is highly seasonal (and essentially
similar for both queenright and worker-reproductive colonies). One
crop of workers and alates is produced each year, the former
eclosing from cocoons between January and June, the latter
between February and April. Most or all alates overwinter in the
nest (along with small to medium-sized larvae), and are released in
the spring (September-November). Similar seasonal patterns are
shown by impressa, purpurea (in Australia), and enigmatica, except
that colonies of enigmatica generally do not retain alates over the
winter.

In the population of chalybaea on the University of Sydney
campus, mating flights took place in early October for 3 consecutive
years. During these flights, flying males became concentrated into
clusters around nest entrances where they attempted to mate with
workers, with males, and with the occasional alate female. Several
worker-male but no queen-male matings were observed in these
nest-associated swarms. Like males, queens appear to disperse some
distance before mating, and possibly utilize mating sites other than
nest entrances.

Acknowledgements

This work was supported by an Australian Commonwealth
Scholarship. Additional support from L. C. Birch and the Univers-
ity of Sydney is gratefully acknowledged. 1 thank D. Feener and A.
Forsyth for comments on the manuscript.

References

Brown, W. L.

1953. Characters and synonymies among the genera of ants. Part I. Breviora,

11, 1 13.

126

Psyche

[Vol. 88

1954. Systematic and other notes on some of the smaller species of the ant
genus Rhytidoponera Mayr. Breviora, 33, 1 11.

1958. Contributions toward a reclassification of the Formicidae. II. Tribe
Ectatommini (Hymenoptera). Bull. Mus. Comp. Zool. Harvard, 118,
175-362.

Haskins, C. P.

1979. Sexual calling behavior in highly primitive ants. Psyche, 85, 407-415.

Haskins, C. P. and W. M. Whelden

1965. “Queenlessness”, worker sibship, and colony versus population structure
in the formicid genus Rhytidoponera. Psyche, 72, 87-1 12.

Holldobler, B. and C. P. Haskins

1977. Sexual calling behavior in primitive ants. Science, 195, 793-794.

Ward, P. S.

1978. Genetic variation, colony structure, and social behaviour in the Rhyti-
doponera impressa group, a species complex of ponerine ants. Ph.D.
Thesis, University of Sydney.

1980. Genetic variation and population differentiation in the Rhytidoponera
impressa group, a species complex of ponerine ants (Hymenoptera:
Formicidae). Evolution, 34, 1060-1076.

1981. Ecology and life history of the Rhytidoponera impressa group (Hymen-
optera: Formicidae). I. Habitats, nest sites, and foraging behavior.
Psyche, 88: 89-108.

THE ONTOGENY OF LYSSOMANES VIRIDIS
(WALCKENAER) (ARANEAE: SALTICIDAE)
ON MAGNOLIA GRANDIFLORA LJ

By David B. Richman and Willard H. Whitcomb

Department of Entomology and Nematology
University of Florida, Gainesville, Florida, 32611

Introduction

Lyssomanes viridis (Walckenaer) is a translucent green spider
found in the southeastern United States from North Carolina to
Florida and Texas (Kaston 1978). It has sometimes been placed in a
separate family, Lyssomanidae, but the most recent taxonomic
study (Galiano 1976) includes it in the Salticidae. This species
commonly lives on the tree Magnolia grandiflora L. in mesic
situations, on palmettoes in various habitats, and on Lyonia sp. and
other shrubs in the sand pine scrub of central Florida. No complete
life cycle has been published for any Lyssomanes species. Crane
(1950) did present descriptions of the early stages, including second
postembryo (her first instar) and first instar (her second instar) of L.
bradyspilus Crane. The current paper is the result of a total of two
and one half years of collection and observation of a natural
population of L. viridis for the purpose of learning about the
ontogeny of this spider in the wild.

Methods

Eggs, immatures, and adults of L . viridis were collected on the
undersides of leaves by turning the leaves and catching the spiders in
vials in a stand of Magnolia grandiflora at Tall Timbers Research
Station, Leon County, Florida (Figure 1). These were collected
monthly from August 1971 to February 1973 and usually twice a
month from June 1977 through June 1978. The population densities
were measured by counting the number of spiders collected per 1000
leaves from September 1977 through the death of the adults and the

'Florida Agricultural Experiment Station Journal Series No. 3059.
Manuscript received by the editor July 10, 1981

127

128

Psyche

[Vol. 88

Figure 1. Magnolia stand at Tall Timbers Research Station, Leon County,
Florida. Figure 2. Gravid female of Lyssomanes viridis (Walckenaer). Figure
3. Eggs of Lyssomanes viridis on underside of magnolia leaf. Figure 4. First
postembryos of Lyssomanes viridis on magnolia leaf.

rise of immatures in June 1978. Leaves were counted arbitrarily as
the stand was circled. Collections were made from 0 to 2 m above
ground level ,on both the outside edge and within the stand. All
spiders with prey were preserved separately and identifications of
the prey obtained from various specialists. Carapace widths were
measured using a dissecting microscope equipped with an ocular
micrometer for 20 specimens per sample, if available. The number
of instars was calculated by using a method of simple regression of
carapace widths developed by Hagstrum (1971). It was assumed that
salticids exhibit a similar mean relationship between logarithms of
carapace width and stadium (log y=0.0871x — 0.2692 where
x=stadium and y=carapace width) as Lycosids, Loxoscelids,
Clubionids, Oxyopids, and Ctenizids. Egg masses collected on the
magnolia trees were allowed to hatch and the time spent in the first

1980]

Richman & Whitcomb — Lyssomanes viridis

129

and second postembryonic stages was measured in the laboratory;
however, these were not raised through the various instars.

Weather information for Tall Timbers was obtained from the
research station. Rainfall during August 1971 to February 1973
averaged 11.4 cm (SD = 7.6 cm) with 33.7 cm falling in June 1972
(hurricane Agnes) and only 1.6 cm falling in September 1972.
Rainfall during July 1977 through June 1978 averaged 10.9 cm(SD
= 4.3 cm) per month with a maximum of 16.9 cm in August 1977
and a minimum of 3.7 cm in October 1977. Relative humidity
almost always reached 100% at some time during the day except for
a few days during the winters.

Results and Discussion

We found that mating took place in May and that the males
disappeared by mid-June. Some females lingered on at least until
August. Gravid females (Figure 2) laid 25-70 eggs (mean 42.7,
SD=11.6, no. =24) at a height of 33-131 cm (mean 87.9, SD=28.0,
no. =12) on the magnolia stand from May 31 to July 6. Second
clutches may have been produced only occasionally as females
usually guarded the eggs until first instar and females laid second
clutches only twice (infertile) in the laboratory (not included in egg
counts). The bright green eggs (Figure 3) were ca. 1 mm in diameter
and were loosely covered by silk (there was no distinct cocoon). The
first postembryonic stage (Figure 4 — chorion molted) lasted 32-35.5
hours (no. of egg masses = 4) and the second postembryonic stage
(legs free of vitelline membrane) lasted 7 days (no. of egg masses =
5). The carapace widths (Figure 5) indicated that there were
probably 7 instars including adult female after second postembryo,
based on Hagstrum’s (1971 Figure 1) data for laboratory reared
Lycosidae, Loxoscelidae, Clubionidae and Oxyopidae and for field
collected Ctenizidae. Males may have one less instar than females.
The immature stage individuals lasted from June to the next May
when most matured (Figure 5). Spiderling first instars occurred
from June to July, most individuals reaching second instar by the
first of August. The majority reached third to fourth instar in
September and passed through the winter as third to fifth instars.
The female’s sixth, or penultimate, instar started to be evident in
March. Courtship was observed by Richman (in press).

Immature spiders, especially early instars, fed primarily on

130

Psyche

[Vol. 88

2.0

□

9 9 9
<3

o°

_ 1.5-

< l.Oh

<
o

□

.5-

jjasondjfmamjj

Figure 5. Mean carapace width of immature and adult Lyssomanes viridis
(Walckenaer) at Tall Timbers, Leon County, Florida 1971-1973 and 1977-1978.
Open circles = 1971-1972, closed circles = 1972-1973, open squares = 1977-1978,
closed square (1) = 1978 broods. Sex symbols indicate males and females for
1971-1972 and 1977-1978.

midges of the family Chironomidae. Adults and large immatures
tended to take larger prey, such as syrphid and dolichopodid flies.
Of 12 prey records, immatures were found with 3 chironomids, (one
identified as Orthocladiini by A. R. Soponis), 1 chaoborid fly, 1
syrphid fly (genus Toxomerus identified by H. V. Weems), 1

1980]

Richman & Whitcomb — Lyssomanes viridis

131

Figure 6. Population densities of Lyssomanes viridis (Walckenaer) in magnolia
stand at Tall Timbers fall 1977 to spring 1978. Data points are means between
surveys made both inside and outside the stand, during the last half of each month.

encyrtid wasp and 1 aphid (genus Macrosiphon identified by H. A.
Denmark). Adult males were collected with an unknown dipteran
and a salticid spider of the genus Hentzia. Adult females were
collected with a dolichopodid fly, an unknown dipteran and a
psocid.

The population density (Figure 6) dropped during the winter, but
rose in the spring nearly to that of the previous fall, probably
reflecting inactivity during the winter, rather than a significant
mortality. The population drop during May is probably a result of
the death of adults. Adults were only found during the spring and
early summer.

Some adult spiders were found in the nest ol a mud dauber of the
genus Trypoxylon by G. B. Edwards at Newnan’s Lake, Alachua
County, Florida. A large Trypoxylon was observed during June at
Tall Timbers and a fresh nest was found on the underside of a
magnolia leaf. The nest in this case was filled with Araneidae. One
adult female L. viridis was collected and found to have a large
mirmithid nematode in its abdomen. No egg parasites were seen.

Complete life histories have been published for several salticids.

132

Psyche

[Vol. 88

notably Philaeus chrysops Poda (Bonnet 1933), three species of
Corythalia (Crane 1948) and Phidippus johnsoni Peckham and
Peckham (Jackson 1978). Female P. chrysops generally had seven
molts (six instars) before maturity, and this was also true of the
three species of Corythalia observed by Crane (1948). Jackson
(1978) reported 6-9 molts for P. johnsoni. Thus, the life cycle of L.
viridis seems to compare well with those of other salticids.

Summary

A population of the salticid spider Lyssomanes viridis (Walcke-
naer) was sampled for two and one half years on a stand of
Magnolia grandiflora L. trees in North Florida. Mating took place
in May and adult males disappeared by mid-June. Females laid
25-70 eggs per clutch mostly during June. These hatched from June
to July and the immatures overwintered in middle instars. After
temperatures increased in the spring the spiders rapidly developed
to adults. Simple linear regression of the carapace widths indicated
that this species has a total of seven instars from the end of second
postembryo through adult female. Males may have one fewer instar.
L. viridis feeds primarily on Diptera in this habitat.

Acknowledgements

We would especially like to thank Dr. Bruce Means and Ed and
Roy Komarek of Tall Timbers Research Station for their help and
support. Also we would like to thank Dr. Robert C. Hemenway, Jr.,
Dr. G. B. Edwards, and Dr. Barbara Saffer for their help in
collecting specimens in the magnolia stand.

Literature Cited

Bonnet

1933. Cycle vital de Philaeus chrysops Poda (Araneide, Salticide). Arch. Zool.
Exp. Gen. 75: 129-44.

Crane, J.

1948. Comparative biology of salticid spiders at Rancho Grande, Venezuela.
Part I. Systematics and life history in Corythalia. Zoologia 33: 1-38.
Comparative biology of salticid spiders at Rancho Grande, Venezuela.
Part V. Postembryonological development of color and pattern. Zoo-
logica 35: 253-61.

1950.

1980]

Richman & Whitcomb — Lyssomanes viridis

133

Galiano, M.E.

1976. Comentarios sobre la categoria sistematica del taxon “Lyssomanidae”
(Araneae). Rev. Mus. Argentino Cienc. Natur. Entomol. 5: 59-70.

Hagstrum, D.W.

1971. Carapace width as a tool for evaluating the rate of development of
spiders in the laboratory and the field. Ann. Entomol. Soc. Amer. 64:
757-60.

Jackson, R.R.

1978. Life history of Phidippus johnsoni (Araneae, Salticidae). J. Arachnol. 6:
1-29.

Kaston, B.J.

1978. How to know the spiders. 3rd Ed. Wm. C. Brown, Dubeque. 272 p.

Richman, D.B.

(in press). Epigamic display in jumping spiders (Araneae, Salticidae) and its use in
systematics. J. Arach.

THE EMIGRATION BEHAVIOR OF TWO SPECIES OF
THE GENUS PHE1DOLE (FORMICIDAE: MYRMICINAE).

By Robert droual1 and Howard Topoff2

Introduction

Colony emigrations are common among ants (Wilson 1971) and
occur for a diversity of reasons. However, except for the legionary
ants, in which colony emigrations are an inherent part of the
foraging ecology (Wilson 1971), and species which inhabit delicate
and easily disturbed nests (Holldobler and Wilson 1977, Moglich
1979), emigrations are thought to occur infrequently. Here we
present evidence that two species of the genus Pheidole, P. deser-
torum Wheeler and P. hyatti Emery, emigrate frequently under
environmentally stable conditions. We further advance the hy-
pothesis that the surplus nests resulting from these emigrations,
reduce the secondary losses which occur as a consequence of the
panic-alarm defense these species employ against army ants of the
genus Neivamyrmex, by serving as temporary shelters and centers
for colony reorganization.

Methods

This investigation was conducted during the months of June, July
and August, 1980, at two different study sites. One site was an oak-
juniper woodland located on the grounds of the Southwestern
Research Station of the American Museum of Natural History near
Portal, Arizona (elev. 1646 m). The other site was a desert-grassland
located 8 km N.W. of Rodeo, Hidalgo Co., New Mexico (elev. 1250
m). In both habitats a winter (Dec., Jan., Feb. and March) and a
summer (July, June and August) rainy season occur. On the oak-
juniper woodland site colonies of both P. desertorum and P. hyatti
were located and marked; on the desert-grassland site only colonies
of P. desertorum were located and marked.

'Biology Program, City College of C.U.N.Y., New York, N.Y. 10031
Psychology Department, Hunter College of C.U.N.Y., New York, N.Y. 10021 and
Department of Entomology, The American Museum of Natural History, New York,
N.Y. 10024

Manuscript received by the editor June 8, 1981

135

136

Psyche

[Vol. 88

Table 1. Emigration characteristics of P. hyatti.

Colony

Days

Observed

Number of
Emigrations

Returns to a
Former Nest

Distance Between First
and Last Observed
Nests (m)

H-Jnl4-1

4.2

H-Jnl4-2

1 1

1.1

H-Jnl4-3

5.4

H-JnI5-l

1.5

H-Jnl5-2

0.0

H-Jnl7-1

H-Jnl7-2

0.8

H-Jnl8-1

3.2

H-Jnl8-2

6.8

H-Jnl9-1

0.0

H-Jnl9-2

0.0

H-Jnl9-3

1.5

H-Jnl9-4

0.0

H-Jnl9-5

2.0

H-Jn21-1

2.6

H-Jn21-2

1.9

H-Jn21-3

0.0

H-Jn21-4

3.0

H-Jn23-1

0.8

H-Jn24-1

0.8

H-Jn26-1

2.5

H-Jn26-2

1.1

H-Jn27-1

0.4

H-Jn28-1

2.4

Total

137

Colony designations were based on the species (D - desertorum,
H - hyatti ), the date when the colony was found (Jn - June, J1 - July,
A - August) and the order in which it was found on that date. For
example, H-Jnl8-2 is the second P. hyatti colony found on June 18.

Activity for both species began at approximately 2000 hr (MST)
and ceased 0500 hr. To determine emigration frequency all colonies
were inspected at least once daily between 2000 and 2400 hr. In
order to avoid disturbing the colony any prolonged observations
were made using red light. About two-thirds of the emigrations for
each species were documented indirectly when a colony occupying a
nest one night was found at another nest the following night. A
colony was assumed to be occupying a nest if 10 foragers were

1981]

Droual & Topoff— Genus Pheidole

137

Table 2. Emigration characteristics of P. desertorum.

Colony

Days

Observed

Number of
Emigrations

Returns to a
Former Nest

Distance Between First
and Last Observed
Nests (m)

D-JnlO-la*

0.5

D-Jnll-la

4.8

D-Jnl l-2a

D-Jnl2-la

15.6

D-Jnl2-2a

2.5

D-Jnl2-3a

0.0

D-Jnl2-4a

3.0

D-Jnl2-5a

4.2

D-Jnl3-la

0.0

D-Jnl4-la

0.0

D-Jnl5-la

6.6

D-Jnl5-2a

0.0

D-Jnl6-1 b

0.0

D-Jnl7-1 b

0.0

D-Jnl7-2b

2.4

D-Jnl8-la

3.1

D-Jn20-1 b

0.0

D-Jn25-la

1.5

D-Jn28-la

3.7

D-Jl 1-la

2.4

D-Jl 13-lb

0.0

D-Jl 13-2b

1.2

D-Jl 15-lb

0.0

D-Jl 30-lb

0.0

D-Jl 30-2a

0.0

D-A 1-la

4.0

Total

101

*a: desert-grassland; b: oak-juniper woodland.

observed leaving the nest during a 1 min period. If this criterion was
not met, or if there was some other reason to doubt the location of
the colony, a small peanut butter bait was used to locate the colony.
To avoid confusion when using this indirect method, an attempt was
made to locate and mark any neighboring conspecific colonies. The
remainder of the emigrations were observed directly when an
emigration was discovered in progress. The nests were marked and
the distance between the old and the new nests measured.

With the statistical tests employed in this paper probabilities of
.05 or less were accepted as significant.

138

Psyche

[Vol. 88

Results

Colonies of both P. desertorum and P. hyatti showed consider-
able variation in their frequencies of emigration (see Tables 1 and 2).
One colony of each species (D-Jnll-2 and H-Jnl7-1) did not
emigrate at all, while one P. desertorum colony (D-Jn20-1) emi-
grated 8 times, and one P. hyatti colony (H-Jnl4-2) emigrated 16
times. Despite this variability, both species showed a clear tendency
to emigrate frequently: over one-half of the P. desertorum colonies
emigrated at least 4 times, and over one half of the P. hyatti colonies
emigrated at least 6 times. To statistically compare the emigration
frequency of the two species, the percentage days for which an
emigration was documented was calculated for each colony, and the
percentages for each species were compared using the Wilcoxon
two-sample test (Sokal and Rohlf 1969). No significant difference
was found in the emigration frequency between the two species
(.1 > P > .05).

This similarity between species in emigration frequency can also
be seen if the frequency of time interval between emigrations is
compared. Figures 1 and 2 show the frequency of emigration
interval for P. hyatti and P. desertorum, respectively. Both distribu-
tions are strongly skewed to the right with a surprisingly high
number of emigrations occurring 1 to 2 days after the previous
emigration. No significant difference was found in the frequency
distribution between the two species (Wilcoxon two-sample test:
.4 > P > .2)

The daily occurrence of emigrations among all colonies is shown
in the form of frequency histograms in Figs. 3, 4 and 5. The upper
line in the graphs outlines the number of colonies which were
included in the sample size each night. Excluded from the sample
were colonies which were raided by army ants, or were still suffering
from the effects of an army ant raid (see Discussion). Superimposed
over the graphs is a bar diagram showing the daily rainfall.

A positive correlation was found to exist between emigration
frequency and daily rainfall in all three cases (Spearman rank
correlation coefficient: P. hyatti : rs = .28, N = 66; P. desertorum in
oak-juniper woodland : rs = .25, N = 64; P. desertorum in desert-
grassland : rs = .32, N = 70). The effect of rainfall on emigration
frequency is most clearly seen in P. desertorum in the desert-
grassland habitat (Fig. 5). During the 29 days before the first heavy

1981]

Droual & Topoff— Genus Pheidole

139

rainfall on July 9 only three emigrations occurred, but within 9 days
after this rainfall 29 emigrations occurred. During this 9 day period
13 of the 15 colonies being observed on this site emigrated at least
once.

The emigration distance for both species was variable. Mean
emigration distance for P. hvatti was 1.8 ± 1.0 m (N=137; range 0.3
— 4.9) and mean emigration distance for P. desertorum was 2.5 ±
1.4 m (N=102; range 0.4 — 6.9). The larger emigration distance of
P. desertorum over that of P. hyatti correlates with the larger size of
this species (mean length of P. hyatti minor = 2.64 ± 0.04 mm,
N=50; mean length of P. desertorum minor = 3.14 ± 0.03 mm,
N=57).

Despite the high emigration frequency of both species, colonies of
neither species tended to move far from the nests at which they were
first discovered. Tables 1 and 2 show the number of times each
colony returned to a former nest, and the distance between the first
and last nests. As can be seen, 49% of P. desertorum ’s emigrations,
and 42% of P. hyatti" s emigrations were to former nest sites, and 1 1
P. desertorum colonies and 5 P. hyatti colonies at the end of the
study were at the nest at which they were first discovered. This
crisscrossing pattern of emigrations is illustrated for three colonies
of each species in Figs. 6 and 7. The relative location of the nests
reveal a clumped rather than a linear arrangement which would be
expected if the colony were emigrating out of an area. The dates of
nest movements for each colony show that the variability of
emigration interval within each colony was considerable. This can
be readily seen by examining the ranges of emigration intervals for
the colonies shown in Figs. 6 and 7: for the P. hyatti colonies the
ranges are, H-Jnl4-2: 1-17 days; H-Jnl4-3: 3-18 days; H-Jnl8-1:
1-8 days; for the P. desertorum colonies the ranges are D-Jn25-1:
1-19 days; D-Jnl2-4: 1-21 days; D-J130-1: 1-4 days.

Because P. desertorum and P. hyatti emigrated so frequently
about 33% of the emigrations of both species were discovered in
progress. These emigrations were readily noticed as hundreds to
thousands of workers, most carrying brood, formed a column
connecting the old nest to the new nest. The width of this column for
P. hyatti was about 3 cm, while for P. desertorum the column
tended to be wider (on one occasion reaching a width of 15 cm).
Laboratory experiments have revealed that P. hyatti's emigrations

140

Psyche

[Vol. 88

EMIGRATION INTERVAL (DAYS)

Figure 1 . Frequency of the time interval between emigrations for Pheidole hvatti.

are totally organized by a substance secreted by the poison gland
(Droual et al., in prep.)- The queen of both species moved inde-
pendently in the emigrations although she was usually surrounded
by a retinue consisting mostly of minor workers (workers of the
genus Pheidole are dimorphic) who tugged her by the mandibles or
antennae if she hesitated en route to the new nest. During June and
the early part of July alates were frequently seen in the column also
moving independently. However, on one occasion, during a P.
desertorum emigration, workers were observed carrying some of the
males.

A number of phenomena related to these species’ high emigration
frequencies were observed. One colony of each species (D-Jn20-1
and H-Jnl9-2) performed what we call an aborted emigration. In
these cases the colony was observed emigrating to a new nest but on
the following night was found to be back at its old nest. One P.
desertorum colony (D-Jn-25-1) appeared to perform two emigra-
tions in one night. On August 17 the colony was observed emigrat-
ing from nest 2 to nest 1 (see Fig. 6). However on the following night
the colony was found at nest 3. On a number of occasions an

1981]

Droual & Topoff — Genus Pheidole

141

EMIGRATION INTERVAL (DAYS)

Figure 2. Frequency of the time interval between emigrations for Pheidole
desertorum.

emigration could be predicted in advance by the colony’s excavation
activity at another site. For example, before colony D-Jnl2-2
emigrated to its second nest site on 8/17, workers from the colony
were observed excavating at the site on 8/4, 8/ 5, 8/7 and 8/ 10-8/ 17.
However two colonies of both P. desertorum and P. hyatti were
observed excavating at sites to which they did not emigrate even
though they emigrated later to other nests.

Discussion

In this paper we have shown that P. desertorum and P. hyatti
emigrate frequently and that the emigration frequencies of the two
species are similar. This similarity in emigration frequency becomes
even more marked when it is taken into account that most of P.
desertorum ’s emigrations in the desert-grassland occurred after the
first rainfall. The sharp increase in emigration activity after the rain
can possibly be explained by the affect of the rainfall upon the soil.
Before the rains began the soil was very hard and compact, but after

142

Psyche

[Vol. 88

6/15 6/20 6/25 6/30 7/5 7/10 7/15 7/20 7/25 7/30 8/5 8/10 8/15

DATE

Figure 3. Daily occurrence of emigrations for Pheidole hvatti. Black bars
indicate the number of colonies which emigrated each night. Upper line outlines the
number colonies included in the sample each night. Right ordinate indicates rainfall
for the superimposed bar diagram showing daily rainfall.

the first heavy rainfall the soil loosened considerably. This un-
doubtedly made the excavation of new nests by the desert-grassland
dwelling colonies much easier. The same reasoning can be applied to
explain the positive correlation between emigration frequency and
daily rainfall in both habitats. However, in the oak-juniper wood-
land, the greater amount of vegetation, the rockier soil and the
generally moister conditions probably account for the relatively
higher emigration activity before the beginning of the rainy season
in this habitat.

The need to perform a colony emigration is a contingency almost
all species of ants can be expected to face (Wilson 1971). However
some species emigrate more than others. Among the legionary ants,
particularly the Ecitoninae and Dorylinae, colony emigrations are
an integral part of the foraging ecology (Wilson 1971). Oppor-
tunistic nesters such as Tapinoma melanocephalum, T. sessile,
Paratrechina bourbonica and P. longicornis occupy ready-made
nests such as the tufts of dead grass and hollow plant stems which

NO OF EMIGRATIONS NO OF EMIGRATIONS

1981]

Droual & Topoff— Genus Pheidole

143

20-|

10-

P DESERTORUM

(OAK -JUNIPER WOODLAND HABITAT)

4U4

"U L

iLI

i ■ ■ j ■

-30

■2 0

6/15 6/20 6/25 6/30 7/5 7/10 7/15 7/20 7/25 7/30 8/5 8/10 8/15

DATE

Figure 4. Daily occurrence of emigrations for Pheidole desertorum in oat
miper woodland (See Fig. 3).

P. DESERTORUM

(DESERT- GRASSLAND HABITAT)

i . . .. i j

6/10 6/15 6/20 6/25 6/30 7/5 7/10 7/15 7/20 7/25 7/30 8/5 8/10 8/15

DATE

Figure 5. Daily occurrence of emigrations for Pheidole desertorum in desert-
grassland (see Fig. 3).

RAINFALL (CM) ' RAINFALL (CM)

144

H - Jn 14 - 2

I METER

H- Jnl4-3

✓ N

Psyche

NEST 7

[Vol. 88

NEST

MOVEMENT DATE

1- 2 6/15

2- 3 7/2

3- 1 7/7

1-4 7/9

4- 3 7/15

3- 4 7/18

4- 5 7/22

5- 1 7/27

1- 4 7/30

4- 6 7/31

6- 5 8/3

5- 4 8/5

4-6 8/6

6- 2 8/11

2- 1 8/13

1-4 8/15

4-6 8/17

NEST

MOVEMENT DATE

1-2 6/18

2- 3 6/22

3- 4 7/10

4- 5 7/20

5- 1 7/27

'1-6 8/1

6- 7 8/11

7 4 8/14

I METER

H - Jn 18-1

/ N

i 1

I METER

NEST 4

NEST 6

NEST 2*^

NEST I

NEST

MOVEMENT DATE

1- 2 7/1

2- 3 7/5

3- 4 7/11

4- 2 7/17

2- 3 7/21

3- 5 7/29

5- 2 8/4

2- 1 8/5

1-3 8/8

3- 6 8/13

Figure 6. Patterns of emigrations for three colonies of Pheidole hyatti. Dates of
nest movements are shown on the right. *This nest had two entrances.

are short-lived. When these nests are disturbed, the colonies quickly
organize emigrations to other such nests (Holldobler and Wilson
1977). Leptothorax acervorum in oak-juniper woodland, construct
delicate nests under stones which can be easily dislodged by large
vertebrates, and are prone to emigrate when their nest is disturbed
(Moglich 1979).

Most species build or choose nest sites which are longer-lived and
less easily disturbed and are thought to emigrate infrequently.
Among these species emigrations can be due to a local factor such as

1981]

Droual & Topoff— Genus Pheidole

145

D-Jn25-I

NEST

MOVEMENT DATE

1- 2 7/12

2- 3 7/15

3- 4 7/16

4- I 8/5

1- 4 8/6

4-2 8/15

2- 3 8/17

I METER

NEST

MOVEMENT

DATE

1-2

7/10

2-1

7/31

1 -3

8/1

3-4

8/4

4-2

8/5

2-4

8/14

D-JI 30-1

/ N

NEST I NEST 4

NEST

MOVEMENT

DATE

1-2

7/31

2-3

8/3

3-1

8/7

1 -3

8/11

3-4

8/15

4-1

8/16

I METER

Figure 7. Patterns of emigrations for three colonies of Pheidole desertorum.
Dates of nest movements are shown on the right.

shading (Brian 1956, Carlson and Gentry 1973), or climatic ad-
versity such as drought or frost (Brian 1952). A colony may also be
forced to emigrate because of some biotic factor such as inter- and
intra-specific competition (Holldobler 1976, Waloff and Blackith
1962, Brian 1952, Brian et. al. 1965) and predation (Gentry 1974).

The view that emigrations occur infrequently among most ants

146

Psyche

[Vol. 88

was recently challenged by Smallwood and Culver (1979). These
investigators conducted a study in which they found that Tapinoma
sessile and Aphaenogaster rudis emigrated frequently. Their study
differs from ours in that colonies were marked and rechecked only
after intervals of 11-21 days and no attempt was made to follow the
behavior of individual colonies. Because T. sessile and A. rudis
choose different nesting sites and have different life styles these
investigators deduced that emigrations occur more frequently
among ants than had been previously thought. However, the fact
that T. sessile, as mentioned above, is an opportunistic nester which
is expected to emigrate frequently weakens their argument.

It is difficult to apply any of the known or previously hypothe-
sized causes of colony emigrations to explain the frequent emigra-
tions of P. desertorum and P. hyatti. The nests of both species are
excavated in the soil to a depth of 30 to 40 cm (based on excavations
in oak-juniper woodland), and hence are not easily disturbed.
Shading is obviously not a factor in the desert-grassland, and is
negligible in the oak-juniper woodland where the canopy is not
extensive. Permanent deterioration of the nest as a cause is elimi-
nated by the fact that colonies return to former nests. Indeed almost
half of all emigrations for both species resulted in a return to a
former nest. This fact, and the patterns of emigrations which tended
to keep a colony in the same area, argue against any hypothesis
which involves a deterioration of some local condition such as might
be due to interference competition or to a decrease in the local food
supply.

However, in one instance the possibility that an emigration may
have been the result of intraspecific competition should be men-
tioned. This involved colony H-Jnl4-1 which was observed emigra-
ting a distance of 4.9 m the day after it was found. This distance is
considerably larger than the mean emigration distance of 1.8 m
found for P. hyatti. Four days later, colony H-Jnl9-1 was dis-
covered emigrating 2.3 m into the nest vacated by colony H-Jnl4-1.
Although large-scale conflicts between these species were never
observed, workers will attack any alien workers of either species
discovered near their nest. Frequent encounters of this sort may
have caused colony H-Jnl4-1 to make its unusually long emigra-
tion.

In discussing the causation of any behavior a distinction should
be made between those hypotheses that invoke a proximate cause

1981]

Droual & Topoff— Genus Pheidole

147

and those that invoke an ultimate cause (Wilson 1971). For
example, it has been hypothesized that the ultimate cause of army
ant emigrations is to prevent a local depletion of food resources
(Wilson 1971). The proximate cause of these emigrations, at least
among the Ecitoninae, was discovered to be recruitment to a new
nest under periods of high colony arousal due to brood stimulation
(Schneirla 1938). However, it has been recently shown that food
supply may also be a proximate factor (Topoff and Mirenda 1980,
Mirenda and Topoff 1980). The hypothesis we are advancing to
explain the frequent emigrations of P. hyatti and P. desertorum
concerns the ultimate cause of these emigrations although both the
ultimate and proximate causes are the subject of further investiga-
tion by us.

Both P. desertorum and P. hyatti , which are small and lack
potent stings, are easy prey for army ants of the genus Neivamyrmex.
Mirenda et. al. (1980) found, in the same desert -grassland site
employed in this study, that P. desertorum was the species most
frequently raided by N. nigrescens. Our own observations also show
that both P. desertorum and P. hyatti are heavily preyed upon by
members of the genus Neivamyrmex (Tables 3 & 4). Some P.
desertorum colonies were raided repeatedly by the same army ant
colony which entered the statary phase in a nearby bivouac. On two
occasions an army ant colony actually bivouacked in the evacuated
nest of a P. desertorum colony. One P. hyatti colony was raided by
two species of Neivamyrmex. Of these colonies only five appeared
to be completely eliminated by the army ants. Part of the reason for

Table 3. Observed army ant raids on colonies of P. hyatti.

Colony

Dates of Raids

Species Raiding

H-Jnl5-1

7/7, 7/8*

Neivamyrmex nigrescens

H-Jnl5-2

7/8

N. nigrescens

H-Jnl9-4

7/28

N. texanus

H-Jn21-3

8/15

N. opacithorax

H-Jn21-4

8/17

N. nigrescens

H-Jn21-5

8/15

N. opacithorax

8/18

N. nigrescens

H-Jn23-1

8/15*

N. opacithorax

H-Jn26-1

8/12

N. nigrescens

H-Jn27-1

8/12*

N. nigrescens

Colony was not seen afterwards.

148

Psyche

[Vol. 88

this is the panic-alarm defense employed by these species against the
army ants.

That defense behavior in ants can be both enemy specific and
complex was established with the discovery of the alarm-recruit-
ment defense of Pheidole dentata against the fire ant Solenopsis
geminata (Wilson 1975 and 1976). Although more evidence is
necessary, the defense behavior of P. hyatti and P. desertorum
appears to be both enemy specific and complex. The defense, which
begins when a Pheidole forager contacts an army ant and runs back
into the nest raising an alarm, occurs in two phases. In the first, or
“milling”, phase, workers carrying brood well out of the nest but
remain in close contact near the nest’s entrance. In the second, or
absconding, phase, the workers flee from the nest. P. desertorum's
flight is protean in nature (Humphries and Driver 1970) with
workers scattering in all directions. In P. hyatti the exodus is more
organized with the workers fleeing in columns which appear to
follow recently-laid chemical trails.

After evacuating from their nest the fleeing workers tend to
concentrate at temporary shelters such as that provided by leaf
litter, fallen branches, rotting logs and tufts of grass. Some workers
eventually find some or all of the former nests and begin to recruit
other workers to them. After the raid is over workers will also start
to return to the evacuated nest. In this manner the colony becomes
fragmented with various proportions of the colony in some or all of
the available nests. The colony then begins the process of reorgan-
izing with segments in one nest emigrating to join segments in
another nest until the colony becomes reunited in one nest. Hence it
appears that the surplus nests resulting from the frequent emigra-
tions of these species serve a dual purpose after an evacuation: they
provide shelter and centers for reorganization.

After a nest evacuation, finding a place of suitable moisture
before the lethal surface temperatures and low surface humidity of
the approaching day is undoubtedly of vital importance for these
nocturnal species. This problem becomes particularly severe in the
desert-grassland where the lack of ground cover makes nests
excavated in the ground the only suitable shelters. Having alternate
nests becomes a necessity when an army ant colony bivouacs in the
evacuated nest. The hypothesis we are proposing then is that the
surplus nests which result from the emigrations of these species
increases the effectiveness of the panic-alarm defense by reducing

1981]

Droual & Topoff— Genus Pheidole

149

Table 4. Observed army ant raids on colonies of P. desertorum.

Colony

Dates of Raids

Species Raiding

H-Jnll-1

8/ 1-8/5

Neivamyrmex nigrescens

H-Jnll-2

7/9

nigrescens

H-Jnl2-1

8/4, 8/6, 8/8

nigrescens

H-Jnl2-4

7/4

nigrescens

H-Jnl2-5

7/2*

nigrescens

H-Jnl3-1

7/12, 8/ 10*

nigrescens

H-Jnl4-1

8/5, 8/6, 8/12-8/15

nigrescens

H-Jnl8-1

7/11, 7/12, 7/16,
7/25, 7/26, 7/28, 8/1

nigrescens

H-Jnl7-1

8/18

nigrescens

H-Jn25-1

8/28

nigrescens

H-Jl 4-1

8/3, 8 / 5—8 / 7, 8/9

nigrescens

H-Jl 30-2

8/13

nigrescens

H-Jl 15-1

7/28

texanus

*Colony not seen afterwards.

the secondary losses which result from the disorganization which
follows the defense. If this hypothesis proves to be correct, the
possibility that the frequent emigrations of these species have
evolved to serve as part of a defense system against the army ants
has to be entertained.

Acknowlegements

This research was supported by a N.I.M.H. training grant (MH
15341), a Theodore Roosevelt Memorial Fund Grant, PSY-CUNY
Grant 13492 and a NSF Grant BNS- 8004565. We thank Dr. G.
Turkewitz for his statistical advice and Roy R. Snelling for identify-
ing the Pheidole specimens.

Literature Cited

Brian, M. V.

1952. The structure of a dense natural ant population. J. Anim. Ecol. 21:
12-24.

1956. Segregation of species of the ant genus Myrmica. J. Anim. Ecol. 25:
319-337.

Brian, M. V., J. Hibble and D. J. Stradling.

1965. Ant pattern and density in a Southern English heath. J. Anim. Ecol. 34:
545-555.

Carlson, D. M. and J. B. Gentry.

1973. Effects of shading on the migratory behavior of the Florida harvester
ant, Pogonomyrmex badius. Ecology 54: 452-453.

150

Psyche

[Vol. 88

Gentry, J. B.

1974. Response to predation by colonies of the Florida harvester ant, Pogono-
myrmex badius. Ecology 55: 1328-1338.

Holldobler, B.

1976. Recruitment behavior, home range orientation and territoriality in
harvester ants, Pogonomyrmex. Behav. Ecol. Sociobiol. 1(1): 3-44.

Holldobler, B. and E. O. Wilson.

1977. The number of queens: an important trait in ant evolution. Natur-
wissenshaften 64: 8-15.

Mirenda, J. T., D. G. Eakins, K. Gravelle and H. Topoff.

1980. Predatory behavior and prey selection by army ants in a desert-grassland
habitat. Behav. Ecol. Sociobiol. 7: 119 127.

Mirenda, J. T. and H. Topoff.

1980. Nomadic behavior of army ants in a desert-grassland habitat. Behav.
Ecol. Sociobiol. 7: 129-135.

Moglich, M.

1978. Social organization of nest emigration in Leptothorax. (Hym. Form.).
Ins. Soc. 25(3): 205-225.

SCHNEIRLA, T. C.

1938. A theory of army-ant behavior based upon the analysis of activities in a
representative species. J. Comp. Psychol. 25: 51-90.

Smallwood, J. and D. C. Culver.

1979. Colony movements of some North American ants. J. Anim. Ecol. 48:
373-382.

SOKAL, R. R. AND F. J. ROHLF.

1969. Biometry. W. H. Freeman and company, San Francisco, xxi + 776 p.

Topoff, H. and J. Mirenda.

1980. Army ants of the move: relation between food availability and emigra-
tion frequency in Neivamyrmex nigrescens. Science 207: 1099-1100.

Waloff, N. and R. E. Blackith.

1962. The growth and distribution of the mounds of Lasius flavus (Fabricius)
(Hym:Formicidae) in Silwood Park, Berkshire. J. Anim. Ecol. 31:
421-437.

Wilson, E. O.

1971. The Insect Societies. Belknap, Harvard Univ. Press, Cambridge, x + 548
P-

1975. Enemy specification in the alarm-recruitment system of an ant. Science
190: 798-800.

1976. The organization of colony defense in the ant Pheidole dentata Mayr
(Hymenoptera: Formicidae). Behav. Ecol. Sociobiol. 1: 63-81.

STATARY BEHAVIOR IN NOMADIC COLONIES
OF ARMY ANTS:

THE EFFECT OF OVERFEEDING

By Howard Topoff, Aron Rothstein*, Susan Pujdak, and
Tina Dahlstrom**

Department of Psychology, Hunter College of CUNY, New York,

N.Y. 10021, and The American Museum of Natural History

Introduction

Nearctic colonies of the army ant Neivamyrmex nigrescens
Cresson (subfamily Ecitoninae) exhibit behavioral cycles consisting
of alternating nomadic and statary phases. During the statary
phase, a colony remains at the same nesting site and forages
irregularly for food. The nomadic phase, by contrast, is character-
ized by night-long raids and frequent emigrations to new bivouacs.
According to Schneirla (1957, 1958), the nomadic phase is triggered
by stimulation arising from newly-eclosed callows, and is subse-
quently maintained by comparable excitation from the developing
larvae. Experimental support for brood-stimulation theory stems
from studies showing: (1) an abrupt reduction in nomadism after
removing a portion of a larval brood (Schneirla and Brown, 1950);
and (2) the eclosion of a pupal brood (in the absence of newly-
hatched larvae) is indeed sufficient to initiate a nomadic phase
(Topoff et al., 1980a). Recent studies have suggested, however, that
brood stimulation may in turn depend on the degree of brood
satiation. Thus, in a preliminary field study involving food aug-
mentation, Mirenda et al. (in press) was able to halt the occurrence
of emigrations during a portion of the nomadic phase in colonies of
N. nigrescens. This was followed by more prolonged laboratory
studies (Topoff and Mirenda, 1980 a,b) showing that the frequency
and direction of nomadic emigrations are indeed influenced by the
amount and location of food.

This paper reports findings from our continued studies of food
augmentation for colonies of N. nigrescens. In previous studies,
larval stimulation was reduced by artificially feeding colonies early
in the nomadic phase, after callow eclosion. Because an additional

*Department of Biology, City College of CUNY, New York, N.Y. 10031

**Department of Zoology, University of California, Riverside, CA 92521

151

152

Psyche

[Vol. 88

goal of the present study was to reduce callow stimulation and
thereby delay the onset of the nomadic phase, overfeeding com-
menced late in the statary period. For the first colony, emigrations
were delayed approximately 6 days. In the second overfed colony,
we were able to virtually eliminate the nomadic phase, together with
all associated patterns of raiding and emigration behavior.

Methods and Procedures

This study was conducted during July and August, 1980, in a
desert-grassland habitat, 8 km east of Portal, Arizona. The site was
chosen because the pavement-like substrate and patchy vegetation
provided us with an excellent view of the ants’ raiding and
emigration activities. Surface soil temperatures averaged 50° C at
1500 hr (MST) and 17° C at 0200 hr throughout the summer
(Mirenda et al. 1980). As a result of the severe daytime temperatures
and aridity, colonies of N. nigrescens were usually active on the
surface only between 1900-0500 hr.

Colonies were located by walking through the study area with
gasoline lanterns or miner’s cap lamps. Colony no. 1 was found on
July 11, at the end of a nomadic phase, and observed nightly
throughout its next statary period. During the subsequent nomadic
phase, the colony was estimated to contain approximately 80,000
adults and 50,000 larvae. Colony no. 2 was collected on July 14,
during its last nomadic emigration, and maintained in the labora-
tory (see Topoff et al. 1980b for details of the rearing procedure)
until the pupal brood was fully pigmented. Prior to release in the
field, the colony was culled to contain 4,000 adults and 4,000 pupae.
By the next nomadic phase, approximately 4,500 larvae were also
present in the colony. This small colony size was chosen for two
reasons: (1) to increase our ability to appreciably overfeed the
colony; and (2) a laboratory colony of comparable size had
previously been released without food supplementation, as part of a
study designed to show that laboratory rearing and population
reduction do not alter qualitative aspects of nomadic behavior. This
colony could therefore serve as a convenient control for our
artificially-fed colony.

Food for both experimental colonies consisted of adult and brood
individuals of the myrmicine ant Novomessor cockerelli, and
workers of the termite genus Gnathamitermes. To collect Novo-

1981] Topoff, Rothstein, Pujdak, & Dahlstrom — Army Ants 153

messor brood we made use of the panic-alarm behavior that this
prey species exhibits when raided by army ants. Accordingly, we
released several hundred adult i V. nigrescens into the nest entrance
of Novomessor, and aspirated the larvae and pupae that were
removed from the nest by their own adult workers. Whenever
colonies were artificially fed, food was given at the start of raiding in
the evening, while the column was within 2 m from their bivouac. If
columns emerged from more than one exit hole, booty was placed at
the front of each ant column.

Results

The raiding and emigration activities of colony no. 1 are summar-
ized in Table 1. This colony was found on July 11, late in its
nomadic phase. It became statary on July 13, after settling into a
kangaroo rat mound ( Dipodomys spectabilis). On the third statary
night, the colony conducted a 3-m long shift to the other side of the
mound. A statary shift differs from a nomadic emigration in that it
is neither preceded nor followed by raiding. It consists instead of a
single, unbranched column, and is presumably caused by a dis-
turbance at the old site. For the next 13 statary days, the colony
remained at the same bivouac, and staged either brief (1-3 hr) or no
predatory raids. On statary day 17, however, the colony conducted a
longer shift to an adjacent mound. During the move, we observed
that all of the pupae were deeply pigmented, and that a few callows
were being transported by mature adults to the new site. As a result
of detecting the onset of eclosion, we started artificial feeding of the
colony on the next night (July 30), and continued to supply food for
a total of six consecutive nights (Table 1).

Each evening, a basal column "appeared on the surface shortly
after sunset (1800-1900 hr). As soon as the ants contacted the food,
the process of mass recruitment resulted in a sharp increase in ant
traffic out of the nest. On the days of heaviest feeding, when more
than 30 g of booty were provided, the army ants required several
hours to transport it back to the bivouac. The colony occasionally
put out additional raiding columns later each night, but all captured
booty was promptly brought back to the original bivouac, and no
emigrations occurred. On the afternoon of August 5, the study area
received 14 mm of rainfall between 1400-1550 hr. The overcast sky,
coupled with cool temperatures late in the afternoon, enabled the

154

Psyche

[Vol. 88

Table 1. Activity schedule for Neivamyrmex nigrescens colony no. 1

Date

Activity

Raid Emigrate

Food

Provided (g)

Proposed

Phase-Day

7/11

N-?

7/12

N-?

7/13

S-l

7/14

S-2

7/15

ss*

S-3

7/16

S-4

7/17

S-5

7/18

S-6

7/19

S-7

7/20

S-8

7/21

S-9

7/22

S-10

7/23

S-l 1

7/24

S-12

7/25

S-l 3

7/26

S-14

7/27

S-l 5

7/28

S-l 6

7/29

ss*

S-l 7

7/30

9.1

S-18

7/31

34.7

N-l

8/1

18.5

N-2

8/2

32.6

N-3

8/3

17.6

N-4

8/4

31.6

N-5

8/5

N-6

8/6

N-7

8/7

N-8

8/8

N-9

8/9

N-10

8/10

-f

N-l 1

8/11

- .

N-12

8/12

N-13

8/13

N-14

8/14

S-l

8/15

S-2

8/16

S-3

8/17

8/18

S-5

8/19

S-6

8/20

S-7

8/21

S-8

8/22

•

S-9

statary shift

1981] Topoff, Rothstein, Pujdak, & Dahlstrom — Army Ants 155

colony to begin raiding earlier than usual. Thus, although we
arrived at the site by 1800 hr, a long (60 m) emigration was already
in progress. Given the large size of the colony, we decided to
terminate food-augmentation. The colony remained nomadic for
the next nine days, during which time it emigrated on six nights.

In order to determine whether we had been successful in delaying
the onset of the nomadic phase, three independent types of evidence
were analyzed: (1) phase length: (2) callow pigmentation; and (3)
larval size. Collectively, our data indicate that the nomadic phase
was indeed delayed for 4-8 days.

Phase Length: Because the colony was temporally anchored, July
13 can be considered the first statary day, August 5 the first nomadic
day. Thus, the statary interval becomes 23 days (Table 1). Accord-
ing to Mirenda and Topoff (1980), the range of statary-phase
duration for N. nigrescens in the same study area is 15-19 days, with
a modal length of 16 days. This suggests that the minimum delay in
nomadic onset for our colony was 4 days. If we use instead Mirenda
and Topoffs modal duration, the delay is calculated as 7 days.

Callow pigmentation: Newly eclosed callows of N. nigrescens are
yellow and acquire adult-like pigmentation between 7-12 days.
Several hundred callows were collected from the colony during its
first emigration on August 5, and compared with preserved samples
collected daily from nomadic colonies in previous years. Although
this form of visual comparison can not always pinpoint the exact
post-eclosion day, callows from the artificially-fed colony were
substantially more pigmented than those typically collected from
other colonies on the first nomadic night. Our comparison between
these callows and previously preserved specimens indicated a post-
eclosion age of between 5-8 days.

Larval size: Several hundred larvae were collected by aspiration
from the first emigration. By visual inspection, we separated the 10
largest and 10 smallest larvae and measured them with the aid of a
dissecting microscope fitted with an ocular micrometer. The mean
length of the large group was 4.0 mm (range = 3. 8-4. 2 mm), as
compared with a mean of 1.5 mm (range = 1.3- 1.7 mm) for the
small group. When these data are compared with Mirenda and
Topoffs (1980) graph of larval growth versus nomadic day, they
correspond to a range of nomadic days between 4-6.

The nightly patterns of activity for colony no. 2 and for the
control colony are summarized in Table 2. For this small colony, we

156

Psyche

[Vol. 88

£ £

t—

£ ^

C/0

c/o

<d o

o ■-

S 2

t ao

a £

O <D

«n

*-> 3

03 i-

Q °

l/~>

r~~

—

r-~

— -

r-~

I-"

c/i

S*.

,&c

-o

aj _2

c/3 -o

Cl, <U

S-15

S-16

N-l

N-2

N-3

N-4

N-5

N-6

N-7

N-8

N-9

S-l

S-2

S-3

■a

^ -C

0- cL

■ C

' ”

-a -a

O OJ

aj
C A

O -a

r-"

Pm >
O

—

•a

-a

C\|

<D o
cj •-

^aj

c q

j- i

>»

I) 2

t. ao

3 6

<x>

C/3

’>

cj <l>

o ^2

w ‘5

03 i-

*-• ,

5 o

Q °

C/3

Tj-

r~-

l/~>

1981] Topoff, Rothstein, Pujdak, & Dahlstrom — Army Ants 157

were able to monitor the time of onset and the duration of each
night’s raid, in addition to the emigration frequency. This colony
was released from a laboratory nest at 1900 hr on August 7. Because
this was statary day 15, most of the pupae were fully pigmented. The
colony promptly moved into a subterranean nest beneath a small
hole in the desert floor. The first raiding column appeared shortly
after 2200 hr, at which time 9.0 g of Novomessor brood and termites
were placed near the raiding front. The army ants removed the
booty in less than 1 hr, after which all surface activity ceased. For
the next seven nights, the colony was either not active on the surface
or, at best, conducted brief raids (each of which was immediately
followed by artificial feeding) but no emigrations. On August 15 we
arrived at the study site after 2200 hr, and found the colony
emigrating 25 m to the NW. Because previously-collected food was
being transported to the new nest, but no larvae had yet appeared,
we considered the emigration to be in an early stage. Accordingly,
10.2 grams of booty were placed near the emigration column, 1 m
from the old bivouac. This resulted in recruitment of ants both from
the short column leading to the old nest, and from the longer
emigration column. All of the artificially-placed food was taken
back to the old nest, and the emigration was aborted.

On August 19 (nomadic day 1 1), after 2 days of not having been
fed, the colony conducted its only successful emigration. The move
took the colony 19 m to the N, beneath an Ephedra bush. On
August 25, we excavated the colony and forced it to shift its statary
bivouac. This procedure verified that the colony’s larvae had
pupated. Thus, throughout a nomadic phase lasting 14 days, the
colony conducted only one completed emigration. On 4 nomadic
nights no raiding occurred. During the 10 nights in which raiding
took place, the median time for raid onset was 2200 hr, and the
median duration of each raid was 1.5 hr.

The control colony, which was also released from the laboratory
at the end of a statary phase, exhibited more typical patterns of
nomadic behavior (Table 2). During a 15-day nomadic phase, the
colony emigrated on 1 1 nights. Some degree of raiding took place
on every nomadic night. The median time of raid onset for the
control colony was 1850 hr, and the median duration of raiding was
9.7 hr.

158

Psyche

[Vol. 88

Discussion

Much of the discussion generated by Schneirla’s brood-stimula-
tion theory concerns the relative degree to which raiding and
emigrations are influenced by interactions between brood and
adults (internal processes), and by external environmental factors.
Theoretical support for emphasizing brood-related processes stems
not only from Schneirla’s own research with army ants (Schneirla,
1957, 1958, 1971), but from studies of other social insects as well.
For example, honeybee workers can collect protein-rich pollen or
carbohydrate-rich nectar. Louveaux (1950) found that the amount
of pollen collected by an incipient colony is small, but increases as
the brood population increases. In another experiment (Louveaux,
1958), he removed the colony queen from a mature colony and
found that pollen collection was unaffected until many of the larvae
had pupated. Further evidence of larval stimulation of adult
foraging came from Fukuda, 1960 (in Free, 1967), who showed that
foraging workers from a recently-divided colony collected very little
pollen until the eggs laid by the new queen hatched into larvae.
Finally, Free (1967) demonstrated that adult worker foraging was
influenced more by direct access to the brood than by brood odor
alone. Perhaps most significant was the additional finding that
artificially feeding a colony with pollen resulted in a decrease in
pollen collection and a corresponding increase in nectar collection.

Although Schneirla was primarily concerned with the role of
callow and larval excitation, he did recognize the role of food as an
ecological parameter. Thus, at an early stage of his field research
with the neotropical genus Eciton, he reported (Schneirla, 1938)
that colonies frequently emigrate along the heaviest raiding route of
that day. Nevertheless, it was Rettenmeyer (1963) who first sug-
gested that the location and amount of captured food might
influence not only the path of colony movements, but the very
tendency to emigrate in the first place. The idea that colony
excitation could be related to brood satiation has received empirical
support from Free’s (1967) study of honeybees and from related
research with the myrmicine ant genus Myrmica (Brian, 1957, 1962;
Brian and Abbott, 1977; Brian and Hibble, 1963). It was therefore
significant that by the time of Schneirla’s last field study, concerning
emigration behavior in the paleotropical army ant genus Aenictus,
he conceded that short-term variations in colony excitation may

1981] Topoff, Rothstein, Pujdak, & Dahlstrom — Army Ants 159

depend upon the “alimentary condition prevalent in the brood”
(Schneirla and Reyes, 1969), and that emigrations are likely to begin
soon after food has run low.

In a recent series of field and laboratory studies of nomadic
behavior in nearctic colonies of N. nigrescens (Topoff and Mirenda,
1980 a,b; Mirenda et al., in press), we demonstrated: (1) that the
location of booty clearly influences the direction of raiding and
therefore of emigrations; and (2) that artificially-fed colonies exhibit
a lower frequency of emigrations. The present study differs from
these in that food augmentation began late in the statary phase,
before most of the callow population had eclosed. In addition to
delaying the onset of the nomadic phase by reducing excitation from
newly-eclosed callows (Topoff et ah, 1980a), this was our first
attempt to eliminate emigrations through a complete nomadic phase
in the field.

During the six days of food augmentation for colony no. 1, we
provided a total of 144 g of booty. Since the colony generated few
additional raiding columns, the artificially-administered booty
represents over 90% of the colony’s total food intake for that period.
According to Mirenda et ah (1980), colonies of N. nigrescens gather
approximately 0.4 mg of booty/ larva/ nomadic night. Thus, on the
average, we provided colony no. 1 each night with an amount of
food that would be collected by a colony containing about 60,000
larvae. Although our estimate of colony size contains an error of ±
20%, we can be reasonably certain of having provided this colony
with about 1.2 times the amount of food it would normally gather.
Although the large size of this colony dictated that we could no
longer supplement its food to the same degree throughout the
remainder of the nomadic phase, the evidence from phase length,
callow pigmentation, and larval size supports the conclusion that
the onset of the nomadic phase was delayed for 4-8 days.

For colony no. 2, which was considerably smaller and more
precisely counted, intensive overfeeding was more feasible. On the
average, 8.8 g of booty were provided on food-supplemented nights.
This is more than 5 times the amount of food that a colony of this
size would collect in the field. In view of this feeding regime, it is not
surprising that the colony conducted only one completed emigration
throughout its 14-day nomadic phase. We must emphasize, how-
ever, that a reduction of the frequency of nomadic emigrations is by
itself not sufficient to infer a relationship between food supply and

160

Psyche

[Vol. 88

colony arousal. In all of our laboratory and field studies, we always
placed food near raiding fronts, within a few meters of the bivouac.
In most cases, the army ants established few or even no additional
raid columns beyond the artificial feeding site. Thus, our feeding
procedure reduces the ability of the ants to locate a suitable nesting
site, which is a prerequisite for a successful emigration (Mirenda et
al., in press). The case for a relationship between food supply and
colony arousal is made considerably stronger by considering, in
addition to emigration frequency, the temporal aspects of the ants’
raiding behavior. Colony no. 2 conducted no raids on 4 nomadic
nights. By comparison, the complete absence of raiding (on
stormless nights) for a nomadic colony of N. nigrescens has never
been reported, although it is a common occurrence for statary
colonies. Finally, when we include the data on raid onset and
duration for colony no. 2, we conclude that overfeeding can
effectively shift the level of overall colony activity from a nomadic
to a statary condition.

Acknowledgments

The base of operations for this study was the Southwestern
Research Station of The American Museum of Natural History.
This research was supported by NSF Grant BNS-8004565, and by
PSC-CUNY Grant 13492.

Literature Cited

Brian, M. V.

1957. Food distribution and larval size in cultures of the ant Myrmica rubra L.
Physiol. Comp. Oecol. 4: 329-345.

Brian, M. V. and A. Abbott

1977. The control of food flow in a society of the ant Myrmica rubra L. Anim.
Behav. 25: 1047-1055.

Brian, M. V. and J. Hibble

1963. Larval size and influence of the queen on growth in Myrmica. Insectes
Soc. 10: 71-81.

Free, J. B.

1967. Factors determining the collection of pollen by honeybee foragers.
Anim. Behav. 15: 134 144.

Louveaux, J.

1950. Observations sur le determinisme de la recolte du pollen par les colonies
d’abeilles. C.R. Acad. Sci., Paris 231: 921-922.

1958. Recherches sur la recolte du pollen par les abeilles ( Apis mellifica L.)
Ann. Abeille 1: 1 13-188, 197-221.

1981] Topoff, Rothstein, Pujdak, & Dahlstrom — Army Ants 161

Mirenda, J. and H. Topoff

1980. Nomadic hehavior by ants in a desert-grassland habitat. Behav. Ecol.
Sociobiol. 7: 129-135.

Mirenda, J., D. Eakins, K. Gravelle, and H. Topoff

1980. Predatory behavior and prey selection by army ants in a desert-grassland
habitat. Behav. Ecol. Sociobiol. 7: 119-127.

Mirenda, J., D. Eakins, and H. Topoff
in press. Relationship between raiding and emigration in the nearctic army ant
Neivamyrmex nigrescens. Insectes Soc.

Rettenmeyer, C. W.

1963. Behavioral studies of army ants. Univ. Kansas. Sci. Bull. 44: 281-465.

Schneirla, T. C.

1938. A theory of army ant behavior based on the analysis of activities in a
representative species. J. Comp. Psychol. 25: 51-90.

1957. Theoretical considerations of cyclic processes in doryline ants. Proc.
Amer. Phil. Soc. 101: 106-133.

1958. The behavior and biology of certain nearctic army ants: last part of the
functional season, southeastern Arizona. Insectes Soc. 5: 215-255.

1971. Army ants: a study in social organization. Freeman, Calif. 349 p.

Schneirla, T. C. and R. Z. Brown

1950. Army-ant life and behavior under dry-season conditions. 4. Further
investigations of the cyclic processes in behavioral and reporductive
functions. Bull. Amer. Mus. Nat. Hist. 95: 265-353.

Schneirla, T. C. and A. Y. Reyes

1969. Emigrations and related behavior in two surface-adapted species of the
Old World doryline ant, Aenictus. Anim. Behav. 17: 87-103.

Topoff, H. and J. Mirenda

1980a. Army ants on the move: relationship between food supply and emigra-
tion frequency. Science 207: 1099-1100.

1980b. Army ants do not eat and run: relationship food supply and emigration
behaviour in Neivamyrmex nigrescens. Anim. Behav. 28: 1040 1045.

Topoff, H., J. Mirenda, R. Droual, and S. Herrick

1980a. Onset of the nomadic phase in the army ant Neivamyrmex nigrescens:
distinguishing between callow and larval excitation by brood substitu-
tion. Insectes Soc. 27: 175-179.

1980b. Behavioural ecology of mass recruitment in the army ant Neivamyrmex
nigrescens. Anim. Behav. 28: 779 789.

LIFE HISTORY OF ANTAEOTR1CHA SP.

(LEPIDOPTERA: OECOPHORIDAE: STENOMATINAE)

IN PANAMA*

By Annette Aiello
Smithsonian Tropical Research Institute
P. O. Box 2072, Balboa, Panama

The subfamily Stenomatinae (Oecophoridae) is a New World
microlepidopteran group of approximately 35 genera and more
than 1200 species. Its range is from the United States through
Argentina; South America is especially rich in species. Little is
known of the biology of these moths, but those that have been
studied include leaf miners, stem borers, and seed eaters. The genus
Antaeotricha Zeller, of similar range, comprises more than 400
species, many of which are leaf tiers.

Three individuals of Antaeotricha sp. near fractilinea (Walsing-
ham) (Figure 1) were reared from larvae collected 29 March through
4 April 1980 on Barro Colorado Island (BCI), Panama. Two
additional individuals were preserved, one in its final instar, the
other as a pupa.

The larvae had constructed tubes (Figure 2) of silk, frass, and cast
head capsules, on the undersides of the leaves of Mascagnia nervosa
(Malpighiaceae).

At the time of collection two of the three larvae were in their final
instar and these pupated five days later. The third individual,
probably a first instar judging from its small size and tiny tube,
molted the day after collection. Due to the uncertainty regarding
instar number, letters instead of numbers are used to refer to instars.

Mascagnia nervosa is a liana which grows into the canopy of the
BCI forest. Seedlings are found frequently around the edges of
clearings and in tree falls. Antaeotricha larvae were common on the
older leaves of plants 10-30 cm tall and bearing three to eight leaves
each; some leaves supported as many as four larval tubes, although
one or two were most common. Possibly Antaeotricha attacks
leaves of this plant in the forest canopy as well.

Head capsule widths (Table 1) ranged from 0.18 mm (instar A) to
1.38 mm (final instar). Instar durations for the larva collected as
instar A were: 5(B), 4(C), 4(D), 3(E), 5(F), and 8(G) days.

♦Manuscript received by the editor September 14, 1981.

163

164

Psyche

[Vol. 88

Figures 13. Antaeotricha sp. (Rearing lot 80-30). 1. Adult (individual no. 4).

Scale = 3 mm. 2. Silk and frass tubes on underside of leaf of Mascagnia nervosa.
Scale = 7 mm. 3. Final instar larva (individual no. 1) reaching out of its tube.
Scale = 2 mm.

Early instars (A, B) had pale-colored heads and green bodies.
They constructed hard tubes of silk and frass, 0.5 mm in diameter,
on the undersides of leaves. Most tubes were initiated along major
leaf veins, a position which may offer protection to young larvae.
The initial tube was dense and the larva inside could not be seen
through its wall.

Subsequent instars extended the tube, each tube addition being
wider and longer than previous portions.

Larvae did not leave their tubes to feed; they reached out (Figure
3) and scraped cells from the leaf surface.

Commencing with instar C, larvae had dark heads and a pink
mesothorax. The remainder of the body was green as before.
Intermediate instars (C, D) continued construction of the hard tube,
and also fed by scraping the leaf surface immediately in front of the
tube opening.

Later instars (E-G) ate whole leaf. Portions of the tube,

1981]

Aiello — Life History of Antaeotricha

165

Table 1. Summary of variables for seven instars of individual no. 4 of Rearing
lot 80-30. (P = pale, pink; D = dark; G = green; S = skeletonizer, soft; H = hard; W =
eats whole leaf).

IN STAR
D

Instar duration (days)

Head capsule width (mm)

0.18

0.25

0.35

0.48

0.63

0.83

1.38

Head capsule color

Mesothorax color

Feeding habit

Tube consistency

constructed by these instars, were soft and less dense than previous
sections. The apical 2-3 cm of these tubes were extremely diffuse
and the larvae inside could be seen clearly. Having consumed all
nearby leaf, hungry larvae dismantled this diffuse portion and
shifted it laterally until additional leaf surface was located, some-
times on other leaves of the same plant.

The final instar larva (G) (Figure 4) was about 7 mm long, lacked
secondary setae; had prolegs on abdominal segments 3-6 and 10;
crochets uniordinal, arranged in a circle; prespiracular wart of
prothorax long, curving part way around spiracle, and bearing three
setae; mesothorax with a single seta on tubercle pi; abdomen with
setae alpha and beta widely separated, setae eta and kappa adjacent,
seta beta on ninth segment placed higher up than alpha, and setae
beta on ninth segment the same distance apart from one another
across the dorsum as on previous abdominal segments; head with
front extending about one-half the way to the vertex. Because the
two sides of the mesothorax bore slightly different setation, on the
specimen studied, both are illustrated in the setal maps (Figure 4).
For ease of comparison, the map of the right mesothorax is shown
in mirror image.

The day before pupation, final instar larvae abandoned their
tubes and dropped to the floor of the cage. No cocoon was
constructed.

The pupa (Figure 5) was ovate in outline, 5 mm long, 2 mm wide;
with wings extending slightly beyond caudal margin of fourth
abdominal segment; apical 1.3 mm of antennae adjacent on the
meson except at their extreme tips, and bearing a distinctive raised

166

Psyche

[Vol. 88

Figure 4. Antaeotricha sp. Final instar head capsule (front and lateral views),
and setal maps. P = prothorax, MS = mesothorax, L = left, R = right, MT =
metathorax, A = abdomen, S = suranal plate. Map of right mesothorax is shown in
mirror image. Scale for head capsule = 0.5 mm.

1981]

Aiello — Life History of Antaeotricha

167

Figure 5. Antaeotricha sp. Pupa (ventral, lateral, and dorsal views). Scale = 1
mm.

structure on the scapes; mesothoracic legs and maxillae ending in
the “V” formed by the meeting of the antennae; prothoracic legs
slightly shorter than mesothoracic legs; labial palpi evident as tiny
triangle between maxillae bases; fronto-clypeal and epicranial
sutures distinct; abdominal segments 1-4 longer than others;
spiracles evident on abdominal segments 2-8; cremaster of 8 weak
setae.

Pupation lasted 10, 11, and 12 days for the three individuals
reared. Total development time, for the individual collected as
instar A, was 40 days. Allowing an additional three days for instar
A, and four days for egg maturation, actual development time was
probably close to 47 days.

Spread adults, pointed head capsules and pupal skins, and a larva
and pupa in alcohol are in the collection of the author, deposited in
the National Museum of Natural History, and are labelled as
Rearing lot 80-30.

I would like to thank the Smithsonian Tropical Research

168

Psyche

[Vol. 88

Institute, Panama, for use of their facilities, J. B. Heppner (SI) for
identification of the moth, and R. Silberglied for reading the
manuscript.

Reference

Watson, A. and P. E. S. Whalley

1975. The Dictionary of Butterflies and Moths in Color, xiv + 296 pp., 405 figs.

POLISTES GALUCUS IN MASSACHUSETTS
(HYMENOPTERA: VESPIDAE)*

By Mary A. Hathaway
Museum of Comparative Zoology,

Harvard University

Cambridge, Massachusetts 02138 USA
Introduction

Polistes gallicus (Linnaeus), a common and widespread paper
wasp in the palearctic region, has been introduced into the United
States in the Boston, Massachusets, area. During 1981 specimens
were collected in Cambridge, Somerville, Belmont, and Newton,
Massachusetts. P. gallicus was also collected in Cambridge in 1980,
but was not seen in Belmont that year (R. J. McGinley, personal
communication). Species identification was verified by Dr. Arnold
S. Menke of the Systematic Entomology Laboratory of the United
States Department of Agriculture. Presumably P. gallicus was only
recently introduced; otherwise it would surely have been reported
before this. It is a brightly colored and conspicuous wasp.

The purpose of this paper is to report the introduction of Polistes
gallicus. The biology of the species in the Old World is reviewed
briefly, and some observations of the wasp in Massachusetts are
reported. Information on how to recognize P. gallicus is also
included.

Biology of Polistes gallicus (Linnaeus)

Polistes gallicus is ubiquitous in the palearctic region, especially
in the south. It is the most common Polistes in Spain (Giner Mari,
1945). The species’ range extends north to Paris, but gallicus
becomes rare in far northern France. It exists in warmer parts of
Belgium and Germany, but does not occur in England, Denmark, or
Scandinavia (Guiglia, 1972). Spradbury (1973) states that occasion-
ally Polistes are introduced into the British Isles, but for some
reason the genus is not able to sustain itself there. To the south, the
range of P. gallicus includes northern Africa, where the species is
known from desert oases (Richards, 1953), and extends east through
Israel and Iran. In Asia P. gallicus has been collected in southern

♦Manuscript received by the editor September 29, 1981.

169

170

Psyche

[Vol. 88

U.S.S.R. and throughout China, east to the Pacific coast (Guiglia,
1972; Yoshikawa, 1962). Generally, the species is found in warmer
and dryer localities within its range. It is not common above 1000 m
elevation, although in southern Spain specimens have been collected
above 2000 m (Guiglia, 1972).

The biology of P. gallicus varies considerably between the
climatic extremes of the area it inhabits. North of the Alps, nests are
built in enclosed places, such as metal containers and gutter pipes.
This type of nest has also been reported by Pardi from the coast of
Tuscany in Italy (Guiglia, 1972, and references therein). Throughout
most of Italy, however, the nests of P. gallicus are built in the open,
and typically hang from eaves, branches, or other protective
horizontal structures. The nest hangs from a slender peduncle with
its disc oriented horizontally, and the cells opening downward.
Often there are several accessory peduncles. Infrequently, nests are
found whose discs are oriented vertically.

In Italy, where the species has been studied extensively, nests of
close to 500 cells have been reported on several occasions (Guiglia,
1972). It is apparently common for colonies of this species to
become quite large in the south.

Polistes gallicus colonies are haplometrotic (with a single found-
ress) in northern Germany and presumably throughout the northern
extent of the species’ range. Further south, for instance in southern
Germany, pleiometrotic colonies (with several foundresses) are
occasionally reported (Richards, 1953). In Italy the species is
typically pleiometrotic, although as with the pleiometrotic colonies
reported from Germany, one queen is clearly dominant and lays
most of the eggs. The subordinate, or accessory females function as
workers in the nest. According to Pardi (1948), after the first
workers emerge the accessory females are chased off the nest or
killed by the queen. This situation resembles nest founding in P.
fuscatus (Fabricius), the common paper wasp in northeastern
United States, except that in P. fuscatus subordinate females are
usually allowed to remain on the nest after workers have emerged
(West, 1967). In Africa P. gallicus colonies reproduce by swarming,
with a reproductive female leaving her nest in the company of
several workers (Richards, 1953).

Local Observations

I report here on 2 nests of P. gallicus in Cambridge, Massachu-

1981]

Hathaway — Polistes gallicus

171

setts. Both were in enclosed situations, similar to those described as
typical in the northern parts of the species’ range in Europe. The
first nest was located inside a metal pole supporting a stop sign. The
pole was IV2 cm in diameter, and open at the top. The single
peduncle of this nest was located 28 cm from the top of the pole. The
nest was suspended from the pole’s side and faced north. This nest
contained 134 cells, and measured 8 cm high and 5 cm across.

A second nest was located inside an open vertical pipe, 35 cm tall
and 8 cm in diameter. The nest was suspended from the side of the
pipe and faced west-north-west. Its peduncle was located 6 cm from
the top of the pipe. This nest contained 153 cells and also measured
8 cm X 5 cm.

P. gallicus does not seem to be an aggressive species. I have been
able to observe a nest from as close as 15 cm, apparently without
disturbing the wasps.

The prognosis for permanent establishment of P. gallicus in the
western hemisphere appears good. The species seems quite able to
withstand the climate in the northeast. 1980-1981 was an unusually
cold year in Boston, with 5,819 degree days accumulated between
June 1 and May 30, as opposed to the 30-year normal of 5,597
(United States National Weather Service statistics, telephone in-
formation).

Recognition of Polistes gallicus

In northeastern United States P. gallicus would, more likely be
confused with a yellow jacket ( Vespula spp.) than with another
paper wasp. Although its shape and flight are similar to the native
Polistes, it is relatively small and its markings and coloring are
strikingly different. P. gallicus is black with bright yellow macula-
tions (see figure 1).

Specimens collected in Massachusetts have varied considerably in
their markings, with some showing more yellow than others,
especially on the clypeus. This has also been true of specimens
collected from the same nest. P. gallicus is known to be quite
variable in Europe (Guiglia, 1972).

Males have completely yellow faces and their antennae are curled
at the tips, a characteristic common in the genus. Their antennae are
quite short, however, compared to males of other species. In other
superficial respects, males of P. gallicus resemble the females.

172

Psyche

[Vol. 88

Figure 1. Polistes gallicus worker (left) and male (right) collected in Cambridge,
Massachusetts. Length of worker, 1.4 cm.

Acknowledgements

I wish to thank Dr. Howard E. Evans and Dr. Ronald J.
McGinley for their helpful comments on the manuscript, Dr. Frank
M. Carpenter for photographing the specimens, and Dr. Arnold S.
Menke for the species identification. I also wish to thank Mr. Loren
Preston of the Cambridge Department of Traffic and Parking, for
the loan of a stop sign pole, and Mrs. Rita Kelley for permission to
work on Her property.

References

Giner Mari, J.

1945. Himenopteros de Espana, fams. Vespidae, Eumenidae, Masaridae,
Sapygidae, Scoliidae, y Thynnidae. Madrid: Instituto Espanol de
Entomologia. 142 pp.

Gijiglia, Delfa

1972. Les Guepes Sociales d’Europe Occidentale et Septentrionale. Faune de
l’Europe et du Bassin Mediterranean, VI. Paris: Masson et Cie. 181 pp.

Pardi, L.

1948. Dominance order in Polistes wasps. Physiol. Zool. 21(1): 1-13.

1981]

Hathaway — Polistes gallicus

173

Richards, O. W.

1953. The social insects. London: Macdonald and Co. 219 pp.

Spradbury, J. Philip

1973. Wasps: an account of the biology and natural history of solitary and
social wasps. Seattle: University of Washington Press. 408 pp.

West, Mary Jane

1967. Foundress associations in polistine wasps: dominance hierarchies and
the evolution of social behavior. Science 157: 1584-1585.

Yoshikawa, Kimio

1962. Introductory studies on the life economy of polistine wasps, VI:
geographical distribution and its ecological significances. J. Biol. Osaka
City University 14: 19-43.

NOTES ON THE POPULATION ECOLOGY OF CICADAS
(HOMOPTERA: CICADIDAE) IN THE CUESTA ANGEL
FOREST RAVINE OF NORTHEASTERN COSTA RICA*

By Allen M. Young
Department of Invertebrate Zoology,

Milwaukee Public Museum
Milwaukee, Wisconsin 53233

Introduction

Several previous field studies of cicadas (Homoptera: Cicadidae)
in Costa Rica have revealed that different sympatric genera and
species often exhibit allochronic (seasonal) annual adult emergence
patterns and habitat associations (Young 1972; 1974; 1975a; 1976;
1980a, b,c; 1981a,b,c). Most of these studies concerned cicadas
associated with lowland tropical forest and the Central Valley
regions of Costa Rica, although one study in particular (Young
1975) examined some aspects of the population ecology of cicadas
in a mountain forest. Because different species, and sometimes
genera, of cicadas are found in different climatic and geographical
regions of Costa Rica (Young 1976), it is necessary to examine the
population ecology of these insects in as many of these ecological
zones as possible. This paper summarizes an ecological survey of the
cicadas thriving in the steep and very rugged forest ravine known as
“Cuesta Angel” in the Central Cordillera of northeastern Costa
Rica. The information reported here complements the studies of
cicadas in other ecological zones of Costa Rica, although by no
means does as extensively owing to the difficulties working on the
very steep slopes of the ravine. It is shown tentatively that (a) the
cicada fauna of this region includes at least two species not
discussed or found in the other regions studied, (b) the resident
species exhibit different annual emergence patterns, and (c) nymphal
skins of several species are distributed at very low densities and in
association with various genera and species of leguminous canopy-
size trees in the ravine habitat.

* Manuscript received by the editor June 12, 1981

175

176

Psyche

[Vol. 88

Fig. 1. The ravine forest at Cuesta Angel, near Cariblanco, Heredia Province,
Costa Rica.

Methods

The Cuesta Angel ravine is an extensive strip of very steep
primary and river-bottom forest (Fig. 1) filtering down from the
highest mountains of Costa Rica’s Cordillera Central and tapering
into the northeastern lowlands known as Sarapiqui. Because of its
rugged profile much of the ravine remains blanketed in forest even
though surrounding level areas have been largely converted to
pastures. This ravine is within the recently extended Carillo Nation-
al Park. There have been relatively few field studies of plants and
animals in the ravine, even though both its invertebrate and
vertebrate faunas contain many forms not found in other parts of
Costa Rica. “Cuesta Angel” is located about 10 km south of the
village of Cariblanco (10° 16'N, 84° 10'W), Heredia Province, and is
classified as montane tropical wet forest (elev. about 1200 m)
(Holdridge 1967). The vertical drop in the ravine is about 300 m.

1980]

Young — Ecology of Cicadas

177

As shown by 1972 and 1973 rainfall data, the region is very wet
and with a short and erratic dry season during January and
February (Fig. 2). For either collections of nymphal skins or
determination of species active by calling songs or collection of
specimens, the locality was visited the following dates: 27-30 June
1972, 14 August 1972, 15-17 February 1973, 20-24 March 1973,
18-20 April 1973 (beginning of nymphal skin regular censuses),
22-25 May 1973, 6 10 June 1973, 4-7 July 1973, 7-9 May 1975, 3
April 1976, 1 and 5 November 1980. Dates of visit included both wet
and dry periods for this region. During the April 1976 visit. Dr.
Thomas E. Moore recorded calling songs of the species active at
that time.

The 1973 visits were concerned primarily with attempting to
census the nymphal skins of various species active at different times
of the year while other dates were devoted to listening and collecting
adult specimens. The nymphal skins of recently emerged cicadas are
relatively easy to distinguish from those of a previous years’
emergence owing to discoloration and disintegration of some parts
(Young 1980a) and therefore provide an accurate record of a recent
or current emergence within the year. The locations of nymphal
skins in the habitat also provide information on the possible feeding
associations of the nymphs in the ground and other aspects of
microhabitat. I censused nymphal skins, with the assistance of at
least one, and usually two, trained student assistants by marking off
rectangular or square plots (usually 5X5 meters) immediately be
neath a tree or other spot where at least one nymphal skin was
found. Initially we crawled through the forest along transects to
determine where nymphal skins were found and then marked off the
trees and places having them. The transect approach was used in the
survey of the very rocky terrain comprising the river-edge forest on
relatively flat ground, but working on the steep slopes entailed spot-
checking various places owing to the difficulty of the terrain and
often very misty conditions. Thus the nymphal skin census program
involved repeated censuses at twelve marked canopy-size trees on
the slopes, and four large river-edge plots of forest, each plot
containing many trees. The four river-edge plots, each one widely
separated from the other by at least 100 meters of forest, ranged in
size from 462m2 to 300m2, the differences being due to rivulet
channels and other interruptions in the forest. With the exceptions
of marked trees 2, 6, and 7 (each of which was a plot of about 90m2),

178 Psyche [Vol. 88

SUCCESSIVE MONTHS

Fig. 2. A sample of three separate years of monthly rainfall patterns at
Cariblanco. In all three years portrayed, a short dry spell occurs between January
and March, although conditions are not completely dry as in other regions of Costa
Rica with distinct dry seasons.

1980]

Young — Ecology of Cicadas

Fig. 3. The forest habitat at the top of the ravine, and above the Sarapiqui
roadcut. The cicada Fidicina n.sp. is abundant here.

most tree plots on the slopes were 25m2. The twelve tree plots gave a
total habitat area of about 484m2 sampled for nymphal skins several
times and a total of 6,957m2) of river-edge forest sampled as well
(total sample area of 7,441m2). The tree plots were widely scattered
with the closest being no less than 30 meters apart. The sample
included the hill-top forest above the Sarapiqui roadcut (Fig. 3) as
well as the forest habitat to either side of the secondary road down
into the ravine (Fig. 4). A census consisted of making an exhaustive
collection of all cicada nymphal skins found within each plot,
including those attached to plants and tree trunks and those lying in
the ground litter. The contents were placed into a plastic bag and
labeled appropriately. Later the skins were determined to species
and sex. The nymphal skins of the cicadas studied were readily
separated to species in my field samples on the basis of marked
differences in size, color, and body profile. Skins were matched with

180

Psyche

[Vol. 88

others obtained from collecting skins when adults were emerging. In
previous studies (Young 1972; 1975a; 1980a, b; 1981a, c) I have
illustrated and discussed distinguishing features of cicada nymphal
skins. Based upon these materials, a key to the Costa Rican cicada
fauna, using both adults and nymphal skins, is being formulated
(T.E. Moore and A.M. Young, in preparation). In the present study,
it was very easy to distinguish nymphal skins of Fidicina species
(three species) on clear-cut differences in color pattern and size; the
Zammara species studied has nymphal skins very different in color
and body profile from the others (see also Young 1972), while the
two species of Carineta species had nymphal skins differing in color,
even though of very similar size. One species has a very dark brown
nymphal skin, and the other, light brown. Based upon matching of
skins with adults done by myself and T.E. Moore, I am reasonably
certain that matches of field collections of skins with adults is very
reliable. Voucher specimens of fruits and leaves of the trees having
nymphal skins beneath them were collected and sent to specialists
for determination.

Other observations included determining the places on the ravine
where adult cicadas were heard chorusing as a means of estimating
preferences among species for the river-edge area and top of the
ravine. In some instances, diurnal patterns of calling were also noted
and the trees used for calling. Once the species were determined,
records of captures of cicadas in other regions of Costa Rica were
checked by examining the University of Michigan collections and
data bank on Neotropical species in other museums, as a means of
determining if the Cuesta Angel species were found elsewhere in
Costa Rica. Because virtually nothing is known about the geo
graphical distributions and habits of Neotropical cicadas in general,
vouchers of both adults and nymphal skins were saved and placed in
collections at the University of Michigan and the Milwaukee Public
Museum.

Owing to the steep terrain and heavy rains of the region, a small
experiment was conducted on estimating the rate of disintegration
of cicada nymphal skins on both forested slope and river-edge
forest. Such a test would tell me how many skins were being missed
between census intervals because they were possibly disintegrated,
particularly on the slopes, before the next census was taken. Thus in
the May 1973 census, two groups of fresh skins of one of the larger
species, each group containing ten skins, were established, one

1980]

Young — Ecology of Cicadas

181

Fig. 4. The forest habitat along the secondary road going to the bottom of the
ravine. Cicadas such as Fidicina sericans, F. mannifera, and two species of Carineta
are heard in the trees along this road.

group of a patch of forest slope where this species emerges, the other
on a level area adjacent to the Sarapiqui River. The skins in each
group were randomly distributed (by throwing) within a one-meter
square area of ground. The numbers of skins remaining in each plot
were then checked in June and July 1973.

Results

The six species of cicadas found and studied at Cuesta Angel are
shown in Fig. 5, and they are: Zammara tympanum Distant,
Fidicina sericans Stal, Fidicina “new species” (n.sp.), a new species,
Fidicina mannifera Fabricius, Carineta postica Walker, and Cari-
neta sp. Three of these, Z. tympanum, F. sericans, and F. manni-
fera, are large-bodied cicadas with very loud calls, while F. n.sp. is
medium-sized, and the two species of Carineta are considered small-
sized (or at the low end of the medium-size range), the latter two

182

Psyche

[Vol. 88

Fig. 5. Cicadas found in the Cuesta Angel ravine forest, top, from left to right:
Zammara typanum, Fidicina mannifera, F. sericans; bottom, left to right: F. n.sp.,
Carineta sp., and C. postica. The vertical black line to the left of each cicada gives the
scale of one cm. relative to the body shown in each photograph.

1980]

Young — Ecology of Cicadas

183

cicadas having very soft calls. Zammara tympanum adults are heard
throughout most of the year and sometimes during the dry season
and they call from the moss and other epiphyte-covered trunks of
forest trees primarily along the river-edge. This cicada is mottled
green and brown and has brown spots on the wings, immediately
distinguishing it from the others. The call is a “winding up-like
pulsating buzz. Adults when calling occur at one per tree, and there
are usually no more than two or three calling males present within
approximately 800m2 parcels of river-edge forest during an optimal
calling period. Males call throughout the day, including overcast
and light drizzle conditions. Males are bright green with brown
markings while females are drab olive green and brown.

Fidicina sericans, both sexes, are black with green markings on
the thorax and smoky wings. The call is a steady rather high-pitched
buzz most frequently heard during sunny weather and during the
dry season. Sometimes several males congregate in the same tree,
particularly if it is along an edge of forest, and sometimes, under
these conditions, several adjacent exposed trees may have males
calling at the same time. The calling males are seen perched on the
upper portions of the trunk and on branches, and they are easily
spotted on light-colored bark species such as Pourouma and
Cecropia. Adult densities, as indicated by calling males, probably
are about 1-20 cicadas per 800m2 of forest during a period of peak
calling, although this may be an underestimate since only a fraction
of males may chorus at any one time. Calling males are heard
primarily on the forest slope and less so at the bottom of the ravine
and at the very top.

Fidicina n.sp., both sexes, possesses a green head and thorax and
black and orange-banded abdomen, sometimes with patches of
silvery hairs laterally. Of all of the cicadas in Costa Rica, this species
is the most difficult one to catch because of their habit to perch very
high in trees and to change trees after one call. Based on compari-
sons with type materials and other specimens, this is most likely a
new species. It has a very distinctive two-part call: the first part is a
series of pulsating chirps followed by a longer period of siren-like
and pulsating calls. Unlike this species, both Z. tympanum and F.
sericans, as well as the other species to be discussed, often make
repeated complete calls from the same perch, even if interspersed
with periods of silence lasting several minutes or an hour or two. F.
n.sp. is heard during the dry season and it occurs in the ravine and

184

Psyche

[Vol. 88

above the Sarapiqui roadcut. Adult densities appear to be very low,
similar to that of Z. tympanum, but difficult to determine due to the
highly mobile habits of males.

Fidicina mannifera, both sexes, is dark brown with some dark
green markings on the head and thorax and with tinges of brown
along the veins of the wings. The body is very pubescent. Males
generally call at dusk and dawn and usually for about 15 20 minutes
during each period. The call is a very intense pulsating shrill buzz.
Based upon observing a total of close to 20 individuals at this
locality, there is about a 50:50 chance that a male just completing a
call will stay in the same tree. Males are heard primarily inside the
forest and on the lower slope and along the river. Densities are very
low with probably only one or two males per 1000m2 of forest
habitat.

Carineta postica, also illustrated in Young (1975), is black with
green markings on the head and thorax and with the entire body
blanketed in setae. The wings are smoky and calling males have the
habit of perchng head-downward on the trunks of forest trees, a
behavioral trait separating the larger-sized members of the genus
from all other Neotropical cicadas. Males sing from moss-covered
tree trunks and branches inside the river-edge forest and along the
river itself. Densities are low, with 1-5 calling males per 500m2 of
forest and with calling limited to the late afternoon or overcast
conditions during the dry season. The call consists of repeated
coarse “zip-zip” sounds, and is reminiscent of a muted version of the
call of the familiar cone-headed grasshopper of North America.
This species may also be C. trivitatta Walker as specimens of both
species are very similar in size and coloration. Clarification awaits
further study.

Carineta sp. is pea-green with clear wings and calls from forest
edge trees such as Ceeropia during the wet season. It is of same size
and profile as C. postica but is most abundant near the top of the
ravine. The call is also similar to that of C. postica but somewhat
louder and calling is generally a dusk phenomenon. Sometimes as
many as eight males have been seen perched at different heights on
the trunk of the same Ceeropia tree.

The data on temporal emergence patterns annually from the
censuses of nymphal skins present a more diffuse picture of
seasonality in the cicadas at Cuesta Angel (Fig. 6). Caution is given
here in that these data are very fragmentary and discontinuous,

ontl

Young — Ecology of Cicadas

185

Carinefa postica

Carineta sp.

iiii 1

«l

Zarorora tympamam

Fictfcina mannifera

Fadicina n. sp.

DATA

— :

Fidicina sericarts

MMj

DATA

Itfl

April

1973

May June July August

Months of census

collections of cicada nymphal skins from tree plots and river-
x Angel.

186

Psyche

[Vol. 88

although the best there are at this time. With the exception of C.
postica, there appears to be a trend for most species to emerge
during both wet and dry seasons, when considering both the
nymphal skin and call records together. Thus although F. sericans is
heard in abundance during the short dry season, there is some
evidence of emergence well into the wet season (Fig. 6). But
examining the 1973 rainfall data shows a marked dip in rainfall
during July (Fig. 2), giving a brief dry spell that month. If it is
assumed that the data are actually representative of emergence
patterns of cicadas at Cuesta Angel, it then appears that another dry
season species, C. postica, did not respond to the July 1973 dry spell
as there was no emergence (Fig. 6). At the same time, the dry spell
was apparently insufficient in intensity to block the emergence of
wet season species such as Z. tympanum. Perhaps even more
interesting is the wet season emergence of another supposedly dry
season species, F. n.sp. (Fig. 6). Adults of such species were not
heard at these times although my sample sizes are very small.
Different patterns of emergence may be associated with different
years in whch monthly rainfall regimes are very different. For
example, during 7-9 May 1975, there was an abundance of F. n.sp
calling in the ravine as was the case for 4-7 July 1973. Both of these
months, in different years, were drier than in other years, and the
rainfall data for 1972 and 1973 clearly show the year-to-year
variation in monthly rainfall patterns at this locality (Fig. 2).
Furthermore, when F. n.sp. emerged during the wet season, calling
was restricted to the dry periods of the day. All of the cicadas
studied exhibit bursts of calling near dusk (see also Young 1981b).

The distribution of nymphal skins for each species studied by
marked trees is given in Table 1. Even though approximately 70
species of canopy-size trees were included along the initial transects
to determine the locations of cicada emergence patches in the
ravine, patches were found to be confined to the species of
Leguminosae listed in Table 1. Note that the estimation of relative
abundance of adults among the species discussed above is confirmed
here in terms of nymphal skins: by far the most abundant species is
F. sericans, whose nymphal skins comprised almost 64% of the total
241 skins collected in the 1973 survey of tree plots alone (Table 1).
F.n.sp., Z. tympanum, and Carineta sp. are about evenly distributed
in terms of abundance of nymphal skins in the tree plots. As in
previous studies of cicadas in Costa Rica, sex ratios are close to

Table 1. Census history of cicada nymphal skins in legume tree plots* in the forest ravine, “Cuesta Angel”, near Cariblanco,
Heredia Province, Costa Rica.

1980]

Young — Ecology of Cicadas

187

w <3

<D ^

c *c:
a o

a to

c c

O O O O

O O <N O O O O

Tf — O O — O

O rM O O

O r’b m O O O — r- — Tf — m

0 — 00

O O O O <N O O

or- — ornniTj-oomo

— — Tt0I^-0^v~>0

§3 ~g

£ s;

a, *5 a.

a <3 53

be be be

c c c

(NmTt^ot^oooNO

O O

O'- Qv
— CN
+1

m in
rn \o
O O

— nj

m r-

r- p
<n —

— <N

+ 1

*Each plot ranged in size from 5X5 meters to 10X9 meters around the base of individual legume trees.

**A11 trees and cicadas were censused 18-20 April, 22-25 May, 4-7 July, 15 August 1973 (total of 14 days), except for trees9, 10,
1 1, and 12, which were added to the census program on 4 July 1973.

188

Psyche

[Vol. 88

unity. Taking the most abundant species, F. sericans, there is
considerable range in numbers of skins found in the different tree
plots, although close to 40% of all skins of this species were found
beneath one individual of Pithecollobium latifolium (Table 1). Yet a
second individual of this tree species yielded only four skins of
cicadas overall and none of F. sericans. Such data, although limited,
indicate the considerable variation encountered over different
patches of the same resource for a cicada species in tropical forests.
Two different individuals of Inga and one P. latifolium together
account for almost 65% of all skins found. That such data may be
underestimates of true values, even for an abundant species such as
F. sericans, is suggested by the results of the estimate of rate of
disintegration of nymphal skins: at the end of a five-week period,
between 50% (level ground) and 80% (slope) of the F. sericans
nymphal skins studied had disappeared. These samples are pitifully
small, but it is the best we have at this time. The intervals between
censuses in my study are of this magnitude and greater, thereby
indicating the likelihood that some skins were missed owing to their
rapid disintegration under very wet conditions. The examination of
nymphal skin distributions by tree plots and river-edge plots
separately provides further confirmation of the data shown in Table
1 (Table 2). Although high percentages, if not all, of plots are
occupied by skins of Z. tympanum, the emergence is one of very low
density since only a small number of skins occur in the plots studied
(Table 2). The tree plots, although only representing an area of
about 6.5% of the combined area of tree plots and river-edge plots,
account for almost 80% of all skins recovered (Table 2). The larger
river-edge plots include a wide variety of tree species whereas the
tree plots each include one individual of a leguminous tree species
and understory plants. Most striking is the relatively high density of
the nymphal skins of F. sericans in the tree plots, almost 0.4
skins/ m2 (Table 2). Yet the same cicada, in a much larger and
representative tract of forest, representing an area about five times
that of the tree plots, has the very low density of about 0.010
skins/ m2 (Table 2). Other patterns of nymphal skin density between
tree plots and river-edge plots are self-evident and support the
pattern discussed for F. sericans (Table 2). From such results, one
can readily appreciate the distortion of density estimates when
different size patches of the environment, with different biological
attributes, are combined to give a summary figure (Table 2). And

Table 2. Some popluation parameters of cicada species in the ravine forest, “Cuesta Angel”, Costa Rica

1980]

189

Young — Ecology of Cicadas

SP- 18 mm 9 3 0.002/ m2 23 8 0.060/ m2 0.005/ m2

*There are four river-edge plots of these sizes: 2,145m2, 1,350m2, 462m2, and 3,000m2 for total area of 6,957m2.

* All estimates of density based on occupied plots only: no empty plots included.

*Most of these plots are 25m2 for a total area of 484m2.

190

Psyche

[Vol. 88

the data also show, that for larger areas of environment as typified
here by the river-edge plots, there are not necessarily going to be
increases in densities of insects recovered.

Discussion

Of the six species studied at Cuesta Angel, none are exclusive to
the locality, but other locality records from Costa Rica indicate
similar elevations and habitat. Fidicina n.sp., Zammara tympanum,
and both species of Carineta have been collected at Turrialba,
Cartago Province as shown by specimens in the collections at The
University of Michigan Museum of Zoology. The species I term C.
postica may also be C. trivitatta Walker, which has also been
collected from the San Jose area, Guapiles (Limon Province) and
Bajo La Hondura (San Jose Province). Two other cicadas, F.
sericans and F. mannifera, have much more extensive ranges in
Costa Rica as both have been collected and studied in premontane
and lowland tropical wet forest regions of the Atlantic coastal
watershed (Young 1972; 1980b), and mannifera also occurs in the
semi-dry to dry forest region of the western provinces of Puntarenas
and Guanacaste (Young 198 la, c). Given the topography of the
Cuesta Angel region relative to the adjacent lowlands, it is not
surprising to find species such as sericans and mannifera along a
more or less continuous elevational gradient within the wet forest
region and over a range of about 90 1 100 meters. Yet this is not true
for the genus Zammara or Carineta since entirely different species
occur in the adjacent premontane and lowland wet forest regions of
northeastern Costa Rica (Young 1972; 1976; 1980b). From both
records of adults calling and nymphal skins, both sericans and
mannifera occur at much lower densities in the Cuesta Angel
montane wet forest than they do in adjacent premontane and
lowland wet forests. Given these records, it is concuded tentatively
that cicadas such as F. n.sp., Z. tympanum, and the two species of
Carineta studied are montane species associated with wet forests
while F. sericans and F. mannifera are lower elevation forms also
associated with generally wet forests and semi-dry forests. Thus the
Cuesta Angel cicada fauna is a mixture of montane and lower
elevation tropical wet forest cicadas.

Both generic and specific richness of cicadas at Cuesta Angel are
not as high as they are in the adjacent lower elevations. There are six

1980]

Young — Ecology of Cicadas

191

genera and about ten species of cicadas found in the adjacent
premontane tropical wet forest (Young 1980b) as studied about 25
km from the Cuesta Angel locality. Young (1975) found only two
genera, each monospecific, at another montane wet forest locality,
Bajo La Hondura. There are also greater numbers of genera and
species found in mid-elevation moist forest (Young 1980a) and
lowland tropical dry forest (Young 1981a) in Costa Rica. Cicadas
such as F. sericans and F. mannifera are tentatively interpreted as
being ecological “leaks” into the forested ravine at Cuesta Angel.
Given the continuous accessibility to lower elevation wet forest
habitats moving along the ravine into the lowlands, it is un-
reasonable to expect some highly mobile insects to colonize at either
end (Young 1975b).

Elsewhere in Costa Rica, cicadas have been found to have distinct
seasonal patterns of adult emergences each year (Young 1972;
1975a; 1980a, b; 198 la, c) with the recognition of usually three kinds
of cicadas: dry season, wet season, and transitional forms between
dry and wet seasons. From the studies of cicadas in premontane
tropical wet forest in particular (Young 1980b), however, it became
apparent that brief spells of wetness in a dry period and of dryness
in the wet season may trigger emergence of wet season and dry
season species respectively. In the premontane tropical wet forest
zone of northeastern Costa Rica, typically wet season cicadas such
as Z. smaragdina Walker will emerge in low numbers during a rainy
spell of about five days or longer within the dry season (Young
1980b: pers. obs.). During a dry spell within the long rainy season at
the same locality, F. sericans, a typical dry season cicada, can also
be heard and fresh nymphal sLins found (Young 1980b; pers. obs.).
Such observations indicate that “seasonality” in tropical cicadas is a
very flexible sort of emergence strategy, perhaps determined by
critical periods of wetness or dryness, depending upon the species
and locality. Such an effect may explain the anolamous emergence
of F. sericans in the wet season at Cuesta Angel. The data from
Cuesta Angel very tentatively provide additional support for this
phenomenon, as shown for species such as F. n.sp. The proximal
cues triggering emergence in tropical cicadas have not been studied
to my knowledge, although some ideas have been suggested for
study (Young 1975; 1980a, b; 1981a). What are also needed are
detailed studies of the effects of small changes in air temperature
and humidity, and light intensity on the behavior of adult cicadas in

192

Psyche

[Vol. 88

the tropics over a typical diurnal cycle. Different species may
possess different levels of physiological capacity to cope with
stressful environmental conditions imposed by either too much
wetness or too little wetness. From my work on cicadas in Costa
Rica over the past eleven years, and particularly from data on
densities of nymphal skins of co-occurring species in the same
patches of habitat, it seems doubtful that seasonal emergence
patterns in cicadas is related to interspecific competition in develop-
ing cicadas. From what little information I have, there is little
reason to suspect competition for oviposition sites. But the great
diversity in the properties of the calling songs among co-occurring
species in tropical forests, and the tendency for several species to
form single species aggregates of chorusing males (Young 1980c)
suggest that there might be competition for optimal calling condi-
tions. In cicadas, the calling song is a major component of fitness
since it presumably functions in mating, and there might be strong
selection to evolve allochronic emergence patterns when the calling
songs of species conflict and reduce mating success. Certain types of
seasonal changes in the environment, yet to be determined, may
provide the most ready cues for these insects to exploit in evolving
allochronic emergence patterns to reduce losses in mating success.
The whole system warrants considerable detailed study as it involves
different stages in the life cycle. Seasonal emergence may or may not
have anything to do with conditions being intrinsically optimal for a
certain species in a certain region at a certain time of the year. If the
latter, the cicada is merely locking in to a convenient cue since,
under this hypothesis, both wet and dry periods provide suitable
resources for adults, including those associated with mating needs.

Under the mating conflict hypothesis, it is implied that cicadas
with very low densities and with unusual calling habits may forego
entering into such a selection arena, thereby circumventing this
adaptive pathway and emerging throughout most of the year, other
things being equal. A typical case in point is the almost catholic
habit of F. mannifera to sing for a brief period at dusk and under
conditions of low population densities in Costa Rica (Young 1972;
1980b; 1981b; this paper). The intensity of the presumed mating
conflict is considered to increase as population densities of co-
occurring species increase individually.

In virtually all other regions studied, the greatest numbers of
cicada nymphal skins occur beneath adult legume trees (Young

1980]

Young — Ecology of Cicadas

193

1972; 1980a, b; 198 la, c) although precise data on the abundances of
skins in legume plots versus non-legume plots is still lacking. At
Cuesta Angel cicada nymphal skin patches too were found beneath
legume trees. If legume trees provide some form of highly suitable
environment for developing cicadas in tropical forests, emerging
populations each year will be spatially disjunct according to the
spatial distribution of the legume trees whose root crowns provide a
primary resource for developing nymphs. The suitability of
Leguminosae for developing cicadas may involve both physical and
chemical properties of the classes of root sizes exploited by various
age-classes of nymphs. The observed low densities of nymphal skins
in all of the plots at Cuesta Angel, relative to previously obtained
densities of the same or similar species in other regions (e.g., Young
1980a, b; 1981a), may therefore be a function of the very dispersed
condition of the legume trees at this locality. A striking contrast is
made with the association of nymphal skins of species such as Z.
smaragdina and F. sericans in relatively large patches of adult
Pentaclethra macroloba in nearby premontane tropical wet forest
(Young 1980b). Densities of these cicadas range from 5.4 to 9.3m2 in
patches of two or more P. macroloba, estimates considerably
greater than for the same species at Cuesta Angel. I interpret such
observations to be the result of P. macroloba occurring as clumps of
several trees thereby increasing the size of a single resource patch for
cicadas, which results in either greater oviposition in the patch or
greater survival of nymphs, or both. The river-edge plots in the
Cuesta Angel study illustrate quite well such an effect. Such plots,
although quite large, only contained one or two widely scattered
legume trees and not clumps of such trees, and some did not contain
legumes at all but were situated near such trees. The observed very
low densities of cicada nymphal skins in these large segments of
forest is due to an absence or scarcity of suitable root crowns for
cicadas. The tree plots, on the other hand, although very small, are
highly suitable for cicadas and therefore densities are high.

The pattern of cicada nymphal skins being associated with legume
trees in tropical forests can have other explanations as well, ones not
involving a presumed coevolved interaction of the sort suggested
above. For example, selective logging of tropical forests may leave
behind the relatively soft-wood legumes thereby increasing their
relative abundance as a resource for insects such as cicadas. Thus
the likelihood for an ovipositing cicada to discover a legume tree

194

Psyche

[Vol. 88

increases greatly over a period of years, even though the root crowns
and other cicada-related characteristics of other trees are equally
suitable for cicada development.

Since all plots were located at or withn the lower one-fourth of
the ravine, the instances in which some species, such as F. n.sp. and
Carineta sp. call primarily from the top of the ravine and not at the
bottom suggests a behavioral response associated with mating
requirements. Such species presumably emerge near the bottom of
the ravine and fly up to the top for courtship. Such species may also
emerge near the top as well although this was not determined in this
study. The observed patterns of adult calling sites within the ravine
are presumably related to the acoustical and thermoregulatory
needs of each species.

Summary

The genera and species of cicadas, their seasonal distributions,
habits, and emergence sites were studied discontinuously over
several years at the Cuesta Angel ravine, a rugged mountain tropical
wet forest locality in the northern portion of the Central Cordillera
of Costa Rica. Emphasis was placed on determining the distribution
of cicadas down one steep forested side of this approximately 300-
meter deep ravine and along a representative portion of its bottom
(Rio Sarapiqui). Some evidence of seasonal fluctuations in abun
dance was obtained for the six species found here, and the greatest
densities of nymphal skins of all species were found in small plots
around individual legume trees. Densities in the large river-edge
plots, containing many different kinds of trees, were relatively very
low. The data are compared to similar data on cicadas from other
regions of Costa Rica. Tropical cicada seasonality, interactions with
Leguminosae, and possible mechanisms underlying population den-
sities, are discussed.

Acknowledgements

This research was originally supported by National Science
Foundation Grant BG-33060 of the United States of America, and
subsequently by the Milwaukee Public Museum. Field assistance
was given (1972-73) by students from Lawrence University. Dr. J.
Robert Hunter, then Director, Costa Rican Field Studies Program
of The Associated Colleges of the Midwest, provided logistical

1980]

Young — Ecology of Cicadas

195

support. Dr. Dieter C. Wasshausen of the Smithsonian Institution
provided identifications of the tree species discussed. Dr. Thomas E.
Moore of The Museum of Zoology at The University of Michigan
assisted with the identification of the cicadas, gave the author access
to the UMMZ cicada collections and other information concerning
cicadas, accompanied the author on one of the visits to the locality
to record the calls of cicadas, and discussed cicada biology with the
author. I thank Dr. Henk Wolda, who, in the capacity of a referee
for this journal, made many helpful suggestions on the manuscript.

Literature Cited

Holdridge, L. R.

1967. Life zone ecology. Tropical Science Center, San Jose, Costa Rica.

Young, A. M.

1972. Cicada ecology in a Costa Rican tropical rain forest. Biotropica
4:152-159.

1974. The population biology of Neotropical cicadas. III. Notes on the
behavioral natural history of Pacarina cicadas in some Costa Rican
grasslands. Entomol. News 85:239-256.

1975a. The population biology of neotropical cicadas. I. Emergences of Procol-
lina and Carineta in a mountain forest. Biotropica 7:248-258.

1975b. “Leakage” of Morpho theseus (Lepidoptera: Nymphalidae) into north-
eastern Costa Rica? Brenesia 6:59-67.

1976. Notes on the faunistic complexity of cicadas (Homoptera: Cicadidae) in
northern Costa Rica. Revista de Biologia Tropical 24:267-279.

1980a. Habitat and seasonal relationships of some cicadas (Homoptera: Cicadi-
dae) in central Costa Rica. American Midi. Naturalist 103:155-166.

1980b. Environmental partitioning in lowland tropical rain forest cicadas. J.
New York Entomol. Soc. 88:86-101.

1980c. Observations on the aggregation of adult cicadas (Homoptera: Cicadi-
dae) in tropical forests. Canadian J. Zoology 58:711-722.

1981a. Seasonal adult emergences of cicadas (Homoptera: Cicadidae) in north-
western Costa Rica. Milwaukee Public Museum Contributions in
Biology and Geology, No. 40.

1981b. Temporal selection for communicatory optimization: The dawn-dusk
chorus as an adaptation in tropical cicadas. American Naturalist: 117:
826-829.

1981c. Notes on seasonality and habitat associations of cicadas (Homoptera:
Cicadidae) in premontane and montane tropical moist forest in Costa
Rica. J. New York Entomol. Soc., 89: 123-142.

CAMBRIDGE ENTOMOLOGICAL CLUB

A regular meeting of the Club is held on the second Tuesday
of each month October through May at 7:30 p.m. in Room 154,
Biological Laboratories, Divinity Avenue, Cambridge. Entomolo-
gists visiting the vicinity are cordially invited to attend.

BACK VOLUMES OF PSYCHE

Requests for information about back volumes of Psyche should
be sent directly to the editor.

F. M. Carpenter

Editorial Office, Psyche
16 Divinity Avenue
Cambridge, Mass. 02138

FOR SALE

Reprints of articles by W. M. Wheeler

The Cambridge Entomological Club has for sale numerous reprints
of Dr. Wheeler’s articles that were filed in his office at Harvard
University at the time of his death in 1937. Included are about
12,700 individual reprints of 250 publications. The cost of the
reprints has been set at 5c a page, including postage; for orders
under $5 there will be an additional handling charge of 50c. A list of
the reprints is available for $1.00 from the W. M. Wheeler Reprint
Committee, Cambridge Entomological Club, 16 Divinity Avenue,
Cambridge, Mass. 02138. Checks should be made payable to the
Cambridge Entomological Club.

ISSN 0033-2615

PSYCHE

A JOURNAL OF ENTOMOLOGY

founded in 1874 by the Cambridge Entomological Club

Vol. 88 1981 No. 3-4

CONTENTS

Dedication: Robert E. Silberglied. Frank M. Carpenter 197

Sound Production by Courting Males of Phidippus mystaceus (Araneae:

Salticidae). G. B. Edwards 199

Maternal Behavior and Alarm Response in the Eggplant Lace Bug, Gargaphia
solani Heidemann (Tingidae: Heteroptera). R. S. Kearns and

R. T. Yamamoto 215

Polymorphism and Division of Labor in the Dacetine Ant Orectognathus

versicolor (Hymenoptera: Formicidae). Norman F. Carlin 231

Trail Communication of the Dacetine Ant Orectognathus versicolor

(Hymenoptera: Formicidae). Bert Holldobler 245

Francis Walker Types of, and New Synonymies for, North American
Hydropsyche species (Trichoptera, Hydropsychidae). Andrew P. Nimmo 259
Territoriality, Nest Dispersion, and Community Structure in Ants.

Sally C. Levings and James F.A. Traniello 265

The Effect of Flower Occupancy on the Foraging of Flower-Visiting Insects.

V. J. Tepedino and F. D. Parker 321

Abdominal Trophallaxis in the Slave-Making Ant, Harpagoxenus americanus

(Hymenoptera: Formicidae). Robin J. Stuart 331

New Name for the Extinct Genus Mesagyrtes Ponomarenko (Coleoptera:

Silphidae S.L.). Alfred F. Newton, Jr 335

Historical Development of Bee Foraging Patterns in Central New York State.

Howard S. Ginsberg 337

Myrmecophilic Relationship of Pella (Coleoptera: Staphylinidae) to Lasius
fuliginosus (Hymenoptera: Formicidae) B. Holldobler, M. Moglich, and

U. Maschwitz 347

Behavioral Origin of Tremulation, and Possible Stridulation, in Green
Lacewings (Neuroptera: Chrysopidae). Peter Duelli and James B. Johnson 375

Arthropods Attracted to Luminous Fungi. John Sivinski 383

Index to Volume 88 391

CAMBRIDGE ENTOMOLOGICAL CLUB
Officers for 1981-1982

President Barbara L. Thorne

Vice-President Frances Chew

Secretary Heather Hermann

Treasurer Frank M. Carpenter

Executive Committee John Shetterly

Mary Hathaway

EDITORIAL BOARD OF PSYCHE

F. M. CARPENTER (Editor), Fisher Professor of Natural History,
Emeritus, Harvard University

W. L. BROWN, Jr., Professor of Entomology, Cornell University and
Associate in Entomology, Museum of Comparative Zoology
P. J. DARLINGTON, Jr., Professor of Zoology, Emeritus, Harvard
University

B. K. HOLLDOBLER, Professor of Biology, Harvard University
H. W. LEVI, Alexander Agassiz Professor of Zoology, Harvard University
R. J. McGlNLEY, Assistant Professor of Biology, Harvard University
ALFRED F. Newton, Jr., Curatorial Associate in Entomology, Harvard
University

R. E. SlLBERGLlED, Smithsonian Tropical Research Institute, Panama
E. O. WILSON, Baird Professor of Science, Harvard University

Checks and remittances should be addressed to Treasurer, Cambridge
Entomological Club, 16 Divinity Avenue, Cambridge, Mass. 02138.

Orders for missing numbers, notices of change of address, etc., should be sent to the
Editorial Office of Psyche, 16 Divinity Avenue, Cambridge, Mass. 02138. For previous
volumes, see notice on inside back cover.

IMPORTANT NOTICE TO CONTRIBUTORS
Manuscripts intended for publication should be addressed to Professor F. M.
Carpenter, Biological Laboratories, Harvard University, Cambridge, Mass. 02138.

Authors are expected to bear part of the printing costs, at the rate of $27.50 per
printed page. The actual cost of preparing cuts for all illustrations must be borne by
contributors: the cost for full page plates from line drawings is ordinarily $18.00 each,
and for full page half-tones, $30.00 each; smaller sizes in proportion.

Psyche, vol. 88, no. 1-2, for 1981, was mailed Deember 28, 1981

The Lexington Press, Inc., Lexington, Massachusetts

Robert Elliot Silberglied

This issue of Psyche is dedicated to the memory of Robert E.
Silberglied, a victim of the Air Florida accident in Washington,
D.C., on January 13, 1982.

Born in Brooklyn, N.Y., in 1946, Bob was already an enthusiastic
naturalist and entomologist even in his school days. He graduated
from Cornell University in 1967 and received his PhD from Harvard
in 1973. He remained at Harvard until July of 1981, as Assistant
Professor and later as Associate Professor in the Department of
Biology, teaching mainly the courses in entomology that I had given
for many years as his predecessor. He was also Assistant Curator
and later Associate Curator in the Entomology Department of the
Museum of Comparative Zoology. During the same period he was
associated with the Smithsonian Tropical Research Institute, spend-
ing about half of each year in Panama or other parts of the
American tropics. At the time of his death he was Staff Scientist
(Research Entomologist) at the Institute.

Bob joined our society on his arrival in Cambridge in 1968 and
for the next 14 years he was one of our most active and enthusiastic
members. He served as vice-president and president, and was a
member of the editorial board of Psyche for the past decade. At our
fall and winter meetings, he could always be depended upon to
relate some unusual collecting experience or to demonstrate with
superb photographs and specimens some of the remarkable insects
that he had collected in the tropics. He combined a warm and
sympathetic personality with a brilliant and imaginative mind. In
both respects he has left a lasting impression on our society and its
members.

The Smithsonian Institution has established the Robert E. Silber-
glied Memorial Fund to support student research and training in
tropical entomology. Those who wish to contribute a gift of any size
may send it to: Robert E. Silberglied Memorial Fund, Accounting
Office, Smithsonian Institution, L’Enfant 3500, Washington D.C.
20560.

Frank M. Carpenter, editor

197

Robert Elliot Silberglied

Photograph taken in 1981

PSYCHE

Vol. 88

1981

No. 3-4

SOUND PRODUCTION BY COURTING MALES OF

PHID1PPUS MYSTACEUS (ARANEAE: SALTICIDAE)i

By G. B. Edwards* 2

The courtship rituals of male salticids generally are considered to
be visually-oriented, despite the fact that a primarily tactile type of
courtship has been demonstrated for 2 species of Phidippus (Ed-
wards, 1975; Jackson, 1977). In addition, chemotactic cues probably
assist a male in locating a female in most species of jumping spiders
(Crane, 1949; Richman, 1977). I report here that males of Phidippus
mystaceus (Hentz) produce sound by means of a palpal stridulatory
mechanism as an integral part of their courtship. This is the first
known case of a salticid producing sound with this type of
mechanism; a similar stridulatory organ has been reported for
lycosid spiders (Rovner, 1975).

Petrunkevitch (1926) reported that the salticid Stridulattus stridu-
lans Petrunkevitch has a stridulatory organ (of type “d”, chelicera-
palpus; Legendre, 1963). However, he did not detect sound produc-
tion. The only other records of a salticid producing sound were by
Bristowe (1958), who reported that Euophrys frontalis (Walckenaer)
made a “distinct sound as the tarsal claws (of the legs I of the male)
hit the ground . . . ,” and by Bristowe and Locket (1926), who had
reported earlier on the same species, but had implicated the legs II
as the sound producers. In either case, it was not clear if the sound
produced by E. frontalis was an integral part of the courtship or
incidentally produced by the movement of the legs.

'Contribution No. 514, Bureau of Entomology, Division of Plant Industry, Florida
Department of Agriculture and Consumer Services, Gainesville, FL 32602.

2Florida State Collection of Arthropods, Division of Plant Industry, P. O. Box 1269,
Gainesville, FL 32602.

Manuscript received by the editor June 7, 1981.

199

200

Psyche

[Vol. 88

Experimental Procedure

Four females and 2 males of P. mystaceus were reared to maturity
from an eggsac containing 12 eggs.3 The spiders were housed
separately in 9 x 1 cm plastic petri dishes; twice a week they were
provided with water by moistening a wad of cotton within the dish
and were fed larvae of the cabbage looper, Trichoplusia ni (Hiibner).

Two different techniques were used for observing courtship. In
one method, the male was placed directly into a female’s petri dish,
on the side opposite the female. In the other method, the male and
female were placed 5-15 cm apart on a 30 x 10 cm section of a live-
oak branch, in order to simulate natural conditions. Temperature
ranged from 24-26° C for all sessions.

Six separate filming and/or recording sessions lasted 10-90 min.
each. Films were made using a Beaulieu Super-8-mm movie camera
and an Auricon Pro 600 16-mm movie camera. Sound recordings
were made with a Sony TC-756-2 reel-to-reel tape recorder and a
Turner S22D microphone. The audiospectrogram was produced on
a Kay 7029A Sound Spectrograph.

Results

Courtships were observed for one of the males (the second male
was killed by the first female with which he was placed). Typically a
male placed into the petri dish housing a female almost immediately
begins palpating the female’s draglines and her abandoned nests,
continuing this palpal exploration until he detects the female
visually.

If the female is not inside a nest when first seen by the male
(usually from 3-6 cm), the male begins producing a soft, audible trill
that is systematically repeated. By apparently engaging the substrate
with enlarged setae (macrosetae, Fig. 1; similar to those observed on
lycosids by Rovner, 1975), leverage is produced enabling a stridu-
latory mechanism on the palpus to be operated. This mechanism
consists of a plectrum-like projection of the tibial apophysis which
fits into a bowl-shaped area on the cymbium containing a compli-
cated file system. The entire mechanism is located laterally (ectally);
in lycosids it is located dorsally. Also, lycosids have the file on the

3Gravid female P. mystaceus collected by Robert Dye, 26 October 1975, 4 miles north
of Texas state line at a rest stop on 1-35 in Oklahoma, under a rock. Eggs were laid
November, 1975.

1981]

Edwards — Phidippus mystaceus

201

tibia facing a cymbial plectrum, the reverse of the condition in P.
mystaceus. The file system of P. mystaceus appears to consist of 2
types of adjacent file fields which blend into one another. Within the
concavity is a fan-shaped file, while along the distal edge of the
concavity is a linearly-arranged file similar to lycosid files. Neither
file is as well-defined as the lycosid files. The individual ridges of P.
mystaceus ’ files are rounded, whereas those of lycosids have distinct
edges; however, in P. mystaceus, both types of file are overlaid with
numerous short ridges of variable length (Fig. 2).

Fig. 1. Distal tip of palpus of male P. mystaceus showing ring of macrosetae (M)
encircling whorled chemotactile setae (W). On extreme left are scale-like setae (S)
which form part of a white and/or yellow spot which probably contributes to the
overall visual stimulus of a courting male (100X). Note the greater number of
macrosetae on the ectal edge (E); see text for explanation. The curved macrosetae
(extreme right) at the tip of the cymbium first contact the substrate and may facilitate
the backwards sliding motion of the palpi by reducing friction with the substrate.

202

Psyche

[Vol. 88

Fig. 2. Left: stridulatory area of male P. mystaceus on ectal edge of cymbium of left palpus, showing fan-shaped file (F), linear
file (L), and plectrum (P), which is a branch of the tibial apophysis (T) (250X); Right: closeup of portion of file indicated by
arrow (1300X).

1981]

Edwards — Phidippus my s face us

203

For each sound sequence, both palpi become engaged nearly
simultaneously by a backward movement in which the palpi appear
to be dragged along the surface of the substrate for a distance of
about 1 mm. Halfway through the backward movement, the cymbia
are bent backward at an angle to the palpal tibiae (Fig. 3). At the
end of the backward movement, the palpi remain stationary for a
fraction of a second while the cymbia are rotated outward (left
palpus clockwise, right palpus counter clockwise). The palpi are
then returned to their most anterior position, apparently by lifting
the palpi from the substrate and moving them forward. When the
palpi are in their most anterior position, they are clearly off the
substrate. A single cycle of palpal movement is approximately 0.8
second (5 frames at 6 frames per second).

Audiospectrograms indicate that 13-20 (x = 17, n = 8) paired
stridulations are made consecutively, separated by pauses subequal
in timing to the sound sequences (Fig. 4). Alternation of stridula-
tions and pauses occurs at the rate of 1.5 sound sequences per
second (at approximately 25° C).

Fig. 3. Diagrammatic illustration of movement of left palpus (ectal view) by male
P. mystaceus during stridulation. A. Anterior position. B. Backward movement,
during which cymbium is bent backward, moving fan-shaped file across plectrum. C.
Rotary movement, during which macrosetae are engaged in substrate and cymbium
is rotated outward, moving linear file across plectrum. F = File cavity, T = Tibial
apophysis.

204

Psyche

[Vol. 88

Simultaneous with the initiation of sound production, the male
extends his legs I forward, positioning them just above and parallel
to the substrate, and spread approximately 40° apart. The tarsi and
metatarsi are turned upward about 15° and occasionally flicked
upward together. On 1 occasion, at a distance of about 1 cm from
the female, the tarsi and metatarsi were flicked continuously for
several seconds at approximately 2 flicks per second.

The male’s approach usually is direct, without the zigzag move-
ment (lateral stepping movement) characteristic of some other
Phidippus species and many other salticids. Forward movement is
slow and halting, the male often remaining in one spot for several
minutes. Total courtship time is long compared to the rapid advance
of the males of some Phidippus species, on 3 occasions lasting
approximately 8 minutes before the female terminated the courtship
by leaving the vicinity. These 3 longest courtships reached an
advanced stage, wherein the male brought his legs I closer together,
touched the female, and attempted to mount her; however, none of
the 4 females allowed their male sibling to mate with them. Instead,
each raised her legs I to repel him, and, if the male was persistent,
lunged sharply forward with open fangs, struck downward with the
legs I, and forced him backward; the female then left the vicinity.

On 2 occasions, the male performed a zigzag display; once prior
to assuming his stridulatory stance, and once in the middle of
courtship after several sequences of stridulation. In the first in-
stance, the zigzag display was brief, lasting less than 30 seconds and
consisting of 4 changes of direction, with a pause between each
lateral move. In the second instance, during mid-courtship, 7
multiple zigzags occurred which included 1-3 changes of direction
during each lateral stepping sequence; total elapsed time was about
3 minutes.

If the female is initially in and remains in a nest when the male is
introduced into the petri dish, the male alternates palpation of the
substrate with sequences of stridulation. Upon finding the nest
sheltering the female, the male attempts to gain entrance by probing
and pulling at the silk with his legs I, interspersing sequences of
palpal vibration on the silk. (Note: other species of Phidippus
known to use a tactile courtship vibrate their entire body). I could
not determine the movement pattern of this palpal vibration (it

1981]

Edwards

Phidippus mystaceus

205

previously reported for any other spider.

206

Psyche

[Vol. 88

appeared to be similar in timing to stridulation), but the palpi were
not in contact with the petri dish, and no audible sound was
produced.

Discussion

Known reproductive behavior of the males of species of Phidip-
pus involves a male locating a female by visual or chemotactic
means (Richman, in press), a visually-oriented courtship by the
male consisting of a series of movements with the legs I and palpi
(usually while advancing in a zigzag path), mounting of the female
by the male, and mating. Typically the male is conspicuously
marked with bright and/or contrasting colors both anteriorly and
dorsally; the anterior patterns are displayed during courtship.
Unlike most other species, both males and females of P. mystaceus
are cryptically-colored gray spiders that live in trees (Specht and
Dondale, 1960; Warren et al., 1967, as P. incertus; see Edwards,
1977, for nomenclatorial comments); males have mostly anteriorly-
oriented modifications (Fig. 5). While anterior modifications are
probably used by each sex to identify the other (especially the
female recognizing the male as a conspecific and potential mate)
from distances of a few centimeters, visual identification at longer
distances might be severely handicapped by cryptic coloration. A
mechanism which increases the chance of one sex locating the other
could be selected for under these circumstances.

The role of acoustic or vibratory signals in the courtship of P.
mystaceus may have co-evolved with cryptic coloration. As selec-
tion for cryptic coloration increased in association with exploitation
of a new microhabitat (most Phidippus species live in the herb-
shrub zone), the role of visual communication might have been in
part supplanted by sound during courtship. The use of sound,
whether airborne or substrate-borne, would have several advantages
over conventional visual courtship, if the sound extended the male’s
communicatory distance from a few centimeters to over a meter (as
it appears to do based on the audible component available to the
human ear). Sound is transmitted well through solids, and consider-
ing that in this case sound is produced on the substrate, vibrations
through this medium may be most important for female-to-male
orientation (as Rovner, 1967, showed to be the case for wolf
spiders). By orienting toward the male upon perception of the

1981]

Edwards — Phidippus mystaceus

207

Fig. 5. Anterior views of male and female P. mystaceus, offspring of female
collected in Oklahoma, which were used in this experiment.

208

Psyche

[Vol. 88

sound, the female might sooner visually detect and be able to
evaluate the male as a prospective mate, and thus sooner choose to
wait for or flee from him. The advantages gained by the male by
increasing his communicatory distance might be: 1) alerting a
receptive female to his presence at a greater distance, possibly
causing her to remain in the vicinity for a longer period of time (and
perhaps inhibiting her predatory instincts), so that the male has a
greater chance of finding and courting her; 2) based on many
observations of P. mystaceus and of other Phidippus species, non-
receptive females usually avoid advancing males; thus, by alerting a
female to his presence at a greater distance, a male would reduce the
chance of stimulating an aggressive response by a non-receptive
female.

As evidence for these probable advantages, analysis of courtships
showed that the male began courting a female from 2-4 times
further away when unconfined (on the live-oak branch) than did
other species of Phidippus when observed under unconfined experi-
mental conditions (Edwards, 1975). At the greater distances, only
sound was used initially by a P. mystaceus male upon sighting a
female, indicating that this form of communication was important
in alerting a female to his presence. Sound was also used alternately
with palpal exploration of the silk when the male was in contact
with the female’s draglines in the petri dish, even though she was not
visible. Under natural conditions, a male likely would often en-
counter a female’s dragline prior to locating her; he could maximize
his chances of mating by beginning to signal immediately, regardless
of whether or not the female was visible to him.

Components of Behavior and Morphology

The male’s initial palpal exploration of the female’s silk draglines
and nests has been noted for other salticids (Richman, 1977). The
presence of a contact pheromone on the silk could indicate to a male
that a female was, or had been, in the vicinity. Contact pheromones
(Hegdekar and Dondale, 1969) and dragline following by males
(Tietjen and Rovner, 1980) have been demonstrated for some
lycosids, but have not yet been conclusively demonstrated for any
salticid. Foelix (1970) demonstrated the presence of chemosensitive
setae in certain araneid spiders and hypothesized that those setae
were contact chemoreceptors. He showed that the suspected chemo-

1981]

Edwards — Phidippus mystaceus

209

sensitive setae in araneids were innervated and structured in essen-
tially the same manner as pheromone receptors of insects. Hill
(1977a, b) noted that the whorled setae on the tarsi and palpal
cymbia of several species of Phidippus also resembled insect
pheromone receptors; male P. mystaceus have the same type of
setae on their palpal cymbia (Fig. 1).

The behavior in P. mystaceus of engaging the palpi against the
substrate is probably derived from similar behavior among its
relatives. Males of other species of Phidippus move their palpi up
and down or back and forth during courtship. This behavior
appears to pre-adapt them for engaging the substrate, since only a
slight change in the amplitude and / or attitude of these movements
would bring the palpi into contact with the substrate. The same
movement occurs more intensely and rapidly when a male encoun-
ters silk made by a female, in association with presumed chemotac-
tile exploration; it is likely that this is the evolutionary pathway of
the development of the use of sound in P. mystaceus.

The shape and arrangement of the macrosetae at the tip of the
cymbium are such that a downward, forward pressure would engage
them with the substrate. By dragging the palpus backward, enough
leverage apparently is produced to move the fan-shaped file across
the relatively stationary plectrum; however, the backward move-
ment and bend of the palpus also may be a prerequisite to
positioning the macrosetae onto the substrate. Once the palpus is
anchored onto the substrate, the cymbium is rotated laterally
outward, then the palpi are returned to their starting position. The
macrosetae are arranged in a circle around the tip of the cymbia in
P. mystaceus, with more macrosetae on the ectal edge than on the
ental edge, which enables the palpus to remain engaged with the
substrate as it rotates outward. The structure involved in sound
production by rotating is the linear file; the cymbium must be
rotated sideways due to the lateral position of the file.

By simulating the direction of palpal movement with a model, it is
apparent that the backward movement would cause the fan-shaped
file to be drawn across the plectrum, while the rotary movement
would bring the linear file into contact with the plectrum. The fan
shape of the proximal file would accommodate the arc-shaped
movement as the palpus is bent on the backward stroke; however,
sound does not seem to be produced by the fan-shaped file. Only

210

Psyche

[Vol. 88

one type of stridulation is produced, evidently from the linear file
(see figure 4); the function of the fan-shaped file remains unclear.
Although the files appear to be oriented so that they could be
stroked from either direction, the timing of a complete palpal
movement indicates that sound is produced only on the backstroke
and not on the return stroke. The mechanics of stridulation by P.
mystaceus are still incompletely understood, and need further study
with more sophisticated filming techniques.

Comparisons to Other Stridulatory Mechanisms

The behavioral application of the palpi to the substrate by P.
mystaceus differs from lycosids in that P. mystaceus moves the tips
of the palpi while stridulating during each brief sound sequence,
whereas the lycosids apparently remain attached in one place to the
substrate for a prolonged sequence of sound production. The
mechanics of sound production with the linear file are similar to
those of the lycosids with respect to the palpus anchored by
macrosetae and the similar file structure, but P. mystaceus differs
from the lycosids in the location of the stridulatory organ, the type
of movement needed to engage the file, and the reversed positions of
the file and plectrum. Rovner (1975) proposed a new category of
stridulatory organ (as an extension of the classification of Legendre,
1963), type “h” to accommodate those types of mechanisms in which
the file and plectrum (scraper) were on adjacent surfaces of a joint
within the same appendage. I propose a subdivision of Rovner’s
category, following Legendre’s method of subdividing categories:
type “h I” in which the file is on the more proximal segment (as in
lycosids), and type “h II” in which the file is on the more distal
segment (as in P. mystaceus).

The stridulatory mechanisms known in other spiders incorporate
plectrum and file systems on opposing faces of the chelicerae and
palpi, legs I and II, carapace and legs I, carapace and abdomen
(Gertsch, 1979), or between palpal tibia and tarsus (Rovner, 1975).
In each of these cases, either the plectrum is moved across a
stationary file or both plectrum and file are moved together. The
stridulatory mechanism of P. mystaceus differs from all of these in
that the primary moving part is the file. Although the plectrum is
passively moved in space during the movement of the palpus to

1981]

Edwards — Phidippus mystaceus

211

engage the substrate, the cymbium containing the files is actively
moved against the plectrum on the tibia. When the palpus is fixed
on the substrate with the macrosetae, again it is the cymbium that is
moved against the stationary plectrum.

Other Types of Vibratory Signaling

A third method of sound production in spiders, vibration (pro-
ducing a “buzz” similar to that of a fly), has been demonstrated for
the sparassid spider, Heteropoda venatoria (L.) (Rovner, 1980). In
the same paper, low amplitude appendage oscillations resulting in a
faint whirring sound were reported for Lycosa rabida Walckenaer.
Phidippus whitmani Peckham and Peckham employs entire-body
(?) vibration (lacking an audible component, but with a widely-
spaced stance similar to H. venatoria ) during its Type I visual
courtship (Edwards, 1980). This is probably an adaptation to its
microhabitat (mesophytic leaf litter), the same substrate used for
vibratory signaling by many lycosids. I have noted another vibra-
tory behavior that also seems similar to that of H. venatoria during
the Type II tactile courtship of Phidippus regius C. L. Koch, while
the male is contacting the nest of the female (Edwards, 1975).
Subsequent laboratory observation showed a similar behavior for
P. whitmani, although the timing of vibratory sequences was
different from those of P. regius, probably a species-specific differ-
ence. Jackson (1977) reported a similar behavior for P. johnsoni
(Peckham and Peckham) and suggested a similarity in some respects
to the vibratory courtships of web-building spiders. I suspect that
the vibratory courtships of Phidippus species, although not pro-
ducing an audible component that I could detect, may be more like
the courtship of H. venatoria than like web-builders, or perhaps all
3 groups produce vibrations in essentially the same way (i.e.,
“juddering” as in araneid males; Robinson and Robinson, 1980). It
is curious that all known forms of non-tactile direct inter-individual
communication not involving vision in salticids are acoustic or
vibratory signals (despite the contention of Crane, 1949, and other
authors, the use of airborne pheromones by salticids has never been
proven). In the case of P. mystaceus and P. whitmani, both visual
and vibratory signals are used simultaneously, although the 2
species produce vibrations in different ways.

212

Psyche

[Vol. 88

Conclusion

The use of stridulation to produce sound by P. mystaceus appears
to represent a third method of communication for salticids (a fourth
method, if the tarsal percussion of Euophrys frontalis is a valid
communicatory process). Despite the fact that females used for the
present research failed to respond favorably to courtship by their
sibling male, the behavioral and morphological evidence in the male
of a functional role for sound production during courtship is
substantial.4

Summary

Males of Phidippus mystaceus have a stridulatory organ located
on the tarsal and tibial segments of the palpi. This organ is
employed by males in the potential or actual presence of adult
females, and forms the most significant part of courtship by males.
The mechanics of stridulation are somewhat similar to those of
lycosids, and as with the lycosids, substrate vibrations may be the
most important component of stridulation. Evolution of sound
production by P. mystaceus is hypothesized to have occurred in
conjunction with the evolution of cryptic coloration. Sound produc-
tion is thought to extend the males’ communicatory distance,
compensating for fewer visual identification opportunities due to
the spiders’ cryptic coloration.

Acknowledgments

I would like to give special thanks to the following: Dr. Robert
Paul for his help in sound recording and producing the audiospec-
trogram; Dr. Jonathan Reiskind for the S. E. M. photomicrographs.
Thanks are also due; Mr. John Thorne and Mr. Stan Blomely for

4Two antepenultimate P. mystaceus were collected by G. B. Edwards, 28 July 1979,
Ocala National Forest, Marion Co., Florida, beating young scrub live oaks, and
reared to maturity (October, 1979). Although these specimens were collected and
reared after the research on the Oklahoma specimens was completed, and the
courtship was neither filmed nor recorded, a courtship and mating was observed for
this pair. Courtship appeared in all respects to be identical to that of the Oklahoma
male, including type of sound, stance, and the rapid upward flicking of the tarsi and
metatarsi at less than 1 cm distance from the female. Mating occurred in the female’s
nest and lasted 87 minutes until the female left the nest. Upon separating, the male
renewed courtship, initially showing a single lateral stepping sequence as in the
Oklahoma male. The female avoided the male, and the pair was separated.

1981]

Edwards — Phidippus mystaceus

213

aid in filming (16-mm); Mr. Lloyd R. Davis, Jr., for obtaining the
gravid P. mystaceus female for me; and Drs. Jerome S. Rovner,
Jonathan Reiskind, Thomas J. Walker, and Robert L. Crocker for
reviewing the manuscript.

Literature Cited

Bristowe, W. S.

1958. The World of Spiders. Collins, London. 305 p.

Bristowe, W. S., and G. H. Locket.

1926. The courtship of British lycosid spiders, and its probable significance.
Proc. Zool. Soc. London 1926(2): 317-347.

Crane, J.

1949. Comparative biology of salticid spiders at Rancho Grande, Venezuela.
Part IV. An analysis of display. Zoologica 34(4): 159-215.

Edwards, G. B.

1975. Biological studies on the jumping spider, Phidippus regius C. L. Koch.

M.S. Thesis, University of Florida. 64 p.

1977. Comments on some genus and species problems in the Salticidae,
including Walckenaerian names. Peckhamia 1(2): 21-23.

1980. Taxonomy, ethology, and ecology of Phidippus (Araneae: Salticidae) in
eastern North America. Ph.D. Dissertation, University of Florida. 354 p.
Foelix, R. F.

1970. Chemosensitive hairs in spiders. J. Morphol. 132: 314-334.

Gertsch, W. J.

1979. American Spiders. Van Nostrand Reinhold Co., Second Edition, New
York. 274 p.

Hegdekar, B. M., and C. D. Dondale.

1969. A contact sex pheromone and some response parameters in lycosid
spiders. Canad. J. Zool. 47(1): 1-4.

Hill, D. E.

1977a. The pretarsus of salticid spiders. Zool. J. Linnean Soc. London 60(4):
319-338.

1977b. Modified setae of the salticid pedipalp. Peckhamia 1(1): 7-8.
Jackson, R. R.

1977. Courtship versatility in the jumping spider, Phidippus johnsoni (Ara-
neae: Salticidae). Anim. Behav. 25(4): 953-957.

Legendre, R.

1963. L’audition et remission de sons chez les araneides. Ann. Biol. 2: 371-390.
Petrunkevitch, A.

1926. Spiders of the Virgin Islands. Trans. Connecticut Acad. Arts Sci. 28:
21-78.

Richman, D. B.

1977. The relationship of epigamic display to the systematics of jumping
spiders (Araneae: Salticidae). Ph.D. Dissertation, University of Florida.

162 p.

214

Psyche

[Vol. 88

1982. Epigamic display in jumping spiders (Araneae: Salticidae) and its use in
systematics. J. Arachnol. 10 (in press).

Robinson, M. H., and Barbara Robinson.

1980. Comparative studies of the courtship and mating behavior of tropical
araneid spiders. Pacific Insects Monograph 36: 1-218.

Rovner, J. S.

1967. Acoustic communication in a lycosid spider ( Lycosa rabida Walck-
enaer). Anim. Behav. 15: 273-281.

1975. Sound production by Nearctic wolf spiders: A substratum-coupled
stridulatory mechanism. Science 190: 1309-1310.

1980. Vibration in Heteropoda venatoria (Sparassidae): A third method of
sound production in spiders. J. Arachnol. 8: 193-200.

Specht, H. B., and C. D. Dondale.

1960. Spider populations in New Jersey apple orchards. J. Econ. Ent. 53:
810-814.

Tietjen, W. J., and J. S. Rovner.

1980. Trail-following behaviour in two species of wolf spiders: sensory and
etho-ecological concomitants. Anim. Behav. 28: 735-741.

Warren, L. O., W. B. Peck, and M. Tadic.

1967. Spiders associated with the fall webworm, Hyphantria cunea (Lepidop-
tera: Arctiidae). J. Kansas Ent. Soc. 40: 382-395.

MATERNAL BEHAVIOR AND ALARM RESPONSE IN
THE EGGPLANT LACE BUG, GARGAPHIA SOLANI
HEIDEMANN (TINGIDAE: HETEROPTERA)1

By R. S. Kearns2 and R. T. Yamamoto

Entomology Department
North Carolina State University
Raleigh, NC 27650

INTRODUCTION

Maternal behavior in the eggplant lace bug, Gargaphia solani
Heidemann (Tingidae: Heteroptera) was first reported by Fink
(1915). He described the female’s guarding of the eggs and shepherd-
ing of the nymphs from leaf to leaf. G. solani is found on the native
horse nettle ( Solanum carolinense) and on the introduced eggplant
(Solanum melongena). Overwintering adults appear in late spring,
and females lay eggs in circular masses on the underside of leaves.
Fink reported that the number of eggs is greater than 100, oviposi-
tion lasts 4 to 5 days, and the incubation period is about 6 days.
Maternal care persists through the development of the nymphs, and
the life cycle is approximately 20 days. Females observed in this
study usually laid less than 100 eggs over a period of 3 to 4 days
(Kearns 1980).

Maternal behavior has been reported for a number of heterop-
terans (Melber and Schmidt 1977) and for two other species of the
genus Gargaphia: Gargaphia tiliae (Weiss 1919, Torre-Bueno 1935,
Sheeley and Yonke 1977) and Gargaphia irridescens (Torre-Bueno
1942). These accounts give few details. The maternal behavior of G.
solani has much in common with that exhibited by treehoppers
(Membracidae: Homoptera) (Wood 1974, 1976a, 1976b, 1977 and
Hinton 1977). The complex behavior patterns of membracids and of
G. solani suggest that aggregations of these insects depend upon a

•Paper number 6950 of the Journal Series of the North Carolina Agricultural
Research Service, Raleigh, North Carolina.

2This work was completed in partial fulfillment of the requirements for the degree of
Master of Science in Entomology.

Manuscript received by the editor August 30, 1981.

215

216

Psyche

[Vol. 88

pheromonal communication system which facilitates group move-
ments. In the membracids, there is indirect evidence for aggregation
pheromones (Hinton 1976, 1977). Alarm pheromones of membra-
cids are released only when the body wall is ruptured, and this type
of release has not been reported for any other insects. Pheromones
causing alarm responses are known to be present in at least 3
membracid species. They are interspecific in action but have not
been identified (Nault et al. 1974).

This study reports an examination of the movements of G. solani
aggregations on host plants with particular emphasis on the female’s
behavior.

MATERIALS AND METHODS

MAINTENANCE OF INSECTS

G. solani was collected from horse nettle growing in or near
Raleigh, N.C., and aggregations were maintained for more than a
year on horse nettle or eggplant either in the laboratory on a 16:8
light: dark cycle or in a greenhouse. The movements of nymphs were
studied after an aggregation consisting of nymphs and a female had
been transferred to a small sprig of horse nettle having an un-
branched stem with 5 or more leaves. A piece of leaf containing a
group was pinned to the upper surface of the second or third leaf
from the bottom of a horse nettle stem. In time, the aggregation
moved off the leaf fragment and onto the fresh leaf.

FEEDING MOVEMENTS

Feeding movements were measured in light, darkness, and with a
light placed below the aggregation. Groups chosen for these studies
were nymphs in the third or fourth instars, with smaller numbers of
the other instars present. For dark conditions, an aggregation on a
horse nettle sprig was placed in a tightly covered metal can which
had been sprayed inside with a dull-finish black paint. Directional
lighting was provided by placing the sprig or plant within a
darkened enclosure and positioning a light about 0.6 m from the
bottom of the plant.

ALARM RESPONSE AND ALARM PHEROMONE

G. solani nymphs exhibited an alarm response after they were
presented with a nymph, freshly squeezed and held by fine forceps,

1981]

Kearns & Yamamoto — Gargaphia

217

Table 1 . Elements of behavior of G. solani females when their broods were responding

to alarm pheromone

% of responses

Behavior

observed (1)

Female positioned slightly below exit axil

Female positioned at exit axil

Female positioned between axils

Female positioned at entrance axil

Female moved off leaf shortly after nymphs left

78*

Female followed nymphs onto new leaf

78*

Female returned to old leaf after group had moved

Female used at least one of the first four elements listed

89*

(1) 18 different trials

* Significant at the 95% confidence level (Binomial distribution. Table 2, partial sums,
Eisenhart 1952. Confidence intervals, Table A-22, Natrella 1963.)

or a nymph pierced with a pin against a small disk of filter paper
(after Nault et al. 1974). Blank disks of filter paper placed near an
aggregation did not elicit an alarm response. Nymphs of G. solani
were also collected and stored in a small container of chloroform.
Several hundred were sufficient to provide a crude extract. Before
the extract was tested for activity, the chloroform was allowed to
evaporate from a small sample contained in a Pasteur pipette. The
tip was then brought close to an aggregation, and the bulb was
pressed gently to force the remaining volatile pheromone from the
tip. As a control, a pipette with evaporating chloroform was also
used. Chloroform vapors disturbed resting aggregations, causing
individuals to move away from the pipette; but chloroform alone
did not elicit a full alarm response with characteristic group
movement to a new leaf. Alarm responses were studied under well-
lighted conditions with and without the female present.

RECORDING OF DATA

Observations were written, tape recorded, or photographed.
Direction of movement of aggregations on the host plant was
recorded as up, down, or “up and down” (part of the group moved
up, part moved down). Movement up or down was considered
“directed”; movement both up and down, “undirected”. The new

218

Psyche

[Vol. 88

position of the group was recorded as 1, 2, 3, or 4 or more leaves
above starting position. The designation “4 or more” included leaf
no. 4 and several small leaves at the growing tip of the plant.

RESULTS

Egg masses of G. solani are deposited on the underside of leaves,
and first instar larvae feed from the leaf surface between the eggs
and then from the areas adjacent to the eggs. The larvae are usually
in a compact circular formation while feeding, and the aggregation
moves away from the oviposition site as leaf tissue is destroyed.
Feeding sites become yellow or brown in color and also brittle. On a
large eggplant leaf, nymphs may pass through several instars before
consuming most of the leafs soft tissue. On the relatively smaller
horse nettle leaf, an aggregation consumes the edible portion of a
leaf more quickly and then moves to another leaf.

MOVEMENTS TO NEW FEEDING SITES

Movements to new sites on the same leaf seemed to proceed
gradually and with little intervention from the female. As individ-
uals in the aggregation withdrew their stylets, they moved away
from where they were feeding, bumping into adjacent nymphs.
These bumped nymphs in turn withdrew their stylets and moved or
milled about, bumping into other nymphs until the entire aggrega-
tion was activated. Movement from the feeding site to another
feeding site on the same leaf then ensued. Movement to a new leaf
usually occurred after 75% or more of the leaf was damaged and
often lasted for about an hour. The parent female, also activated by
the milling nymphs, usually moved slowly down the petiole while
keeping close physical contact with the nymphs immediately behind
her. If there was any break in contact, the nearest nymphs moved
forward and touched the tips of the female’s wings with their
antennae, or the female turned around and touched her antennae to
those of the nearest advancing nymphs. During one group move-
ment, the female waited first at the axil of the new leaf and then on
the underside of the petiole, as nymphs filed by her. This behavior
was identical to that observed during alarm responses. On two
occasions, the female seemed to initiate movement of the nymphs by
forcing her way into the cluster of feeding nymphs; but this occurred
only when the nymphs were in the earlier instars and were moving

1981]

Kearns & Yamamoto — Gargaphia

219

Table 2. Direction of movement of G. solani aggregations on a host plant

Female Present Female Absent

% of groups observed % of groups observed

Upward

Directed

Upward

Directed

Feeding Movements

movement

In light

82*

96* (1)

100* (2)

In darkness

89*

100* (2)

100*

100* (3)

Light source below

aggregation

78 (2)

67 (3)

Alarm Response

In light

90*

100* (4)

67 (5)

(1) 23 observations

(2) 9 observations

(3) 6 observations

(4) 19 observations

(5) 12 observations

* Significant at the 95% confidence level

from one surface of the leaf to the opposite surface. At no time was
“herding” by the female observed as described by Fink (1915).

Movements to a new leaf were difficult to predict and lengthy to
monitor. Because of time considerations, it was not feasible to make
a statistical study of the female’s total behavior pattern during these
group movements. When a female was present, she led the group to
a new leaf. In the absence of a female, the nymphs moved on their
own. Females sometimes wandered about on adjacent leaves but
usually returned to their aggregations.

ALARM RESPONSES

When an aggregation of the third through fifth instar nymphs of
G. solani was alarmed with a squashed fifth instar nymph, the group
responded quickly, usually within 10 seconds. The duration of the
response was from 4 to 20 minutes. If the nymphs were on the top
surface of the leaf they moved to the underside, and conversely. In
either case, at least some of the nymphs moved quickly to the midrib
and from there to the petiole of the leaf. At the exit axil, the nymphs
moved up or down the stem; but they were more likely to move up
the stem (Table 2, Female Present). During this activity, the female

220

Psyche

[Vol. 88

Figure 1. Adult female of G. solani positioned at the exit axil during an alarm
response.

moved quickly to the axil of the leaf and oriented herself a little to
one side of the nymphs’ path (Fig. 1). When she was in this position,
the nymphs moved up the stem rather than down.

As more of the nymphs left the leaf, the female sometimes moved
about the axil and positioned herself along the side of the stem as
the nymphs moved past her. When she was in these positions, she
seemed to have little physical contact with the nymphs which filed
past her unless they happened to bump into her in passing. Part way
through an alarm response, the female moved quickly up the stem,
usually to the axil of the first leaf above the previously occupied
leaf. Many of the nymphs had already reached this axil and had
moved down the petiole onto the new leaf. A few nymphs often
proceeded above this axil and continued up the plant before
returning to the group. The female oriented herself at the axil of the
new leaf (Fig. 2) and waited there as more nymphs arrived. After
most of the nymphs had passed along the petiole, the female joined
the aggregation on the new leaf.

1981]

Kearns & Yamamoto — Gargaphia

221

There were always a few nymphs that remained behind or that
failed to keep up with the bulk of the aggregation. These slower
individuals wandered out onto other leaves but did not settle down
until they found the group. Apparently the nymphs maintain
locomotor activity unless they have sufficient physical or chemical
contact with other nymphs. There were variations in the female’s
behavior (Table 1), but it was not obvious what environmental
conditions might cause the female to include or change a particular
element of her behavior. When an aggregation of first and/or
second instars was alarmed, the group was likely to relocate on the
same surface of the leaf rather than to move to the opposite surface
or off the leaf.

Certain elements of the female’s behavior were clearly recogniz-
able and repeated more than once. These elements are recorded in
Table 1 with the frequency of their occurrence in 18 different
experimental responses to alarm pheromone. In 89% of the alarm
responses, the female exhibited at least one of the first 4 elements

Figure 2. Adult female of G. solani positioned at the extrance axil during an alarm
response.

222

Psyche

[Vol. 88

listed. The fourth element, female positioned at the entrance axil
(Figure 2), was repeated most frequently, although the element was
not itself significant. There was more variation in the female’s posi-
tion on the stem at the beginning of the alarm response than at the
end.

DIRECTION OF MOVEMENT

Aggregations with and without the female present were studied in
order to determine the significance of the female’s role in feeding
movements or alarm responses. It was hypothesized that the
female’s presence would keep the aggregation together and inhibit
random movements on the plant. The relocation of the group was
studied in terms of the direction of movement on the plant (up,
down, or in both directions) and the choice of a new leaf on which to
feed. Movement was considered directed if a group moved in one
direction or the other, but not in both. A table for a binomial
distribution was used to evaluate significance at the 95% confidence
level (Table 2, partial sums, Eisenhart 1952; Table A-22, confidence
intervals, Natrella 1963). The results for upward and directed
movements of aggregations, with and without the female present,
are recorded in Table 2 as percentages of groups observed. Those
results significant at the 95% confidence level are marked with an
asterisk.

Feeding Movements

In light or in darkness, feeding movements with or without the
female present were directed rather than random, and the group
usually moved upward. When the source of light was 180° away
from the usual direction, neither moving upward nor directed
movement was significant; but the aggregations did not reverse their
direction of movement and move toward the light. It is possible that
the abnormal position of the light source acted as a conflicting
stimulus which confused some of the aggregations.

A small field sample of horse nettle plants (13) showing damage
from G. solani was examined for evidence of group movements.
Eighty-five percent of the groups had moved upward on the plants
from leaves containing the remains of egg masses. Moving up was
significant at the 95% confidence level and closely matched the
results obtained in the laboratory.

1981]

Kearns & Yamamoto — Gargaphia

223

Alarm Response

The female’s presence or absence made a significant difference
when the aggregation was alarmed. When the female was present,
90% of the groups moved upward, and 10% moved downward; but
none split and moved in both directions. When the female was
absent, 67% of the groups moved either upward or downward; one-
third of the groups split.

POSITION OF AGGREGATIONS FOLLOWING FEEDING
MOVEMENTS OR ALARM RESPONSES

The results of experiments for both feeding and alarm movements
were combined, and a comparison was made between female
present and female absent.

Choice of Leaf

With a choice of 4 leaf positions above the one occupied by the
aggregation, the probability of an aggregation’s reassembling on
any one of the leaves was 0.25. Using the binomial distribution
(Eisenhart 1952), we compared the choice of leaf no. 1 with the
choice of any other leaf (Table 3). Whether females were present or
absent, the aggregations were more likely to move to leaf no. 1 than
to any of the other leaves. If the aggregations split between leaves,
the split usually included leaf no. 1. This behavioral pattern of the
nymphs increased the likelihood that the group would remain
together following movements on the host plant. When the female
was present, the group was more likely to move as a unit to leave
no. 1.

Choice of Single or Multiple Leaves

Movement to a single leaf was compared to movement to multiple
leaves (Table 4). The probability associated with this choice was 0.5.
When females were present, aggregations usually moved as a unit to
a single leaf on the host plant but when females were absent,
aggregations split up as often as they chose a single leaf.

WING FANNING BY THE FEMALE

Fink (1915) reported that on one occasion he saw an adult female
G. solani chase a ladybeetle (Hippodamia convergens Guer.) away

224

Psyche

[Vol. 88

Table 3. Choice of leaf position by G. solani aggregations following movements
(1) on a host plant

% of groups observed
Female Present Female Absent

Leaf no. 1

Leaf no. 1 in combination with one or

more leaves

Total positions including leaf no. 1

72* (2)

71* (3)

(1) Feeding movements and alarm responses combined

(2) 36 observations

(3) 14 observations

* Significant at the 95% confidence level

from an aggregation of feeding nymphs: the female “with out-
stretched, slightly raised wings suddenly darted toward the intruder,
driving it from the leaf.” In the laboratory, adult females of G.
solani responded similarly (Fig. 3) to ladybeetles, anthocorids, ants,
the tip of a brush, and a tomato pinworm caterpillar which was
spinning a cocoon. Beamer (1930) and Wood (1976a, 1976b, 1977,
and 1978) reported wing fanning in a total of 4 species of mem-
bracids. In each of these species, wing fanning was used by the adult
female as a response to a predator (Beamer 1930; Wood 1976a,
1976b, 1977) or a threatening stimulus, such as a pencil used to prod
the female (Wood 1978). Sheeley and Yonke (1977) observed wing
fanning by the tingid Corythucha bulbosa when a jumping spider

Table 4. Choice of single or multiple leaves by G. solani aggregations following
movements (1) on a host plant

% of groups observed
Female Present Female Absent

Choice

Single leaf

81* (2)

50 (3)

Multiple leaves

19 (2)

50 (3)

(1) Feeding movements and alarm responses combined

(2) 36 observations

(3) 14 observations

* Significant at the 95% confidence level

1981]

Kearns & Yamamoto — Gargaphia

225

approached, and they reported that the spider’s response to touch-
ing the tingid suggested the presence of a defensive chemical.

Wing fanning in G. solani occurred not only in response to a
predator, but also under other circumstances. It was often associ-
ated with alarm responses and was directed toward the nymphs as
well as toward a possible predator. For 27 brooding females, 143
occurrences of fanning were recorded in 2 categories: deterring a
predator (26%) and controlling the nymphs in one of several ways
(74%).

Deterring Predators

The brooding female responded to predators quickly after she
detected their presence. The relatively large coccinellids ( Hippo -
damia convergens, Olla abdominalis) were detected more readily
than smaller predators such as Pharaoh ants ( Monomorium phara-
onis) or the anthocorid, Orius insidiosus. Attacks by ants and
anthocorids were observed with a dissecting microscope. Females

Figure 3. Adult female of G. solani fanning her wings.

226

Psyche

[Vol. 88

and nymphs failed to respond when a single ant removed an egg or
ate a newly hatched first instar nymph. Ants carried tiny nymphs
away from the brood, and the release of alarm pheromone was
apparently not detected. When two or more ants moved in front of a
brooding female, she responded with wing fanning and moved her
body over the egg mass. Orius insidiosus nymphs, which were about
the size of second instar G. solani nymphs, attacked their victims by
penetrating intersegmental membranes. O. insidiosus was not al-
ways detected by the brooding female or nymphs, perhaps because
the site of penetration was often in the coxal area rather than on the
abdomen. When an attack occurred in front of a brooding female,
she responded by fanning her wings and prodding the anthocorid
with her head. Anthocorids responded by remaining motionless for
periods of up to 55 minutes in length.

An attendant female responded to a coccinellid by rushing at it,
fanning her wings, and, occasionally, by prodding it with her head.
In 5 experiments with the adult coccinellid Hippodamia convergens,
first and second instar nymphs were killed each time. In 3 of those
encounters, the adult female lace bug was successful in driving away
the coccinellid, preventing further loss of nymphs. In 2 encounters
with starved coccinellids, the female lace bug was not able to drive
the attacker away. The remaining nymphs survived because they
fled apparently in response to an alarm pheromone released by
crushed nymphs. In 3 encounters with the coccinellid Olla abdomi-
nalis, the female lace bug chased the approaching beetle successfully
(Fig. 4); however, the beetle did not attack any nymphs or show
much interest in them.

Controlling Nymphs

Females used wing fanning in their interactions with the nymphs.
On at least 6 occasions, the attendant female went ahead of the
moving aggregation and waited on the new leaf for the nymphs to
arrive. While waiting for the nymphs, the females fanned their wings
repeatedly.

There were a number of instances in which wing fanning was used
to quiet a restless aggregation or one which had recently dispersed
to a new leaf. The adult female circled the group with rapid, jerky
movements and stopped occasionally to fan her wings. For 2
different females and aggregations, the female backed up to the

1981]

Kearns & Yamamoto — Gargaphia

227

Figure 4. Adult female of G. solani responding to coccinellid, Olla abdominalis.

nymphs and pointed the tip of her abdomen toward them as she
fanned her wings. This behavior suggests that the fanning may be
used to propel a pheromone toward the group. One female was
observed to use wing fanning to prevent the movement of an
aggregation. The female was oriented at the base of the leaf, headed
toward the group of nymphs. When 2 nymphs left the group, moved
down the mid-vein, and approached her, she fanned her wings. The
nymphs’ response was a retreat.

AGGREGATION

There is some indirect evidence for an aggregation pheromone or
for the nymphs’ need for physical contact with each other. Upon
hatching, nymphs feed near the egg mass for a short period and then
move away as an aggregation. Older nymphs wander away from
their own aggregations and join others, stray fifth instars being
particularly conspicuous when they join groups of first and second

228

Psyche

[Vol. 88

instars. Following one experiment on the alarm response, 5 fifth
instar nymphs from a second aggregation were released, one at a
time, slightly above the leaf just vacated by the first aggregation.
Four of the fifth instars moved directly up the stem and onto the
newly occupied leaf, and the remaining nymph wandered about,
first on higher leaves and then on the vacated leaf before moving to
the occupied leaf. Other experiments showed that nymphs would
reaggregate after they were separated by the experimenter. Need for
physical contact might explain this adequately, but the presence of
an aggregation pheromone should not be ruled out. Reaggregation
is essential if the alarm response is to occur repeatedly.

female’s behavioral maturation

Preliminary experiments (Kearns 1980) indicate that females
undergo behavioral maturation from the time of oviposition through
egg hatch and early development of the nymphs. Females at
different stages of development were substituted for females which
were attending aggregations of nymphs. Only those substitutes
which had attended aggregations of their own behaved normally
during an alarm response. When females which were still oviposit-
ing were used as substitute mothers, they either avoided the alarmed
nymphs or failed to interact with them.

CHLOROFORM EXTRACT

A chloroform extract of G. solani nymphs proved to be as
effective in eliciting an alarm response as a fifth instar nymph
squeezed with forceps or squashed on filter paper. Crushed adults
also released the alarm pheromone, but the nymphal response was
slower by a minute or less, to a crushed adult than to a crushed
nymph. Preliminary attempts were made to test for alarm pher-
omones in 3 other tingids available locally: Corythucha ciliata, the
sycamore lace bug; Corythucha cydoniae, the hawthorne lace bug;
and Corythucha marmorata, the chrysanthemum lace bug. Nymphs
of each species showed an alarm response to a crushed nymph of
the same species. There were also cross responses between G. solani
and each of the three species of Corythucha. Since not all three
species of Corythucha overlap in time, it will be necessary to rear the
insects in the laboratory or to make extracts of each for testing cross
responses.

1981]

Kearns & Yamamoto — Gargaphia

229

DISCUSSION

Gargaphia solani and some of the membracids (Wood 1974,
1976b) are unusual in having maternal care extend from the time of
oviposition through the maturation period of the nymphs. This long
brooding period appears to be an adaptation to environments in
which predation is an important factor. The host plants of G. solani
grow close to the ground, and ants appear to be the most numerous
predators. Maternal care in this tingid may have evolved as a
response to ants or to low-flying predators or to both. Sheeley and
Yonke (1977) were unable to find predators for some of the 7 species
of tingids studied, perhaps because the host plants of 6 species are
trees rather than small annuals. Gargaphia tiliae, having maternal
care, might be expected to live close to the ground, but it is a tree-
dwelling species. Sheeley and Yonke found no natural enemies of
this insect, but the predators could have included tiny anthocorid
nymphs which escaped detection.

It seems worthwhile to compare G. solani with some of the
membracid species since there are striking similarities, including
wing fanning by the attendant female and the release of an alarm
pheromone when the body wall is ruptured. If G. solani and the
membracids represent examples of parallel evolution, they may be
responding to similar environmental stresses.

Literature Cited

Beamer, R. H.

1930. Maternal instinct in a Membracid ( Platycotis vittata) (Homopt.) Ento-
mol. News 41(10): 330-331.

Eisenhart, C.

1952. Tables of the Binomial Probability Distribution. National Bureau of
Standards Applied Mathematics Series 6. U. S. Govt. Printing Office,
Washington.

Fink, D. E.

1915. The eggplant lace-bug. Bull., U. S. Dept. Agricult. 239: 1-7.

Hinton, H. E.

1976. Maternal care in the Membracidae. Proc. Roy. Entomol. Soc. London
(C). 41: 3-4.

1977. Subsocial behavior and biology of some Mexican membracid bugs.
Ecological Entomology 2: 61-79.

Kearns, R. S.

1980. Maternal behavior in the eggplant lace bug Gargaphia solani Heide-
mann (Tingidae: Heteroptera). M.S. thesis, North Carolina State Uni-
versity, Raleigh.

230

Psyche

[Vol. 88

Melber, A., and G. H. Schmidt

1977. Sozialphanomene bei Heteropteren. Sonderdruck aus Zoologica. 127:
19-53.

Natrella, M. G.

1963. Experimental Statistics. National Bureau of Standards Handbook 91. U.
S. Govt. Printing Office, Washington.

Nault, L. R., T. K. Wood, and A. M. Goff

1974. Treehopper (Membracidae) alarm pheromones. Nature 249: 387-388.
Sheeley, R. D. and T. R. Yonke

1977. Biological notes on seven species of Missouri tingids (Hemiptera:
Tingidae). J. Kansas Entomol. Soc. 50: 342-356.

Torre-Bueno, J. R.

1935. Notes on Gargaphia tiliae. Bull. Brooklyn Entomol. Soc. 30: 78.
1942. Maternal solicitude in Gargaphia iridescens Champion. Bull. Brooklyn
Entomol. Soc. 37: 131.

Weiss, H. B.

1919. Notes on Gargaphia tiliae Walsh, the linden lace-bug. Proc. Biol. Soc.
Wash. 32: 165-168.

Wood, T. K.

1974. Aggregating behavior of Umbonia crassicornis (Homoptera: Membra-
cidae). Can. Ent. 106: 169-173.

1976a. Alarm behavior of brooding female Umbonia crassicornis (Membra-
cidae: Homoptera). Ann. Entomol. Soc. Amer. 69: 340-344.

1976b. Biology and presocial behavior of Platycotis vittata (Homoptera: Mem-
bracidae). Ann. Entomol. Soc. Amer. 69: 807-811.

1977. Role of parent females and attendant ants in the maturation of the
treehopper, Entylia bactriana (Homoptera: Membracidae). Sociobiol-
ogy 2: 257-272.

1978. Parental care in Guayaquila compressa Walker (Homoptera: Membra-
cidae). Psyche 85: 135-145.

POLYMORPHISM AND DIVISION OF LABOR IN THE
DACETINE ANT ORECTOGNATHUS VERSICOLOR
(HYMENOPTERA: FORMICIDAE)*

By Norman F. Carlin

Department of Organismic and Evolutionary Biology,
Harvard University, Cambridge, Mass. 02138

Introduction

The ants of the myrmicine tribe Dacetini exhibit a primary
evolutionary trend from primitive epigaeic and subarboreal foragers
to advanced cryptobiotic forms; in association with this trend are a
number of secondary tendencies, including reduction in body size
and mandible length, increasing specialization on collembolan prey,
and loss of worker caste differentiation (Brown and Wilson 1959).
The subarboreal and impressively long-mandibulate subtribe Orec-
tognathiti, comprising the genera Orectognathus and Arnoldidris,
occupies an intermediate position between the primitive polymor-
phic genus Daceton and the largely monomorphic higher subtribes
Epopostrumiti and Strumigeniti. All but one of the twenty-nine
known species of Orectognathus are monomorphic, the exception
being O. versicolor , which possesses a distinctive major caste
(Taylor 1977, 1979). Caste differentiation in this species is con-
sidered to have evolved secondarily, from the monomorphic generic
stock (Brown and Wilson 1959).

The extreme polymorphism of Daceton armigerum, the only
lower dacetine whose behavior has been studied, is put to work in an
equally extreme division of labor (Wilson 1962). The minor workers
are strictly limited to brood care tasks (in which they are aided by
callows of larger castes), and to regurgitation with other adults.
Small medias forage widely and actively, but larger medias and
majors tend to rest in “way-stations” some distance from the nest.
These large workers take prey away from returning smaller foragers,
bringing it into the nest themselves, so that little prey is carried back
by those that hunt for it. The species takes a broad variety of prey
items; it has been suggested that the dietary specialization on
collembolans seen in higher dacetines might account for their

♦Manuscript received by the editor December 1, 1981.

231

232

Psyche

[Vol. 88

surrendering the polymorphism and polyethism of Daceton (Wilson
1971).

Orectognathus versicolor , as the sole polymorphic intermediate
dacetine, is of special interest for polyethism analysis. The species is
also an easy one to study, its slow-moving habits and small colony
size making possible the recording of nearly every behavioral act
performed by each individual worker. The minor workers possess
the same long, slender mandibles, with pointed apical teeth, that
their congeners bear. Majors, however, have massive, relatively
short mandibles, with apical teeth thick, blunt and recessed; their
large occipital lobes contain disproportionately developed mandible
adductor muscles (figs. 1 and 2). In mandible allometry, at least, this
species may be the most exaggeratedly polymorphic of all dacetines.
The division of labor by which such morphologically divergent
forms are utilized, particularly since the major caste is a secondary
development, may shed light on the advantages of specialized castes
in the context of dacetine evolution. To what use are the singular
majors put? Does the polyethism of O. versicolor in any way
resemble that of Daceton , or is it entirely independent? Has the
return to polymorphism been accompanied by a return to the
polyphagy of Daceton , or is O. versicolor a collembolan specialist,
as the rest of its genus is thought to be (Brown 1953)? An
opportunity to address these questions in the laboratory arose when
Bert Holldobler brought a live queenright colony of these ants from
North Queensland, Australia; the results of observation of this
colony are reported below.

Materials and Methods

The O. versicolor colony was settled in a glass test tube (2 cm in
diameter), with water trapped at its end behind a tight cotton plug.
The tube was placed in a plaster-floored clear plastic container (18
cm by 12 cm by 6 cm), and a dissecting microscope was set over it on
a moveable mount to permit viewing of ants both inside the nest
tube and out on the container floor. A total of 45 hours of
observation were made over a period of five weeks, during which
7,891 separate behavioral acts were recorded. Estimation of the
completeness of caste behavior repertories was made by fitting the
data to a lognormal Poisson distribution, following the method of
Fagen and Goldman (1977). The ants were offered various food

1981]

Carlin — Polymorphism in Orectognathus

233

Figure 1: An Orectognathus versicolor colony. The queen is at the left; to her right are two major workers — note their
mandibles and head size and shape. To their right are another major (top), minor (middle) and media (bottom) workers.

234

Psyche

[Vol. 88

items; to examine their defensive behavior, small Solenopsis invicta
workers were introduced into their container.

The two morphological castes were easily distinguished on the
basis of mandible thickness. In order to record division of labor
among individuals of different sizes, yet similar proportions — so
critical in weakly allometric species such as Daceton — the minor
workers were arbitrarily divided into small and medium size classes,
also distinguishable by eye. For convenience, these subcastes will be
referred to as “minors” and “medias”, as in Wilson 1978. By-eye
assignment of caste to preserved specimens, subsequently measured,
produced the following definitions of size classes and castes: minors,
head width less than 1.12 mm; medias, head width between 1.13 and
1.64 mm; majors, head width greater than 1.65 mm. After some
initial die-off, the colony contained fifty-two adults for the duration
of the study: one queen, thirty minors, fifteen medias and six
majors.

Results

O. versicolor is in fact polyphagous. Live flightless Drosophila
were readily accepted, and young were successfully raised on this
diet. The ants also accepted Drosophila larvae, and, not surprisingly,
collembolans. (Alternative foods were not offered simultaneously to
test preferences; however, most collembolan specialist species would
not touch other prey even if starving.) The same colony had been fed
mealworm and cockroach fragments, various diptera and honey-
water in Australia (B. Holldobler, pers. comm.).

The ethogram or behavioral catalogue of workers and queen is
presented in table 1, which gives both numbers of individual acts
performed and the relative frequencies of acts in the total repertory
of each caste. The colony repertory consisted of twenty-seven
categories of behavior. (Worker regurgitation with the queen was
added as a twenty-eighth because it was seen twice during prelimi-
nary observations, though never during the study.) The observed
minor and media repertories both contained twenty-seven behavior
categories; the observed major repertory contained twenty-four.
Using the Fagen-Goldman statistical method, the estimated total
repertory size for minors — the observed repertory plus an estimate
of the number of categories not observed — was calculated to be
twenty-nine, with a 95% confidence interval of (27,32) acts. The

1981]

Carlin — Polymorphism in Orectognathus

235

236

Psyche

[Vol. 88

Table 1 : Ethogram of Orectognathus versicolor. The values given are numbers of
individual acts performed by members of each caste. In parentheses are given relative
frequencies of performance of each act in the total repertory of the caste.

Minor

Media

Major

Queen

Self-groom

1370 ( .3481)

845 ( .2996)

462 ( .4306)

45 ( .7258)

Allogroom minor

437 ( .1110)

93 ( .0330)

18 ( .0168)

1 ( .0161)

Allogroom media

144 ( .0366)

224 ( .0794)

16 ( .0149)

Allogroom major

100 ( .0254)

52 ( .0184)

15 ( .0140)

Allogroom queen

27 ( .0069)

20 ( .0071)

8 ( .0075)

Regurgitation

with minor

48 ( .0122)

15 ( .0053)

3 ( .0028)

2 ( .0323)

with media

10 ( .0025)

21 ( .0074)

7 ( .0065)

with major

13 ( .0033)

2 ( .0007)

2 ( .0019)

with queen

Carry or manipu-

late egg

4 ( .0010)

2 ( .0007)

1 ( .0009)

Lick egg

18 ( .0046)

5 ( .0018)

1 ( .0009)

1 ( .0161)

Carry or manipu-
late larva

53 ( .0135)

37 ( .0131)

2 ( .0019)

Lick larva

602 ( .1529)

520 ( .1844)

155 ( .1445)

9 ( .1425)

Regurgitate with
larva

4 ( .0010)

1 1 ( .0039)

2 ( .0019)

Feed larva solids

19 ( .0048)

20 ( .0071)

Carry or manipu-
late pupa

7 ( .0018)

9 ( .0032)

6 ( .0056)

Lick pupa

54 ( .0137)

63 ( .0223)

18 ( .0168)

Forage

364 ( .0925)

356 ( .1262)

126 ( .1174)

Capture prey

19 ( .0048)

26 ( .0092)

2 ( .0019)

Return prey to
nest

19 ( .0048)

6 ( .0021)

Process prey

45 ( .0114)

25 ( .0089)

4 ( .0037)

Eat prey

132 ( .0335)

109 ( .0387)

19 ( .0177)

2 ( .0323)

Guard

313 ( .0795)

280 ( .0993)

186 ( .1733)

Manipulate nest

material

67 ( .0170)

13 ( .0046)

2 ( .0019)

1 ( .0161)

Lick tube wall

27 ( .0069)

26 ( .0092)

12 ( .0112)

1 ( .0161)

Remove refuse
(in tube)

9 ( .0023)

1 ( .0004)

Remove refuse
(out of tube)

12 ( .0030)

19 ( .0067)

4 ( .0037)

Carry dead ant

19 ( .0048)

20 ( .0071)

2 ( .0019)

Total # acts

3936(1.0 )

2820(1.0 )

1073(1.0 )

62(1.0 )

# categories

# individuals

1981]

Carlin — Polymorphism in Orectognathus

237

estimated total repertory size for medias was twenty-eight, the 95%
confidence interval (27,33); for majors, twenty-seven, with a confi-
dence interval of (24,37).

Minor and media workers engaged in the same tasks with
essentially similar frequencies, while majors, with a smaller reper-
tory, also performed certain acts with quite different frequencies.
Self-grooming was the commonest act in all castes. Allogrooming
and regurgitation occurred freely among all castes, with a tendency
among minors and medias to interact with their own class. After
self-grooming, brood care and foraging were the most frequently
performed acts in the minor and media repertories. An ant was
scored as “foraging” any time it left the nest tube - an act that does
not necessarily signify hunting for food. Though majors did “forage”
by this definition, they captured almost no prey and returned none
to the nest. “Processing”, in which workers tore at, dismembered
and occasionally stung prey that had been brought inside the tube,
was rarely performed by majors, despite the seeming usefulness of
their heavy mandibles for such a task.

The province of the majors was “guarding”: walking to the tube
mouth and facing outward without setting foot on the container
floor; after self-grooming, it was their most frequent act. A guarding
ant might station itself at the opening for less than a minute or up to
half an hour. That this is in fact a defensive behavior will be shown
below. Minors and medias also guarded in large numbers, but less
frequently than they foraged or attended brood.

Nest maintenance was undertaken almost exclusively by the small
size classes. Carrying refuse down the tube, to be dropped inside or
just outside the entrance, was defined as “in-tube refuse removal”,
while carrying trash out to corner refuse piles on the container floor
(to which dead ants were also brought) was defined as “out-of-tube
refuse removal.” “Manipulation of nest material”, that is, of the
fibers of the cotton plug, may not be an actual maintenane behavior
used in natural colony sites (under stones, in rotting wood);
similarly, ants may lick the tube wall only to drink condensation on
the glass, and not exhibit any such behavior in the wild.

The division of labor among minor and media size classes, and
the role of the major caste, were better elucidated by constructing
polyethism curves, depicting the percent contributions of each caste
to the total colony performance of behaviors (figs. 3 and 4). For
simplicity, certain behavioral categories from the ethogram were

238

Psyche

[Vol. 88

NEST MAINTENANCE

ATTACK ALIEN

Ml ME MA

1981]

Carlin — Polymorphism in Orectognathus

239

combined, so that the polyethism curves indicated represent groups
of tasks. There was a tendency to divide those tasks performed
primarily by small workers among the size classes on the basis of
size of objects handled and task location (fig. 3). Minors performed
most in-nest maintenance; medias performed somewhat more out-
of-nest maintenance than did minors. Minors contributed most to
egg care. While both size classes attended larvae and pupae, minors
contributed less to larva care than to egg care, still less to pupa care,
medias compensating by putting more effort into care of larger
brood.

On the introduction of Solenopsis workers, the function of the
guarding majors became apparent. As an alien ant approached, they
spread their mandibles about 120° apart. When the tip of the
invader’s head was within a major’s gape, the mandibles snapped
shut, pinching the invader’s extremity with sufficient force to shoot
it away like a squirted watermelon seed. This very effective defensive
behavior, which was termed “bouncing”, kept nearly all alien ants
from gaining entrance to the nest. Only majors, with their large
mandibles and powerful adductor muscles, are equipped to do this
properly (fig. 3). Major bouncers, guarding the tube mouth, could
propel invaders backward for up to 8 or 9 cm; a single large media
was able to bounce an invader, but not for very far. The blunt apical
teeth of majors pinched but did not penetrate — invaders were not
injured at all, just repelled. Ants of all castes struck at invaders that
managed to get past the bouncers, majors contributing most to these
attacks (fig. 3). They did not attempt to bounce a successful invader,
but instead grabbed it in their mandibles and dragged it out,
unharmed, after which they resumed the guarding position.

Minor and media workers foraged in nearly equal numbers, but
did not participate equally in predatory behavior. More prey was

Figure 3: Polyethism curves of nest-centered activities, showing the percent
contribution of workers of each caste to the total colony performance of given tasks.
MI = Minor worker; ME = media; M A = Major. Some tasks are composites of several
behavior categories in theethogram (table 1): Egg care = carry or manipulate egg + lick
egg; larva care = carry or manipulate larva + lick larva + regurgitate with larva + feed
larva solid food; pupa care = carry or manipulate pupa + lick pupa. In-tube nest
maintenance = manipulate nest material + lick tube wall + remove refuse (in tube);
out-of-tube maintenance = remove refuse (out of tube) + carry dead ant. Attacking
alien ants and “bouncing” described in the text.

240

Psyche

[Vol. 88

captured by medias, while most was returned to the nest by minors
(fig. 4); minors also contributed most to processing, an in-nest
activity. Medias brought back only about one-fifth of the prey they
caught. It is possible that minors play a role similar to that of majors
in Daceton , bringing in food captured by foragers of another caste,
not themselves hunting as actively. However, minors were never
observed to take prey away from medias. They simply retrieved prey
that medias had dropped, a rather slipshod method of transferring

FORAGING CAPTURE PREY

RETURN PREY PROCESS PREY

Figure 4: Polyethism curves of predatory behavior. Behavior categories are the
same as in the ethogram (table 1).

1981]

Carlin — Polymorphism in Oreetognathus

241

food. Alternatively, the medias may have been killing flies as
trespassers approaching the nest too closely, rather than as prey,
whereupon minor foragers picked up the remains. Majors could not
leave their post at the entrance to engage in defense of the nest
vicinity without exposing the opening to invaders. Besides, the
bouncing strategy would be less effective in the open; it requires an
invader to walk directly into the defender’s mandibles.

Callow workers being easily recognizable by their lighter body
color, the repertories of age groups within castes were examined for
age polyethism. Callows exhibited fewer categories of behavior than
older adults. As in Daceton and many other ant species (Wilson
1971) they tended to concentrate on safe, in-nest tasks. Callow
majors were notably more involved in brood care than older majors.

As Brown (1957) had reported the genus to be nocturnal,
observations were taken both during the day and, under red light, at
night. Most foraging did indeed occur at night, but the ants engaged
in a greater total number of acts, in more behavior categories,
during the day, due to a diurnal rise in brood care and in-nest
maintenance activity. This result suggests that more complete
behavioral repertories can be compiled in the laboratory by studying
ants during their periods of “inactivity”, when they are not investing
so much of their effort in foraging.

Discussion

Polymorphic workers of Oreetognathus versicolor exhibit, all in
all, a fairly elementary division of labor: Minor and media reper-
tories are predictably similar, while majors constitute a distinct caste
on behavioral as well as morphological grounds. The minor size
class contributes most to in- and near-nest activity, including prey
retrieval; the medias have a somewhat greater tendency to perform
out-of-nest tasks and care for large brood; and the majors defend.

Even if the medias are capturing prey and dropping it for minors
to bring in, the resemblance to the polyethism pattern of Daceton is
convergent at most. Daceton majors, not minors, return prey to the
nest; Oreetognathus majors are bouncers. Daceton minors are
restricted to brood care, while medias perform in-nest processing
and refuse disposal (Wilson 1962); Oreetognathus minors attend all
these tasks. The polyethism of O. versicolor is entirely unrelated to
that of Daceton , having apparently arisen de novo along with its
secondary polymorphism.

242

Psyche

[Vol. 88

O. versicolor has also returned to polyphagy along with poly-
morphism, consistent with the general correlation seen in its tribe —
the only higher dacetine to return secondarily to polymorphism,
Strumigenys loriae , is also polyphagous (Brown and Wilson 1959).
The degree of dietary specialization in the genus Orectognathus as a
whole may have been overestimated: A colony of the monomorphic
species O. clarki , collected by Holldobler in New South Wales,
Australia, was maintained at a subsistance level on a diet of
cockroach and mealworm fragments and honey- water (Holldobler,
pers. comm.). However, this colony did not thrive, while the O.
versicolor colony on the same diet flourished, raising many new
workers and even males. Clearly O. versicolor does take non-
collembolan prey more readily; what is not clear is the causality
behind this correlation. The polyethism of Daceton, at least, is
associated with predatory behavior. I had speculated that the O.
versicolor majors might serve as “arthropod millers”, analogous to
the seed-miller majors of Solenopsis geminata (Wilson 1978), their
heavy mandibles used in processing a variety of prey with hard
exoskeletons. Instead, they proved to be soldiers; perhaps in
defending so efficiently, they somehow free smaller workers to
forage for different prey items, which might require wandering
further from the nest vicinity than would foraging for abundant
collembolans. But this reasoning is vague at best and requires
further investigation.

It is the major caste and its role that make this species noteworthy,
among dacetines and among ants in general. “Bouncing” is a new
kind of nest defense strategy, ideally suited for repelling enemies in a
species whose modified mandibles, designed for impaling soft-
bodied prey, are of no use in fighting. Minors and medias can be
seriously injured, in attacking invaders they are unable to harm.
Bouncing minimizes contact between defenders and invaders, expel-
ling the latter without a fight. Presumably, large workers of the
monomorphic species ancestral to O. versicolor , modifying slightly
the prey-capturing strike to pinch an extremity rather than pierce,
found themselves able to shoot enemies away for short distances.
This defense was so advantageous that heavier mandibles with
blunt, pinching teeth were strongly selected for, along with guarding
behavior, eventually producing the modern majors. Generally,
major castes in ants serve as soldiers. In a few species, they specialize
in physically blocking the nest opening with their large heads

1981]

Carlin — Polymorphism in Orectognathus

243

(certain Camponotus species, Wilson 1971; Zacryptocerus texanus,
Creighton and Gregg 1954). In Zacryptocerus varians, which also
has modified mandibles useless for fighting, majors use their saucer-
shaped heads to actively “bulldoze” invaders out (Wilson 1976).
Major bouncers of O. versicolor are unique in using their mandibles
to expel invaders without injury.

To produce a caste so specialized for this form of defense,
colonies must be under considerable pressure from ant species
approximately the same size as Solenopsis (it would be hard to
shoot a larger ant). When bouncing fails, majors do attack in a more
conventional manner, as is seen in their response to successful
invaders. (Bouncing might accidentally shoot these further into the
nest.) It has recently been shown (Holldobler 1982) that majors also
respond to alarm-recruitment pheromones.

Other dacetines, including O. clarki, the monomorphic species
most closely related to O. versicolor , often post “sentinels” at nest
entrances (Brown 1953; he also observed occasional “retrosalience”,
an ant striking at a hard surface and shooting itself backward — the
same motor act as bouncing, but apparently accidental). The O.
clarki colony, when subjected to size class polyethism analysis,
revealed a weak division of labor very similar to that of O.
versicolor minors and medias. It is easy to conceive of these size
classes as the “primitive caste” (Wilson 1980) typifying the mono-
morphic ancestor of both species, from which increasing defensive
specialization turned the sentinels still seen in the former into the
bouncers of the latter.

Acknowledgments

I am very grateful to Dr. B. Holldobler and Dr. E. O. Wilson, for
the use of materials and the suggestion of methods, for helpful
advice, for criticizing the manuscript, and for allowing me this entry
into the insect societies. I would also like to thank Mark Moffett for
suggestions, assistance, comments on the manuscript and moral
support, Dr. R. Taylor for identifying the ants, David S. Gladstein
for help with the repertory size estimations and polyethism curves,
Dr. Holldobler for the photograph in figure 1, Edward Seling for
the electron photomicrographs, and Kathleen Horton for the word
“bouncer”.

This work was supported in part by grants from the National

244

Psyche

[Vol. 88

Science Foundation, #8NS80-02613, and from the National Geo-
graphic Society, both to B. Holldobler.

References

Brown, W. J., Jr.

1953. A revision of the dacetine ant genus Orectognathus. Mem. Queensland
Mus. 13:84-104.

1957. A supplement to the revision of the dacetine ant genera Orectognathus
and Arnoldidris, with keys to the species. Psyche 64(1): 17-29.

Brown, W. L. Jr. and E. O. Wilson

1959. The evolution of the dacetine ants. Quart. Rev. Biol. 34(4):278-294.

Creighton, W. S. and R. E. Gregg

1954. Studies on the habits and distribution of Cryptocerus texanus Santschi
(Hymenoptera: Formicidae). Psyche 61(2):4 1 —57.

Fagen, R. and R. Goldman

1977. Behavioral catalogue analysis methods. Anim. Behav. 25:261-274.

Holldobler, B.

1982. Trail communication in the dacetine ant Orectognathus versicolor.
Psyche 88:245-257.

Taylor, R. W.

1977. New ants of the genus Orectognathus, with a key to the known species.
Austr. J. Zool. 25:581-612.

1979. New Australian ants of the genus Orectognathus, with summary descrip-
tion of the twenty-nine known species (Hymenoptera: Formicidae).
Austr. J. Zool. 27:773-788.

Wilson, E. O.

1962. Behavior of Daceton armigerum (Latreille), with a classification of self-
grooming movements in ants. Bull. Mus. Comp. Zool. 127(7):403-421.

1971. The Insect Societies. Belknap Press of Harvard Univ. Press, Cambridge,
Mass.

1976. A social ethogram of the Neotropical arboreal ant Zacryptocerus varians
(Fr. Smith). Anim. Behav. 24:(2):354-363.

1978. Division of labor in fire ants based on physical castes (Hymenoptera:
Formicidae: Solenopsis ). J. Kansas Entom. Soc. 51(4):6 15-636.

1980. Caste and division of labor in leaf-cutting ants (Hymenoptera: Formici-
dae: Atta). Behav. Ecol. Sociobiol. 7:143-156.

TRAIL COMMUNICATION IN THE DACETINE ANT
ORECTOGNATHUS VERSICOLOR
(HYMENOPTERA: FORMICIDAE)*

By Bert Holldobler

Department of Organismic and Evolutionary Biology,
Harvard University, Cambridge, Mass. 02138

Although division of labor within two dacetine species has been
studied at length (Wilson 1962; Carlin 1982), very little has hitherto
been reported on social communication in the Dacetini, a myrmi-
cine tribe of nearly 200 known species (Brown and Wilson 1959;
Wilson 1962). Foraging habits have also been studied in several
species (for review see Brown and Wilson 1959; Wilson 1962). As
now known, the dacetines seem to be individual foragers; recruit-
ment to food sources and cooperation during retrieval of prey have
not been observed. It is therefore of some interest that we have
recently discovered trail laying and trail following in the dacetine
species Orectognathus versicolor. Experiments in the laboratory
further indicate that trail communication may play an especially
important role during nest emigrations.

Material and Methods:

A queenright colony of O. versicolor was collected from rotting
wood near Eungella, North Queensland (Australia) and housed in a
glass tube (</> 1 cm), with water trapped at its bottom behind a
cotton plug. The nest tube was laced into an arena (45 X 30 cm) in
which small pieces of cockroaches ( Nauphoeta cinerea), chopped
meal worms ( Tenebrio molitor ), several species of small flies and
honey water were provided as food. The colony developed very well
under these conditions, and when the experiments began (4 weeks
after collection) it contained one queen, 80 workers (42 minors, 27
medias, 11 majors (see Carlin, 1982)), 14 freshly eclosed males, and
brood of all stages.

For histological investigations live specimens were fixed in Car-
noy (Romeis 1948), embedded in methyl-methacrylate and sectioned
8 fj, thick with a Jung Tetrander I microtome (Rathmayer 1962). The

♦Manuscript received by the editor December 1, 1981.

245

246

Psyche

[Vol. 88

staining was Azan (Heidenhain). The SEM pictures were taken with
an AMR 1000 A scanning electron microscope.

Additional methodological details will be given with the descrip-
tion of the individual experiments.

Results:

As demonstrated by Carlin (1982) most of the foraging in O.
versicolor is conducted by the minor and medium worker castes; the
majors function primarily or entirely as a defense caste, for which
they have unique modifications in the form of the mandibles.
Although workers of O. versicolor seem to forage individually, our
observations in the laboratory indicate that some sort of social
facilitation might be involved in stimulating foraging activity in the
colony.

Often not more than 1-3 workers roamed the foraging arena. But
when suddently 30-50 flightless Drosophila flies were released into
the arena, and the first one or two foragers had returned with
captured prey to the nest, the number of workers leaving the nest
tube and venturing into the foraging arena increased markedly. We
did not, however, observe the foragers performing any motor
display inside the nest, which might have stimulated the nestmates.

Fig. 1. Part of the colony of Orectognathus versicolor, the three worker castes
(minors, medias, majors), males, and different developmental stages.

1981] Holldobler — Communication in Orectognathus

247

nor did it appear that workers leaving the nest followed chemical
trails.

On the other hand, trail following was very obvious when the
colony or fragments of the colony were forced to move to a new nest
site. For example, when we shook the colony out of the nest tube
into the arena, which had been provided with a new papered floor
before each experiment, the “homeless” colony soon gathered at one
spot, where it was closely guarded by members of the major worker
caste (Fig. 1). After varying intervals (sometimes lasting more than
one hour), some of the minors and medias began exploring the
arena, and eventually they discovered a nest tube that had been
provided at the edge of the arena (usually 30-35 cm away from the
displaced colony). After exploring the nest tube, some of the ants
returned to the colony, and after a while they often moved again to
the nest tube to continue to explore it thoroughly.

Usually this procedure was repeated several times, before the first
signs of a colony movement could be observed. It occurred when
several additional minors and medias departed from the colony and
traveled directly to the new nest. Their straight orientation and the
fact that during running they kept the tips of their antennae close to
the ground, suggested that these ants were following a chemical
trail. Soon afterwards the traffic between the “homeless” colony and
the newly discovered nest tube increased leading finally to a full-
scale colony emigration.

All three worker castes were involved in transporting brood,
callow workers, and males to the new nest, although the minors
handled eggs and small larvae preferentially while the medias and
majors concentrated on large larvae, pupae and adults (Fig. 2).
Usually the queen moved during the early phase of the colony
movement and always traveled on her own. On the other hand, the
males were always carried by the workers (Fig. 2), usually not before
most of the brood had already been moved. Only once did we see a
fully pigmented worker being carried by a nestmate. The trans-
ported individual was grasped dorsally at the head and lifted
upwards with gaster tip pointing forwards; it had the appendaes
folded in the pupal position. All ants traveled along a relatively
narrow route between colony and new nest site. This strongly
suggested that O. versicolor employs chemical trail communication
during the process of colony migration. The following experiments

gpggp

248

Psyche

[Vol. 88

Fig. 2. Colony emigration in Orectognathus versicolor, (a) major transporting larva; (b) major transporting pupa; (c) media
transporting larva, accompanied by a minor; (d) media transporting male.

1981] Holldobler — Communication in Orectognathus

249

were designed to localize the anatomical source of a possible trail
pheromone in O. versicolor.

Close-up cinematography and photography revealed that many
Orectognathus workers, when moving back and forth between the
displaced colony and the nest tube, touched their abdominal tips
intermittently to the ground, presumably depositing droplets of trail
pheromone. Three major exocrine glands open at or near the
abdominal tip of O. versicolor workers: the poison gland and
Dufour’s gland, both of normal size, and a relatively large pygidial
gland, which opens between the 6th and 7th abdominal tergites (Fig.

3).

Most myrmicine ants have a more or less well developed pygidial
gland (Holldobler et al. 1976, Holldobler and Engel 1978; Kugler
1978), but in O. versicolor this gland is more complex than usually
found in Myrmicinae. It more closely resembles the pygidial gland
of some ponerine species, for example Pachycondyla laevigata , in
which it serves as the source of a trail pheromone. The paired
reservoir sacs (invaginations of the intersegmental membrane be-
tween the 6th and 7th tergites) are filled with a clear, lightly
brownish liquid. Several ducts lead from paired clusters of glandular
cells into the reservoir, penetrating the intersegmental membrane
(Fig. 4). The cuticle of the 7th tergite has a grooved structure (Fig.
5), underneath of which is a large glandular epithelium (Fig. 4).
Orectognathus versicolor workers, when engaged in trail laying
behavior, usually hold their gaster in an almost vertical position.
This brings the opening of the pygidial gland very close to the floor
so that part of the grooved structure on the 7th tergite can be easily
put in contact with the surface of the ground.

In the next series of experiments we tested the trail following
response of O. versicolor to artificial trails drawn with glandular
secretions of the poison gland, Dufour’s gland and pygidial gland.
The glands were dissected out of freshly killed workers and for each
trail test one gland of a kind was crushed on the tip of a hardwook
applicator stick and smeared once along a 20-cm-long pencil line.
The trails were made to originate either from the entrance of the
nest tube or from the periphery of the clustered colony, which had
previously been shaken out of their nest tube into the test arena. As
a control a second trail was offered simultaneously which was
derived either from a droplet of water or from one of the other

250

Psyche

[Vol. 88

Fig. 3. Gaster of a media of Orectognathus versicolor, (a) SEM picture; arrow
indicates opening of pygidial gland, (b) Sagittal section through gaster of a media.
PG = pygidial gland; R = reservoir of pygidial gland; D = Dufour’s gland; S =
stinger.

1981] Holldobler — Communication in Orectognathus

251

Fig. 4. (a) Sagittal section through pygidial gland of a media of Orectognathus
versicolor. PG = pygidial gland; R = reservoir of pygidial gland; GE = glandular
epithelium, (b) Close-up of glandular epithelium (GE) under the cuticular structure
(CS) of pygidium.

252

Psyche

[Vol. 88

Fig. 5. Above: SEM pictures of the gaster tip of a media of Orectognathus
versicolor. The slightly extruded stinger is visible. It is surrounded by long sensory
setae (confirmed by histology; probably mechanoreceptors) on the edge of the
pygidium and 7th sternum.

Below: Groved cuticular structures on the pygidium associated with the pygidial
gland. This structure is usually covered by the preceding 6th tergite.

1981] Holldobler — Communication in Orectognathus

253

glands. All ants following the trails to the end during a 5-minute
period were counted.

As can be seen from table 1, trails drawn either with crushed
pygidial glands or poison glands elicited a precise trail following
behavior in all three worker castes (Fig. 6), but the ants did not
respond to trails drawn with crushed Dufour’s glands. We noticed,
however, several differences in the reaction of the ants to poison
gland trails and pygidial gland trails. (1) When both trails were
offered simultaneously, starting at the periphery of a “homeless”
colony, significantly more workers (Tab. 1) carrying brood moved
along the poison gland trail. (2) In all tests the poison gland trail was
the more effective one and lasted over a longer period of time. After
5 minutes the ants’ response to pygidial gland trails had almost
vanished, whereas they were still strongly following the trail drawn
with poison gland material. In fact, poison gland trails presented to
the ants 24 hours after they were drawn were still effective as
orientation cues for emigrating O. versicolor workers. (3) Although
we could not detect a preference for either trails drawn with poison
glands or pygidial glands, ants moving along the pygidial gland trail
seemed to gape their mandibles more frequently than ants moving
along poison gland trails.

From these observations we conclude that the trail pheromones
serve different functions. The poison gland trail is obviously
employed during nest emigrations, where it serves as a stimulative
recruitment signal as well as a longer lasting orientation cue. On the
other hand, the pygidial gland trail probably functions as a relatively
short lasting alarm-recruitment signal, channeling workers to areas
of disturbance near the nest. It is also possible that the pygidial
gland pheromone is discharged by successful foragers when they
return to the nest, which might cause the social facilitation of the
foraging activity mentioned above. In fact, when a crushed pygidial
gland is presented inside the nest tube, it elicits more excitement in
the workers than any other glandular secretions (mandibular gland,
poison gland, Dufour’s gland), causing several workers to move
toward the nest entrance.

All three worker castes have the same glandular equipment and
their secretions release the same behavioral responses.

254

Psyche

[Vol. 88

Fig. 6. Trail test with Orectognathus versicolor. Artificial trails drawn with
secretions from the poison gland (PoG) and Dufour’s gland (C), both originating at
the opening of the nest tube, are offered simultaneously. All workers follow the
poison gland trail.

1981] Holldobler — Communication in Orectognathus 255
Discussion:

It has been well documented that many species of the sub-family
Myrmicinae employ secretions from the glands associated with the
sting apparatus (poison gland, Dufour’s gland) for chemical trail
communication and orientation (for review see Wilson 1971; Holl-
dobler 1978). This paper presents the first evidence of the phenom-
enon in the myrmicine tribe Dacetini.*

In the dacetine species Orectognathus versicolor trails laid with
poison gland secretions function both as recruitment and orienta-
tion signals during nest emigration. In fact, many dacetine species
seem to construct relatively simple nests in soil or rotting wood and
it is easily conceivable that colonies frequently abandon their nests
and move to new nest sites. More surprising, however, was the
discovery that this species possesses a pygidial gland whose struc-
ture closely resembles that of the pygidial gland of some ponerine
species. The secretions of this gland can also function as a recruit-
ment trail pheromone in O. versicolor.

Table 1. Number of workers following artificial trails within 5 min. periods. The
means and standard deviations are given.

Trails presented at nest entrance (n = 4)

Dufour’s

water

Poison

water

Pygidial

water

gland

control

gland

control

gland

control

12.7 ± 3.8

8.3 ± 2.8

Trails presented simultaneously at periphery of clustered colony (n = 5)

Dufour’s gland

Poison gland

Pygidial gland

with

brood

without

with brood without

with brood

without

incl.

males

brood

inch males brood

incl. males

brood

7.8 ± 3.8 7.0 ± 2.2

2.0 ± 1.6

9.8 ± 3.1

*Blum and Portocarrero (1966) demonstrated that three attine ant genera follow
trails drawn with poison gland secretions of Daceton armigerum, but they could not
demonstrate trail following behavior in Daceton.

256

Psyche

[Vol. 88

From recent investigations we know that the pygidial gland is
quite common in the Myrmicinae (Kugler 1978; Holldobler and
Engel 1978). In at least two species its secretions serve as an alarm
pheromone (Holldobler et al. 1976; Kugler 1979). On the other
hand, in several ponerine species the pygidial gland secretions have
been demonstrated to function as a recruitment pheromone during
tandem running (Holldobler and Traniello 1980a) or trail communi-
cation (Maschwitz and Schonegge 1977; Holldobler and Traniello
1980b). From our findings in O. versicolor it appears now that this
ponerine trait might have been preserved in the Dacetini, whose
origin presumably dates back to early Tertiary times (Brown and
Wilson 1959). If this speculation is right, we should expect that the
most primitive dacetine species, Daceton armigerum (Brown and
Wilson 1959; Wilson 1962), has a well developed pygidial gland,
resembling closely that found in many ponerine ants, and its
secretions presumably serve as an alarm-recruitment pheromone. In
fact, Wilson (1962) observed that workers of D. armigerum often
moved to areas of excitement and when a worker just had dis-
covered prey it moved in “excited broken running patterns” by
which other ants in the vicinity might be attracted. Wilson (1962,
1971) hypothesized that this running pattern might serve as a
communicative signal of the kind of “Stager’s kinopsis”, i.e. the
large-eyed Daceton workers might respond to the visual stimuli
produced by the excitedly moving nestmate. We have now to
consider the possibility that a Daceton huntress which pursues a
prey, discharges a short-range recruitment pheromone from the
pygidial gland, and that consequently the attraction of other
huntresses in the close vicinity is caused by this chemical signal.

Acknowledgments

I would like to thank E. O. Wilson for reading and commenting
on the manuscript. Hiltrud Engel and E. Seling (SEM-Lab of the
MCZ, Harvard) provided valuable technical assistance. Many
thanks to R. W. Taylor and the Division of Entomology of CSIRO,
Canberra (Australia) for their support and hospitality. R. W. Taylor
identified the ants; voucher specimens were deposited in the Austral-
ian National Insect Collection, Canberra. This work was supported
by grants of the National Geographic Society, National Science

1981] Holldobler — Communication in Orectognathus

257

Foundation (BNS 80-02613) and by a fellowship from the John
Simon Guggenheim Foundation.

References

Blum, M. S. and C. A. Portocarrero

1966 Chemical releases of social behavior. X. An attine trail substance in the
venom of a non-trail laying myrmicine, Daceton armigerum. Psyche
(Cambridge) 73: 150-155.

Brown, W. L. and E. O. Wilson

1959 The evolution of the dacetine ants. Quarterly Rev. Biology 34, 278-294.

Carlin, N. F.

1982 Polymorphism and division of labor in the dacetine ant Orectognathus
versicolor (Hymenoptera: Formicidae) Psyche (Cambridge) 88:231-244.

Holldobler, B. and H. Engel

1978 Tergal and sternal glands in ants. Psyche (Cambridge) 85, 285-330.

Holldobler, B. and J. Traniello

1980a Tandem running pheromone in ponerine ants. Naturwissenschaften 67,
360.

Holldobler, B. and J. F. A. Traniello

1980b The pygidial gland and chemical recruitment communication in Pachy-
condyla (= Termitopone ) laevigata. J. Chem. Ecology 6, 883-893.

Holldobler, B., R. Stanton and H. Engel

1976 A new exocrine gland in Novomessor (Hymenoptera: Formicidae) and
its possible significance as a taxonomic character. Psyche 83, 32-41.

Kugler, C.

1978 Pygidial glands in the myrmicine ants (Hymenoptera: Formicidae).
Insectes sociaux 25, 267-274.

1979 Alarm and defense: a function for the pygidial gland of the myrmicine
ant, Pheidole biconstricta. Annals Entomological Society America 72,
532-536.

Maschwitz, U. and P. Schonegge

1977 Recruitment gland of Leptogenys chinensis. Naturwissenschaften 64,
589-590.

Rathmayer, W.

1962 Methylmetacrylat als Einbettungsmedium fur Insekten. Experientia
(Basel) 18, 47-48.

Romeis, B.

1948 Mikroskopische Technik, Miinchen 1948.

Wilson, E. O.

1962 Behavior of Daceton armigerum (Latreille), with a classification of self-
grooming movements in ants. Bull. Mus. Comp. Zool. Harvard Univ.
127, 401-422.

1971 The insect societies. The Belknap Press of Harvard University Press,
Cambridge, Mass.

FRANCIS WALKER TYPES OF, AND NEW SYNONYMIES
FOR, NORTH AMERICAN HYDROPSYCHE SPECIES
(TRICHOPTERA, HYDROPSYCHIDAE)*

By Andrew P. Nimmo

Department of Entomology, University of Alberta
Edmonton, Alberta, Canada T6G 2E3

Introduction

Recently, while assembling the manuscript of a handbook to the
Arctopsychidae and Hydropsychidae of Canada, I had occasion to
examine the female holotypes of three species of Hydropsyche
described from North America by Francis Walker (1852). Betten &
Mosely (1940) record these types, but in line with the still all too
prevalent practice regarding unassociated female Trichoptera, they
did not illustrate the genitalia. If they had provided figures for these
types the true identities of at least two North American species of
Hydropsyche would have been long since established. The genital
segments of these types are illustrated here for the first time.

Hydropsyche confusa (Walker)

Philopotamus confusus Walker, 1852: 103.

Hydropsyche confusa : Milne, 1936: 61; Betten & Mosely, 1940: 21.
Hydropsyche separata Banks, 1936: 129; Ross & Spencer, 1952: 46
(as synonym of H. guttata Pictet); Smith, 1979: 10. new

SYNONYMY.

Hydropsyche guttata Pictet: Schuster & Etnier, 1978: 126.
Hydropsyche corbeti Nimmo, 1966: 688; Schuster & Etnier, 1978:
126 (as synonym of H. guttata Pictet), new synonymy.

Fig. 2 depicts the genitalia of the female holotype of H. confusa
(Walker). Fig. 1 is of the genitalia of a female which has been
recognised as belonging to H. separata Banks. The rather obscure
locality information recorded by Betten & Mosely (1940) indicates
that the type of confusa was collected in the western Northwest
Territories, adjacent to the northern boundary of Alberta. The
female of separata was collected at Empress in southeastern Alberta.

* Manuscript received by the editor October 14, 1981.

259

260

Psyche

[Vol. 88

Figures 1-5. Fig. 1. Hydropsyche separata Banks [=confusa (Walker)] — genital
segments of female, lateral aspect. Fig. 2. H. confusa (Walker) — genital segments
of female holotype, lateral aspect. Fig. 3. H. recurvata Banks [ =alternans (Walker)]
— genital segments of female, lateral aspect. Fig. 4. H. alternans (Walker) — genital
segments of female holotype, lateral aspect. Fig. 5. H. reciproca (Walker) — genital
segments of female holotype, lateral aspect.

1981]

Nimmo — Walker Types of Hydropsvche

261

While not absolutely identical (the differences may be attributed
to geographic variation, laboratory treatment, and observer vari-
ables), these two specimens are much more similar to each other
than either is to the females of the most nearly related species ( H .
betteni Ross), and I judge them to be conspecific. H. separata is
therefore synonymised with H. confusa which has clear priority.

Smith (1979) quotes me as considering the possibility that H.
corbeti Nimmo may be a synonym of separata. Prior to my
examination of the type of confusa I had decided that it was.
However, it must now be entered as a synonym of confusa.

In view of the taxonomic history of this species Walker must be
attributed with remarkable insight in naming it confusa.

Hydropsyche alternans (Walker)

Philopotamus alternans Walker, 1852: 104.

Hydropsyche alternans: Vorhies, 1909: 707 (sp.indet.); Betten, 1934:
185 (prob. H. bifida ).

Philopotamus indecisus Walker, 1852: 104.

Hydropsyche indecisa: Betten & Mosely, 1940: 20 (as synonym of
H. alternans ).

Hydropsyche slossonae var. recurvata Banks, 1914: 253.
Hydropsyche recurvata : Betten, 1934: 190; Milne, 1936 73 (as
synonym of H. slossonae ); Ross, 1944: 99. new synonymy.
Symphitopsyche recurvata : Schuster & Etnier, 1978: 34.
Hydropsyche codona Betten, 1934: 187; Milne, 1936: 73 (as syn-
onym of H. slossonae ); Ross, 1938: 18 (as synonym of H.
recurvata).

Fig. 4 depicts the genitalia of the holotype female of H. alternans
(Walker) (from the Albany R., far northern Ontario). Fig. 3 depicts
the genitalia of a female (from Wandering River, northeastern
Alberta) which has been recognised as belonging to H. recurvata
Banks. Again, these two females are not precisely identical, for
possible reasons similar to those given under H. confusa above. I
judge these two specimens to be conspecific. H. recurvata is
therefore synonymised with H. alternans which has priority.

262

Psyche

[Vol. 88

Hydropsyche reciproca (Walker)

Philopotamus reciprocus Walker, 1852: 104.

Hydropsyche reciproca : Betten & Mosely, 1940: 22.

The genitalia of the female holotype do not correspond to those
of any other species known to me. They are illustrated here, for the
first time, for the future reference of students of North American
Hydropsyche species. The type locality is given simply as ‘North
America’.

Summary

The female holotypes of Hydropsyche confusa (Walker), H.
alternans (Walker), and H. reciproca (Walker), all from North
America, were examined, and illustrations of the genitalia are
provided for the first time. It is concluded that H. separata Banks is
conspecific with H. confusa (Walker), and that H. recurvata Banks
is conspecific with H. alternans (Walker). The Walker names have
priority. H. reciproca (Walker) cannot yet be equated with any
other known species.

Acknowledgments

Loan of the three types examined here was very kindly arranged
by Peter Barnard of the Entomology Dept, British Museum (Na-
tural History). This paper is an offshoot of work on preparation of a
Handbook to the Arctopsychidae and Hydropsychidae of Canada,
which was supported by a contract from the Canada Dept of
Agriculture. The work was carried out in the Dept, of Entomology,
University of Alberta. Steve Ashe of that Department read, and
commented on the manuscript. Publication costs were met by
George E. Ball of the same Department, from grant NRC A- 1399
held by him.

To all, my sincere thanks.

1981]

Nimmo — Walker Types of Hydropsyche

263

References

Banks, N.

1914. American Trichoptera — notes and descriptions. Can. Ent. 46: 252-258.
1936. Notes on some Hydropsychidae. Psyche, Camb. 43: 126-130.

Betten, C.

1934. The Caddis Flies or Trichoptera of New York State. Bull. N.Y. St. Mus.,
Albany 292:1-576.

Betten, C. and M. E. Mosely

1940. The Francis Walker Types of Tricoptera in the British Museum. British
Museum (Natural History), London, vi+248 pp.

Milne, L. J.

1936. Studies in North American Trichoptera. 3:56-128, Cambridge, Mass.
Nimmo, A. P.

1966. A list of Trichoptera taken at Montreal and Chambly, Quebec, with
descriptions of three new species. Can. Ent. 98:688-693.

Ross, H. H.

1938. Lectotypes of North American Caddis Flies in the Museum of Compara-
tive Zoology. Psyche, Camb. 45:1-61.

1944. The Caddis Flies, or Trichoptera, of Illinois. Bull. 111. St. Nat. Hist.
Surv., Urbana, 23, 326 pp.

Ross, H. H. and G. J. Spencer

1952. A preliminary list of the Trichoptera of British Columbia. Proc. Ent.
Soc. Br. Columb. 48:43-51.

Schuster, G. A. and D. A. Etnier

1978. A manual for the identification of the larvae of the Caddisfly genera
Hydropsyche Pictet and Symphitopsyche Ulmer in eastern and central
North America (Trichoptera: Hydropsychidae). Environmental Moni-
toring and Support Laboratory, Cincinnati EPA-600/ 4-78-060: xii+129

pp.

Smith, D.

1979. The larval stage of Hydropsyche separata Banks (Trichoptera: Hydro-
psychidae). Pan-Pacif. Ent. 55:10-20.

Vorhies, C. T.

1909. Studies on the Trichoptera of Wisconsin. Trans. Wis. Acad. Sci. Arts
Lett. 16:647-738.

Walker, F.

1852. Catalogue of the specimens of neuropterous insects in the collections of
the British Museum. Part 1. 192 pp. London.

TERRITORIALITY, NEST DISPERSION,

AND COMMUNITY STRUCTURE IN ANTS.

By Sally C. Levings1 and James F. A. Traniello2
Introduction

The dispersion patterns of ant colonies have been reported for a
variety of species having very different ecological characteristics
(Pontin 1961; Yasuno 1963, 1964a, b, 1965; Brian 1964; Brian et al.
1965, 1966; Greenslade 1971; Room 1971, 1975a, b; Bernstein and
Gobbel 1979; Levings and Franks 1982), and typically, spacing
studies involve discussions of territoriality. Recently, Holldobler
and Lumsden (1980), using a cost/ benefit approach, examined the
importance of the economic defensibility of territories and its
influence on the use of space and dispersion patterns. Holldobler
(1974, 1976a, 1979a, b) demonstrated the relationship between re-
source distribution, territory shape and nest spacing. These studies
also emphasize that in order to understand thoroughly territoriality
and other intra- and interspecific relationships, it is necessary to
comprehend the role of social design in the establishment and
maintenance of territory. Without such a combined approach of
behavior and ecology, it is difficult to assess accurately the signifi-
cance of territoriality in social species such as ants.

In many studies there have been problems in the application of
the term territoriality and discrepancies in the identification of
territorial phenomena. Terms describing the use of foraging area
such as territory and home range have been rather poorly defined
and vary in meaning between authors. Territory to some authors
denotes a defended area (Baroni-Urbani 1979; Holldobler 1974,
1976a; Holldober and Wilson 1977a, b; Holldobler and Lumsden
1980) whereas to others it is synonymous with home range or is
casually used (Dobrzanska 1958, 1966). There are also problems
with the application of information on territoriality in the interpre-
tation of spacing patterns. For example, mathematical evidence of

Museum of Comparative Zoology Laboratories, Harvard University, Cambridge,
Mass. 02138.

department of Biology, Boston University, Boston, Mass. 02215 [To whom reprint
requests should be sent].

Manuscript received by the editor June 19, 1981.

265

266

Psyche

[Vol. 88

nest overdispersion is frequently confused with, or taken as evidence
for, territoriality although crucial behavioral patterns are not
considered. However, sufficient information is available in the
literature to suggest some of the behavioral and ecological factors
important in the regulation of nest distribution.

With the above cautions in mind, we here present a simple model
of predicted spatial distributions of colonies under different ecologi-
cal conditions. We then survey the literature to examine the fit of
available data to our predictions. Finally we discuss the general
problem of the form of interactions between colonies and some of
the implications of this for both field and theoretical considerations.

Theoretical Aspects of Nest Distribution Patterns.

We would first like to develop a set of biologically realistic
predicted spatial distributions of colonies. We begin by positing
some simple assumptions about a hypothetical ant population:

1. Nest sites are unlimited.

2. The habitat is homogenous and inhabited by a single species.

3. Each colony forages symmetrically around the nest to some
distance r, which forms the radius of a circle. Within this circle, no
other colonies can forage or become established.

Simberloff (1979) derives the maximum foraging distance, r, as

sly

4\/3 \/p

where p is the density of nests. In this case, nests are hexagonally
packed and the array of nests is overdispersed (more regularly
spaced than expected if random; Figure 1, case 1). Nests are spaced
2 r apart and have 6 equidistant nearest neighbors.

Under different ecological conditions, the expected spatial dis-
tribution of nests will change. In low density populations, nest
distribution should reflect the best foraging or nest sites; nests may
be dispersed in any way and should tend towards a random
distribution (Figure 1, case 2). Internest distance should on the
average be at least twice r and usually more; its variance should be
high. If nest sites are not uniformly available, then nest spacing will

1981]

Levings & Traniello — Territoriality in Ants

267

depend upon whether or not nest sites are farther apart or closer
together than this distance. We predict one of 2 patterns: (1) nests
will be more overdispersed than potential nest sites (Case 3a) or (2)
although nests may be clumped in space, foraging ranges which are
asymmetric and which partition foragers will develop (Case 3b). If
potential nest sites are farther apart than twice r, then nests will be
distributed only with respect to potential nest sites. The effects of
habitat heterogeneity will depend upon the scale and extent of the
patchiness in relation to the foraging range of a species. If patches
hold several to many colonies, then clumps of nests which are
overdispersed within the clump are predicted. Smaller patches in
complex mosaics will not generate predictable nest distributions
unless the arrays of patches are very regularly distributed.

The effect of adding more species to the system will depend upon
the species. Generally, in multi-species systems, the level of repul-
sion observed between co-occurring species should be directly
proportional to the amount of overlap in resource use. Species
utilization curves can range in overlap from 0 to essentially com-
plete ecological identity (100% overlap). Predicted spatial patterns
will clearly depend on the actual distribution of species. If two or
more species with identical requirements and foraging radii occur in
the same area, interactions within and between species should be
equally strong. In this case, the pattern of nest distribution predicted
is random for any one species (Franks 1980; Levings and Franks
1982). Nests should be overdispersed, but each species is distributed
with respect to every other species (i.e., nests of all species are
treated as equivalents). In addition, there should be no pattern in
the species identity of nearest neighbors (Case 4). Removal of any
one species should have the effect of the removal of a nest at random
from an overdispersed array; the degree of observed overdispersion
should decrease. The spatial dispersion of any one species in such an
array should tend to look like a low density nest population (Case
2), but the history of the area may cause any type of pattern under
different conditions.

If two or more species have the same foraging radius but do not
overlap 100% in resource requirements, intraspecific interactions
should be stronger than interspecific interactions (Case 5). We
predict that (1) the entire array will be overdispersed and (2) each
species will also be overdispersed from itself. Franks (1980) and

268

Psyche

[Vol. 88

FIGURE 1

Case 1 High density population
Assumptions: 1. Single species

2. All nests have the same r

3. Unrestricted nest sites

Predictions: 1. Overdispersed nest array

2. Nest to nest distance ~ 2 r

Case 2 Low density population
Assumptions: 1, 2, 3

Predictions: 1. Nest distribution will tend to randomness

2. Average nest to nest distance >2 r

3. High variance in nest to nest distance

Case 3 Limited nest sites

a. Assumptions:
Predictions:

b. Assumptions:
Predictions:

1, 2

1. Nests more overdispersed than potential nest sites

2. Nest spacing will vary with nest site location, minimum nest
to nest distance = 2 r, average nest to nest

distance >2 r

3. High variance in nest to nest distance
1

1. Nests distributed as nest sites

2. Asymmetric foraging ranges

Case 4 Intraspecific = interspecific interactions
Assumptions: 2, 3

Predictions: 1. Entire nest array overdispersed

2. Individual species are more randomly dispersed than the
total array

3. No pattern in the identity of nearest neighbor

4. High variance in nest to nest distances within a species,
average nest to nest distance >2 r

5. Low variance in nest to nest distances for the entire array,
average nest to nest distance = 2 r

Case 5 Intraspecific interactions > interspecific interactions
Assumptions: 2, 3

Predictions: 1. Entire nest array overdispersed

2. Individual species within the array are also overdispersed

3. Nearest neighbors tend to be members of other species

4. Low variance in nest to nest distances within species, average
nest to nest distance >2 r

5. Low variance in internest distances for the entire array,
average nest to nest distance = 2 r

1981]

Levings & Traniello — Territoriality in Ants

269

CASE I

CASE 2

CASE 3a

CASE 3b

CASE 4

CASE 5

Figure 1. Theoretical nest dispersion patterns under different ecological condi-
tions. Additional details in text.

270

Psyche

[Vol. 88

Levings and Franks (1982) have reviewed the relevant statistical
literature and give a suggested procedure for examining this
problem.

In addition to changes in the observed spatial array of any one
species, in multi-species populations, there should be correlated
changes in expected internest distances under different competitive
regimes. If intra- and interspecific interactions are equally strong,
the average internest distance within any one species should be
longer than twice the species’ average r and the variance in between
nest distances within any one species should be high (essentially a
low density population, Case 2). If intraspecific interactions are
more important than interspecific interactions, then internest dis-
tance within any one species should be greater than twice the
species’ average r and their variance should be relatively low. The
exact predicted distance will be a function of the number of
interacting species and their relative abundances. It may be possible
to use the degree of departure from predicted intraspecific spacing
patterns as a measure of competition between species in homog-
enous habitats. If intranest distances within a species are 2 r, then it
does not appear to be interacting significantly with sympatric
species, at least not in ways which affect its spatial distribution.

Detection of Overdispersion and Methodological Problems

There are certain methodological difficulties in applying any sort
of spatial analysis to previously published data on nest distribu-
tions. In particular, the complicated structure of the nests of many
species has confused workers, especially when many nest entrances
are present. In Lasius neoniger, Headley (1941) assumed that the
species was unicolonial, since he could only occasionally elicit
aggression between adjacent nest entrances. In fact, L. neoniger
colonies are distinct and well organized, but extensive field tests are
required to delineate colony boundaries (Traniello 1980). Simple
mapping of nest openings may reflect the distribution of colonies
fairly well (as it does for many species in the ground ant community
in Panama, Levings and Franks 1982; Levings, personal observa-
tions), but may lead to confusion unless sufficient data on the
species are available (see, for example, Brough 1976). Whitford et
al. (1980) assumed that workers of Novomessor cockerelli were
entering an alien nest because they did not return to the same nest

1981]

Levings & Traniello — Territoriality in Ants

271

entrance from which they departed. However, Holldobler et al.
(1978) and Davidson (1980) documented that this species has nests
with multiple entrances.

Although there are several methods for the detection of over-
dispersion (Pielou 1977), we have chosen to apply Clark and Evans’
(1954) nearest neighbor (NN) technique wherever possible. It is
based upon the ratio between the observed mean nearest neighbor
distance and the expected distance when a population is distribution
at random. The index R can range from 0 (perfect aggregation) to
2.1491 (perfect hexagonal overdispersion). A value of 1 indicates a
random dispersion pattern. The significance of R is tested using the
z transformation. In an overdispersed population, the observed
mean nearest neighbor distance is larger and the variance in nest to
nest distance is lower than it would be in a randomly distributed
population. Thus a population which is significantly overdispersed
using this measure confirms 2 of our predictions (overdispersion
and low variance in NN distance). Other methods do not have this
property.

In our evaluation of spacing information in the literature, if we
were unable to apply nearest neighbor methods, but complete
quadrat counts were published, we calculated variance/ mean ratios
and tested them for significance using X2 statistics (Pielou 1977). A
V/M ratio of less than 1 indicates overdispersion while values
greater than 1 indicate clumping. Cases are included in which data
are not sufficient to test for statistical overdispersion, but informa-
tion on partitioning of resources or area was published. We have
organized the available data by geographic region, habitat and food
types (Table 1). Methods used in gathering previously unpublished
data will be described with the specific set of data. In testing our
model and spatial predictions from the literature, we are limited by
the previous interests and focus of other authors. We are able to test
the spatial predictions far more thoroughly than the hypotheses
about the actual expected distances between nests, but there is no
empirical reason that they cannot be experimentally verified in the
field (see discussion).

Data are discussed by subdividing reported cases into groups
according to foraging type: (1) species which do not defend re-
sources although they may or may not recruit to food, (2) species
which defend randomly and unpredictably distributed resources
(e.g., dead insects, which are patchy in both space and time), (3)

272

Psyche

[Vol. 88

species which defend predictable and persistent resources (e.g.,
honeydew from aphids, resources which are patchy in space but not
in time) and (4) truly territorial species which defend area which has
potential food resources. These divisions mark some ecologically
important foraging types within communites.

Observed Patterns

1. Nest Defense

The data suggest that species which display only nest defense fall
into 4 major groups, depending upon the details of their foraging
biology. First, some species forage only as solitary individuals for
food items which a single forager can subdue and retrieve (Group I
foragers, Oster and Wilson, 1978). Examples of this group include
most Dacetini, many Ponerinae, and some of the non-leaf cutting
Attini (Brown and Wilson 1959, Wilson 1971, Oster and Wilson
1978).

There is very little applicable data on this group. The frequency of
dacetine nests in extensive Berlese sampling of a tropical deciduous
forest fit a Poisson distribution indicating a random distribution
(Levings, unpublished data), but this sort of data does not differen-
tiate between the suitability of the site or other important factors in
the distribution of nests. Certainly there was no indication that nests
were clumped. The maximum number of nests found was 6 in 84
0.25 m2 samples. When a truncated Poisson was fit (0 class
excluded), the distribution did not differ from Poisson expectation
(p > 0.5, x2 test).

Second, some species may recruit nestmates to food resources,
but make no attempt to defend them, decamping if another, more
aggressive, species arrives before the food is retrieved (Group II, in
part, Oster and Wilson 1978). These species specialize in the rapid
location and removal of food. Examples include Paratrechina
longicornis and Tapinoma melanocephalum (Wilson 1971). No data
on their nest distribution is available, but many are known to form
small fragmented colonies which move frequently between ephem-
eral nest sites.

The third set of species have developed mechanisms for feeding at
the same resources as other, more aggressive ants, without eliciting
defensive reactions (Groups I & II, in part, Oster and Wilson 1978,
Wilson 1971). It is not known how much of a colony’s food intake

1981]

Levings & Traniello — Territoriality in Ants

273

results from such theft and how much is independently gathered.
Examples include Leptothorax acervorum and various Cardio-
condyla species (Brian 1955; Wilson 1959a, 1971). These species
usually recruit few other workers to the food item; many of these
species recruit only one other nestmate using tandem running
(Wilson 1959a). No spacing information is available for these
species.

The fourth set of species include the legionary ants (true group
foragers) and most of the specialists on extremely difficult prey
(Groups IV and V, Oster and Wilson 1978). These species defend
only their nest sites (which may move often) and forage in various
sized groups. The most spectacular examples of this type of foraging
are the army ants (Schneirla 1971). Specialists on difficult prey
occur in several genera (examples, Pachycondyla ( =Termitopone ),
Leptogenys, Gnamptogenys)\ specialized retrieval methods may
involve extensive cooperative foraging (Wilson 1971). Little nest
spacing information is recorded about these groups. Army ants of
several genera have been observed to avoid each other when they
meet in the field, but no similar information is available for related
groups (Schneirla 1971). Other legionary groups are relatively rare
on BCI and, in 4 years of field work, no interactions were observed
(Levings, personal observation).

In general, information on spacing patterns of ants which defend
only their nests is extremely difficult to gather, since the investigator
must usually depend upon luck to locate colonies and will never be
certain that all colonies in an area have been found. Because
information on foraging ranges for most species is unavailable, we
are unable to test those aspects of our hypotheses. Many species
which are now assumed to defend only their nest sites may well be
found to defend either resources or a foraging territory.

2. Resource defense
a. short term

The defense of unpredictable resources occurs on varying time
scales. Resources which persist for very short periods (i.e., minutes
for most dead insects) are defended by many generalist or scavenging
ants during the recruitment/ retrieval process (Groups II & III, in
part, Oster and Wilson 1978). Spatial overdispersion in densely
populated areas has been shown in one complex tropical com-
munity (Levings and Franks 1982). It is probably typical of many

274

Psyche

[Vol. 88

reported cases of overdispersion in temperate ground ant com-
munities dominated by relatively few generalist species (most species
of the genera Myrmica, Tetramorium, Lasius, Aphaenogaster, some
Formica; Table 1). Some species are placed here somewhat ar-
bitrarily because good foraging ecology data are not available.

In more complex (i.e., non-uniform) habitats, the pattern of nest
spacing is reported to be directly related to environmental condi-
tions. Lasius flavus, which has been intensively studied in several
European habitats, displays different nest distributions between
locations. Waloff and Blackith (1962; Table 1) found that nests were
overdispersed in a high density population and tended toward
randomness in a low density population. In a wet, low pasture with
limited nest sites, nests were also overdispersed (Blackith et al.,
1963, Table 1). With Myrmica rubra present in a low density
population, L. flavus was randomly distributed (Elmes 1974).
However, the partial segregation of species indicated that both
intra- and interspecific interactions were present; M. rubra nests
were more overdispersed than potential nest sites (Table 1). Similar
patterns have been noted in other species. Petal (1972) showed that
the pattern of distribution in Myrmica laevinodis depended upon
the scale with which the species was examined. Within the habitat,
nests were clumped, but within clumps of nests on a small scale,
nests were either overdispersed or randomly distributed. In another
study, Petal (1977) linked observed nest distribution and the avail-
able food supply in Myrmica lemanica. In a year with low food
abundance, nests were overdispersed; when food was abundant, nest
distribution was random, tending to aggregation. Petal did not state
if she distinguished between nests and nest openings by testing
aggressive responses between colonies. However, overall nest density
remained approximately the same. Most other studies have assumed
but not demonstrated the correlation between food abundance and
nest dispersion patterns.

Within colonies with multiple nest entrances, the distance be-
tween nest entrances should be approximately 2 r and nest entrances
should be overdispersed if avoiding redundant search is the under-
lying cause of polydomy. This appears to be the case in Lasius
neoniger. Each nest is composed of a series of nest entrances which
are overdispersed within a colony (Traniello 1980). L. neoniger is
unable to retrieve prey effectively further than approximately 15 cm
from any given nest opening due to interference from other species

1981]

Levings & Tranie/lo — Territoriality in Ants

275

or congeners (Traniello, 1980). Inter-opening distances are not
statistically different from 30 cm in a set of 12 nests with varying
numbers of nest openings (P > 0.10, t test, 11/12 cases; range 2-27
nest entrances), fitting our predictions quite well. The only nest with
consistently closer inter-opening spacing was hemmed in by 3 larger
nests; its openings occupied essentially the entire available area (18
cm between entrances, 4 entrances). Although this species fits our
predictions, we are unable to test them further with other species,
either within species between nest openings or between separate
nests. Nest entrance patterns of Paltothyreus tarsatus, which is also
a polydomous species, appear to be similar in function to those of L.
neoniger (Holldobler, personal communication). However, in poly-
domous species of Camponotus, Atta and Pheidole, nest entrances
are often much less than 2 r apart (Yasuno 1964a; Holldobler and
Moglich 1980). Therefore, the association between foraging ecology
and nest structure probably depends on the details of the biology of
individual species.

When resources persist for slightly longer time periods (patches
that can be exploited in a few days such as rotting fruit), we also
expect overdispersion of nests. This pattern has been confirmed in
several species. Myrmecocystus mimicus nests in desert areas and
exploits patchy, unpredictable concentrations of termites which
form a major part of its diet (Table 1, Holldobler 1976b, 1979a,
Holldobler and Lumsden 1980). During the retrieval of these
patches of food, a nest will defend the area by engaging surrounding
nests in a complex ritualized display and battle (“tournamenting”)
which may result in the destruction of incipient colonies. Normally,
the tournamenting behavior persists until the patch is exploited;
searching in the area continues during this time. Nests are overdis-
persed (G. Alpert, personal communication). Nests of Prenolepsis
imparis are overdispersed (Table 1), and workers defend pieces of
fruit for 1 or more days. This species has also been observed to
tournament as Myrmecocystus mimicus does (Traniello, unpub-
lished observations). It appears that in these species the cost of
allocating a portion of the worker force to engage foragers from a
neighboring nest in tournaments that prevent their access to a
resource is less than the benefits obtained from these patchily
distributed food sources (Holldobler and Lumsden 1980).
b. persistent resources

Persistent resources vary in their importance to colonies, depend-

276

Psyche

[Vol. 88

ing on their nutritional value, and can differ from a branch with a
few aphids to a large homopteran population which provides most
of a colony’s food intake. The degree to which a colony depends
upon persistent resources will approximately determine the intensity
of their defense. Formica fusca tends only a very few aphids and can
be chased from them relatively easily (Brian 1955), while F. rufa
colonies regularly destroy each other in battles for the control of
specific trees (Elton 1932, Skinner 1980). Therefore, the removal of
persistent resources can affect colonies differently; some nests will
die if they are deprived of them (Elton 1932). In this section we will
only consider species which are dependent, at least in part, upon
such resources (Groups II & III, in part, Oster and Wilson 1978).

Most studies on the defense of persistent resources concern the
genera Formica and Pogonomyrmex. The patterns of their nest
distribution depends upon colony structure, nest site requirements
and habitat complexity. Most Formica nest distributions are the
result of the interaction between the need for high insolation of the
nest and the proximity of trees or bushes which are suitable for
tending aphids. Many species nest along the ecotones between fields
and forests, in forests, and in forest clearings (F. lugubris, F.
schaufussi, F. exsectoides, F. polyctena, F. rufa, F. ulkei). These
species will be found in overdispersed arrays only if habitat patches
are found in rather predictable patterns. These are clearly special
cases and explain some of the variation between authors for some
species (see for example, F. lugubris, Table 1). We expect the linear
distance along the ecotone to be relatively even in this case, but we
have no data to test this hypothesis. Casual observations on F.
schaufussi tend to support this (Traniello, personal observations).
Formica species which nest in fields should be found in overdis-
persed arrays; the few reports that exist indicate that they are (F.
uralensis, F. opaciventris, F. fusca, F. pratensis, Table 1). In
addition, Pogonomyrmex species which defend patches of seeds are
found in overdispersed arrays. These and other species that defend
persistent resources and are not nest site limited are in general found
in overdispersed arrays ( Atta spp., Acromyrmex octospinosus,
Lasius niger, etc., Table 1).

Colonies which depend upon persistent resources frequently
organize resource defense and utilization with trunk trails. Trunk
trails are long term routes which are marked with persistent trail

1981]

Levings & Traniello — Territoriality in Ants

277

pheromones (Holldobler, 1974, 1976a; Traniello, 1980; Group III,
Oster and Wilson, 1978). Thus both the track to the resource and
the resource itself constitute the defended area. These foraging
ranges are highly asymmetric — foragers from different colonies are
only likely to interact when trail systems overlap. Essentially all
foragers follow these trails in some species (Holldobler, 1976a), but
this varies a great deal from group to group. In general, we expect
that nest to nest distances will be shorter than the distance to the
defended resource if colonies have highly skewed foraging. This
prediction is born out in a study of three sympatric species of
Pogonomyrmex (Holldobler, 1976a). Between nest distances are
shorter in the two interspecifically defending species which forage
on trunk trails than between nests of the individually foraging P.
maricopa.

3. Defense of area

We consider defense of space larger in area than nest yards or
core areas (Holldobler 1976a) to be true territoriality. This defense
of area is, in essence, defense of potential foraging grounds. Only a
few ant species, characterized by complex mechanisms of mass
recruitment and high levels of intra- and interspecific aggression, are
therefore truly territorial in our classification. Most dominant
tropical canopy ants (some members of the genera Azteca, Oeco-
phylla, Crematogaster, Camponotus, Monads, Polyrachis, Anoplo-
lepis, Table 1) and at least one member of the genus Solenopsis are
truly territorial. We must point out that in some cases the distinc-
tion between true territoriality and resource defense is not perfectly
clear, and that strategies of territorial defense and resource defense
are at times difficult to distinguish.

Solenopsis invicta, an introduced species from South America,
has been extensively studied in the southern United States where it
may form monocultures in fields (Wilson et al. 1972). Extensive
mapping of one population showed overdispersion of nests main-
tained over time despite frequent nest movement (Eisenberg 1972,
Table 1).

Maps of intercolony dispersion have been published for a number
of ant species in tree crops in tropical areas (Table 1). Individual
colonies hold territories in the canopy both intra- and interspecifi-
cally. The distribution of the canopy mosaic of dominants can have
a very complex structure (Way 1953; Greenslade 1971; Majer

278

Psyche

[Vol. 88

1976a,b). Individual colonies are clearly separated from each other,
frequently by a no ant’s land between defended areas (Holldobler
1979b).

However, the statistical dispersion of these colonies is difficult to
assess. Territory size varies a great deal between species because
population structure has very strong effects on colony size and
organization. Only a few polydomous, polygynous colonies may
occupy extensive areas (Steyn 1954, Greenslade 1971, Leston 1973,
Majer 1976a, b). The dispersion of volumes in space is difficult to
treat statistically from published data although 3-dimensional meth-
ods exist (Clark and Evans 1979, Simberloff 1979). Luckily, the
biological evidence for dispersion and nonoverlap of area is over-
whelming. Territorial battles are commonly observed and, in popu-
lations followed over a number of years, control of a given area
shifts from colony to colony and species to species. Although we
predict statistical overdispersion, we are unable, for both statistical
and biological reasons, to test for it in these cases. There is,
however, no question about the existence of true territorial defense
and the spatial separation of colonies.

Ant plants are a special set of cases of true territoriality. Several
tropical tree species (Table 1) are coevolved with certain species of
ants (some members of the genera Pseudomyrmex, Azteca or
Pachysima ) which protect the tree from herbivores or overgrowth in
return for food and nest sites. Few other animals of any species are
tolerated on the plant; the ant species are characteristically extreme-
ly aggressive. The mutualism is sufficiently old than at least one
species parasitizes it by using the plant without protecting it in
return (Janzen 1975). These ant colonies are thus distributed with
respect to the distribution of their host and form distinct territories
within the canopy mosaic. They may also help “grow” new host
plants by affecting the survival of nearby seedlings (Janzen 1967,
1973).

Intercolony Spacing Effects

Interspecific overdispersion is regularly reported in almost all
habitats (Table 1). However, detailed ecological studies indicate that
different mechanisms operate in different habitats. In part, this is
due to the fact that the only necessary characteristic required to
generate overdispersed arrays is the ability of a colony to interfere

1981]

Levings & Traniello — Territoriality in Ants

279

with colony foundation of potential competitors. Few studies have
examined the pattern of species mingling. Brian and his co-workers
have shown that nest sites probably limit many species in England,
which apparently is a rather poor habitat for ants. The pattern of
dominance over nest sites determines the location and abundance of
many species (Brian 1952, 1956b, Brian et ah 1965). Nest density
could be increased by providing new nest sites, and once established,
populations remained relatively stable over long periods (Brian et
al. 1966). Competition between species where nest sites were not as
limiting tended to restrict individual species to areas which were
close to optimal species requirements (Brian et al. 1966, Elmes 1971,
1974). These studies indicate that the details of species biology and
physical tolerances may be critical, even in very simple habitats like
heath. However, even in these systems, species are definitely not
distributed independently of their competitors.

Tropical canopy dominants are associated with certain canopy
conditions, and tend to be found mostly in shade or under certain
other limited environmental states (Majer 1976a). Removal experi-
ments indicate that colony foraging areas are competitively com-
pressed; when a dominant is removed, surrounding colonies expand
to fill the available space. Species usually found in one kind of
canopy may expand into other types of foliage if adjacent domi-
nants are extirpated (Majer 1976a,b). This pattern is similar to that
found in far simpler grassland communities.

In a complex tropical ground ant community with at least 16
ecologically similar species, Levings and Franks (1982) have shown
that new nests are not added at random to the nest array. Grouping
all species, nests are overdispersed from each other. Each common
species considered independently was also overdispersed. This is
interpreted as evidence that species are interacting more strongly
intra- than interspecifically, but that interspecific effects were still
important in determining nest distributions. Similar patterns in
simpler communities indicate that this may be common (Table 1).
The worst neighbor in a competitive sense should be identical to
oneself. In any case, the simplifying assumption that species have
identical requirements is almost infinitely unlikely to apply; even
small differences in requirements or tolerances can be important in
determining colony distributions. However, few adequate tests have
been done and, in one published case, two closely related congeners,

280

Psyche

[Vol. 88

Pogonomyrmex rugosus and P. barbatus, essentially act like exact
ecological equivalents (Holldobler 1976a). Davidson (1977a) has
suggested that the distribution of several individually foraging
Pogonomyrmex (maricopa, californicus, desertorum and magni-
canthus) is consistent with the hypothesis that they replace each
other between habitats. The pattern could also occur in some other,
less completely documented cases, perhaps in Atta (Rockwood
1973).

Population Structure and Its Effects on the Spatial
Distribution of Colonies

Monogynous, queenright colonies are almost innevitably aggres-
sive to conspecific nests or foragers, regardless of how territorial
they are (Holldobler and Wilson 1977c). Polygynous colonies may
or may not display internest aggression. Holldobler and Wilson
(1977c) point out the importance of queen number in the mainte-
nance of clear territorial borders. Species which commonly have
polygynous colonies and/or those which adopt newly fertilized
females to augment or replace females already in the nest do not
always have strong intraspecific interactions; some do not form
distinct colonies ( Formica yessensis, F. lugubris, Table 1). In these
cases the location of nests should be predominantly determined by
ecological factors, in particular the kind of resource defense the
colony shows. Thus some species should retain overdispersed
patterns of nest distribution while other show clumped or random
patterns (see model and predictions).

Examining this issue is complicated by the lack of population
structure data for many species. Several Formica species which
form unicolonial populations, but depend upon randomly and
unpredictably distributed resources, are found in overdispersed
arrays [those species found in fields: F. pratensis (provisionally), F.
uralensis (provisionally), F. opaciventris, F. exsectoides, Table 1];
those which nest along the margins of a habitat and/or which
defend persistent resources tend to have more random or clumped
distributions ( F . ulkei, F. rufa, F. lugubris, Table 1). Territory size
in some tropical tree ants is partially a result of population
structure. Many dominant species are polygynous and are able to
expand their territories almost indefinitely under good ecological
conditions (Greenslade 1971, Majer 1976a,b). In some cases, single

1981]

Levings & Traniello — Territoriality in Ants

281

queen species like Oecophylla may be at a disadvantage. Resistance
to invasion or persistance of the nest may be limited by the female’s
egg production under some conditions, although this does not
usually seem to be the case (Holldobler and Wilson 1977c; Holl-
dobler and Lumsden 1980). We must emphasize that in populations
with complex or variable structure it may be very difficult to
determine the factors which are controlling distributions. Spacing
may reflect foraging ecology as well as being an aspect of territorial-
ity. More data are needed before good generalizations can be made.

Behavioral and Ecological Aspects of Spacing

For the cases we have been able to examine statistically, 67 out of
80 show overdispersed nest distributions or tend toward overdisper-
sed nest distributions. The other 80 cases, which cannot be treated
statistically, mainly have either overdispersed nest distributions or
tend toward overdispersed nest distributions. Thus the majority of
species studied tend to have regular nest arrays. This pattern holds
despite the large number of species, food types and habitats
considered. Species which defend only their nests are too rare to
consider in our sample.

Our basic assumption is that no colony can become established or
forage within some radius r of another colony. There is a biological
basis for this assumption in the patterns of interference with colony
establishment and foraging patterns. Therefore, to understand nest
spacing it is important to understand the different levels of competi-
tion in ant communities. Fertilized females or incipient colonies are
usually destroyed when they are encountered by foragers from
established colonies (Wilson 1971). The specificity of this behavior
varies between species depending in part on population structure
(Holldobler and Wilson 1977c, DeVroey 1979). There is some
evidence that workers are more likely to attack females from
conspecific nests or closely related species, especially in monogy-
nous, queenright colonies, as has been shown in Pogonomyrmex
(Holldobler 1976a) and Myrmecocystus (Holldobler, personal com-
munication). The studies of Pontin (1960) and others (reviewed in
Wilson 1971) suggest that such behavior is more often directed
toward queens of the same species as the attacking workers.

Another factor which may operate during this period is resource
depletion mediated by either direct interference or exploitation.

282

Psyche

[Vol. 88

Within the foraging radius of an established colony, there is likely to
be less food available, even if the established colony ignores
incipient colonies. The amount of depletion will depend on the
amount of resource overlap. Because destruction of females and
incipient colonies is frequently reported and resource depletion
probably also affects colony persistence, the chance of a small
colony becoming established is low. Wilson (1971) estimates that
only 0.01% of all fertilized females survive to found successful nests.
Therefore, established colonies tend to persist and interact over long
periods, insofar as is known (Wilson 1971). Given this pattern, what
is the form of the interaction and why are patterns of interspecific
overdispersion so common?

According to current theory, species can segregate a habitat to
avoid or lessen competition in several ways: microhabit partition-
ing, food size or type, and activity period (Pianka 1978). Further,
equilibrium theory generally asserts that only a limited amount of
overlap is tolerated on any given niche axis (Mac Arthur and Levins
1967; Colwell and Futuyma 1971). Species which are too similar
should not be able to coexist and, over a long enough period, the
superior competitor in the overlapping pair will drive the other
species extinct. Although there are many problems with the assump-
tions of this argument, we will use its basic divisions to examine the
patterns of overlap between co-occurring ant species. Ant species
may be specialized along these three major axes. We will consider
each potential kind of specialization in turn and evaluate the
evidence that segregation of species along that factor is usually
sufficient to prevent strong competitive interactions.

Species may be considered specialized on food types in 3 major
ways: (1) restricted prey types (i.e., only centipedes), (2) specific size
ranges of prey (i.e., only prey 1-3 mm in length) or (3) some
combination of (1) and (2) (i.e., centipedes between 5 and 8 mm
long). Different kinds of specializations will have different effects on
colony structure, nest size and foraging strategy. Resource restric-
tion is frequently based on the matching of mandible or head size to
food size or type (the trophic appendage, Schoener 1971, see below).
Resources which are retrieved by individual workers, not by
coordinated action, are especially likely to be treated in this manner
(for example, seeds for desert ants, collembolans for dacetines). The
resistance of the food item to recovery is also important; items

1981]

Levings & Traniello — Territoriality in Ants

283

which do not resist (seeds) are more likely to be size matched than
items which require more complex treatment from the ants. Nests of
specialist species may be restricted to areas which contain concen-
tration of suitable prey (and as such violate the assumptions of our
model). If resource size is matched with worker size class, then size
polymorphism is one way to expand the resource spectrum of the
colony without any changes in individual retrieval patterns (Oster
and Wilson 1978). The development of coordinated retrieval mech-
anisms can further expand the accessible resource spectrum.

Almost all specialists, by definition, have less harvestable energy
available to them than generalists. Thorne and Sebens (1981)
suggest that species with low habitat quality (i.e., low food density)
will have smaller nests than species with high quality areas (high
food density). We extend this argument to predict that once a
species has broadened its diet, it will include essentially all retriev-
able food types encountered. Such an increase in diet breadth is
needed to support large colony sizes, based on almost any simple
foraging efficiency model. Although specific prey types, especially
those with noxious chemical defenses, require special handling
methods, many prey types may be captured and/or retrieved by
species with a limited behavioral repertoire. Certainly scavenged
material can be handled by all but the most specialized mandibular
types. Since ant colonies persist over years, they more or less
continually require resources. Resource distributions are highly
variable over time; prey types appear and disappear seasonally
(Mabelis 1979; Levings and Windsor 1982). It is a general con-
sequence of this that once a species generalizes its diet, it is likely to
overlap strongly with one or more sympatric species. The value of
large colony size is reflected in reproductive output. Numbers of
reproductives usually increase with colony size to some upper limit
(Wilson 1971). Since the chance of success for any given reproduc-
tive is low, high production will be likely to correlate with the largest
probability of leaving successful offspring. Colonies which bud will
tend to have higher rates of success if the new buds have large
worker forces; this is also a function of energy intake. Colonies
almost all require protein to raise brood (usually from insect prey or
seeds) and many accept or require sugar to maintain adult workers
(usually from Homoptera, fruit or nectaries). In general, large
colony size is strongly associated with the maintenance of sugar

284

Psyche

[Vol. 88

resources. It appears that when adults can be fueled from sugar,
more intensive foraging is possible and more brood and workers can
be supported (Greenslade 1971; Leston 1973).

We do not deny that species which are specialized on prey types
evolved resource segregation from competitive pressure. In fact,
among specialized species which forage individually for prey, we
expect some equilibrium co-existence theory to apply (see for
example, Davidson 1980). We assert that there is no support for the
contention that generalists segregate the resource spectrum to
reduce competition (Wilson 1971). Available empirical studies
indicate that high or essentially complete overlap in food type is
frequent (Brian 1956a, b; Pontin 1961, 1963, 1969; Yasuno 1964a, b);
Abe 1971; Holldobler 1976a; Levieux 1977, Levings and Franks
1982). At best, partitioning of food type can account for only a
small part of the observed pattern of species distribution in most
habitats.

Habitat partitioning is a second possible method of limiting
competitive interactions. Even in simple temperate grassland com-
munities, co-occurring species forage at slightly different heights in
the grass or tend to move more or less beneath the surface (Brian
1952, 1955, 1956b; Brian et al. 1966). However, all these species are
usually described as being interspeciflcally territorial. Tropical
faunas are well divided into arboreal and terrestrial components;
many specialized species are further restricted to logs, rotting leaves
or other microhabitats (Wilson 1959b, 1971; Carroll and Janzen
1973). Within these strata, high overlap between species resulting in
intra- and interspecific aggression is frequently described (Carroll
and Janzen 1973; Leston 1973; Greenslade 1975; Room 1975a, b).

Faunas may be further subdivided by time of foraging, if by
foraging at different times, different resources are harvested. Time
of foraging can differ daily (nocturnal vs. diurnal, Carroll and
Janzen 1973), seasonally ( Prenolepis imparts which forages in early
spring and late fall, Talbot 1943, Lynch et al ., 1980) or may track
environmental cues, such as desert species that forage after rains
(Bernstein 1974, 1979). Although it has not been proven, it is
probable that generalist and scavenging species forage on different
resources if they forage at different times, if there are temporal
components to food availability. Most dead or readily captured
prey do not remain available for long periods, few probably persist
even hours (Carroll and Janzen 1973, Culver 1974, Traniello 1980).

1981]

Levings & Traniello — Territoriality in Ants

285

Other resources may be similarly affected — for example, winds may
cover and uncover seeds in the desert (Reichman 1979). The option
to forage at different times is not uniformly available to ant species;
thermal tolerances may severely limit foraging time in both cold and
hot climates or may affect the outcome of foraging contests (Hunt
1974; Davidson 1977a, b,; Holldobler and Moglich 1980). Many
species change the time of their foraging in the presence of
competitors (Hunt 1974, Swain 1977). Thus foraging times may
separate some species, but as in the case of food or habitat, high
overlap between sets of sympatric species in foraging time is the
norm, not the exception, in ant communities. The evolution of
intra- and interspecific behaviors incuding complex patterns of food
retrieval and defensive strategies has resulted from such high
overlap.

The form and outcome of interactions between species is de-
termined in large part by the mechanisms of recruitment and
communication within species. The subtleties of recruitment com-
munication and their effects on foraging ecology and interference
competition are not appreciated by most ecologists. Behavioral
mechanisms are so critical that we suggest that when examining the
diet of a species, an investigator first ask why more items are not
included. For many years harvester ants were considered to forage
individually for seeds, but the field and laboratory studies of
Holldobler (1976a), Holldobler and Wilson (1970) and Holldobler
et al. (1978) unequivocally demonstrated that species of Pogon-
omyrmex and Novomessor rely on a sophisticated array of recruit-
ment behaviors in foraging. In Novmessor cockerelli, short-range
recruitment, mediated by both chemical and vibrational signals,
allows workers to move food sources quickly and thereby enables
them to compete with sympatric species (Holldobler et al. 1978;
Markl and Holldobler 1978).

Behavioral interactions, not food choice, seem to partition food
resources among generalists. Protein foods (arthropod prey) tend to
be highly unpredictable in time, space and size; adaptations to this
resource distribution among generalists may be behavioral rather
than morphological. Monomorium pharonis and Solenopsis fugax
employ a chemical interference technique both defensively and
offensively during foraging (Holldobler 1973). Adams and Traniello
(1981) have documented the ecological effects of recruitment and
resource defense in Monomorium minimum, a north temperate

286

Psyche

[Vol. 88

open field ant. Monomorium minimum is a small (head width 0.47
mm), monomorphic species. Workers are successful at retrieving
food particles which are either extremely small (less than 0.5 mg in
weight) or large (greater than 450 mg in weight). Most items of
intermediate size are lost due to either exploitative or interference
competition from other species. Detailed laboratory and field
experiments on the organization of foraging showed that M.
minimum recruits other workers to food sources using trail pher-
omones. The quantity of pheromone determines the foraging re-
sponse of the colony. As trail pheromone concentration increases,
more workers are recruited. The amount of trail pheromone
deposited is dependent upon resource quality (in this case, measured
by the investigators as weight). Large food items induce trail laying
by many workers and therefore result in strong recruitment. If there
is interference from another species while prey is being dissected,
workers display a specific posture (gaster flagging) while extruding
the sting and discharging a droplet of poison gland secretion. This
secretion has a repellent effect on intruding ants and causes them to
recoil and vigorously groom. The effectiveness of this defensive
behavior is dependent on the number of workers recruited. There-
fore, large prey, which elicit strong trail pheromone deposition,
induce strong recruitment responses and this results in a worker
force which can successfully defend the item during retrieval. The
result of this feedback between prey size, pheromone concentration
and colony response is a diet composed of small individually
retrieved items and large items recovered by recruitment and
successful defense.

Perhaps the best evidence for the importance of behavioral
parameters in species interactions is the phenomenon of alarm
specification. Certain ant species which interact strongly with other
species may respond specifically to the presence of the competitor.
The best studied case is that of Pheidole dentata and Solenopsis
geminata (Wilson 1975). Pheidole dentata colonies respond to the
presence of Solenopsis by a strong recruitment of major workers.
Major workers proceed to attack and kill all Solenopsis encountered
and to search thoroughly the area near where the Solenopsis
workers were found. They do not respond to the odors or presence
of other species with major worker recruitment. A similar pattern of
response is indicated in the interactions between Oecophylla longi-

1981]

Levings & Traniello — Territoriality in Ants

287

noda and Camponotus sp. in Kenyan forest; alarm/ recruitment
specification may be the behavioral mechanism responsible for the
structure and maintenance of the tropical canopy ant mosaic (Holl-
dobler 1979b).

In general, to defend any resource or area, including the nest, an
ant must be able to summon her nestmates to a particular location.
Within the nest, even quite primitive ants are able to communicate
alarm and attract reinforcements (Robertson 1971; Traniello un-
published data). Outside the nest, recruitment is a necessary com-
ponent of effective defense.

Ant species possess a wide diversity of recruitment communica-
tion techniques that are ecologically significant (see review by
Holldobler 1977). It is important to understand the ethology of
social design to comprehend its role in ecological interaction. There
are definite phylogenetic constraints and/or trends in the form of
recruitment communication within the various subfamilies of ants
(Holldobler 1977). More primitive groups (some Ponerinae) usually
recruit few workers to food sources; some group raiding species are
exceptions. Mass recruitment is characteristic of some groups of
Myrmicinae, Dolichoderinae and Formicinae. Each lineage has
developed within certain paths involving specific glandular, physical
and behavioral trends. These pathways are important in considering
the evolution and development of ant community structure.

Summary and Conclusions

We have argued that a very simple hypothesis is sufficient to
generate predictions of spatial distributions of colonies under a
variety of ecological settings. The majority of cases in the literature
(Table 1) support the hypothesis that most ant species are regularly
distributed with respect to conspecifics and other co-occurring
species. We assert that this is a natural outcome of high overlap in
food utilization in many species, and in particular, among general-
ists. We have suggested that departures from expected spatial
patterns be used as a measure of competition between species, but
too little information on colony foraging radii in relation to spacing
patterns exists to test our hypotheses critically. More field measure-
ments of colony foraging distances in relation to intercolony spacing
are needed. Measurements of potential foraging distances when

Nest Defense Forager Nest

Species site1 type2 type3 spacing4 Evidence5 Source

288

Psyche

[Vol. 88

s ? g

w SO
- rX Crt

C X} a >

.2 25 g

CQ W

<N VO

u-> ^ On
ON —

^ NO

C JQ
.2 2
'C
CQ

.£ o
(6

60 3
cd X

O e

S 4> «

£ o

O • -

E 1
♦2 o
8 ^
z

S? g ts

cd X • —

^ -C *«

Cd
O X

c *2

S *S

2 ‘o

1 ^ ^

T3 <u

.s § *

T3 t«

<u .£ C

& £ S>

& o «

,§!3

« 2 «

j? <u

O H

<L>

00"

a -S

a .5

.»3

o 2

53 "E*

•Si g

■g 2

.2 C

.Si C

S -o

C Q

5 1

Nj C
1 1

•JC

X g

X %

X -2

■j

O S

acervorum distribution 1956a)

Tetramorium A b,c?,e?,f? II o/+ Morisita’s index, tend to Baroni-Urbani (1969)

caespitum overdispersion

Tetramorium A,B b,c,e,f II o/+ Intra- and interspecifically Brian et al. (1966)

caespitum territorial

Lasius A,B b,c,e,f II o/+ R = 1.22-1.27, dense popula- Waloff and Blackith

flavus tion overdispersed, low dens- (1962)

ity population tends to
randomness

1981]

Levings & Traniello — Territoriality in Ants

289

, ,

• p

•p

>w'

C -2
> ^

<L>

w 1

•o

T3 3

'-3

<L>

n-4

3 lH

|?3 . o

'53

C/3

'vi

1/3

C/3

CL>

3 00

c/3

<L)

U g

’5

c/3

fl)

<L>

C/3

£ ts

C/3

O (U

■3

c/3

C/3

•3

° 3

A S

C/3

•3

■3

C/3

3 >

'g ^

•3 s

3 >

g ^

3 >

•3 ^

3 0$

g 5

S5 <L

3 Oc

g S

Lasius A b,c?,e?,f? II o/+ Morisita’s index, tend to Baroni-Urbani (1969)

alienus overdispersion

Lasius A,B b,c,e,f II o/+ Intra- and interspecifically Brian et al. (1966)

alienus territorial

Nest Defense Forager Nest

Species site1 type2 type3 spacing4 Evidence5 Source

290

Psyche

[Vol. 88

</-) so

m so

Os Os

rsi

>rs

£ &
2 «
w o

s 5

■§ i

(50

1956a

t-l

«3

2 ®
13 o

> -3
2 5

2 o

■S ftS

> -s

« N
2 o

« f*

1 s

•g £

«3

ffl

.5 _o

£ 2
I 1

o O

O >
M O

fc „
« >

X) ”D

•C o g

.2 ^§8
T3 o «

„ >
T3 ‘35

(L> CO

*2 D
3 i_
u 00
5 60
CL c«
o
C/2

CL 2
c « co
3 <u

.2 0) O)

> 0)

T3 ‘5S £

0) C/5 *■>

o o)

cd ln x

u. GO

(< GO «

CL & g

CO £

■ (fl

* •§

T3 3
d> O

3 6

o .2

O C)

>s g.

g S3
a "S 3
SP o o

<u 0)

o o'

X x"

■Si

•2

3 :§

•2

Jk.

•2

<L>

ft,

<L>

e<5

5 *§

«2

Jk.

<L)

t£

l. JS

Jk.

ft,

1981]

Levings & Traniello — Territoriality in Ants

291

r~ o

ON 3

O 6

3 "O

1 JS

_ vi

3 00

"* X
o

v~i ©

m oo

ON ON

■a

i.s

3 <D

a* o-

•5S-

I S3 .3

o £ x>

.3 <D

* . &

T3 5

« 3

i3 E

H «5
§•8
if I- §>

(u c

■t ’5
cs 6 si
?? o o

.2 x)

oo a,

00 Vi
3 ^

w *-

Vi O

3 O

•<9 3

O 14)

o =

•2

0£

§. a

’o

<4)

!■*

*->

«>5

CXh

lugubris smaller colonies, all (1977)

mutually aggressive

Formica A b,c,e III + R = 1,32, p <C 0.001 Breen (1979)

lugubris

OLD WORLD I

Nest Defense Forager Nest

Species site1 type2 type3 spacing4 Evidence5 Source

292

Psyche

[Vol. 88

Myrmicinae

Messor A b?,c?,e?,f II? o/+? Nonoverlap of foraging trails. Pickles (1944)

barbatus interspecific aggression

Messor A b?,c?,e?,f II? o/+? Nonoverlap of foraging trails, Pickles (1944)

aegypticus interspecific aggression

1981]

Levings & Traniello — Territoriality in Ants

293

. - cr>

vo 2
r- w

—i e
W o

'e? J

. „ m

? 2
ON ^

? 2

—

2 c

w o

. „ r*->

S 2

<U D

•J? J

£ ^
£ 13

u 3

'a,

'O.

’o,

c/5

'Oh

C/5

t-H

"c/5

C/5

c/5

<L>

(S3

•2

& 2

Si -2

a 2

O 3

1 ^
£

a .a

5 S

a -

§ 'C
S -2

castanea aggressive

Pheidole A,C b,c,d,e,f II o/+? Nonoverlapping territories, Greenslade (1971)

megacephala also interspecifically

aggressive

isjiiiiUtMiii |ii|lj||i |i 1« li Ihi

D b,c,d,e,f II, III? o/+?
D b,c,d,e,f 11,111? o /+?
C b.c.d.c.f 11,111? o/+?

A,C b,c,d,e,f II o/+?

ft*

Nest Defense Forager Nest

Species site1 type2 type3 spacing4 Evidence5 Source

294

Psyche

[Vol. 88

Camponotus A b,c,d,e,f II? o/+? Nonoverlapping territories, Majer (1976a);

acvapimensis also interspecifically Leston (1973)

aggressive

Polyrachis ? b,c,d,e,f II, III? o/+? Intra- and interspecifically Steyn (1954)

schitacea aggressive

1981] Levings & Traniello — Territoriality in Ants 295

Nest Defense Forager Nest

Species site1 type2 type3 spacing4 Evidence5 Source

296

Psyche

[Vol. 88

Formica A b?,c,e,f? III? o Mounds located along the edges Scherba (1958);

ulkei of fields Talbot (1961)

Formica A b?,c,e,f? Ill o Mounds located along the edges Nipson (1978)

exsectoides of fields

Lasius A b,c,e,f III + Nest craters overdispersed Traniello

1981]

Levings & Traniello — Territoriality in Ants 297

fulva

Stenamma A b,e,f? ? o/+(!) V/M = 0.79, N.S. Headley (1952)

brevicorne

jijliiliilli jfifl W i* j jl^l1 j ih'M

V/M = 0.85, N.S.

S’

Nest Defense Forager Nest

Species site1 type2 type3 spacing4 Evidence5 Source

298

Psyche

[Vol. 88

rl •§
r-~ S

W >>

>.= 00

73 s

cd CQ

t-i

3 _

e 2

cd w

s' <8

cd T3

S "O

S C
8 *
S* •a

s S

cd ^

iS ^3
S C

8 «*
a -o

•— .2:

c g

• DO

2 5510

s s

cd -a

£ -s

S-.-2

5 2

- ¥ 3

DO cd

eo <u
cd C

«5 P

s S

s e

O o
\ n i:

c/3 ^

Si ^

DO 2
00 >
cd cd

S c

Oh ^3

,S2 T3 ^
^ X C/3
TD (D

£ ” *
° o o

C c/3
cd z, «
cd X)

^ On Cd

on .ti

E 2

» o £

2 I

O V.

X O

2 I
it

rr Do

2 c
S.O
4

* w
5 §
2 .1

■T

<U Co

r* h*

<3 ^

•2 o

Formica A b,c,e,f III +? Intraspecific aggression; Bradley (1972, 1973);

obscuripes nest movement away from (1968)

competitors Bradley and Hinks

(1968)

Prenolepis A b,e,f III o/+(?) V/M = 0.54, N.S. Headley (1952)

imparis

Prenolepis A b,e,f III o/+? Nests spaced out under fruit Talbot (1943);

imparis trees, aggression over food Lynch et al. (1980)

sources

1981]

Levings & Traniello — Territoriality in Ants

299

s •§

•1 &
5 s

° e

T3 -^3

° 2
T3

2 °

1 1
3

5 1
o
c

o C

a, c

I S

■2

<4<

a, 2

•a "5

5 a,

«U a

.2 1

i *

a §

oo a

Solenopsis A b?,c?e?,f? II? o R = 0.6-2. 1 Bernstein and Gobbel

xyloni (1979)

Nest Defense Forager Nest

Species site1 type2 type3 spacing4 Evidence3

Intraspecific aggression, nest
movement away from competit'

Bradley (1972, 1973);
Bradley and Hinks
(1968)

Francoeur and Pepin
(1978)

Bradley (1972, 1973);
(1968)

Nest Defense Forager Nest

Species site1 type2 type3 spacing4 Evidence5 Source

300

Psyche

[Vol. 88

■0

1 1

‘C

’C

C/3

S-i

C/3

’>

ec3

<U On

'C On

a -

- >

O to
§ 8
s 3

T3 is

>> g

•o

Cl.

e<J

O g
cu o

o g
o. .2
„ o
00 u
o’ 03

\© 2
©

II %

« « <u

xT xf xf

9k.

ik.

O 5
§ |

Jk. ._

O ^

§ g

■S

03 ?

<43 r

5 3
^ -a
5 5

2* ^
£ 3

k .a

1 s

S •§

5 |

0 *■»

1 1

§<§.
rs ^

p c

Jk.

ft,

<43

ft,

? p

> 0
£ ^

-s:

CU V

-s: *

1 8

ft.

ft,

ft.

Dgonomyrmex A b,c?,e?,f? II? + R = 1.14 Bernstein and Gobbel

occidentalis (1979)

1981]

Levings & Traniello — Territoriality in Ants

301

£ 2
ON ^
w C

’5

’S3

5— <

’>

C/3

’>

C/3

s_x

5—i

s_^

v ✓

v — /

« *8

C £

x Ck

c .2

cd X

'

£s o
c

D-

<*>

« *8

c 2

'"H <u
X Cl,
C .2

Cd TO
k.

' <u

5 >

£2 O
C

C .2
cd to
■ 
£3 o
e

5 £

o -c
R ?\
O £

£ 5

5 §

5 3

O «T

c* O
§ ^
S’ 2

* 3
5 |

o «

R *£

o 2
r -£

<to

o 2

R *P

<4;

jS 5
**•

5 2
o t:

£* dj

§ 2

fti

k a

b (3
S p
o .2

o a

S> S

X
O
X
o

Q °

onomyrma A b?,c?,f? II? o/ + R = 0.8-1. 2 Bernstein and Gobbel

insana (1979)

Nest Defense Forager Nest

Species site1 type2 type3 spacing4 Evidence5 Source

302

Psyche

[Vol. 88

f> <U

a. c

< 5-§

« K. §
I s !

ON 5

C bQ zl

W = 0

<U M

3Q 2

m Os

a £

Os & Os

p- ~ r-

ON g ON

_c ^

'C On

a> on

I 2

5 o

Oh O
O o

II u

11 </3

S 3
c >»
jd 3

' 1/3

-S O

3 Oh

b .2

g o

!■§

O o

8 |
o

b c

o ^

c a

s s, §. s a

2 ^
g 3

2 .c

1 1

1°

| t

S 5

« .3

.o ^3

I 1

Myrmecocystus A b?,c? II? o/+ R = 0.9-1. 6, depending upon Bernstein and Gobbel

kennedei density (1979)

Myrmecocystus A b,c II + R = 1 .3-1.7 Bernstein and Gobbel

mexicanus (1979)

1981]

Levings & Traniello — Territoriality in Ants

303

S' -2

so 3

r- 3

2 o.

3 tc

O <u

=3 <
PC

-a

C ON

&p oo
c on

w. o

<L>

J5 |

g o
o °
U

4: -2

■g s

o. >-

TO u

i: -a
c

> 1

W w

T3 O
c x>
ed

Ofi

^ --N
t-, <U
ed C

J3 "P

T3 O g
ed £ T3
^ .2? &

•3 "P
C <U

o .-s

3 S

O -J-

O £»
•35 £

K to

2, 3

O §

o ^

& .s>

1 s

1 U

X s

v. ^
^3

I S

<3 3

s g

-§ -§

a a
-s: £•

$■

Pheidole A,C b,c?,e,f II + Clark and Evans’ nearest Levings and Franks

biconstricta neighbor analysis, over- (1981)

group species dispersed

Defense Forager Nest
tyPe: type3 spacing

NEW WORLD Tropical forests: generalists

Clark and Evans* nearest
overdispersed

neighbor analysis, nests
overdispersed

Holldobler (1976b);
Alpert (unpublished)

Levings and Franks
(1981)

Nest Defense Forager Nest

Species site1 type2 type3 spacing4 Evidence5 Source

304

Psyche

[Vol. 88

£4

i-i

O. ^
„ C

•P

r- o

r- m

ON Cd

as cT

as rT

W t

ftp oo

op oo

c ^

.£

> w
<U

£ ^
’> w

C/3

^ c
r-~ o
r- •£
On cd

" t

C as

1 »

C/3

« >
« O

> |

v- T3
T3 O
C X) £
ed X3 <u

2? b*

£ C .2

03 C T3

C/3 1)

Si >

£ o

C as

cd g

> s

w ”

u. T3
"O ° <U
C £ £
cd .c <d

£ u .2

cd C -O

■O >

<P '5

_>> cd
"£ af

o -a

C £
2^
as .
ed >»
£X

*5 C

* S

«* as
as ,a
4> Cm
'o O
D <L>

o. o.

X3 ed ^

1 6 3

*J d T3
ed <D w
.S . g

s « .s

C cj —

0 S! O
^ ■*->

1 s' I

5 ^

X) C

■° o
>» -o

•S cd

O ©

"S °

£ U

£:

C/J

£ S

eu «J

^ rn
£

p ^

1 1

C ?3

.o >

.2 g

S’*

NEW WORLD Tropical forests: army ants

NEW WORLD Tropics: leaf cutters (Attini)

1981] Levings & Traniello — Territoriality in Ants 305

Azteca spp. C b,c,d,e,f III? As Found only on Cordia spp. Janzen (1969)

plants

srljtWi'litj >t{ ||i*

Nest Defense Forager Nest

Species site1 type2 type3 spacing4 Evidence5 Source

306

Psyche

[Vol. 88

-2

.g,

o £
o 73

o o

<+H

d, o

c n 1)
C/3 Q,

<3 E

5 a

8 .s

e 5 'S3
cd c/3
' <u

•a d u

c is M

is s

< jd

’d

■§

cd u

g s 8.

c/3

iii.

^ fc 3 T
8 ^ S S
ft,

•2 o

« &

? 1/5

v r; ^
c J2 +
£ g CJ/

•jc

!§

macrops (in prep.)

Myrmecia A b?,e I + Intraspecifically territorial Muir (1974)

tarsata and overdispersed

Myrmecia A b?,e I + Intraspecifically territorial Muir (1974)

simillima and overdispersed

Myrmecia A b?,e I + Intraspecifically territorial Muir (1974)

decipiens and overdispersed

1981]

Levings & Traniello — Territoriality in Ants

307

c m

ca in

S 2

— ' C/5

ca • —
o TJ
Eh

'o >
x> O
cx

c/5 "O

nj c

rz o

ca T3

C/5

ac ci;
ca ^

~ *o g-
<u c t3

s i g

S £ o
c

_c« x)
S ca

C t_

X3 O

C/5

'•3 2

e -
E CX

S £

E ’S

S -o

s t

> g

[jq ca

T3 O
E X)
aj X

^ .SP

t- X)

ca c

c/5 ‘5b ^

60 2 X)
C o X)

•- V C/5

e e

S. g> 8.

° leg

c/3 O Jr;
<L> O S>
c >

u u °

<« O

X) T3 O

u ea
ea ex

E c/5

ca c ca

CX 3 X)

<U XX E

c« o ca

.o -2

S S
s -a

X 5,

fc 5»

-2 -S;
^3 ^3

^ 3

£ S-

H
! =

! t

Nest Defense Forager Nest

Species site1 type2 type3 spacing4 Evidence5

308

Psyche

[Vol. 88

-rt >-*

« O

c/3 O

1 e

X 5

o ^

S O

4> O
2 O
u C

& a

w £

« to
<u

s £

O 4>

;r! c/3

^ 3

4> i— i

Oh 2

C/2 1—1

*3 £

<L) ‘-3
Ch 3

G C
o ca
o

. „ to
4)

o 2
" o

-T-l CO
^ 4)

O i-
O

£ s

w 4)

^ 2
"3 c/3

> t-i
<3 4>

O CU
X .
4) CJ

Uh _ „

O c/3

<+-i 4>

4> O

Oh 3

t 2

2 o

C/3

- £

03 !>

a x.

C/3 C

OJO

s § !

4> Jr
c o

° X °

.-a ,u c
3 5t3 o
o •§ g

^ 6 o,

_J~ W ' C/3

60 >
3 O

Oh C/3

3 X
O 3

l~ QJ
60 -*

I 5 o

* > i

4> O

2 J3

g .3 $ -S

.3 &3 ici Chj

•a -3 "3 -3 ^

6 £ o, o .o

8 £ $

s .3 8

§ J

C/3

<L>

4) fv 4)
3^1-

3 — 4>

O 3 C4

’* .a .2

O H-I X

£ ^3 £

t-i

c/3

&■» >
3 3°

-3

C*H

^ ° +

C/3

§.j

u cd r

‘0

O ’o 4)

5 0.
2 g<£

4/^3

C/3

60 - —

2 >

C/3

<2 ~ r

° c«
X «

c -o
5

O C/3
I- 4»
60 C

c«

- | * 6

. « .’2 2 c

4> > *- X

C X ^ «
,4> C 3 t-

5V/M, variance to mean ratio; !, our measurements or statistics based on published data; R, Clark and Evans’ nearest
neighbor analysis, R = 1, random dispersion, R > 1, overdispersed, R < 1, clumped
6Mabelis (1979) studied this species in dunes; it is usually found in woodland.

1981]

Levings & Traniel/o — Territoriality in Ants

309

competitors are removed can provide additional evidence for com-
petitive effects on spacing and foraging patterns.

Acknowledgements

This paper is a synthesis of portions of the doctoral dissertations
of both authors. We would like to thank the following people for
their assistance and helpful criticism: G. Alpert, F.M, Carpenter,
N.R. Franks, S.D. Garrity, B. Holldobler, R. Levins, R. Lewontin,
B.L. Thorne, and E.O. Wilson. Supported by the Anderson and
Richmond Funds of Harvard University, NSF Grant BNS 80-02613
(B. Holldobler, sponsor), and NSF predoctoral grants to both
authors.

Literature Cited

Abe, T.

1971. On the food sharing among four species of ants in a sandy grassland, I.
Food and foraging behavior. Japanese Journal of Ecology 20: 219-230.
Adams, E., and J. F. A. Traniello.

1981. Chemical interference competition by Monomorium minimum.
Oecologia in press.

Alloway, T. M.

1980. The origin of slavery in leptothoracine ants (Hymenoptera: Formicidae).
American Naturalist 115: 247-261.

Ashton, D. H.

1979. Seed harvesting by ants in forests of Eucalyptus regnans F. Muell in
central Victoria. Australian Journal of Ecology 4: 265-277.

Autori, M.

1941. Contribuicao para o conhecimento da sauva ( Atta spp. Hymenoptera:
Formicidae) ( Atta sexdens rubripilosa Forel 1980). Arquivos do Institu-
te Biologico 12: 197-228.

Baroni-Urbani, C.

1969. Ant communities of the high altitude Appenine grasslands. Ecology 50:
488-492.

Baroni-Urbani, C.

1979. Territoriality in social insects. In: The Social Insects, H. R. Hermann,
ed. Academic Press, N.Y.

Bernstein, R. A.

1974. Seasonal food abundance and foraging activity in some desert ants.
American Naturalist 108: 490-498.

Bernstein, R. A.

1975. Foraging strategies of ants in response to variable food density. Ecology
56: 213-219.

Bernstein, R. A.

1979. Schedules of foraging activity in species of ants. Journal of Animal
Ecology 48: 921-930.

310

Psyche

[Vol. 88

Bernstein, R. A., and M. Gobbel.

1979. Partitioning of space in communities of ants. Journal of Animal Ecology
48: 931-942.

Blackith, R. E., J. W. Siddorn, N. Waloff, and H. F. van Emden.

1963. Mound nests of the yellow ant, Lasius flavus L. on waterlogged pasture
in Devonshire. Entomologist’s Monthly Magazine 99: 48-49.

Bradley, G. A.

1972. Transplanting Formica obscuripes and Dolichoderus taschenbergi (Hy-
menoptera: Formicidae) colonies in jack pine stands of southeastern
Manitoba. Canadian Entomologist 104: 245-249.

Bradley, G. A.

1973. Interference between nest populations of Formica obscuripes and Do-
lichoderus taschenbergi (Hymenoptera: Formicidae). Canadian Ento-
mologist 105: 1525-1528.

Bradley, G. A., and J. D. Hinks.

1968. Ants, aphids and jack pines in Manitoba. Canadian Entomologist 100:
40-50.

Breen, J.

1979. Nest sites of Formica lugubris (Hymenoptera: Formicidae) in Irish
plantation woods. Journal of Life Sciences, Royal Dublin Society 1:
13-32.

Brian, M. V.

1952. The structure of a dense natural ant population. Journal of Animal
Ecology 21: 12-24.

Brian, M. V.

1955. Food collection by a Scottish ant community. Journal of Animal
Ecology 24: 336-351.

Brian, M. V.

1956a. The natural density of Myrmica rubra and associated ants in west
Scotland. Insectes Sociaux 3: 473-487.

Brian, M. V.

1956b. Segregation of species of the ant genus Myrmica. Journal of Animal
Ecology 25: 319-337.

Brian, M. V.

1964. Ant distribution in a southern English heath. Journal of Animal
Ecology. 33: 451-461.

Brian, M. V., J. Hibble, and D. Stradling.

1965. Ant pattern and density in a southern English heath. Journal of Animal
Ecology 34: 544-555.

Brian, M. V., J. Hibble, and A. F. Kelly.

1966. The dispersion of ant species in a southern English heath. Journal of
Animal Ecology 35; 281-290.

Brough, E. J.

1976. Notes on the ecology of an Australian desert species of Calomyrmex
(Hymenoptera: Formicidae). Journal of the Australian Entomological
Society 15: 339-346.

1981]

Levings & Traniello — Territoriality in Ants

311

Brown, W. L., and E. O. Wilson.

1959. The evolution of the dacetine ants. Quarterly Review of Biology 34:
278-294.

Byron, P. A., E. R. Byron, and R. A. Bernstein.

1980. Evidence of competition between two species of desert ants. Insectes
Sociaux 27: 351-360.

Carroll, R., and D. H. Janzen.

1973. Ecology of foraging by ants. Annual Review of Ecology and Systematics
4: 231-257.

Cherix, D., and G. Gris.

1977. The giant colonies of redwood ant in the Swiss Jura ( Formica lugubris
Zet.). Proceedings of the Eighth International Congress of International
Union for the Study of Social Insects. Wageningen, the Netherlands.
Centre for Agricultural Publishing and Documentation, Wageningen.

Clark, P. J., and F. C. Evans.

1954. Distance to nearest neighbor as a measure of spatial relationships in
populations. Ecology 35: 445-453.

Clark, P. J., and F. C. Evans.

1979. Generalization of a nearest neighbor measure of dispersion for use in K.
dimensions. Ecology 60: 316-317.

Colwell, R., and D. Futuyma.

1971. On the measurement of niche breadth and overlap. Ecology 52: 567-576.

Corn, M. L.

1976. The ecology and behvior of Cephalotes atratus; a Neotropical ant
(Hymenoptera: Formicidae). Unpublished Ph.D. dissertation, Harvard
University.

Culver, D.

1974. Species packing in Caribbean and north temperate ant communities.
Ecology 55: 974-988.

Davidson, D.

1977a. Species diversity and community organization in desert seed-eating ants.
Ecology 58: 711-724.

Davidson, D.

1977b. Foraging ecology and community organization in desert seed-eating
ants. Ecology 58: 725-737.

Davidson, D.

1980. Some consequences of diffuse competition in a desert ant community.
American Naturalist 116: 92-105.

DeVita, J.

1979. Mechanisms of interference and foraging among colonies of the har-
vester ant Pogonomyrmex calif ornicus in the Mojave desert. Ecology 60:
729-737.

DeVroey, C.

1979. Aggression and Gausse’s Law in ants. Physiological Entomology 4:
217-222.

Diver, C.

1935. A population of wood ants ( Formica sp.) at Abernethy. Journal of
Animal Ecology 4: 32-34.

312

Psyche

[Vol. 88

Dobrzanska, J.

1958. Partition of foraging grounds and modes of conveying information
among ants. Acta Biologiae Experimentalis, (Warsaw) 18: 55-67.

Dobrzanska, J.

1966. The control of territory by Lasius fuliginosus Latr. Acta Biologiae
Experimentalis, (Warsaw) 26: 193-213.

Duncan-Weatherly, A. H.

1953. Some aspects of the biology of the mound ant Iridomyrmex detectus.
Australian Journal of Zoology 1: 178-192.

Eisenberg, R. M.

1972. Partitioning of space among colonies of the fire ants, Solenopsis
saevissima, I. Spatial arrangement. Texas Journal of Science 24: 39-43.

Elmes, G. W.

1971. An experimental study on the distribution of heathland ants. Journal of
Animal Ecology 40: 495-499.

Elmes, G. W.

1974. The spatial distribution of a population of two ant species living in
limestone grassland. Pedobiologia 14: 412-418.

Elton, C. S.

1932. Territory among wood ants ( Formica rufa L.) at Picket Hill. Journal of
Animal Ecology 1: 69-76.

Fluker, S. S., and J. W. Beardsley.

1970. Sympatric associations of three ants: Iridomyrmex humilis, Pheidole
megacephala and Anoplolepis longipes in Hawaii. Annals of the Ento-
mological Society of America 63: 1290-1296.

Fra 'coeur, A., and D. PeppiN.

i978. Productivitep de la four mi Formica dakotensis dans la pessiepre tour-
beuse. 2. Variations annuelles de la densitep des colonies, de l’occupation
des nids et de la repartition spatiale. Insectes Sociaux 25: 13-30.

Franks, N.

1980. The evolutionary ecology of the army ant Eciton burchelli on Barro
Colorado Island, Panama. Unpublished Ph.D. dissertation, The Univer-
sity of Leeds, Leeds, England.

Greenslade, P. J. M.

1971. Interspecific competition and frequency changes among ants in the
Solomon Islands coconut plantations. Journal of Applied Ecology 8:
323-352.

Greenslade, P. J. M.

1975. Dispersion and history of a population of the meat ant Iridomyrmex
purpureus (Hymenboptera: Formicidae). Australian Journal of Zoology
23: 495-510.

Headley, A. E.

1941. A study of the nest and nesting habits of the ant Lasius niger subsp.
alienus var. americanus Emery. Annals of the Entomological Society of
America 34: 649-657.

Headley, A. H.

1952. Colonies of ants in a locust woods. Annals of the Entomological Society
of America 45: 435-442.

1981]

Levings & Traniello — Territoriality in Ants

313

Higashi, S., and K. Yamauchi.

1979. Influence of a supercolonial ant Formica (Formica) yessensis on the
distribution of other ants in Ishikari Coast. Japanese Journal of Ecology
29: 257-264.

Holldobler, B.

1973. Chemische strategic beim nahrungserwerb der Diebsameise ( Solenopsis
fugax Latr.) und der Pharaoameise ( Monomorium pharoanis L.) Oeco-
logia 11: 371-380.

Holldobler, B.

1974. Home range orientation and territoriality in harvesting ants. Proceed-
ings of the National Academy of Sciences, USA 71: 3274-3277.

Holldobler, B.

1976a. Recruitment behavior, home range orientation and territoriality in
harvesting ants. Behavioral Ecology and Sociobiology 1: 3-44.

Holldobler, B.

1976b. Tournaments and slavery in a desert ant. Science 192: 912-914.

Holldobler, B.

1977. Communication in social Hymenoptera. Pp. 418-471 in T. A. Sebeok,
ed., How animals communicate. Indiana University Press, Bloomington.

Holldobler, B.

1979a. Territoriality in ants. Proceedings of the American Philosophical So-
ciety 123: 211-218.

Holldobler, B.

1979b. Territories of the African weaver ant ( Oecophylla longinoda (Latreille)):
A field study. Zeitschrift fur Tierpsychologie 51: 201-213.

Holldobler, B., and C. Lumsden.

1980. Territorial strategies in ants. Science 210: 732-739.

Holldobler, B., and M. Moglich.

1980. The foraging system of Pheidole militicida. Insectes Sociaux 27: 237-
264.

Holldobler, B., and E. O. Wilson.

1970. Recruitment trails in the harvester ant Pogonomyrmex badius. Psyche
77: 385-399.

Holldobler, B., and E. O. Wilson.

1977a. Weaver ants: social establishment and maintenance of territory. Science
195: 900-902.

Holldobler, B., and E. O. Wilson.

1977b. Colony-specific territorial pheromone in the African weaver ant Oeco-
phylla longinoda (Latreille). Proceedings of the National Academy of
Science, USA 74: 2072-2075.

Holldobler, B., and E. O. Wilson.

1977c. The number of queens: an important trait in ant evolution. Naturwissen-
schaften 64: 8-15.

Holldobler, B., Stanton, R. C., and H. Markl.

1978. Recruitment and food-retrieving behavior in Novomessor (Formicidae,
Hymenoptera) I. Chemical signals. Behavioral Ecology and Socio-
biology 4: 163-181.

314

Psyche

[Vol. 88

Hunt, J.

1974. Temporal activity patterns in two competing ant species (Hymenoptera:
Formicidae). Psyche 81: 237-242.

Ito, M., and S. Imamura.

1974. Observations on the nuptial flight and internidal relationship in a
polydomous ant, Formica (F.) yessensis Forel. Journal of the Faculty of
Science, Hokkaido University, Series VI, Zoology 19: 681-694.

Janzen, D. H.

1967. Interaction of the bull’s-horn acacia ( Acacia cornigera L.) with an ant
inhabitant ( Pseudomyrmex ferruginea F. Smith) in eastern Mexico.
Kansas University Science Bulletin 47: 315-558.

Janzen, D. H.

1969. Allelopathy by myrmecophytes: the ant Azteca as an allelopathic agent
of Cecropia. Ecology 50: 147-153.

Janzen, D. H.

1972. Protection of Barteria (Passifloraceae) by Pachysima ants (Pseudo-
myrmicinae) in a Nigerian rain forest. Ecology 53: 885-892.

Janzen, D. H.

1973. Evolution of polygynous obligate acacia ants in western Mexico.
Journal of Animal Ecology 42: 727-750.

Janzen, D. H.

1975. Pseudomyrmex nigropilosa: A parasite of a mutualism. Science 188:
936-937.

Leston, D.

1973. The ant mosaic in tropical tree crops and the limiting of pests and
diseases. Pest Articles and News Summaries 19: 311-341.

Levieux, J.

1971. Mise en epvidence de la structure des nids et de l’implantation des zones
de chasse de deux espeQces de Camponotus (Hym., Form.) aQ l’aide de
radio-isotopes. Insectes Sociaux 18: 29-48.

Levieux, J.

1977. La nutrition des fourmis tropicales: V. Elepments de syntheQse. Les
modes d’exploitation de la biocoenose. Insectes Sociaux 24: 235-260.

Levieux, J., and D. Louis.

1975. La nutrition des fourmis tropicales. II. Comportement alimentaire et
regime de Camponotus vividus (Smith) (Hymenoptera: Formicidae).
Comparison intragenerique. Insects Sociaux 22: 391-404.

Levings, S. C., and N. R. Franks.

1982. Patterns of nest dispersion in a tropical ground ant community. Ecology,
63 in press.

Levings, S. C., and D. M. Windsor.

1982. Seasonal and annual variation in litter arthropod populations. In E. G.
Leigh, A. S. Rand, and D. M. Windsor, eds., The ecology of a tropical
forest: seasonal rhythms and longer term changes. Smithsonian Institu-
tion Press, Washington, D.C. In press.

Lewis, T.

1975. Colony size, density and distribution of the leaf cutting ant, Acromyr-

1981]

Levings & Traniello — Territoriality in Ants

315

mex octospinosus (Reich) in cultivated fields. Transactions of the Royal
Entomological Society of London 127: 51-64.

Lewis, T., G. V. Pollard, and G. C. Dibley.

1974. Rhythmic foraging in the leaf cutting ant Atta cephalotes. Journal of
Animal Ecology 43: 129-141.

Lynch, J., E. C. Balinsky, and S. G. Vail.

1980. Foraging patterns in three sympatric forest ant species, Prenolepis
imparts, Paratrechina melanderi and Aphaenogaster rudis (Hymenop-
tera: Formicidae). Ecological Entomology 5: 353-371.

Mabelis, A. A.

1979. The relationship between aggression and predation in the red wood ant
(Formica polyctena FoRrst.). Netherlands Journal of Zoology 29:
451-620.

MacArthur, R. H., and R. Levins.

1967. The limiting similarity, convergence and divergence of coexisting species.
American Naturalist 101: 377-385.

Majer, J. D.

1976a. The maintenance of the ant mosaic in Ghana cocoa farms. Journal of
Applied Ecology 13: 123-144.

Majer, J. D.

1976b. The ant mosaic in Ghana cocoa farms: further structural considerations.
Journal of Applied Ecology 13: 145-153.

Markl, H., and B. Holldobler.

1978. Recruitment and food-retrieving behavior in Novomessor (Formicidae,
Hymenoptera) II. Vibration Signals. Behavioral Ecology and Sociobiol-
ogy 4: 183-216.

Marikovsky, P. I.

1963. The ant Formica sanguinea Latr. as pillagers of Formica rufa Lin. nests.
Insectes Sociaux 10: 119-128.

Morglich, M., and G. Alpert.

1979. Stone-dropping by Conomyrma bicolor: a new kind of interference
behavior in ants. Behavioral Ecology and Sociobiology 6: 105-114.

Muir, R. J.

1974. A study of the coexistence of four sympatric species of Myrmecia
(Formicidae). Unpublished Ph.D. dissertation, University of New Eng-
land, Armidale, New South Wales, Australia.

Nipson, H. E.

1978. Inbreeding in the ant species Formica exsectoides. Unpublished Ph.D.
dissertation, Harvard University.

Oster, G., and E. O. Wilson.

1978. Caste and ecology in the social insects. Princeton University Press,
Princeton, N.J.

Petal, J.

1972. Methods of investigating the productivity of ants. Ekologia Polska 20:
9-22.

Petal, J.

1977. The effect of food supply and intraspecific competition in an ant

316

Psyche

[Vol. 88

population. Proceedings of the Eighth International Congress of the
International Union for the Study of Social Insects, 5-10 September
1977, Wageningen, The Netherlands, pp. 60-61.

PlANKA, E. R.

1978. Evolutionary ecology, 2nd ed. Harper and Row, New York.

Pickles, W.

1944. Territories and interrelations of two ants of the genus Messor in Algeria.
Journal of Animal Ecology 13: 128-129.

PlELOU, E. C.

1977. Mathematical ecology, 2nd ed. John Wiley, New York.

Pisarski, B.

1972. La structure des colonies polycaliques de Formica (Coptoformica)
exsect a Nyl. Ekologia Polska 20: 113-116.

Pontin, A. J.

1960. Field experiments on colony foundation by Lasius niger (1.) and Lasius
flavus (F.) (Hym., Formicidae). Insectes Sociaux 7: 227-230.

Pontin, A. J.

1961. Field experiments on colony foundation by Lasius niger (L.) and L.
flavus (F.) (Hym.: Formicidae). Journal of Animal Ecology 38: 47-54.

Pontin, A. J.

1963. Further consideration of competition and the ecology of the ants Lasius
flavus (F.) and L. niger (L.). Journal of Animal Ecology 32: 565-574.

Pontin, A. J.

1969. Experimental transplantation of nest mounds of the ant Lasius flavus
(F.) in a habitat containing also L. niger (L.) and Myrmica scabrinodis
Nyl. Journal of Animal Ecology 38: 747-754.

Poole, R. E.

1974. An introduction to quantitative ecology. McGraw-Hill, New York.

Reichman, O. J.

1979. Desert granivore foraging and its impact on seed densities and distribu-
tion. Ecology 60: 1085-1092.

Robertson, P. L.

1971. Pheromones involved in aggressive behavior in the ant Myrmecia
gulosa. Journal of Insect Physiology 17: 691-715.

Rockwood, L.

1973. Distribution, density and dispersion of two species of Atta (Hymenop-
tera: Formicidae) in Guanacoste Province, Costa Rica. Journal of
Animal Ecology 42: 803-807.

Room, P. M.

1971. The relative distributions of ant species in Ghana’s cocoa farms. Journal
of Animal Ecology 40: 735-751.

Room, P. M.

1975a. Relative distributions of ant species in cocoa plantations in Papua, New
Guinea. Journal of Applied Ecology 12: 47-61.

Room, P. M.

1975b. Diversity and organization of the ground foraging ant faunas of forest,
grassland and tree crops in Papua, New Guinea. Australian Journal of
Zoology 23: 71-89.

1981]

Levings & Traniello — Territoriality in Ants

317

Sanders, C. J.

1970. The distribution of carpenter ant colonies in the spruce fir forests of
northwestern Ontario. Ecology 51: 865-873.

Scherba, G.

1958. Reproduction, nest orientation and population structure of an aggrega-
tion of mound nests of Formica ulkei, Emery (Formicidae). Insectes
Sociaux 5: 201-213.

Scherba, G.

1964. Species replacement as a factor affecting distribution of Formica
opaciventris Emery (Hymenoptera: Formicidae). Journal of the New
York Entomological Society 72: 231-237.

Schneirla, T. C.

1971. Army ants: a study in social organization. W. H. Freeman and Co., San
Francisco.

Schoener, T.

1971. Theory of feeding strategies. Annual Review of Ecology and Systematics
2: 369-404.

SlMBERLOFF, D. S.

1979. Nearest neighbor assessments of spatial configurations of circles rather
than points. Ecology 60: 679-685.

Skinner, G. H.

1980. Territory, trail structure and activity patterns in the wood ant, Formica
rufa (Hymenoptera: Formicidae) in limestone woodland in northwest
England. Journal of Animal Ecology 49: 381-394.

STEBAEV, I., AND J. I. REZNIKOVA.

1972. Two interaction types of ants living in steppe ecosystem in South
Siberia, USSR. Ekologia Polska 20: 103-109.

Steyn, J. J.

1954. The pugnacious ant ( Anoplolepis custodiens Smith) and its relation to
the control of citrus scales at Letaba. Memoirs of the Entomological
Society of South Africa 3: 1-96.

Sudd, J. H., T. M. Douglas, T. Gaynard, D. M. Murray, and J. M. Stockdale.

1977. The distribution of wood ants ( Formica lugubris Zetterstedt) in a
northern English forest. Ecological Entomology 2: 301-313.

Swain, R.

1977. The natural history of Monads, a genus of Neotropical ants. Unpub-
lished Ph.D. dissertation, Harvard University.

Talbot, M.

1943. Population studies of the ant Prenolepis imparis. Ecology 24: 31-44.

Talbot, M.

1951. Population and hibernating conditions of the ant Aphaenogaster (At-
tomyrma) rudis Emery (Hymenoptera: Formicidae). Annals of the
Entomological Society of America 44: 302-307.

Talbot, M.

1954. Populations of the ant Aphaenogaster (Attomyrma) treatae Forel on
abandoned fields on the Edwin S. George Reserve. Contribution from
the Laboratory of Vertebrate Biology of the University of Michigan, no.
69, pp. 1-9.

318

Psyche

[Vol. 88

Talbot, M.

1957. Populations of ants in a Missouri woodland. Insectes Sociaux 4:
375-384.

Talbot, M.

1961. Mounds of the ant Formica ulkei at the Edwin S. George Reserve,
Livingston County, Michigan. Ecology 42: 202-205.

Talbot, M.

1967. Slave raids of the ant Polyergus lucidus Mayr. Psyche 74: 299-313.

Talbot, M. and C. H. Kennedy.

1940. The slave making ant, Formica sanguinea subintegra Emery, its raids,
nuptial flights and nest structure. Annals of the Entomological Society
of America 32: 304-315.

Thorne, B. L., and K. P. Sebens.

1981. Regulation of social insect colony size: a model based on foraging
energetics. Unpublished manuscript.

Traniello, J. F. A.

1980. Studies on the behavioral ecology of north temperate ants. Unpublished
Ph.D. dissertation, Harvard University.

Waloff, N., and R. E. Blackith.

1962. The growth and distribution of mounds of Lasius flavus (F.) (Hymenop-
tera: Formicidae) in Silwood Park, Berkshire. Journal of Animal
Ecology 31: 421-437.

Ward, P. S., and R. W. Taylor.

1981. Allozyme variation, colony structure and genetic relatedness in a
primitive ant Nothomyrmecia macrops (Hymenoptera: Formicidae).
Unpublished manuscript.

Way, M. J.

1953. The relationship between certain ant species with particular reference to
biological control of the coreid, Theraptus sp. Bulletin of Entomological
Research 45: 93-112.

Whitford, W. G., E. Depree, and P. Johnson.

1980. Foraging ecology of two Chihuahuan desert ant species: Novomessor
cockerelli and N. albisetosus. Insectes Sociaux 27: 148-156.

Whitford, W. G., P. Johnson, and J. Ramirez.

1976. Comparative ecology of the harvester ants Pogonomyrmex barbatus and
P. rugosus. Insectes Sociaux 23: 117-132.

Wilson, E. O.

1959a. Communication by tandem running in the ant genus Cardiocondyla.
Psyche 66: 29-34.

Wilson, E. O.

1959b. Some ecological characteristics of ants in New Guinea rain forests.
Ecology 40: 437-447.

Wilson, E. O.

1971. The insect societies. Belknap Press of Harvard University Press, Cam-
bridge, Mass.

Wilson, E. O.

1975. Enemy specification in the alarm-recruitment system of an ant. Science
190: 798-800.

1981]

Levings & Traniello — Territoriality in Ants

319

Wilson, N. L., J. H. Diller, and G. P. Markin.

1972. Foraging territories of imported Fire ants. Annals of the Entomological
Society of America 64: 660-665.

Yasuno, M.

1963. The study of the ant population in the grassland at Mt. Hakkoda. I. The
distribution and the abundance of ants in the grassland. Ecological
Review 16: 83-91.

Yasuno, M.

1964a. The study of the ant population in the grassland at Mt. Hakkoda, II. The
distribution pattern of ant nests at the Kayano grassland. Science
Reports of the Tohoku University, Sendai, Japan, ser. 4 (Biol.) 39:
43-55.

Yasuno, M.

1964b. The study of the ant population in the grassland at Mt. Makkoda, III.
The effect of the slave making ant, Polyergus samurai, upon the nest
distribution pattern of the slave ant, Formica fusca japonica. Science
Reports of the Tohoko University, Sendai, Japan, ser. 4 (Biol.) 30:
167-170.

Yasuno, M.

1965. Territory of ants in the Kayano grassland at Mt. Hakkoda. Science
Reports of the Tohoko University, Sendai Japan, ser 4 (Biol.) 31:
195-206.

THE EFFECT OF FLOWER OCCUPANCY ON THE
FORAGING OF FLOWER-VISITING INSECTS*

By V. J. Tepedino and F. D. Parker
Bee Biology & Systematics Laboratory, Agricultural Research,
Science & Education Admin.,

USDA

Utah State University, UMC 53
Logan, Utah

Introduction

To locate flowers, insects use a variety of visual and olfactory cues
such as flower color, shape, movement and scent (Faegri and van
der Pijl 1971). In addition, other insects on the flowers may also
serve as cues that either attract or repel prospective foragers. First,
foragers might avoid occupied inflorescences because: 1) there is a
high probability that other flowers on the inflorescence have been
recently exploited (Pleasants and Zimmerman 1979, Zimmerman
1981); 2) of the potential loss of time and energy due to aggressive
encounter with the occupant (Kikuchi 1963, Decelles and Laroca
1979); 3) the occupant might be an enemy (e.g., thomisids, phyma-
tids, etc.). Thus, when flowers are abundant, unoccupied inflor-
escences may yield a greater quantity of energy and/or nutrients per
unit effort. If so, the distribution of foragers across inflorescences
should be regular or underdispersed, i.e., there should be more
inflorescences with only one insect than expected on the assumption
of a random distribution.

Existing evidence also suggests that a second hypothesis is
tenable. Prospective foragers may be attracted by floral occupants
because: 1) the presence of other foragers indicates that resources
are available on the inflorescence; 2) the occupants themselves are
sources of pollen to some foragers (Laroca and Winston 1978,
Thorp and Briggs 1980). If insects are attracted to occupied
inflorescences, then their distribution across inflorescences should
be over-dispersed.

* Manuscript received by the editor October 20, 1981

321

322

Psyche

[Vol. 88

In this paper we use data for insects foraging on plantings of
commercial sunflowers ( Helianthus annuus L.) and onions {Allium
cepa L.) to test these hypotheses.

An additional question of interest is whether bee species differ in
their distribution across flowers. For example, Benest (1976) has
suggested that honeybees {Apis mellifera L.) are more tolerant of
joint foraging than are bumblebees {Bombus sp.) and Kalmus (1954)
reported that honeybees tend to form clusters at artificial feeding
sites. Group foraging, leading to clumped distributions on flowers
has also been reported for several tropical bee species (Frankie and
Baker 1974). To ascertain if the distribution of the multispecies
assemblage obscured differences among the component species, we
compared the distributions of the more abundant species with the
balance of other foraging individuals on the inflorescences.

Methods

Five cultivars of sunflower and 2 of onions were grown at the
Greenville Farm Agricultural Research Station in North Logan,
Utah. Sunflowers were planted in 5 adjacent 40m rows, 1 row per
cultivar. The 2 onion cultivars were planted alternately in 4 adjacent
rows, 2 rows per cultivar.

Counts of floral visitors were made several times during the
flowering period as 1 observer (FDP) walked along each row. A
tape recorder facilitated observations. Only heads with some open
flowers were censused.

The data were transcribed to number of flower heads with zero,
one, two, etc. insects and then compared with values expected on
the assumption of a Poisson distribution (Southwood 1978). The
Poisson series describes a random distribution and is written Px(k) =
e"x(xk/K!) where e = base of Napierian logarithms, and Px is the
expected number of flower heads with k insects (k = 0, 1, 2,—). The
parameter x is estimated by the mean number of insects per flower
head. For the Poisson distribution, the mean and variance are
equal, and an indication of the dispersion of insects across flowers is
given by the coefficient of dispersion (C.D. = s2/x). When C.D. is
>1.0 the dispersion is clumped or contagious; and when <1.0
dispersion is regular or repulsed (Southwood 1978). The expected

1981]

Tepedino & Parker — Flower-Visiting Insects

323

and observed distributions were tested for significance using the x2
test (Zar 1974).

The distributions of more abundant species across sunflower
heads was compared with the balance of the foraging assemblage as
follows: each individual recorded was assigned to one of two
mutually exclusive categories, according to whether it foraged alone
or with at least one other insect (irrespective of species) on the
inflorescence. A chi-square test of independence was used to com-
pare each species represented by >10 individuals with the balance of
the assemblage.

Results

Bees were the predominant visitors to sunflowers; we recorded 15
species in 5 families (Appendix). The species were similar to that
reported previously by Parker (1981) for the same study site. Onion
visitors included many species of wasps and flies that did not forage
on sunflowers. In contrast to sunflowers, there were more non-bee
than bee visitors to onions.

For all sunflower censuses the distribution of total insects across
flower heads did not differ significantly from a Poisson distribution,
i.e., insects appeared to be foraging independently of other insects.
The coefficients of dispersion were mostly around 1.0. There was no
tendency for C.D.’s to be greater or less than 1; for 8 censuses C.D.
was >1.0 and for 6 censuses C.D. <1.0. (Table 1).

Only 2 of 7 censuses of onions deviated significantly from a
random distribution (Table 1). Both deviations occurred on the
same day and were in the direction of under-dispersion; more heads
with single visitors were recorded than expected. There was a
general tendency for insects visitors to be under-dispersed on
onions; in all tests C.D. > 1.0.

There was no indication that any particular species foraged other
than randomly, with respect to other occupants of sunflower heads.
The results of 34 comparisons of the distribution of individuals of
abundant species with the balance of foragers for the single and
joint foraging categories are shown in Table 2. Only one comparison
yielded significant results; another closely approached significance
(7/31 Peredovik, AM, Halictus ligatus , P. = 0.051). It is likely that
these two instances were due to chance.

Table 1. Total flower heads, mean insects per head, coefficient of dispersion (C.D. = s2/x) and probability levels (X2 test) for
insects visiting sunflower and onion heads. All counts made between 1000-1 100 hrs except those with asterisks which were made

324

Psyche

[Vol. 88

A A A v V A A

OO Tf Tf — ' Tfr SO <N

Os OK ON Os OO OS Os

N 'O ^ O O O W

r- ->3- < <n u-s os

r-~ oo ^ so m t}- m

K -rt ( N 't (N t
N IT) CL n fl, IT)

> CU CU -1 Cu -tL Cu

osno^nsriosnsnoosnoo

AAAAAAAAAAAAA

_„,_0_000„0„0__

«__0— 'OOOOOOOO

cd -o

2 8

f— 1 JS

r-Ttommoo— 'OOTtr-^so
sfinnxici-o-<doso0'o
<N — - <N — I <N CN

M M
‘> ‘>
o o

S T3 T3

•Si a> <u

U L* •— M

Q D « r/s a

> CU CU Oh

^ m o —i

0-0 0
00 <L> oo o
00

5 S - *

<L> iP o o

S3 £ % £ »

:> oo

C/3 .3* CU C/3 <N

C/3 — —

> *

I |

/ 254* 160 0.45 1.03 >0.25

/ W501* 195 0.61 1.04 >0.50

1981]

Tepedino & Parker — Flower-Visiting Insects

325

Discussion

In this study, foraging insects did not appear to react to the
presence of other insects in choosing flowers. Only 2 of the censuses
on onions and none of the censuses on sunflowers displayed a
significant departure from a random distribution (Table 1). Sun-
flower foragers (Apis, Perdita, Halictus) frequently entered the
flower by landing on the back of the petals or on the involucral
bracts and then crawling onto the head. If occupancy by another
insect were important, this would be an inefficient method of
choosing a flower. In a similar study Waddington (1976) also
concluded that halictid bees were foraging independently on bind-
weed ( Convolvulus arvensis). None of the abundant species
present appeared to forage other than randomly with respect to
other flower occupants. This was especially surprising for honeybees
which have been reported to more readily tolerate, or even form,
clumped distributions (Kalmus 1954, Benest 1976). However, con-
tagious distributions of honeybees may occur only under unusual
circumstances; the data of Kalmus (1954) were gathered from a
small number of feeding dishes and are quite artificial. Benest’s
(1976) suggestion that honeybees are more tolerant of joint foraging
than bumblebees does not stand close examination. Additional
study is required before such conclusions are warranted.

Instead of using the presence of insects on inflorescences as cues,
some flower-visiting insects may make selections based on the
number of open flowers or the amount of nectar or pollen available.
Although all heads censused had some open flowers, some had more
open flowers than others and insects may have been choosing those
heads with more flowers irrespective of other visitors. Even if heads
were equivalent in number of flowers, continuous removal of nectar
and pollen by foragers would cause variation in resource availability
between heads (e.g., Pleasants and Zimmerman 1979) and insects
may be responsive to such variation prior to landing on a flower.
For example, Thorp et al. (1975) have suggested that the fluorescent
nectar (and perhaps pollen) of many species with open flowers may
be used as a cue by foraging insects (see also Kevan 1976, Thorp et
al. 1976); and onion nectar is intensely fluorescent (Thorp et al.
1975). Recently Heinrich (1979) has shown that bumblebee foragers
reject many more nectar depleted (recently visited) white clover
( Trifolium repens) heads than heads with abundant nectar. Rejec-

326

Psyche

[Vol. 88

■© o,

5 £

3: a

~S |
I1 a3

•2

ci,

1 1
^ -

“3

§ -2

I I ll l 2 l I I l I l l I i

I I I I I

l I I

l l I

I o | «n <n
I O loo

— « "3- O oo r*->
O o’ <N m' <N

I I I

I II II M I I

m — o
O o' o'

I I I I I I

I I

I II I I I

"NO

o' o' d

M M
> >
o o
•o "O

u «U

M
’>
o

o -g
60 .S3 „

> *

o o g ®
60 S3 £

<UCL» 4J 4> <D <U ^

Lp <0.025

1981]

Tepedino & Parker — Flower-Visiting Insects

327

tion was accomplished without landing and the cue was probably
scent of nectar (Heinrich 1979). Future field studies should explore
the use of these more subtle cues by foraging insects.

Acknowledgments

We thank Drs. Ivan Palmbad and Richard Rust for reviewing the
manuscript.

Appendix

Insect taxa visiting sunflower and onion plantings.

Sunflowers

Hymenoptera:

Bees — Andrenidae [ Andrena helianthi Robert-
son, Perdita sp., Pseudopanurgus sp., Ptero-
sarus sp.]; Anthophoridae [Melissodes agilis
Cresson, Svastra obligua (Say), Triepeolus heli-
anthi (Robertson)]; Apidae [ Apis mellifera Lin-
naeus, Bombus spp.]; Halictidae [Agapostemon
sp., Dialictus sp., Halictus farinosus Smith,
Halictus ligatus Say]; Megachilidae [Megachile
paralella Smith, Megachile pugnata Say].

Diptera:

Lepidoptera:

Syrphidae

Hesperiidae

Onions

Hymenoptera:

Bees — Apidae [Bombus sp.]; Halictidae [Evy-
laeus sp, Halictus farinosus Smith, Halictus
ligatus Say]; Megachilidae [Hoplitis fulgida
(Cresson), Megachile pacifica (Panzer), Mega-
chile sp.].

Wasps — Eumenidae [Euodynerus sp., Ptero-
cheilus sp.]; Ichneumonidae; Sphecidae [Am-
mophila sp., Astata sp., Cerceris sp., Philanthus
sp., Podalonia sp., Sphex sp., Tachytes sp.]

Diptera:

Muscidae; Nemestrinidae; Sarcophagidae; Syr-
phidae; Tachinidae.

Lepidoptera:

Hesperiidae

328

Psyche

[Vol. 88

Literature Cited

Benest, G.

1976. Relations interspecifiques et intraspecifiques entre butineuses de Bom-
bus sp. et d 'Apis mellifica L. Apidologie, 7:113-127.

Decelles, P. and S. Laroca

1979. Behavioral interactions among solitarily foraging bees (Hymenoptera:
Apoidea). J. Kansas Entomol. Soc., 52:483-488.

Faegri, K. and L. Van Der Pijl

1971. The Principles of Pollination Ecology. 2nd Ed. Rev. London. Pergamon
Press. 291 pp.

Frankie, G. W. and H. G. Baker

1974. The importance of pollinator behavior in the reproductive biology of
tropical trees. An. Inst. Biol. Univ. Nal. Auton. Mexico 45 Ser.
Botanica, ( 1 ): 1 — 10.

Heinrich, B.

1979. Resource heterogeneity and patterns of movement in foraging bumble-
bees. Oecologia, 40:235-245.

Kalmus, H.

1954. The clustering of honeybees at a food source. Brit. J. Anim. Behav.,
2:63-71.

Kevan, P. G.

1976. Fluorescent nectar. Science, 194:341-342.

Kikuchi, T.

1963. Studies on the coaction among insects visiting flowers. III. Dominance
relationship among flower-visiting flies, bees and butterlies. Sci. Rep.
Tohoku Univ. Ser. IV (Biol.), 29:1-8.

Laroca, S. and M. L. Winston

1978. Interaction between Apis and Bombus (Hymenoptera: Apidae) on the
flowers of tall thistle: honeybees gather pollen from bodies of bumble-
bees. J. Kansas Entomol. Soc., 51:274-275.

Parker, F. D.

1981. Sunflower pollination: abundance, diversity and seasonality of bees and
their effect on yields. J. Apic. Res., 20:49-61.

Pleasants, J. M. and M. Zimmerman

1979. Patchiness in the dispersion of nectar resources: evidence for hot and
cold spots. Oecologia, 41:283-288.

Southwood, T. R. E.

1978. Ecological Methods, 2nd Ed. Rev. London: Chapman & Hall. 524 pp.

Thorp, R. W., D. L. Briggs, J. R. Estes, and E. H. Erickson

1976. Fluorescent nectar. Science, 194:341-342.

Thorp, R. W., D. L. Briggs, J. R. Estes and E. H. Erickson

1975. Nectar fluorescence under ultraviolet radiation. Science, 189:476-478.

Thorp. R. W. and D. L. Briggs

1980. Bees collecting pollen from other bees (Hymenoptera: Apoidea). J.
Kansas Entomol. Soc., 53:166-170.

1981]

Tepedino & Parker — Flower-Visiting Insects

329

Waddington, K. D.

1976. Foraging patterns of halictid bees at flowers of Convulvulus arvensis.
Psyche, 83:112-119.

Zar, J. H.

1974. Biostatistical Analysis. Englewood Cliffs, N. J.: Prentice-Hall.
Zimmerman, M.

1981. Patchiness in the dispersion of nectar resources: Probable causes.
Oecologia, 49:154-157.

ABDOMINAL TROPHALLAXIS IN THE SLAVE-MAKING
ANT, HARPAGOXENUS AMERICANUS
(HYMENOPTERA: FORMICIDAE)*

Robin J. Stuart

Department of Zoology, Erindale College,

University of Toronto,

Mississauga, Ontario, L5L 1C6 Canada

Abdominal trophallaxis refers to the passage of fluids from the
abdominal tip of one individual to the mouthparts of another. It is
common among lower termites (Kalotermitidae and Rhinotermiti-
dae) where it functions in the vital transmission of intestinal
flagellates to newly molted individuals. However, it has rarely been
documented among ants (Wilson, 1971, 1976). By strict definition,
the term abdominal trophallaxis should be applied only when
alimentary fluid is being transmitted (Wilson, 1971). Nevertheless,
in practice, the origin of the fluid is often unknown, at least initially.
Indeed, in all cases where this behavior has been described in ants,
the fluid is either suspected of being, or has since been shown to be,
ovarian in nature. For example, workers of certain Eciton species
(Dorylinae) readily feed from droplets secreted from the tip of the
queen’s abdomen, but this behavior has been observed only during
egg-laying bouts (Schneirla, 1944; Rettenmeyer, 1963). So-called
“proctodeal feeding” has also been described among the Doli-
choderinae ( Dolichoderus quadripunctatus, Tapinoma erraticum
and Iridomyrmex humilis) (Torossian, 1958, 1959, 1960, 1961).
However, at least in the case of D. quadripunctatus, the fluid has
been identified as the yolky remnants of abortive trophic eggs
(Torossian, 1978, 1979). Among the Myrmicinae, Zacryptocerus
varians exhibits a similar behavior which is also thought to be
associated with egg-laying (Wilson, 1976). This paper reports an
unusual and interesting case of abdominal trophallaxis in colonies
of the socially parasitic myrmicine ant, Harpagoxenus americanus. I
have followed Wilson (1976) and tentatively applied the term
abdominal trophallaxis, because the origin of the fluid is unknown.

H. americanus is an obligatory slave-maker and forms mixed
colonies with members of certain Leptothorax species in eastern

* Manuscript received by the editor November 10, 1981.

331

332

Psyche

[Vol. 88

North America (Alloway, 1979). The observations reported here
took place in artificial nests in the laboratory (Alloway, 1979) and
utilized colonies of H. americanus containing one or both of its host
species, L. ambiguus and L. longispinosus. The colonies were
collected in the regional municipalities of Halton and Peel, Ontario.

Intermittent observations of activity inside H. americanus nests
revealed that slave-maker queens and workers occasionally convey
fluids to their slaves by means of abdominal trophallaxis. Donors
characteristically raise their abdomens and assume a stereotyped
posture, similar to that seen during “sexual calling” (“Locksterzeln”)
in other leptothoracine ants (Buschinger and Alloway, 1979). I was
unable to ascertain whether the sting is exposed in the present
context, as it is during “sexual calling” and “tandem calling”, a
similar behavior used during nest-mate recruitment in some lepto-
thoracine ants (Moglich et al., 1974). While maintaining this
posture, the donor secretes a droplet of clear fluid from the tip of
her abdomen, and holds it there, at times for several minutes. Slaves
do not seem to be attracted from any appreciable distance by this
behavior, but those close by turn and antennate the donor’s
abdomen, apply their mouthparts to the tip, and consume the
droplet. As many as three slaves have been observed to attend a
donor simultaneously in this manner, clustered about her abdominal
tip and attempting to consume the droplet. On one occasion, the
droplet was removed from the donor’s abdomen by three workers in
concert, held between their mandibles momentarily, and then
consumed. Once the droplet is removed, the donor lowers her
abdomen, and both donor and recipients appear to resume normal
activities. There is no indication that slaves ever solicit this fluid;
and to date, the reverse, slaves donating to slave-makers, has not
been observed. Similarly, this behavior has never been observed in
laboratory colonies of the host species. The nature of the fluid
transmitted is unknown. It may be ovarian in origin, since H.
americanus workers will lay eggs, even in queenright colonies
(Buschinger and Alloway, 1977). The frequency of this behavior is
uncertain. It appears to be rare, since frequent observations of
colonies for other purposes have seldom encountered it. However,
no detailed behavioral repertoire study of this ant has been con-
ducted.

The fact that H. americanus employs a characteristic posture
during abdominal trophallaxis suggests that this behavior may have

1981]

Stuart — Trophallaxis in Harpagoxenus

333

important biological consequences. Furthermore, the apparent ab-
sence of this behavior in free-living leptothoracine ants, and the fact
that transmission is consistently from slave-maker to slave, suggests
that abdominal trophallaxis may in some way contribute to this
species’ particular socially parasitic relationship. The discovery of
this behavior in a slave-making ant opens a previously unknown
avenue for consideration in discussions of the means by which slave-
makers may affect the behavior of their slaves.

Acknowledgments

I thank T.M. Alloway for his comments on the manuscript.
Financial assistance was provided by an Ontario Graduate Scholar-
ship to the author, and a grant from the Natural Sciences and
Engineering Research Council of Canada to T.M. Alloway.

Literature Cited

Alloway, T. M. 1979. Raiding behaviour of two species of slave-making ants,
Harpagoxenus americanus (Emery) and Leptothorax duloticus Wesson
(Hymenoptera: Formicidae). Anim. Behav. 27: 202-210.

Buschinger, A., and T. M. Alloway. 1977. Population structure and poly-
morphism in the slave-making ant Harpagoxenus americanus (Emery)
(Hymenoptera: Formicidae). Psyche 83: 233-242.

1979. Sexual behaviour in the slave-making ant Harpagoxenus cana-
densis M. R. Smith, and sexual pheromone experiments with H. canadensis,
H. americanus (Emery), and H sublaevis (Nyl.) (Hymenoptera: Formicidae).
Z. Tierpsychol. 49: 113-119.

Moglich, M., U. Maschwitz, and B. Holldobler. 1974. Tandem calling:
a new kind of signal in ant communication. Science 186: 1046-1047.

Rettenmeyer, C. W. 1963. Behavioral studies of army ants. Kan. Univ. Sci. Bull.
44: 281-465.

Schneirla, T. C. 1944. The reproductive functions of the army-ant queen as
pace-makers of the group behavior pattern. J. N. Y. Entomol. Soc. 52: 153-192.

Torossian, C. 1958. L’aliment proctodeal chez la fourmi Dolichoderus quadri-
punctatus (Dolichoderidae). C. R. Acad. Sci. 246: 3524-3526.

1959. Les echanges trophallactiques proctodeaux chez la fourmi Doli-
choderus quadripunctatus (Hymenoptere-Formicoidea). Ins. Soc. 6: 369-374.

1960. Les echanges trophallactiques proctodeaux chez la fourmi Tapi-

noma erraticum. Ins. Soc. 7: 174-175.

1961. Les echanges trophallactiques proctodeaux chez la fourmi d’ Ar-
gentine: Iridomyrme x humilis (Hym. Form. Dolichoderidae). Ins. Soc.

8: 189-191.

334

Psyche

[Vol. 88

1978. La ponte d’oeufs abortifs chez les ouvrieres de la fourmi Doli-

choderus quadripunctatus. Soc. Hist. Nat. Toulouse Bull. 114: 207-211.

1979. Importance quantitative des oeufs abortifs d’ouvrieres dan le

bilan trophique de la colonie de la fourmi Dolichoderus quadripunctatus.
Ins. Soc. 26: 295-299.

Wilson, E. O. 1971. The insect societies. Harvard Univ. Press, Cambridge, Mass.

x + 548 pp.

1976. A social ethogram of the neotropical arboreal ant Zacryptocerus

various (Fr. Smith). Anim. Behav. 24: 354-363.

NEW NAME FOR THE EXTINCT GENUS MESAGYRTES
PONOMARENKO (COLEOPTERA: SILPHIDAE S.L.)

By Alfred F. Newton, Jr.

Museum of Comparative Zoology
Harvard University, Cambridge, Mass. 02138

Mesagyrtes communis Ponomarenko, a new beetle genus and
species attributed to Silphidae, has recently been described from
fossil-bearing beds of Jurassic age from the locality of Novospassk,
USSR (Arnoldi et al., 1977: 117). Unfortunately the generic name is
preoccupied by Mesagyrtes Broun (1895: 95), proposed for a Recent
New Zealand species originally placed in Silphidae; this genus is
now considered a subgenus of the genus Colon Herbst of the family
Leiodidae (Szymczakowski 1964).

I have brought the homonymy to the attention of Dr. Ponoma-
renko, who has kindly allowed me to propose a replacement name
for use in publications on the family Silphidae now in preparation.
Accordingly, I propose Mesecanus, new name, to replace Mesa-
gyrtes Ponomarenko (not Broun). The new name alludes to the
resemblance in habitus between the extinct genus and the Recent
agyrtine silphid genus Ecanus Stephens.

Literature Cited

Arnoldi, L. V., V. V. Zherikhin, L. M. Nikritin and A. G. Ponomarenko
1977. “Mesozoic Coleoptera” [in Russian]. Trudi Paleont. Inst. Akad. Nauk
SSSR 161, 204 pp.

Broun, T.

1895. Descriptions of new Coleoptera from New Zealand. Ann. Mag. Nat.
Hist. (6)15: 67-88.

Szymczakowski, W.

1964. Revision des Colonidae (Coleoptera) des regions orientale et austra-
lienne. Acta Zool. Cracov. 9(8): 1e59.
lienne. Acta Zool. Cracov. 9(8): 1-59.

335

HISTORICAL DEVELOPMENT OF BEE FORAGING
PATTERNS IN CENTRAL NEW YORK STATE

By Howard S. Ginsberg*

Department of Entomology
Cornell University
Ithaca, New York 14853

Introduction

The bee fauna of the northeastern United States has changed
markedly in the past few centuries. The impetus for this change
came largely from human activities, notably from introductions of
foreign species and modifications of the regional flora. Several bee
species, most notably the honey bee ( Apis mellifera), were intro-
duced into this region (Crane 1975; Linsley 1958). Honey bees can
powerfully influence the foraging patterns of native bees (Pearson
1933; Eickwort and Ginsberg 1980). Replacement of forests over
large areas by cities and farms (Ferguson and Mayer 1970; Vaughan
1929) and numerous introductions of alien plant species (Wiegand
and Eames 1925) have resulted in major changes in northeastern
plant communities.

How broad were these changes and how have they influenced the
foraging ecology of northeastern bees? What was this area like
before the European settlers arrived? The answers to these questions
are vital to an understanding of contemporary bee foraging patterns
and of community level interactions between flowers and their
pollinators. The purpose of this paper is to describe some general
trends in the foraging patterns of Apoidea in central New York
State, and to interpret them in terms of the historical development
of the flora and bee fauna of the region.

Materials and Methods

The study site was a 5.8 hectare abandoned field (last cultivated
about 1956) located near Ithaca, New York. It was bordered by

♦Present Address: Department of Ecology and Evolution, State University of New
York at Stony Brook, Stony Brook, New York 11794
Manuscript received by the editor August 8, 1981

337

338

Psyche

[Vol. 88

wooded areas and cultivated fields. The soils were well-drained and
flower bloom was profuse. More than 150 entomophilous species
bloomed on the field. The most common woody plants were red
maple (Acer rubrum), staghorn sumac (Rhus typhina ), and various
willows (Salix spp.), dogwoods (Cornus spp.), and brambles (Rubus
spp.). The dominant herbaceous plants included several entomo-
philous species and the grasses timothy (Phleum pratense) and
orchard grass (Dactylis glomerata).

I sampled Apoidea by walking transects and capturing bees from
flowers. There were 10, 30 m transects randomly-placed on the field.
I took transect samples during times of maximum foraging activity
(1000-1600 hours) throughout the season (at least 3 samples in each
2-week period, late May-October, 1974 and 1975). I used all-day
samples from randomly-selected patches of common flower species
(throughout the growing season, 1975 and 1976) to confirm the
results from the transect samples and to study spatial distributions
of foraging bees. Voucher specimens of the bee species are placed in
the Cornell University Insect Collection, lot number 1039.

I counted the number of flowers of each species at anthesis in 100,
lm2 subquadrats. The subquadrats were arranged in groups of 10,
randomly-placed within 30 m X 30 m quadrats (the bee transects
were also within these quadrats). There were 10 quadrats randomly-
placed on the field. Flowers were sampled once every 2 weeks
throughout the season. Voucher specimens of the plant species are
placed in the Bailey Hortorium Herbarium, Cornell University.
Details of the field techniques are given by Ginsberg (1979).

I used the records of Fernald (1950) and Wiegand and Eames
(1925) to determine whether flower species were native or were
introduced into the area. Their determinations were based largely
on the records of early botanical explorers (e.g. Pursh 1923) and on
previous species lists for the area (e.g. Dudley 1886). Admittedly,
there is some margin for error in these judgements, but because of
the large number of entomophilous species on the sample site,
mistakes about the points of origin of a few species should not
influence the major arguments.

Results

Red maple was the first abundant flower species to bloom on the
field in spring. Several willows and rosaceous trees (Prunus cerasus,

1981]

Ginsberg — Bee Foraging Patterns

339

Pyrus malus ) bloomed soon after, as did several roadside weeds
such as dandelion ( Taraxacum officinale ) and yellow rocket (Bar-
barea vulgaris). The spring species were typically clustered in
distribution at roadsides and forest edges, and the woody species
had relatively short blooming times. Of 16 species recorded on the
field in spring (late April and early May in 1975) half were native
and half were introduced. I do not include any of the several species
that bloomed in the woods nearby.

Flower bloom increased on the field to a maximum in early
summer (late June, early July). Most of the species in bloom at this
time of the year were introduced (Fig. 1). Table I lists the most
common of these species and gives their frequencies of occurrence in
the subquadrats. Note that the most common flowers at this time
were those of introduced herbaceous species. Most flowers of these
species were past blooming by midsummer.

In August, goldenrods ( Solidago spp.) predominated on the field.
These late summer flowers are native to this region (Table I). Aster,
another native genus of composites, predominated after goldenrod
passed bloom in the fall. Late season flowers, therefore, were mostly
native species (Fig. 1).

DATE

Fig. 1. Number of introduced and native flower species blooming over the summer,
1974, in an old field near Ithaca, New York.

340

Psyche

[Vol. 88

Table I. Frequencies of common flower species, 1974

Flower species

Origin1

Time of
maximum
bloom

Frequency2

No.

inflores-
cences/m2^

Hieracium pratense

mid June

12.55 ± 2.06

Chrysanthemum

leucanthemum

late June

2.21 ±0.32

Cornus racemosa

late June

2.10 ± 1.10

Satureja vulgaris

late July

12.11 ± 3.08

Achillea millefolium

late August

1.04 ±0.34

Daucus carota

late August

0.87 ± 0.20

Solidago juncea

late August

12.86 ±2.08

S. graminif olia

early Sept.

9.16 ± 1.85

S. rugosa

early Sept,

8.42 ± 2.03

S. altissima

early Sept.

14.69 ± 3.05

1 N = native species; I

= introduced

species

2 Number of 1 m2 subquadrats (out of 100) in which species was flowering during

period of maximum bloom.

3 Mean number of inflorescences (sprays for Solidago ) per subquadrat during
period of maximum bloom ± standard error.

This flowering trend of early-summer introduced species and late-
summer-fall native species probably holds for central New York as a
whole. In Figure 2 I plotted the number of open-habitat, entomo-
philous species blooming in the entire Cayuga Lake Basin during
each 2-week period over the season (compiled from Wiegand and
Eames 1925). Again, introduced species predominate in early
summer. Later in the summer, native and introduced species are
about equal in number, but the tremendous abundance of goldenrod
(Table I; also Ginsberg 1979, Hurlbert 1970) results in a preponder-
ance of native flowers late in the season.

Foraging phenologies of the most common bee species indicate a
partitioning of the season according to foraging times. Native wild
bees (mostly primitively social halictines) predominated in early
summer, while Apis mellifera predominated in late summer (Table
II). This presents the interesting situation that native bees foraged
primarily on introduced flowers in early summer, while the intro-
duced honey bees foraged on native flowers in late summer and fall
(Table III).

341

1981] Ginsberg — Bee Foraging Patterns

M J J A S O
DATE

Fig. 2. Number of introduced and native flower species blooming in the Cayuga
Lake Basin (compiled from Wiegand and Eames 1925).

Table II. Percent of honey bees in transect samples, 1974

Period

Dates

% honey bees1

22 May-4 June

2.1

5 June-18 June

1.9

19 June-2 July

9.0

3 July- 16 July

7.6

17 July-30 July

13.8

31 July-13 August

15.9

14 Aug-27 August

79.8

119

28 Aug-10 September

95.2

230

11 Sept. -24 Sept.

89.7

Percent of bees captured in transect samples that were Apis mellifera. Other
bees in these samples were native wild bees (except for 3 individuals of Andrew
willcella captured on 28 May, 12 June, and 8 July — this species was probably
introduced into the region).

342

Psyche

[Vol. 88

Table III. Flower species most commonly visited by bees during the summer, 1974

Bee species1

Flower species1

% of visits2

sample size3

Apis mellifera (I)

Solidago altissima (N)

26.2

409

S. graminifolia (N)

24.0

S. juncea (N)

18.8

Ceratina 4 (N)

Rubus allegheniensis (N)

28.0

Halictus ligatus (N)

Chrysanthemum leucan-

themum (I)

51.2

Halictus confusus (N)

Potentilla recta (I)

47.6

Auguchlorella striata (N)

Chrysanthemum leucan-

themum (I)

37.5

Dialictus rohweri (N)

Potentilla recta (I)

47.4

1 Point of origin given in parentheses; N = native to North America; I = introduced.

2 Percent of individuals of that bee species in samples that were on named
flower species.

3 Number of bees of that species in transect samples, 1974.

4 Includes Ceratina dupla and C. calcarata. Females of these species are indis-
tinguishable at present.

Spring-flying bees were not included in Table II because they
foraged on flowers that were most common off the field and could
not be sampled by the transect technique. All-day samples from
patches of common spring flowers revealed a great diversity of
native bees, primarily solitary, univoltine species of Andrena,
Dialictus, and Ceratina. Honey bees were also common in spring,
especially on willows, rosaceous trees, and on large clusters of
dandelion and yellow rocket.

Discussion

The fact that native bees foraged on introduced flowers in early
summer, while introduced bees predominated on native flowers in
late summer, suggests that this type of old field association is quite
recent in origin. Indeed, the development of this curious pattern can
be clarified by tracing the recent biotic history of the Ithaca area.

Early explorers in the region (up until the early 1800’s) reported
extensive forested areas that were thickest near the head of Cayuga
Lake and to the south of Ithaca (Dudley 1886). The Indians cleared
considerable acreages for villages, corn fields, etc. (Day 1953) and
kept corridors of land clear for stalking deer by annual burning

1981]

Ginsberg — Bee Foraging Patterns

343

(Dudley 1886). These cleared areas were probably far less extensive
than present-day open habitats. Also, the deer-stalking grounds
differed from modern old fields because they were burned each year,
and because they lacked many of the introduced flower species that
are now common. Some of these species were introduced by 1807,
when the explorer Frederick Pursh passed through Ithaca (Dudley
1886; Pursh 1923).

The first settlers arrived in Ithaca about 1789 (Dudley 1886). By
the mid 1800’s extensive areas of land had been cleared for farming
and settlements. Total acreage used for farming reached a peak in
New York State (approximately 23,780,754 acres) about 1880. Since
then, gradual abandonment of farmland has given rise to many
abandoned fields. By 1925, only 19,269,926 acres of farmland
remained (Vaughan 1929). By the late 1960’s the area of crop and
pasture land in New York State totalled only about 8,771,800 acres
(Ferguson and Mayer 1970). Much of this farm land was lost to
villages and cities, but a considerable amount was left as abandoned
fields. In the late 1800’s and early 1900’s several weedy species were
introduced, and many others increased in abundance in central New
York. Among the species that became common at this time were
Hieracium pratense and Potentilla recta (Wiegand and Eames
1925), both important species at my sample site (Tables I and III).
Taken together, these facts suggest that the current floral composi-
tion of old-field communities in central New York is on the order of
100 years old.

As a result of these changes in the local flora, at least three new
classes of abundant flower forage have become available to bees. In
spring, the introduced rosaceous trees and roadside weeds provide
considerable forage. Second, the increased acreage of abandoned
fields, along with introductions of several plant species, results in an
historically novel flower bloom in early summer. Finally, the large
acreage of open fields results in an unprecedented profuse bloom of
goldenrod in late summer.

The honey bee was introduced into North America by the early
colonists (Crane 1975). The Italian strain {Apis mellifera ligustica),
which now predominates in New York State, was not introduced
until 1859 (Ruttner 1975). Some more recent introductions into the
Ithaca area include the megachilids Megachile rotundata (Mitchell
1962) and Anthidium manicatum (Pechuman 1967), and the andre-

344

Psyche

[Vol. 88

nid Andrena wilkella (Linsley 1958). At my study site, the honey bee
is far the most abundant of these species (Ginsberg 1979). In the
1950’s, honey bee populations declined sharply in New York State
due to the increased use of pesticides and the decline in farm acreage
devoted to buckwheat, an important food source for honey bees
(Morse 1975). Before 1950, therefore, honey bees were even more
common than at present.

Apis mellifera is a high-density specialist in flower foraging. Its
large colony size and recruitment capabilities facilitate this special-
ization (Eickwort and Ginsberg 1980; Sakagami 1959). In spring,
honey bees forage on high-density resources such as rosaceous trees,
willows, and clusters of roadside herbs. In late summer, honey bees
forage on the super-abundant goldenrods, also high-density re-
sources.

In early summer, honey bees are relatively rare on the old field
(Table II). At this time of season they forage primarily off the field
on high-density resource species in forests and on cultivated fields
(Farrar 1944; Ginsberg 1979). The introduced herbs that bloom at
this time are exploited by primitively social halictines (Table III).
The multivoltine seasonal cycles of these bees allow them to build
up their populations over the season, thus they can exploit the
recently introduced flower species that are now abundant in early
summer. Ceratina, which is probably univoltine in the Ithaca area,
is also common in early summer, but it forages somewhat earlier
than the halictine bees, and is most common on native flowers such
as Rubus spp. (Table III).

An interesting result of this analysis is that each of the major
historically novel instances of resource abundance is exploited by
social bees. Honey bees forage on rosaceous trees and roadside
weeds in spring, and on goldenrods in late summer. Native bees
forage on these flowers also, but honey bees predominate because of
their high populations and recruitment ability, both features related
to their social behavior. Social halictines predominate on intro-
duced herbs in early summer because of their broad host ranges and
their multivoltine seasonal cycles, also related to their sociality. Ap-
parently, the ability to adapt to landscape-level changes in resource
availability is an important advantage that accompanies social
behavior in bees. This does not mean that only social insect species
can adapt rapidly to changes in resource levels. It does suggest that
in bees, sociality facilitates this rapid adaptability.

1981]

Ginsberg — Bee Foraging Patterns

345

Conclusions

In an abandoned field in central New York State, native bees
foraged predominantly on introduced flower species in early sum-
mer, while the introduced honey bee predominated on native
goldenrods in late summer. This situation results from recent
changes in the flora and fauna of the region.

The activities of European settlers have caused large-scale changes
in the flora of the northeastern United States. These changes result
primarily from introductions of alien species, and from clearing of
land for farming with subsequent abandonment. At present, there
are at least three instances of profuse flowering over the season that
are historically novel to this area. These are the abundant bloom of
introduced trees and roadside weeds in spring, the flowering peak of
introduced weeds in early summer, and the profuse flowering of
native goldenrods in late summer. In all three of these cases, the
predominant foragers are social bees; honey bees in spring and late
summer, and social halictines in early summer. The ability of these
bees to exploit historically novel pulses of flowering results from
features related to their social behavior; large colony size and
recruitment ability in Apis mellifera, and the multivoltine seasonal
cycle in the social halictines.

Acknowledgments

I thank G. C. Eickwort, F. C. Evans, R. A. Morse, and R.
Nowogrodzki for constructive comments on early drafts of the
manuscript. F. J. Rohlf provided useful advice. I also thank W.
Denison and family for providing their land as a study site. This
study was conducted in partial fulfillment of the requirements for
the Doctor of Philosophy at Cornell University. Funding was
provided, in part, by NSF grant no. BMS-72-02386 to G. C.
Eickwort.

Literature Cited

Crane, E.

1975. The world’s beekeeping — past and present. In Dadant and sons (eds.)
The Hive and the Honeybee. Hamilton, 111.: Dadant and sons, pp. 19-38.

Day, G. M.

1953. The Indian as an ecological factor in the northeastern forest. Ecology
34:329-346.

346

Psyche

[Vol. 88

Dudley, W. R.

1886. The Cayuga flora. Bull. Cornell Univ. (Science) 2:1-132.

Eickwort, G. C. and H. S. Ginsberg

1980. Foraging and mating behavior in Apoidea. Ann. Rev. Entomol. 25:421-
446.

Farrar, C. L.

1944. Productive management of honeybee colonies in the northern states.
USDA circ. No. 702. 28 pp.

Ferguson, R. H. and C. E. Mayer

1970. The timber resources of New York. USDA Forest Service Resource
Bull. NE-20, 193 pp.

Fernald, M. L.

1950. Gray’s Manual of Botany. 8th ed. New York: D. Van Nostrand. 1632 pp.
Ginsberg, H. S.

1979. Foraging ecology of pollen utilizing insects on an old field in central New
York State. Ph.D. Thesis, Cornell Univ. Ithaca, N.Y. 221 pp.
Hurlbert, S. H.

1970. Flower number, flowering time, and reproductive isolation among ten
species of Solidago (Compositae). Bull. Torrey Bot. Club 97:189-195.
Linsley, E. G.

1958. The ecology of solitary bees. Hilgardia 27:543-599.

Mitchell, T. B.

1962. Bees of the eastern United States, Vol. II. N. Carolina Agr. Exp. Sta.,
Tech. Bull. No. 152. 557 pp.

Morse, R. A.

1975. Bees and Beekeeping. Ithaca, N.Y.: Cornell Univ. Press. 295 pp.
Pearson, J. F. W.

1933. Studies on the ecological relations of bees in the Chicago region. Ecol.
Monogr. 3:375-441.

Pechuman, L. L.

1967. Observations on the behavior of the bee Anthidium manicatum (L.). J.
N. Y. Entomol. Soc. 2:68-73.

PURSH, F.

1923. Journal of a botanical excursion in the northeastern parts of the states of
Pennsylvania and New York during the year 1807. Dehler Press. 1 13 pp.
Ruttner, F.

1975. Races of bees. In Dadant and sons (eds.) The Hive and the Honeybee.
Hamilton, 111.: Dadant and sons, pp. 19-38.

Sakagami, S. F.

1959. Some interspecific relations between Japanese and European honeybees.
J. Animal Ecol. 28:51-68.

Vaughan, L. M.

1929. Abandoned farm areas in New York. Cornell Univ. Agr. Exp. Stat. Bull.
490:1-285.

WlEGAND, K. M. AND A. J. EAMES

1925. The flora of the Cayuga Lake Basin. Cornell Univ. Agr. Exp. Sta. Mem.
92:1-491.

MYRMECOPHILIC RELATIONSHIP OF PELLA
(COLEOPTERA: STAPHYLINIDAE) TO
LASIUS FULIGINOS US (HYMENOPTERA: FORMICIDAE)

By B. Holldolber*, M. Moglich**, U. Maschwitz***
Introduction

A large number of staphylinid beetles are closely associated with
ants and termites (for review see Wilson 1971, Kistner 1979). Those
species living with ants are commonly called myremcophiles. At
least a few ( Atemeles , Lomechusa ) have “broken” the communica-
tion code of their host species and are thereby able to become
completely integrated in the social system of the ants (Holldobler
1967, 1970, 1971). In an attempt to understand the evolutionary
pathways of this highly specialized social parasitic behavior, we
studied closely related staphylinid species that do not live within the
ant society but instead occupy the foraging trails and garbage
dumps of an ant nest.

Many of such myrmecophilous staphylinids can be found with the
formicine ant Lasius fuliginosus and most of them belong to the
genus Pella. Apparently these beetles are not endowed with the
behavioral repertory that would enable them to live within the ant
colony, although they seem to have a close ecological association
with ants (Holldobler 1972).

Kistner (1971) redefined the genus Zyras and raised the former
subgenus Pella to generic rank. The first behavioral observations
concerning Pella ( =Zyras , Myrmedonia ) were published by Was-
mann (1886, 1930). He stated that these beetles feed on dead or
disabled ants, but that they also lie in wait near the entrance and
hunt ants returning to the nest. Furthermore, Wasmann pointed out
that because of their generalized and primitive structure these
beetles can be regarded as close to the ancestral forms from which
some of the more specialized staphylinid myrmecophiles were
derived.

♦Department of Biology, Harvard University, Cambridge, Mass., USA.
♦♦Present address: Am Lowentor 15, Darmstadt, W. -Germany.
♦♦♦Fachbereich Biologie (Zoologie) der Universitat Frankfurt, W. -Germany.
Manuscript received by the editor October 19, 1981.

347

348

Psyche

[Vol. 88

Material and Methods

At our major study sites near Ochsenfurt, Riederau (both Bavaria,
W. -Germany) and Gravenbruch (Hessen, W. -Germany) we found
12 staphylinid species associated with Lasius fuliginosus (Tab. 1).
Our investigations concentrated on the genus Pella (mainly P.
funesta, P. laticollis and P. cognata ). Since the myrmecophilous
behavior of these species was found to be very similar we will refer
to the individual species only where necessary. In fact, when
observing the beetles in the field it was usually not possible to
identify the species precisely. We made additional observations with
Pella humeralis, which can be found with L. fuliginosus, but which
often also occurs near the nests of Formica polyctena (Wasmann
1920; Kolbe 1971).

The field observations were conducted throughout the years
1967-1969, and sporadically in 1970-1973. In an attempt to follow
the life cycle of the beetles in the laboratory we set up a large colony
of Lasius fuliginosus in a laboratory nest. The culture and mainte-
nance of these ants over a longer period of time was particularly
difficult, because L. fuliginosus constructs carton nests with the aid
of a special symbiontic fungus ( Cladosporium myrmecophilum) . A
detailed description of the nest building behavior of L. fuliginosus
and of the laboratory nest is given in Maschwitz and Holldobler
(1970).

In order to measure quantitatively possible trophallactic feeding
of the myrmecophiles by their host ants, tracer experiments were
carried out using the radioisotope 32P mixed with honey-water. The
quantity of marked food taken up by the ants was reflected in the
counts per minute which were determined with a standard Geiger-

Table l

Staphylinids found near one nest of Lasius fuliginosus.

Pella laticollis
Pella lugens
Pella cognata
Pella funesta

Pella humeralis

Oxypoda vittata
Rugilus rufipes
Thiasophila inquilina
Homoeusa acuminata
Sipalia circellaris
A theta fungi
A theta sodalis

1981]

Holldobler, Moglich, & Maschwitz — Pella

349

Muller counter combined with an automatic sample changer
(Philips). For further information concerning the tracer techniques
applied in this study see Gosswald and Kloft 1958; Kloft 1959.

For histological investigations live specimens were fixed in alco-
holic Bouin (Dubosq Brasil) or Carnoy (Romeis 1948), embedded in
Methyl Methacrylate, and sectioned 5-8^ thick with a Jung Tet-
rander microtome (Rathmayer 1962). The staining was Azan
(Heidenhain).

For the chemical analysis of the defensive secretions of Pella ,
liquid material was collected with glass capillaries from the dissected
glandular reservoirs or washed with water from the surface of the
irritated beetles. The quinones were identified by thinlayer chroma-
tography as 2,4 - dinitrophenylhydrazine in 2 N hydrochloric acid or
by reduction with sulphurous acid. The dinitrophenylhydrazones
were separated on alumina F 254 (Merck) with chloroform meth-
anol (19:1) as mobile phase and on silica gel F 254 (Merck) with
benzene-ethyl acetate (4:1) as mobile phase (Moore 1968). The
hydroquinones were separated on silica gel F254 with benzene
dioxane (3:1) as mobile phase and then sprayed with a solution of
0.5% hydrogen peroxide and a solution of peroxidase. The newly
formed quinones were made visible by spraying with DNP and
treating with ammonia vapour (Schildknecht and Kramer 1962).
Hydrocarbons, terpenes and carbonic acids were analyzed by GLC.
We used a Perkin Elmer chromatograph, model 300, equipped with
a flame ionization detector. Columns: 1.8m X 2.7mm stainless steel,
packed with a) 4% polypropylene glucol on Chromosorb G (100°C
column temperature); b) 4% polyethyleneglycol 1500 on Chromo-
sorb G (70° C); c) 25% diethylhexyl sebacinate plus sebacinic acid
on Kieselgur 60-100 (140° C) (30 ml N2/min; FID).

Results

The life cycle of Pella funesta

The following description of the life cycle of Pella funesta is based
on field observations and on data obtained from laboratory cul-
tures. Pella laticollis appears to have a similar life cycle, but our
observational data are not as complete as for P. funesta.

In late March and early April a large number of P. funesta beetles
were typically found in the excavation material on the base of the
trunks of L. fuliginosus nest trees. At this time most of the beetles

350

Psyche

[Vol. 88

were lying motionless in the loose material and showed a kind of
“dormance posture”: the abdomen was bent over its back, with the
legs and antennae folded tightly to the body. On warmer days,
however, the beetles exhibited high locomotory and flight activity,
and in the laboratory they showed a strong positive phototaxis.
During this period we frequently observed beetles copulating in the
laboratory nests. Toward the end of April the sexual behavior and
flight activities ceased. In the laboratory as well as in the field the
beetles were now active primarily during the night, while during the
daytime they clustered under shelters near the Lasius fuliginosus
nest. Only occasionally were we able to spot a beetle outside the
shelters at daytime.

Also near the end of April we found the first beetle eggs in the
“garbage dumps” of the laboratory nest of L. fuliginosus , and by
early May the first Pella larvae had hatched. The larvae developed
quite rapidly, so that in mid-May we found the first pupae in the
“garbage dumps” of the ant nests, even though larvae could still be
found throughout the months of June and July. In June the
mortality of adult beetles in our laboratory nest increased markedly
and in late July and August the first young beetles eclosed from
their pupae. These beetles, as well as those collected in the field in
early August, exhibited strong positive phototaxis and high flight
activity for a few days. After this short period, however, the beetles
were primarily active at night and during the day they stayed in
shelters. Finally, in October, the number of beetles found outside
the ant nest declined markedly and by November no more beetles
could be found outside the nest. In December we excavated to L.
fuliginosus nests. In both nests we found several Pella beetles in
“dormance position” covered by loose nest material of the peripher-
al nest chambers and on the ground inside the nest tree trunk.
Presumably these beetles were overwintering within the Lasius
fuliginosus nest until their activity period would start again in early
spring the coming year.

From these observations we propose the following life cycle for
Pella funesta : in early spring the adult beetles deposit eggs near the
ants’ “garbage dump” area. The larvae develop in the “garbage
dump”, pupate during the period from May to July and between
July-August the adult beetles eclose. After eclosion the young
beetles apparently migrate, as indicated by the short period of high
diurnal locomotory and flight activity. After this period the beetles

1981]

Holldobler, Moglich, & Maschwitz — Pella

351

forage near the L. fuliginosus nest during the night and stay in
shelters during the day. They overwinter in dormancy inside the L.
fuliginosus nest. With the end of winter the beetles enter a second
diurnal activity phase during which mating takes place. After repro-
duction the beetles die, normally a few weeks before the new beetle
generation ecloses in June.

The behavior of the larvae of Pella

The description of the behavior of the larvae is primarily based on
observations in the laboratory. In the field and in the laboratory
nest, the larvae were almost exclusively found near or in the
“garbage dumps” of the L. fuliginosus nests. We frequently ob-
served the larvae feeding on dead ants (Fig. IB). During feeding the
larvae always attempted to stay “out of sight” either by remaining
entirely beneath the booty and devouring it from below or by
crawling inside the dead ant’s body. Occasionally 2-4 larvae could
be observed feeding on the same ant cadaver. But when they became
too crowded they frequently attacked each other, sometimes result-
ing in one larvae eating the other (Fig. 1C).

When ants encountered the larvae they usually attacked them.
Almost invariably the larvae reacted with a typical defense beha-
vior. They raised their abdominal tip towards the head of the ants.
Usually the ant responded by stopping the attack and palpating the
larva’s tip (Fig. 1A). In most cases this short interruption was
enough to allow the larvae to escape. We observed hundreds of such
encounters between ants and Pella larvae; only a few ended fatally
for the larvae.

Histological invesigations revealed that the Pella larvae have a
complex dorsal glandular structure with a reservoir near the
abdominal tip in the second last segment. In addition we found
single cell glands positioned dorso-laterally in each segment. Similar
glandular structures had previously been found in larvae of the
myrmecophile staphylinids Atemeles and Lomechusa, and circum-
stantial evidence strongly indicated that in these species the glands
produce so-called pseudopheromones which release adoption beha-
vior in the host ants (Holldobler 1967). We have no evidence to
suggest that these glands have a similar function in Pella. However,
it is possible that the more complex glandular structure at the
abdominal tip, produces an appeasement secretion by which the
aggressiveness of attacking ants can be briefly blunted.

352

Psyche

[Vol. 88

Fig. 1 Behavior of the larvae of Pella. A) Larva presenting abdominal tip to an
attacking worker of Lasius fuliginosus. The ant interrupts attack and licks at the
larva. B) Larva feeds on dead ant. C) Cannibalistic behavior of Pella larvae.

1981]

Holldobler, Moglich, & Maschwitz — Pella

353

In any case, Pella larvae are able to come into close contact to
workers of L. fuliginosus without being attacked, especially when
the temperature is low (14-17° C). Under those circumstances we
have seen the beetle larvae licking the cuticle of live ants, including
even the mandibles and mouthparts. This led us to the question of
whether the beetle larvae occasionally solicit regurgitation in ants.
In order to investigate this possibility ants were fed with honey-
water labeled with the isotope 32P and then housed together with
beetle larvae. For each sample we kept 30 radioactive ants with 5
beetle larvae in plastic containers (10 X 15 cm) with a moist gypsum
bottom. One experimental series was conducted in a temperature
range of 14.5-16.5° C, the other in 20-23° C. After 24 hours we
measured the amount of radioactivity in each individual ant and
larva. No significant amount of radioactivity had been transferred
from the ants to the larvae, except in container 6, where one ant was
found dead and obviously partly eaten by the larvae. Since the
amount of radioactivity carried by some of the larvae was only very
slightly above the background activity, we concluded that it was
transferred by contamination. From this experiment it appears that
the Pella larvae do not solicit regurgitation in ants. Their main food
source seems to be dead ants or debris of the ants. In fact, they can
easily be raised by keeping them entirely separated from living ants,
just by feeding them regularly with dead ants.

Predatory behavior of adult beetles
Since Wasmann’s early observations (1886, 1920, 1925) very little
has been reported concerning the biology of the myrmecophilous
Pella. Wasmann reported that all species he had studied (P.
humeralis, P. funesta, P. cognata, P. similis, P. lugens and P.
laticollis ) live with Lasius fuliginosus , and only P. humeralis can
also be found with species of the Formica rufa group. According to
Wasmann all these Pella species prey on ants, concentrating espe-
cially on disabled ants. In addition Wasmann observed that the
beetles are active primarily during the night. In a more recent
publication Kolbe (1971) failed to find a predatorial behavior in P.
humeralis and concluded that this species primarily feeds on dead
ants. Similar observations were made with Pella japonicus, which
lives with Lasius spathepus (= L. fuliginosus var. spathepus
Wheeler) (Yasumatsu 1937; Kistner 1971). Kistner also observed
that these beetles “ate small insects that are being transported by the

354

Psyche

[Vol. 88

ants”. However, he could not “see the Pella eating live ants or
fighting any of the ants on the trail”.

Our observations of Pella funesta, P. laticollis and P. humeralis
confirmed that these species live as scavengers, feeding on dead or
disabled ants and debris discarded by the ants. However, we also
observed these beetles acting as very effective predators on the ants.
Most of the following studies were made with P. laticollis and P.
funesta.

During the main foraging season from May to October Lasius
fuliginosus is active day and night. Foragers travel along well
established trunk trails to feeding sites which are sometimes more
than 40 m distant. At daytime we only occasionally saw Pella beetles
moving along or nearby the trail. However, when we watched the
trunk trails with a flash light at night many Pella were seen running
along the ants’ foraging routes. Although most beetles were found
within a range of 5 m from the nest tree of L. fuliginosus , we also
found beetles on the trunk trail as far as 22 m away from the nest.

On 6 different occasions we witnessed Pella beetles hunting L.
fuliginosus workers at night on the foraging trail. When an ant was
killed it was dragged a few centimeters away from the trail and eaten
under a shelter, sometimes by several beetles simultaneously.

More detailed observations on the behavioral interactions of
Pella and L. fuliginosus were made in the laboratory. As long as
enough dead ants were available at the ants’ nest midden, the beetles
showed no predatory behavior at all, limiting themselves to a diet of
ant cadavers (Fig. 2A). But when the beetles were starved for a few
days and then placed together with ants in an observation arena, the
predation by Pella became strikingly prominent, although the time
of onset was often very unpredictable. We saw the beetles hunting
during the daytime, but we observed such activity most frequently in
the evening or at night. The beetles chased after individual ants and
pursued them through approximately 2-6 cm (very rarely through
longer distances than that). When the beetle moved directly behind
the ant it attempted to mount it and insert its head between the ant’s
head and thorax. When attacked the ant usually reacted by
suddenly stopping and pressing the femur rapidly and tightly to its
body (Fig. 5). Often this reaction threw the beetle off the back of the
ant, allowing the ant to escape. In one series of observations we
counted 178 beetle onslaughts on L. fuliginosus workers within a
period of 3 hours; of these, only 9 attempts (5%) were successful.

1981]

Holldobler, Moglich, & Maschwitz — Pella

355

Fig. 2 A) Pella beetle feeding on dead L. fuliginosus worker. Frequently the
beetles lick first the mouth parts of the ants, before tearing the cadavers apart. They
might be attracted to mouthparts by sweet remainders of honeydew. B) Cluster of
Pella beetles around a prey object.

356

Psyche

[Vol. 88

Fig. 3 Photograph and drawing of Pella attacking live worker of L. fuliginosus.

1981]

Holldobler, Moglich, & Maschxvitz — Pella

357

The hunting behavior of the beetles was always the same in P.
laticollis, P. funesta and P. humeralis : the beetle attacked from
behind and always attempted to insert its head between the head
and thorax of the ant. We inspected several ants which had just been
immobilized by an attack of a beetle and found that in most cases
the pronotum was widely separated from the head and usually the
oesophagus and connectives of the nervous system were cut.

Occasionally we observed 2-3 beetles chasing behind one ant
(Fig. 4). Once the ant was caught by a beetle the other beetles joined
in subduing and killing the ant. Although individual beetles often
tried to drag the prey away from the rest of the “hunting pack”,
usually several beetles fed on the prey simultaneously. No aggres-
sion among the beetles was observed in this situation. However,
when the beetles were starved for several days and were kept
without ants, they occasionally chased each other, jumping on each
other’s back as they normally did when hunting ants. But we never
saw cannibalistic behavior among the adult beetles, even when the
beetles were densely crowded around a prey object (Fig. 2B).

Defense and appeasement behavior in adult beetles
Defense with tergal gland secretion:

Usually the Pella beetles run around with their abdomen curved
slightly upwards. When encountering an ant, the beetles flex the
abdomen even more strongly. This is a typical and frequently
described behavior of many staphylinid myrmecophiles and is
commonly considered a defense response (Wasmann 1886, 1920;
Jordan 1913; Patrizi 1948; Koblick and Kistner 1965; Pasteels 1968;
Holldobler 1970, 1972; Kolbe 1971). It has been suggested that
during this abdominal flexing the beetles discharge secretions from
their tergal gland (Jordan 1913; Kistner and Blum 1971).

The tergal gland is located between the sixth and seventh
abdominal tergites (Fig. 6), and is unique to the subfamily Aleo-
charinae (Jordan 1913; Pasteels 1968). The chemistry of the tergal
gland secretions of several species has been investigated and found
to be extraordinarily diverse (Blum et al. 1971; Brand et al. 1973;
Kolbe and Proske 1973).

Kistner and Blum (1971) suggested that Pella japonicus and
possibly also P. comes , both of which live with Lasius spathepus,
produce citronellal in their tergal glands. This substance is also a

358

Psyche

[Vol. 88

Fig. 4 Sequence of group hunting behavior of Pella. A) Two beetles chase a
forager of L. fuliginosus. One of the beetles is jumping on the back of the ant. B) The
ant has been captured and subdued by both beetles. C) A third beetle is joining the
hunting group.

1981]

Holldobler, Moglich, & Maschwitz — Pella

359

Fig. 5 A) Pella in “death feigning” position. B) A “death feigning” beetle is
carried around by a L. fuliginosus worker. C) Pella presents abdominal tip to
attacking L. fuliginosus worker. The ant licks at the abdominal tip.

360

Psyche

[Vol. 88

major compound of the mandibular gland secretions of their host
ants, for which it functions as an alarm pheromone. Although no
Pella tergal gland contents were available for chemical analysis,
because irritated beetles seemed to smell like the ants’ mandibular
gland secretion, Kistner and Blum speculated that Pella produce in
their tergal glands citronellal and thereby mimic the alarm phero-
mone of their host ants. They suggested that in this way the beetles
can “cause the ants to reverse their direction; a reaction which
allows the myrmecophiles to escape”.

Our investigations of the defensive strategy employed by the
European Pella towards their host ants Lasius fuliginosus led to

m ]j

Fig. 6 Schematical drawing of a Pella beetle indicating the position of the
exocrine glandular complexes. TG: tergal gland; AG: appeasement gland complex.

1981]

Holldobler, Moglich, & Maschwitz — Pella

361

different results. Pella laticollis , when irritated mechanically dis-
charges a pungent smelling brownish secretion from its tergal gland
that shows acidic reactions. Only when the beetles were severely
attacked and firmly grasped on their appendages by the ants, could
we smell the tergal gland secretion. We never observed the beetles
employing tergal gland secretion when they were attacking ants.
Ants contaminated with tergal gland secretion usually exhibited a
repellent reaction, releasing the grip on the beetles and grooming
and wiping their mouth parts and antennae on the substrate. But the
beetles had to escape quickly, because other ants close by became
alerted and were rapidly approaching the scene, apparently alarmed
by the ants’ alarm pheromone. We noticed that beetles that were
attacked by L. fuliginosus workers often smelled somewhat like the
ants’ mandibular secretions, but the beetles’ tergal gland secretions
clearly smelled differently. Conceivably, some of the attacked
beetles were contaminated with the ants’ strongly smelling mandib-
ular gland secretions.

Our chemical analysis of the tergal gland secretions of P. laticollis
did not reveal a resemblance to the mandibular gland secretions of
L. fulginosus, whose major compounds are farnesal, 6-methyl-5-
hepten-2-one; perillene and dendrolasin, a furan (Quilico et al 1957;
Bernardi et al 1967). When we treated the tergal gland section with
2,4 — dinitrophenyl-hydrazine, we obtained an orange-yellowish pre-
cipitate. This was subjected to thinlayer chromatography in two
separate systems. In each system we obtained two spots. The Rf
values and the color reaction, when treated with ammonia vapour,
identified them as dinitrophenylhydrazones of p-benzoquinone and
p-toluquinone. Furthermore, the chromatography of the hydro-
quinones obtained from the secretion by reduction with SO2 also
demonstrated the presence of p-benzo- and p-toluquinone in the
tergal gland secretion.

For comparison we used thinlayer chromatography to analyze the
dinitrophenylhydrazones of the tergal gland secretion of several
other aleocharine staphylinds found near the nests of L. fuliginosus.
Pella humeralis, A theta fungi and Sipalia cireellaris also produce
benzo- and toluquinone; in Oxypoda vittata we found only tolu-
quinone.

In addition Kolbe and Proske (1973) identified isovaleric acid in
the tergal gland secretion of P. humeralis , and with the aid of gas
chromatography we detected saturated hydrocarbons and short

362

Psyche

[Vol. 88

,o O
ir C

>-> D

3 ’rt

Xi >

c C

1981]

Holldobler, Moglich, & Maschwitz — Pella

363

chained fatty acids in the secretion of P. laticollis (Tab. 2). However,
in none of the species could we find citronellal.

The common presence of quinones in the tergal gland secretions
of Pella and the related aleocharine species agree with the previous
findings by Blum et al (1971), who found the tergal gland secretion
of Lomeehusa strumosa to contain benzoquinone, methyl-benzo-
quinone, ethyl-benzoquinone and n-tridecane, the latter substance
accounting for more than 80% of the volatiles detected in the
secretion. In addition Brand et al. (1973) analyzed the tergal gland
secretion of Drusilla canaliculata, also an aleocharine beetle, finding
quinones and hydroquinones together with alkanes, saturated and
unsaturated aliphatic aldehydes. Pasteels (1968) demonstrated that
D. canaliculata effectively employs the tergal gland secretion as a
repellent-defense weapon against ants in a similar fashion as we
described it for Pella.

Although we could not find any resemblance of the Pella tergal
gland secretions to the mandibular gland secretions of Lasius
fuliginosus, it was noteworthy that the Pella secretions contained
undecane, a hydrocarbon commonly found in the Dufour’s glands
of formicine ants (for review see Blum and Hermann 1978) and
considered to be an alarm pheromone in L. fuliginosus (Dumpert
1972). However, isolated tergal gland secretions of P. laticollis
elicited a repellent reaction rather than an alarm response in L.
fuliginosus. Apparently the repellent effect of the quinones in the
secretions is stronger than a possible alarming effect released by
undecane. In fact, when the ant’s antennae were directly contam-
inated with the beetles’ tergal gland secretions the antennae were
hanging almost motionless and flabby and the ant appeared dis-
oriented for several minutes. From all our laboratory tests it
appears obvious that the tergal gland secretions of Pella functions as
a powerful chemical defense weapon against attacks by ants.

Appeasement behavior:

When foraging on the ants’ “garbage dumps” or running along
the ants’ trails, Pella frequently encounter ants. Yet we were
impressed by the scarcity of their application of the tergal gland
defensive system. Much more frequently the beetles employed an
appeasing defensive strategy, and the repellent defense seemed to be
employed only as a last resort.

364

Psyche

[Vol. 88

Fig. 7 A,B) Transversal section through glandular epithelium in 9th sternum of
P. humeralis. C) Sagital section through sternal gland in 7th sternum of P.
humeralis. GC: glandular cell; P: pore in cuticle.

1981]

Holldobler, Moglich, & Maschwitz — Pella

365

It was especially common in early spring, when most of the
beetles were found close to the entrance of the ants’ nest, that the
beetles showed “death feigning” behavior, when attacked by ants.
They fell to the side, the legs and antennae folded tightly to the body
and the abdomen curved upwards (Fig. 5A). The ants either ignored
these motionless beetles or carried them around and finally dis-
carded them on the “garbage dump”. But only rarely did they injure
the beetles (Fig. 5B).

Later in the year, when the activity of ants and beetles was much
higher, the beetles employed a different appeasement technique. As
mentioned before, we only very rarely saw the discharge of tergal
gland secretions by the beetles, although every time they en-
countered ants they flexed their abdomen and pointed with the
abdominal tip toward the head of their adversaries. Usually the ants
responded by antennating the tip and briefly licking it (Fig. 5C).
This ordinarily slowed down the ants’ aggression and the beetles
used the ants’ distraction to escape. Occasionally, when the ants
remained very persistent, a white, viscous droplet appeared at the
abdominal tip, whereupon the ants usually very eagerly licked it up.
This appeasing defensive behavior was much more common during
the interactions between Pella and Lasius fuliginosus than the
repellent defense. For a series of simulation experiments we cut off
the last 3 segments of the abdomen of freshly killed P. laticollis,
sealed the cut with wax, pinned the segments on dissecting needles
and presented these “dummies” to the ants. In a total of 60 tests
(using 3 different dummies) the ants interrupted their run in 47 cases
(78%) and licked the abdominal tip briefly.

Histological investigations revealed that the abdominal tips of
Pella are batteries of exocrine glandular structures, all of which
together we call the appeasement gland complex. In the following
section we give a brief description of the glands which could be
involved in the appeasement behavior.

The most comprehensive study of the glandular morphology of
some termitophilous and myrmecophilous aleocharine beetles has
been published by Pasteels (1968). From this work we learned that
these beetles possess a surprising variety of exocrine glandular
structures and that various species can differ considerably in their
glandular systems. In the four species of Pella ( P . cognata, P.
funesta, P. humeralis, P. laticollis ) we investigated, we did not find
major differences, although P. humeralis appeared to be somewhat

366

Psyche

[Vol. 88

Fig. 8 Sagital section through abdominal tip of P. humeralis. GC: glandular cells
under 10th tergite near anus. A: anus.

more richly endowed with hypodermal glandular cells, especially in
the area of the paratergites.

In staphylinid beetles the first fully developed abdominal seg-
mental ring (tergite plus sternite) is usually considered to be the 3rd
abdominal segment (Blackwelder 1936). All Pella species have a well
developed compound tergal gland between the 6th and 7th tergites
(Fig. 6) as described by Jordan (1913), Pasteels (1968) and Holl-
dobler (1970). We have also detected glandular cells located pri-
marily in the 7th segment, which Pasteels (1968) calls postpleural
glands. According to Pasteels the glandular channels associated
with these cells open dorsolaterally through the pleural membrane
between the 7th and 8th segments. Pasteels could clearly see these
openings in several species (for example in Gyrophanaena affinis ),
but not in Pella (Zyras) humeralis. In a series of longitudinal, trans-
versal and frontal sections, we too were unable to detect the external
openings of these glandular cells.

At the anterior edge of the 4th, 5th, 6th and 7th sternites are
found clusters of glandular cells that open through pores in the
cuticle (Fig. 7A). They are especially well developed in the 7th
sternite. Pasteels (1968) assumes that the secretions of these glands

1981]

Holldobler, Moglich, & Maschwitz — Pella

367

Fig. 9 Sagital sections through abdominal tip of Pella laticollis female. GC:
glandular cell clusters; M: membrane; CH: glandular channels.

368

Psyche

[Vol. 88

serve primarily for lubrication to alleviate friction between the
sternal sclerites when the beetles flex their abdomen. But more than
any other of the abdominal segments the last 3 tergites (VII, IX, X)
(Fig. 6) and the 8th and 9th sternites are richly endowed with
glandular epithelia, the individual cells of which open through pores
in the cuticle (Fig. IB, 1C). The last two segments can be telescoped
with especial ease into the preceding segments, and during the
appeasement process the beetles often move them slightly back and
forth. Furthermore, there are clusters of glandular cells with longer
channels under the 10th tergite near the anus (Fig. 8). They resemble
the type of cells that Holldobler (1971) located in the same position
in Atemeles and called pygidial glands. We have, however, aban-
doned this term, because it is very confusing, especially in the
Aleocharinae, where the last visible tergite is usually not the 8th
tergite (often called pygidium in the Coleoptera) but the 10th tergite.

In addition to these hypodermal glandular structures, females and
males possess special exocrine glandular complexes that might be
involved in the reproductive processes but which could also play a
role in the myrmecophilous behavior of the beetles. In the 9th
sternite of females there are several clusters of glandular cells, the
channels of which open through the intersegmental membrane at
the tip of the abdomen and near the oviduct (Fig. 9). Males have
similar glands in the 9th sternite which also open through the
intersegmental membrane near the posterior part of the genital
chamber (Fig. 10). Furthermore, males possess a very large gland-
ular complex, consisting of numerous tightly packed glandular cells
each connected with a long channel that open dorsally in bundles
through a membrane at the genital chamber (Fig. 11). We assume
that the secretions of this gland flow into the genital chamber.
Females do not have this gland, but the spermathecal gland has a
very smilar appearance.

Finally, the hindgut might also be involved in the appeasement
process. On several occasions we observed that beetles, upon
presenting their abdominal tip to the ants, released a droplet at the
anus that was licked up by the ants.

Discussion:

Some of the most advanced myrmecophilic relationships are
found in the aleocharine beetles Lomechusa and Atemeles. We

1981]

Holldobler, Moglich, & Maschwitz — Pella

369

Fig. 10 A) Sagital sections through the abdominal tip of P. humeralis male. T:
9th and 10th tergite; S: 9th sternite; AE: aedeagus. B) Close-up of sagital section
through 10th tergite and 9th sternite. GC: glandular cell clusters.

370

Psyche

[Vol. 88

Fig. 1 1 A) Glandular complex in males of Pella humeralis, located dorsally of the
genital chamber. B) Opening of the bundles of glandular channels through mem-
brane; presumably into the genital chamber.

1981]

Holldobler, Moglich, & Maschwitz — Pella

371

know from observations by Wasmann, made more than 60 years
ago, that these beetles are both fed and reared by their host ants.
Both chemical and mechanical interspecific communication is in-
volved in these unusual relationships. These aleocharines have
broken the communication code of their host ants and are thereby
able to live as parasites within the social system of the ant colony
(Holldobler 1967, 1970, 1971, 1972).

Species of the genus Pella are less advanced in their myrmeco-
philic relationships. Rather than occupy the brood chambers of the
ant nest, they live as scavengers and predators in the peripheral
zones around the nest, at the garbage dumps, and on the trunk
routes. Some of the behavioral features of Pella , however, seem to
be very similar to those of Atemeles and Lomechusa. In fact, these
behavioral patterns might be preadaptations for the evolution of a
highly advanced myrmecophilic relationship in the aleocharine
beetles. In particular, the appeasement behavior appears to be an
important prerequisite for living closely with ants. This “gentle”
defense technique does not cause excitement in the ants, as a
repellent defense would do.

Indeed, our observations indicate that Pella only rarely employ
their strongly smelling tergal gland secretions when they are near the
host ant colony. This defense system might be used more during the
migration phase, when the beetles can be attacked by individual
foraging ants. Similar results were previously obtained with Ate-
meles (Holldobler 1970) and Lomechusa (Holldobler unpublished).
In the presence of their host ants these species use the appeasement
defense almost exclusively.

The appeasement behavior also plays an important role during
the adoption of Atemeles by their host ants. When encountering a
worker of the host species near the ant’s nest, the beetle first offers
the appeasement gland complex (at the abdominal tip) to the ant.
This apparently suppresses aggressive behavior in the ant; only then
does the beetle lower its abdomen to permit the ant access to the
adoption glands, which are located in the paratergites. The glandu-
lar openings are surrounded by bristles. These are grasped and used
by the ant to carry the beetle into the brood chamber. While being
carried, Atemeles adopts the same posture as that used by Pella
during the “death feigning” behavior. As we have noted, the initially
aggressive ants respond by either ignoring the beetles or else picking

372

Psyche

[Vol. 88

them up and carrying them around until they eventually discard
them, usually unharmed, at the garbage dump. It is conceivable that
the carrying posture of Atemeles has evolved from a defensive
“death feigning” behavior employed by less advanced ancestral
species.

Finally, Pella beetles do not have adoption glands associated with
trichrome bristles. It is most likely, however, that the small clusters
of glandular cells in the paratergites (for example in P. humeralis )
represent morphological precursors of the massively developed
adoption glands in Atemeles and Lomechusa.

Acknowledgements

We would like to thank Hiltrud Engel for technical assistance,
Dr. V. Puthz for identifying the beetles, Ruiko Pierce for translating
the work of Yasumatsu from Japanese into English, and A1 Newton
and Margaret Thayer for helping in disentangling the segmental
morphology of Pella. This work was supported by grants from the
Deutsche Forschungsgemeinschaft and National Science Founda-
tion.

References

Bernardi, R., C. Cardani, D. Ghiringhella, A. Selva, A. Baggini, M. Pavan
1967. On the components of secretion of mandibular glands of the ant Lasius
(Dendrolasius) fuliginosus. Tetrahedron Lett. 40, 3893-3896.

Blum, M. S., R. M. Crewe, J. M. Pasteels

1971. Defensive secretions of Lomechusa strumosa a Myrmecophilous Beetle.
Ann. Entomol. Soc. America 64, 975-975.

Blum, M. S., H. R. Herman

1978. Venoms and Venom Apparatuses of the Formicidae: Myrmeciinae,
Ponerinae, Dorylinae, Pseudomyrmecinae, Myrmicinae and Formi-
cinae; in Handbook of Experimental Pharmacology 48, pp. 801-869.
Springer Verlag, Berlin, Heidelberg, New York.

Brand, J. M., M. S. Blum, H. M. Fales and J. M. Pasteels

1973. The Chemistry of the Defensive Secretion of the Beetle Drusilla
canaliculata. J. Insect Physiol. 19, 369-382.

Dumpert, K.

1972. Alarmstoffrezeptoren auf der Antenne von Lasius fuliginosus (Latr.)
(Hymenoptera, Formicidae). Z. Vergl. Physiol. 76, 403-425.

Gobwald, K., Kloft, W.

1958. Radioaktive Isotope zur Erforschung des Staatenlebens der Insekten.
Umschau, 58, 743-745.

1981]

Holldobler, Moglich, & Maschwitz — Pella

373

Holldobler, B.

1967. Zur Physiologie der Gast-Wirt-Beziehungen (Myrmecophilie) bei Amei-
sen. I. Das Gastverhaltnis der Atemeles- und Lomechusa Larven (Col.
Staphylinidae) zu Formica (Hym. Formicidae). Z. Vergl. Physiol. 56,
1-21.

1970. Zur Physiologie der Gast-Wirt-Beziehungen (Myrmecophilie) bei Amei-
sen. II. Das Gastverhaltnis des imaginalen Atemeles pubicollis Bris.
(Col. Staphylinidae) zu Mvrmica und Formica (Hym. Formicidae). Z.
Vergl. Physiol. 66, 176-189.

1971. Communication between ants and their guests. Sci. American, March
1971. 86-91.

1972. Verhaltensphysiologische Adaptationen an okologische Nischen in
Ameisennestern. Verhandlungsbericht der Dtsch. Zool. Ges. 65. Jah-
resvers. 137-143.

Jordan, K. H. C.

1913. Zur Morphologie und Biologie der myrmecophilen Gattungen Lome-
chusa und Atemeles und einiger verwandter Formen. Z. Wiss. Zool. 107,
346-386.

Kistner, D. H.

1971. Studies of Japanese Myrmecophiles Part I. The Genera Pella and
Falagria (Coleoptera, Staphylinidae) in Entomological Essays to Com-
memorate the Retirement of Professor K. Yasumatsu, pp. 141-165.
Hokurynkan Publ. Co. Ltd. Tokyo.

1979. Social and Evolutionary Significance of Social Insect Symbionts. In
Social Insects (vol I) (ed. H. R Hermann) pp. 340-413, Academic Press,
Inc. New York, San Francisco, London.

Kistner, D. H., M. S. Blum

1971. Alarm Pheromone of Lasius (Dendrolasius) spathepus (Hymenoptera:
Formicidae) and Its Possible Mimicry by Two Species of Pella (Cole-
optera: Staphylinidae). Ann. Entomol. Soc. America 64, 589-594.

Kloft, W.

1959. Direktes und indirektes Verfahren zur Messung der Beta-Strahlenab-
sorption von kleinen Gewebeschichten an Insekten. Glas-Intr. Techn. 3,
79-82.

Koblick, T. A., D. H. Kistner

1965. A revision of the species of the genus Myrmechusa from tropical Africa
with notes on their behavior and their relationship to Pygostenini
(Coleoptera, Staphylinidae). Ann. Ent. Soc. Amer. 58, 28-44.

Kolbe, W.

1971. Untersuchungen iiber die Bindung von Zyras humeralis (Coleoptera,
Staphylinidae) an Waldameisen. Entomol. Blatter 67, 129-136.

Kolbe, W., M. G. Proske

1973. Iso-Valeriansaure im Abwehrsekret von Zyras humeralis (Col. Staphy-
linidae). Entomol. Blatter 69, 57-60.

Maschwitz, U., B. Holldobler

1970. Der Kartonnestbau bei Lasius fuliginosus Latr. (Hym. Formicidae). Z.
Vergl. Physiol. 66, 176-189.

374

Psyche

[Vol. 88

Moore, B. P.

1968. Studies on the chemical composition and function of the cephalic gland
secretion in Australian termites. J. Insect Physiol. 14, 33-39.

Pasteels, J. M.

1968. Le systeme glandulaire tegumentaire des Aleocharinae (Coleoptera,
Staphylinidae) et son evolution chez les especes termitophiles du genre
Termitella. Arch. Biol. (Liege) 79, 381-469.

Patrizi, S.

1948. Contribuzioni alia conoscenze delle formiche e dei mirmicofile dell’
Africa orientale. V. Note etologiche su Myrmechusa Wasmann (Cole-
optera, Staphylinidae). Bull. 1st. Entomol. Univ. Bologno 17, 168-173.

Quilico, A., F. Piozzi, M. Pavan

1957. The structure of dendrolasin. Tetrahedron 1, 177-185.

Rathmayer, W.

1962. Methylmethacrylat als Einbettungsmedium fur Insekten. Experientia
(Basel) 18, 47-48.

SCHILDKNECHT, H., H. KRAMER

1962. Zum Nachweis von Hyrdochinonen neben Chinonen in den Abwehr-
blasen von Arthropoden. Z. Naturforsch. 176, 701-702.

Wasmann, E.

1886. Uber die Lebensweise einiger Ameisengaste. Dtsch. Entomol. Zeitschr.
30, 49-66.

1920. De Gastpflege der Ameisen. Gebrtider Borntraeger, Berlin.

1925. Die Ameisenmimikry. Gebriider Borntraeger, Berlin.

1930. Zur Biologie von Myrmedonia (Zyras). Entomol. Berichte 8, 150-151.

Wilson, E. O.

1971. The Insect Societies. The Belknap Press of Harvard University Press,
Cambridge, Mass.

Yasumatsu, K.

1937. Lasius fuliginosus (Latreille) var. spathepus (Wheeler and its synecht-
trans Zyras comes Sharp and Zyras cognathus Markel var . japonicus
Sharp. Nippon no Kochu 1, 47-51.

BEHAVIORAL ORIGIN OF TREMULATION,

AND POSSIBLE STRIDULATION,

IN GREEN LACEWINGS
(NEUROPTERA: CHRYSOPIDAE)1

By Peter Duelli2 and James B. Johnson3

University of California, Berkeley,

Division of Biological Control
1050 San Pablo Ave.

Albany, CA 94706 USA

Introduction

Abdominal vibration or “jerking” in connection with courtship
behavior has been described for several green lacewing species (e.g.
Smith 1922; Toschi 1965; Tauber 1969; Sheldon and MacLeod
1974) and explored in detail by Henry (1979, 1980a, b, c). In
Chrysoperla carnea (Stephens) isolated individuals produce long,
patterned sequences of discrete short bursts of rhythmic vibration of
the abdomen in the vertical plane. The wings may also vibrate.
Sexually receptive pairs establish duets of reciprocal abdominal
jerking. Actual drumming of the abdomen on the substrate does not
occur. It had been assumed that abdominal vibration produces
high-frequency sounds by stridulation (Adams 1962, Riek 1967,
Eichele and Villiger 1974, Henry 1979) and acoustical communica-
tion was discussed in connection with the tympanal ultrasound
receptor organ described by Miller (1970, 1971). Courtship and
copulation take place on the vegetation, usually on the underside of
leaves. Henry (1980a, c) in his work with Chrysoperla spp. demon-
strated that communication is performed via low-frequency sub-
strate vibration and not by airborne sound. Males were able to
establish duets with females within a range of 15 cm. According to
Henry (1980a, b, c), differences in the vibration patterns of Chry-

1 Published with the approval of the Director of the Idaho Agricultural Experiment
Station as Research Paper No. 81613.

2Present address: Zoologisches Institut, Universitat Basel, Rheinsprung 9, 4051 Basel
Switzerland.

3Present address: Department of Entomology, University of Idaho, Moscow, Idaho
83843 U.S.A.

Manuscript received by editor December 15, 1981.

375

376

Psyche

[Vol. 88

soperla rufilabris (Burmeister), C. downesi (Banks) and C. carnea
suggest that “acoustical” communication may help to reproduc-
tively isolate sympatric lacewing species.

Since the vibration produced by abdominal jerking in lacewings
seems to be propagated in a transverse wave (perpendicular to the
plane of the substrate) we prefer to call this type of communication
“tremulation”, following Busnel et al. (1956), Henry (1980c) and
Morris (1980). On the other hand, sound in the form of longitudinal
waves, is produced by stridulation and percussion. Possible stridula-
tory structures in lacewings were first described for the chrysopid
Meleoma schwarzi (Banks) by Adams (1962) and later for other
Neuroptera (Riek 1967). However, to date, there is no reported
record of any sound produced by these organs (Henry 1980c). In M.
schwarzi, sound may be produced when the second abdominal
sternite, with its regular striae of microtrichia, is rubbed against the
femora by abdominal vibration (Adams 1962). C. carnea and some
other species of Chrysopidae may stridulate using microtrichia on
the venter of the anal lobe of the forewings and dorsolaterad on the
metanotum (Riek 1967; Henry 1979). Alternatively, these paired
areas of microtrichia may function to hold the wings in place when
at rest (Henry 1980c). Thus, tremulation and possible stridulation
are both produced by vibrating the wings and abdomen.

Methods and Materials

Observations of free flight and mating behavior were made on the
following species: C. carnea, Eremochrysa punctinervis McLachlan,
E. tibialis Banks, Mallada basalis (Walker), Meleoma hageni Banks
and Nodita n. sp. The only specialized technique required for this
study was the use of a strobe light to illuminate lacewings on a flight
mill (Duelli 1980). By varying the frequency of the strobe flashes, it
was possible to determine the rate of the wing beats and other body
movements, as the highest flash frequency at which the motion
appeared to be “frozen” and each structure was seen in only one
position. A multiple of this frequency again produces a frozen
image, but the body is seen in multiple positions. C. carnea and M.
basalis were examined in this manner.

Results

Among the species observed, there appeared to be great variation
in the patterns and intensities of vibration of the wings and
abdomen during courtship, but this was not quantified. The beha-

1981]

Duelli & Johnson — Green Lacewings

377

vior was strongly developed in species of the genera Meleoma and
Eremochrysa, but was even more conspicuous in the Indo-Pacific
lacewing M. basalis, as observed on the island of American Samoa.
In this species, the males flapped their wings so vigorously that they
hit the substrate and produced sounds easily perceptible to the
human ear. Heavily developed pterostigmata in the hind wings of
the male may enhance substrate vibration and protect the wings
from damage (Fig. 1). During courtship, the males moved forward

Figures 1-5. Fig. 1. Forewing and hind wing of female (left) and male (right)
Mallada basalis. The arrow indicates the heavily developed pterostigma in the
hindwing of the male. Fig. 2. Stationary flight of a tethered Chrysoperla carnea
male. Strobe flashes (60 Hz, exposure 0.25 sec) show the extent of the abdominal
motion. Fig. 3. Chrysoperla carnea male mounted horizontally on a flight mill.
Strobe flashes (54 Hz) show the exact antiphase of abdominal and wing vibration.
Fig. 4. Chrysoperla carnea male mounted on a flight mill in “natural” flight position
as shown in figure 2. Any forced deviation from the “natural” body angle leads to an
increased amplitude of the abdominal vibration (See figure 3 for comparison).
Fig. 5. Same specimen and same position as in Fig. 3. 400 Hz strobe flashes show
the full flow of the movements of wings and abdomen.

378

Psyche

[Vol. 88

and backward in front of the female, and sometimes even sideways.
Especially vigorous males were seen to perform small jumps,
reminiscent of take-off behavior.

A chrysopid usually flies with its head higher than its abdomen. If
the insect is mounted on a flight mill and illuminated with a strobe
light, the abdomen can be seen moving up and down in the same
way as described for the courtship behavior (Fig. 2). When mounted
horizontally, the abdominal movements were exaggerated (Fig. 3).
In both orientations the strobe flashes revealed that the frequencies
of the wing beat and of the abdominal vibration were the same.
With each down-stroke of the wings the abdomen was lifted (Figs. 3
and 4). The flow of the movements can be seen in Fig. 5.

Observations made during this study indicate that the wing beat
frequency was positively correlated with temperature and, in gen-
eral, negatively correlated with wing length. At 23° C, a wing beat
frequency of 27 Hz (strokes /sec) was recorded for C. carnea and 38
Hz for the smaller M. basalis. Miller (1975) reported similar results,
25 Hz at 21-24°C in tethered flying C. carnea.

Discussion

The frequencies of abdominal vibration during courtship have
been reported for three species of Chrysoperla. For C. rufilabrus the
rates were 14-18 Hz (Henry 1980a). No temperature data were
given. In C. carnea the frequencies varied from 30 to 100 Hz at
24-28° C (Henry 1980c), while the courtship behavior of C. downesi
included volleys of abdominal vibration with a frequency of 60-80
Hz, with a mean of approximately 73 Hz, at 24-29° C (Henry
1980b).

The greater variability in the frequencies of abdominal vibration
during courtship, relative to flight, is probably related to two
factors. First, there is no minimum rate of wing beats necessary to
maintain flight. Second, the maximum possible rate is increased,
since the wings merely vibrate rather than making full strokes.
These would open a wide range of frequencies for chrysopids to use
in tremulation. If character displacement occurred, as hypothesized
by Henry (1980b), this would tend to expand the range of frequen-
cies actually used by chrysopids.

Tremulation has also been reported in the courtship behavior of
other groups of insects. Plecoptera communicate via the substrate
by drumming with their abdomens (Rupprecht 1968). Similar

1981]

Duelli & Johnson — Green Lacewings

379

drumming and/or abdominal vibration is known from certain
Psocoptera (Pearman 1928), Orthoptera (refs, in Rupprecht 1968
and Morris 1980), Megaloptera (Rupprecht 1975) and Mecoptera
(Rupprecht 1974). Wing fluttering is also involved in courtship of
Panorpa spp. (Mecoptera) (Rupprecht 1974) and three genera of
Coniopterygidae (Johnson and Morrison 1979).

The function of the abdominal motion in flight is unknown. In
the Diptera, the halteres (modified second pair of wings) have been
shown to act as specialized organs to maintain flight stability
(Pringle 1948). They vibrate in a vertical or nearly vertical plane
and, as gyroscopic indicators, reveal any change in the spatial
orientation of the thorax via sensors at their bases. The halteres
vibrate with the same frequency as the wings, but in antiphase. Since
the same is true for the abdominal movements in lacewings, it is
tempting to regard their abdominal vibration as an analogous
gyroscopic mechanism to stabilize the orientation of the thorax
during the slow hovering flight, thus keeping the insect in an upright
position with regard to the horizontal plane. This possibility is
supported by the similar orientation and abdominal movements of
flying Plecoptera, Megaloptera ( Sialis spp. and Neohermes sp.) and
Mecoptera ( Panorpa spp.) as observed in the field.

Indirect morphological evidence also supports this possibility.
Whereas most other nocturnal insects have large ocelli, chrysopids
and most other Neuroptera lack ocelli. An important function of the
ocelli in locusts and other insects is to recognize relative changes in
the height of the horizon (Taylor 1981) and thus to stabilize the
flight position.

Based on the similarities between abdominal vibration during
flight and courtship behavior, we suggest that tremulation behavior
in lacewings and perhaps other slow-flying insects may have evolved
from a particular “pre-adapted” feature in the take-off and flight
behavior, where its main function might be flight stabilization.

Acknowledgments

We wish to thank Dr. P. A. Adams for verifying the identifica-
tions of the species studied, Dr. K. S. Hagen for his helpful
discussions and Dr. F. M. Carpenter for his suggestions regarding
the manuscript.

380

Psyche

[Vol. 88

Literature Cited

Adams, P. A. 1962. A stridulatory structure in Chrysopidae (Neuroptera). Pan
Pac. Entomol. 38(3): 178-180.

Busnel, R. G., B. Dumortier and M. C. Busnel. 1956. Recherche sur le
comportement acoustique des ephippigeres (Orthoptera: Tettigoniidae). Bull.
Biol. Fr. Belg. 3: 219-286.

Duelli, P. 1980. Preovipository migration flights in the green lacewing, Chryso-
pa carnea. Behav. Ecol. Sociobiol. 7: 239-246.

Eichele, G. and W. Villiger. Untersuchungen an den Stridulationsorganen der
Florfliege Chrysopa carnea Steph. (Neuroptera: Chrysopidae). Int. J. Insect
Morphol. Embryol. 3(1): 41-46.

Henry, C. S. 1979. Acoustical communication during courtship and mating in
the green lacewing, Chrysopa carnea. Ann. Entomol. Soc. Amer. 72(1): 68-79.

Henry, C. S. 1980a. Acoustical communication in Chrysopa rufilabrus (Neurop-
tera: Chrysopidae), a green lacewing with two distinct calls. Proc. Entomol. Soc.
Wash. 82(1): 1-8.

Henry, C. S. 1980b. The courtship call of Chrysopa downesi Banks (Neuroptera:
Chrysopidae): Its evolutionary significance. Psyche 86: 291-297.

Henry, C. S. 1980c. The importance of low-frequency, substrate-borne sounds in
lacewing communication (Neuroptera: Chrysopidae). Ann. Entomol. Soc. Amer.
73(6) 617-621.

Johnson, V. and W. P. Morrison. 1979. Mating behavior of three species of
Coniopterygidae (Neuroptera). Psyche 86: 395-398.

Miller, L. A. 1970. Structure of the green lacewing tympanal organ ( Chrysopa
carnea, Neuroptera). J. Morphol. 131: 359-382.

Miller, L. A. 1971. Physiological responses of green lacewings (Neuroptera:
Chrysopidae) to ultrasound. J. Insect Physiol. 17: 491-506.

Morris, G. K. 1980. Calling display and mating behavior of Copiphora rhinoc-
eros Pictet (Orthoptera: Tettigoniidae). Anim. Behav. 28: 42-51.

Pearman, J. V. 1928. On sound production in the Psocoptera and on a presumed
stridulatory organ. Entomol. Monthly Mag. 64: 8-11.

Pringle, J. W. S. 1948. The gyroscopic mechanism of the halteres in Diptera.
Phil. Trans. Roy. Soc. B. 233: 347-384.

Riek, E. F. 1967. Structures of unknown, possible stridulatory, function on the
wings and body of Neuroptera; with an appendix on other endopterygote
orders. Austr. J. Zool. 15: 337-348.

Rupprecht, R. 1968. Das Trommeln der Plecopteren. Z. Vergl. Physiol. 59:
38-71.

Rupprecht, R. 1974. Vibrationssignale bei der Paarung von Panorpa. Experien-
tia 30: 340-341.

Rupprecht, R. 1975. Die Kommunikation von Sialis (Megaloptera) durch Vi-
brationssignale. J. Insect Physiol. 21: 305-320.

Sheldon, J. K. and E. G. MacLeod. 1974. Studies on the Biology of the Chry-
sopidae. IV. A field and laboratory study of the seasonal cycle of Chrysopa
carnea Steph. in Central Illinois. Trans. Amer. Entomol. Soc. 100: 437-512.

Smith, R. C. 1922. The biology of the Chrysopidae. Mem. Cornell Univ. Agr.
Exp. Sta. 58: 1291-1372.

1981]

Duelli & Johnson — Green Lacewings

381

Tauber, C. A. 1969. Taxonomy and biology of the lacewing genus Meleoma
(Neuroptera: Chrysopidae). Univ. Calif. Publ. Ent. 58: 1-62.

Taylor, C. P. 1981. Contribution of compound eyes and ocelli to the steering of
locusts in flight. I. Behavioral analysis. J. Exp. Biol. 93: 1-18.

Toschi, C. A. 1965. The taxonomy, life histories, and mating behavior of the
green lacewings of Strawberry Canyon. Hilgardia 36: 391-431.

ARTHROPODS ATTRACTED TO LUMINOUS FUNGI

By John Sivinski

Department of Entomology and Nematology
University of Florida
Gainesville, Florida 32611

Some fungi emit light. Luminescence may be present in mycelia
[e.g. a number of Mycena species (Wassink 1978)] or in both
mycelia and fruiting bodies [e.g. North American populations of
Panellus (= Panus ) stypticus, Buller 1924]. Lights have been
described as blue, white, or green depending on the species (Buller
1924, Wassink 1978). Emission intensities vary considerably. In the
forests of Borneo Mycena (= Poromycena ) manipularis are visible
at ca. 40 meters (Zahl 1971). An Australian species 1 “pours forth its
emerald green light” with sufficient intensity to read by (Lauterer
1900 in Buller 1924). North American forms, such as examined here,
tend to be dimmer. The eye often requires several minutes of dark
adaptation before their glows become visible.

The receiver(s) toward which fungi direct their luminous signals
are unknown. Lights have been supposed to lure spore dispersing
insects (Ewart 1906), but such an argument fails to account for
mycelial lights (Ramsbottom 1953). There has apparently been no
conjecture on the benefits mycelia accrue by glowing. The different
environments of mycelia and fruiting bodies make it questionable
whether their lights are directed at identical receivers or even serve
similar functions.

Until this time any proposed reactions of animals to fungal lights
have been speculative. I here present evidence that certain arthro-
pods are more likely to be captured in traps baited with light-
emitting mycelia and fruiting bodies than in controls containing
fungus-free substrate or dead and dark specimens of luminous
species. Several possible interactions between fungi and attracted
arthropods are discussed.

■Described as Panus incandescens, a name of doubtful taxonomic value (see
Wassink 1978).

Manuscript received by editor September 29, 1981.

383

384

Psyche

[Vol. 88

Methods

Test tubes (10 X 75 mm) were covered with Tack Trap®, a sticky
trapping compound, and capped with a cork. Luminous twigs,
conifer needles and leaf fragments, covered with mycelia of a
Mycena sp., were put into 31 such tubes. An identical number of
control tubes contained similar but nonluminous forest litter. Tubes
with glowing fungi were placed as closely as possible to the original
position of their contents (note that mycelia are most abundant deep
in litter, but traps were placed on the litter surface). Controls were
set ca. 80 mm to the side. Glass screw top vials (14 X 40 mm) were
also coated with Tack Trap®. From 3-6 fruiting bodies of the
luminous mushroom Dictyopanus pusillus were put into 72 such
vials. An identical number of controls contained 3-6 D. pusillus,
killed and rendered nonluminous by bathing in alcohol. Luminous
and control vials were alternately placed, ca. 80 mm apart, on and
by rotting logs on which D. pusillus had been found. Traps were put
out at night, gathered the following morning, and arthropods stuck
on their surfaces removed.

All specimens were captured during August in Alachua County,
Florida.

Results

More arthropods were captured on traps baited with glowing
fungal mycelia ( Mycena sp.) and luminous fruiting bodies ( D .
pusillus ) than their respective controls (x2 = 10.14, p< .001; \ =
6.41, p< .01, see Table 1). Taxa significantly more abundant on
luminous traps in the summed samples are Collembola (x2 — 12.81,
p < .001), and Diptera = 5.54, p < .025). It is of interest that
Collembola are not attracted to the bioluminescence of a sedentary
luminous predator, larvae of the fungus gnat Orfelia fultoni (Sivin-
ski 1982). Predators, i.e. spiders, ants, earwigs occur in a luminous:
dark ratio that borders on significance (x2 = 3.76, p < .10). Groups
captured in statistically indistinguishable numbers on luminous and
control traps are Isopods (x2 — 0.78, p > .25) and Amphipods (x2 =
0.59, p > .25). An unusual set of captures is the 5 crickets,
Eunemobius carolinus, taken only with luminous mycelia.

1981]

Sivinski — Arthropods and Luminous Fungi

385

Table 1. The numbers of Arthropods captured on traps containing luminous
mycelia ( Mycena sp.), luminous fruiting bodies ( Dictyopanus pusillus), and their
respective controls.

Mycena

sp.

Control

pusillus

Control

Summed

Fungi

Summed

Control

Collembola

Isotomidae/

Entomobryidae

Sminthuridae

Diptera

Phoridae

Sphaeroceridae

Cecidomyiidae

Ceratopogonidae

Psychodidae

Mycetophilidae

Predators

Araneida

Formicidae

Carabidae

Dermaptera

Hymenoptera

Isopods

Amphipods

Acari

Orthoptera

Gryllidae

Blattellidae

Cicadellidae

Thysanoptera

Unidentified

All Arthropods

113

199

127

386

Psyche

[Vol. 88

Discussion

Attraction of insects to fungal lights does not demonstrate that
luring arthropods is the function of the bioluminescence. With this
caveat in mind, note that an acceleration in the rate of certain
fungus /insect interactions even as an effect of a bioluminescent
signal is apt to influence the evolution of luminous fungi. In
particular, the argument that fungal lights are functionless, and by
implication harmless by-products of metabolism, loses force (see
also Lloyd 1977). Bearing a light near arthropods is unlikely to be
selectively neutral (for counterviews, see Buller 1924; Prosser and
Brown 1961).

Some possible functions of fungal glows become more plausible
with, or fail to find support in, the presented data. Both are
discussed below.2

Attraction of spore dispersers: Stinkhorn fungi (Phallales) use
odor, and perhaps color, to attract spore dispersing insects. Diptera,
in particular, consume a sweet malodorous spore-containing mu-
cous smeared on the fungal surface. Spores develop after being
discharged in the insect feces (discussed in Ramsbottom 1953). An
early conjecture on the function of fruiting body luminescence was
that lights, like odor and color in stinkhorns, lure spore dispersers
(Ewart 1906; see also Lloyd 1974, 1977). 3

A large proportion of the animals attracted to luminous fungi are
potential consumers of its spores. Many Collembola feed on fungal
spores, mycelia, and fruiting bodies. Some members of captured
Diptera families breed in fungi. The phorid Megaselia halterata, for
instance, is a pest of cultivated mushrooms (Oldroyd 1964). Whether
spores of D. pusillus pass unharmed through the insect gut is

2The following functions concern heterospecific receivers; however, biolumines-
cence is often intimately associated with mating (see Lloyd 1977). Sexual congress in
relevant Basidiomycetes consists of exchange of nuclei between haploid mycelia. Is it
possible that glows might direct the growth of photo-sensitive hyphae at this stage
and so serve as mating signals? Such an explanation fails to account for luminosity in
diploid mycelia or the fruiting body.

3Insects may evolve an affinity for fungal lights due to “rewards,” in food, shelter,
etc., the fungus provides. An alternative is that attraction is due to fungal
exploitation of arthropod “phototropisms.” The function of “phototropisms” are
often obscure. Some are apparently effects of orientation systems based on the
relative position of celestial objects (see Lloyd 1977).

1981]

Sivinski — Arthropods and Luminous Fungi

387

unknown. Nor is it known if attracted flies, such as phorids and
cecidomyiids, would be useful agents of dispersal. Vagile adults may
not feed on fungal materials. Protein consumption by cecidomyiids
is particularly rare (see Sivinski and Stowe 1981). Spores may be
moved, however, by attachment to the surface of a passing insect.

The topography and timing of luminous displays are often
suggestive of guiding dispersers. In Mycena pruinosa-viscida and
M. rorida from the Far Eastern tropics only the spores emit light
(Haneda 1955). Most fruiting body lights are restricted to, or
brighter in, the spore bearing hymenium (Wassink 1978) and
Panellus stypticus glows most strongly at the time of spore matura-
tion (Buller 1924). Conscription of dispersal agents is less likely to
account for light-emitting mycelia, unless mycelial cells pass safely
through the gut or can be carried to new locations on an arthropod’s
exoskeleton.

Attraction of carnivores: Predaceous arthropods were found on
glowing traps in numbers that border on significance, and fungus/
predator interactions can be imagined as important in the evolution
of bioluminescence. Luminous fungi might concentrate carnivores
about them by exploiting their “phototropisms.” If predators arrive
at rates effectively greater than lured fungivores, the resulting
predator:prey ratio may favor the fungus (an argument similar to
but more evolutionarily feasible than the “burglar alarm” theory of
Dinoflagellate luminescence; Burkenroad 1943; see Buck 1978).
Such an advantageous ratio is not obvious in my sample. Alterna-
tively, carnivores could seek out luminous fungi as locales of high
prey density. Glowing mushrooms might be mistaken for lumines-
cent animal prey.

Attraction of fungivores: If luminous mycelia are unpalatable, or
otherwise difficult to ingest, then fungivores attracted to lights
might consume adjacent competitors.

Attraction of fertilizers: Lloyd (1974) suggests that arthropods
lured by luminescent fungus might excrete beneficial materials and
so aid growth. Any nutritional gain must be balanced by the
metabolic expense of the signal.

Repulsion of negatively phototropic fungivores: Bioluminescence
might repel an organism’s negatively phototropic enemies or com-
petitors (Nicol 1962; see also Sivinski 1981 and citations). Repulsion
is particularly plausible in explaining luminous mycelia, some of

388

Psyche

[Vol. 88

which occur buried in litter, inside rotting logs, or on roots deep
underground where the opacity of the environment precludes
attraction as a function of light.

Among surface dwelling arthropods, there is no indication of a
light-avoiding taxon. This does not preclude repulsion. A rare, but
dangerous, enemy could keep fungal lights burning but escape
inclusion in the present sample, especially since mycelia baited traps
were not placed in the area of greatest mycelial abundance, deep in
the leaf litter. The intended receiver may not be an arthropod or
even macroscopic. Protozoa sometimes respond to lights. A glow
could repel certain pathogens and keep the fungus free of particular
diseases.

Light as a warning signal: Lights emitted by unpalatable fungi
might serve as warning signals directed towards nocturnal fungi-
vores (a similar function has been hypothesized for ancestral
flowers, Hinton 1973). Of North American fungi with luminous
fruiting bodies, one, P. stypticus, is a bitter tasting purgative, while
another, Omphalotus olearius, is a toxic hallucinogen (Miller 1979;
the palatability of D. pusillus is unknown). Pleurotus japonicus, a
luminescent Japanese species, is deadly poisonous (Buller 1924).
However, the luminous fruiting bodies of Malaysian Mycena
manipularis are quickly attacked by fungus gnats (Corner 1954;
gnats could be specialists, immune to toxins). Again there is no
evidence of arthropods avoiding fungal lights. My traps, of course,
would fail to quantify the discouragement of deer or other large
fungivores.

Like aposematic insects, luminous mushrooms often occur in
clumps (kin groups?) (see illustrations in Buller 1924, Harvey 1957;
also descriptions in Wassink 1978). Aggregations might intensify
warning signals (Cott 1957) and be instrumental in the evolution of
conspicuousness (Fisher 1930, for arguments concerning the kin
selection of aposematism). Several tropical light emitters, however,
apparently occur singly (see Wassink 1978).

White fungi can reflect enough celestial light to be surprisingly
obvious at night (noticed at twilight by Lloyd 1977). An assumption
of similar receivers for the bright white and luminous signals of
fruiting bodies allows the nocturnal aposematic signal hypothesis to
be tested with a larger sample. Mushrooms that appear to me to be
uniformly bright white include 6 toxic species, 13 edible and 5 whose

1981]

Sivinski — Arthropods and Luminous Fungi

389

palatability is unknown (color and palatabilities from photos and
text of Miller 1979). This distribution does not support the aposem-
atism argument (in comparison with a random sample of 41 non-
poisonous and 9 poisonous species x2 = 0.80 p > .25).

Summary

Arthropods, principally Collembola and Diptera, are attracted to
the lights of luminous fungal mycelia ( Mycena sp .) and fruiting
bodies ( Dictyopanus pusillus ). Such attraction does not prove that
bioluminescence has evolved to lure insects but does affect the
plausibility of hypotheses concerning the function of fungal glows.
The possibilities of lights being used to lure spore dispersers, attract
consumers of fungivores and competing fungi, repel negatively
phototropic fungivores, and serve as warning signals, are discussed.

Acknowledgments

Comments by J. E. Lloyd, T. J. Walker, T. Forrest, S. Wing, and
P. Sivinski improved the paper. B. Hollien professionally prepared
the manuscript. Dr. J. W. Kimbrough identified D. pusillus and J.
Sivinski helped gather specimens. Florida Agricultural Journal
Series No. 3280.

Literature Cited

Buck, J.

1978. Functions and evolutions of bioluminescence. Pages 419-460 in P. J.
Herring (ed.). Bioluminescence in action. Academic Press, New York.
Buller, A. H.

1924. The bioluminescence of Panus stypticus. Researches on fungi. Vol. 3. pp.
357-431.

Burkenroad, M. D.

1943. A possible function of bioluminescence. J. Mar. Res. 5: 161 164.
Corner, E. J. H.

1974. Further descriptions of luminous agarics. Trans. Br. Mycol. Soc. 37:
256-271.

Cott, H. B.

1957. Adaptive coloration in animals. Methuen and Co., Inc., London.
Ewart, A. J.

1906. Note on the phosphorescence of Agaricus ( Pleurotus ) candenscens. Bull.
Viet. Nat. 23: 174.

Fisher, R. A.

1930. The genetical theory of natural selection. Oxford Univ. Press, London.

390

Psyche

[Vol. 88

Haneda, Y.

1955. Luminous organisms of Japan and the Far East. Pages 335-386 in F. H.
Johnson (ed.). The luminescence of biological systems. Am. Assoc. Adv.
Sci., Wash., D. C.

Harvey, E. N.

1952. Bioluminescence. Academic Press, New York.

Hinton, H. E.

1973. Natural deception. Pages 96-159 in R. L. Gregory and E. H. H.
Gombrich (eds.). Illusion in nature and art. Charles Scribner’s Sons,
New York.

Lloyd, J. E.

1974. Bioluminescent communication between fungi and insects. Fla. Ento-
mol. 57: 90

1977. Bioluminescence and communication. Pages 164-183 in T. A. Sebeok
(ed.). How animals communicate. Ind. Univ. Press, Bloomington, Ind.

Miller, O. K., Jr.

1979. Mushrooms of North America. E. P. Dutton, New York, N. Y.
Nicol, J. A. C.

1962. Animal luminescence. Adv. Comp. Physiol. Biochem. 1: 217-273.
Oldroyd, H.

1964. The natural history of flies. W. W. Norton and Co., Inc., New York.
Prosser, C. L., and F. A. Brown, Jr.

1961. Comparative animal physiology. W. B. Saunders Co., Philadelphia, Pa.
Ramsbottom, J.

1953. Mushrooms and toadstools. Collins, London.

SlVINSKI, J.

1981. The nature and possible functions of bioluminescence in Coleoptera
larvae. Coleopts. Bull, (in press)

1982. Prey attraction by luminous larvae of the fungus gnat Orfelia fultoni.
(submitted)

Sivinski, J., and M. Stowe.

1981. A kleptoparasitic cecidomyiid and other flies associated with spiders.
Psyche. 87: 337-348.

Wassink, E. C.

1978. Luminescence in fungi. Pages 1 71-197 in P. J. Herring (ed.). Biolumines-
cence in action. Academic Press, New York.

Zahl, P. A.

1971. The secrets of nature’s night lights. Natl. Geogr. 140: 45-70.

PSYCHE

INDEX TO VOLUME 88, 1981

INDEX TO AUTHORS

Annette, Aiello. Life History of Antaeotricha sp. (Lepidoptera: Oecophoridae:
Stenomatinae) in Panama. 163

Carlin, Norman F. Polymorphism and Division of Labor in the Dacetine Ant Orec-
tognathus versicolor (Hymenoptera: Formicidae). 231
Carpenter, Frank M. Dedication: Robert E. Silberglied. 197
Dahlstrom, Tina. See Topoff, Howard.

Droual, Robert and Howard Topoff. The Emigration Behavior of Two Species of the
Genus Pheidole (Hymenoptera: Formicidea). 135
Duelli, Peter and James B. Johnson. Behavioral Origin of Tremulation, and Possible
Stridulation, in Green Lacewings (Neuroptera: Chrysopidae). 375
Edwards, G. B. Sound Production by Courting Males of Phidippus mystaceus (Ara-
neae: Salticidae). 199

Ginsberg, Howard S. Historical Development of Bee Foraging Patterns in Central
New York State. 337

Hathaway, Mary. Polistes gallicus in Massachusetts (Hymenoptera: Vespidae). 163
Haverty, Michael I. See Howard, Ralph W.

Holldobler, Bert. Trail Communication of the Dacetine Ant Orectognathus versi-
color (Hymenoptera: Formicidae) 245

Holldobler, B., M. Moglich, and U. Maschwitz. Myrmecophile Relationship of Pella
(Coleoptera: Staphylinidae) to Lasius fuliginosus (Hymenoptera: Formicidae).
347

Howard, Ralph W., Eldon J. Mallette, Michael /. Haverty, and Richard V. Smythe.
Laboratory Evaluation of Within-Species, Between-Species, and Parthenogenetic
Reproduction in Reticulitermes flavipes and Reticulitermes virginicus. 75
Johnson, James B. See Duelli, Peter.

Kearns, R. S. and R. T. Yamamoto. Maternal Behavior and Alarm Response in the
Eggplant Lace Bug, Gargaphia solani (Heidemann) (Tingitidae: Heteroptera).
215

Levings, Sally C. and James F. A. Traniello. Territoriality, Nest Dispersion, and
Community Structure in Ants. 265

MacKay, William P. A Comparison of the Nest Phenologies of Three Species of Po-
gonomyrmex Harvester Ants (Hymenoptera: Formicidae). 25
Mallette, Eldon J. See Howard, Ralph W.

Maschwitz, U. See Holldobler, B.

Moglich, M. See Holldobler, B.

391

Newton, Alfred F. Jr. New Name for the Extinct Genus Mesagyrtes Ponomarenko
(Coleoptera: Silphidae S.L.). 335

Nimmo, Andrew P. Francis Walker Types of, and New Synonymies for, North Amer-
ican Hydropsyche species (Trichoptera: Hydropsychidae). 259
Parker, F. D. See Tepedino, V. J.

Pujdak, Susan. See Topoff Howard.

Richman, David B. and Willard H. Whitcomb. The Ontogeny of Lyssomanes viridis
(Walckenaer) (Araneae: Salticidae). 127
Rothstein, Aaron. See Topoff, Howard.

Sivinski, John. Arthropods Attracted to Luminous Fungi. 383
Smythe, Richard V. See Howard, Ralph W.

Steiner, A. L. Anti-predator Strategies. II. Grasshoppers (Orthoptera, Acrididae)
Attacked by Prionyx parkeri and Some Tachyspex Wasps (Hymenoptera, Sphe-
cinae and Larrinae): A Descriptive Study. 1
Stuart, Robin J. Abdominal Trophallaxis in the Slave-Making Ant, Harpagoxenus
americanus (Hymenoptera: Formicidae). 331
Tepedino, V. J. and F. D. Parker. The Effect of Flower Occupancy on the Foraging
of Flower-Visiting Insects. 321
Topoff, Howard. See Droual, Robert.

Topoff, Howard, Aaron Rothstein, Susan Pujdak, and Tina Dahlstrom. Statary Be-
havior in Nomadic Colonies of Army Ants: The Effect of Overfeeding. 151
Traniello, James F. A. See Levings, Sally C.

Ward, Philip S. Ecology and Life History of the Rhytidoponera impressa Group
(Hymenoptera: Formicidae). I. Habitats, Nest Sites, and Foraging Behavior.
89

Ward, Philip S. Ecology and Life History of the Rhytidoponera Group (Hymenop-
tera: Formicidae). II. Colony Origin, Seasonal Cycles and Reproduction. 109
Whitcomb, Willard H. See Richman, David B.

Yamamoto, R. T. See Kearns, R. S.

392

INDEX TO SUBJECTS

All new genera, new species and new names are printed in capital type.

A comparison of the nest phenologies of
three species of Pogonomyrmex har-
vester ants, 25

Abdominal trophallaxis in Harpagoxe-
nus, 33 1

Alarm response in Gargaphia, 215
Antaeotricha, life history, 163
Anti-predator strategies. II. Grasshop-
pers attacked by Prionyx parkeri and
some Tachysphex wasps: a descriptive
study, 1

Ants, territoriality, nest dispersion, and
community structure, 265
Apis mellifera, 337
Army ants, 151

Arthopods attracted to luminous fungi,
383

Bee foraging patterns, 337
Behavioral origin of tremulation in green
lacewings, 375
Chrysopidae, 375
Cicadas, population ecology, 175
Dedication: Robert E. Silberglied, 197
Ecology and life history of the Rhytido-
ponera impressa group (Hymenoptera:
Formicidae). I. Habitats, nest sites,
and foraging behavior, 89
Ecology and life history of the Rhytido-
ponera impressa group. II. Colony,
seasonal cycles, and reproduction,
109

Effect of flower occupancy on the forag-
ing of flower-visiting insects, 321
Eggplant lace bug, 215
Emigration behavior of two species of
Pheidole, 135
Flower-visiting insects, 321
Francis Walker types of Hydropsyche,
259

Gargaphia solani, 215
Grasshoppers, 1
Green lacewings, 375
Harpagoxenus americanus, 33 1

Harvester ants, 25

Historical development of bee foraging
patterns, 337

Hydropsyche, new synonymies, 259
Hydropsyche alternans, 261
Hydropsyche confusa, 259
Hydropsyche reciproca, 262
Laboratory evaluation of within-species,
between-species, and parthenogenetic
reproduction in Reticulitermes flavipes
and Reticulitermes virginicus, 75
Lasius fuliginosus, 347
Life history of Antaeotricha sp. in Pan-
ama, 163

Luminous fungi, 383

Lyssomanes viridis, 1 27

Maternal behavior in Gargaphia, 215

Mesagyrtes, 335

Mesecanus, 335

Myrmecophilic relationships of Pella to
Lasius, 347

Neivamyrmex nigrescens, 151
New name for extinct genus Mesagyrtes,
335

New Synonymies for Hydropsyche, 259
Notes on the population of ecology of
cicadas in the Cuesta Angel forest ra-
vine of Northeastern Costa Rica, 175
Ontogeny of Lyssomanes viridis, 127
Orectognathus versicolor, 231, 245
Parthenogenetic reproduction in Reticu-
litermes, 75
Pella, 347
Pheidole, 135
Phidippus mystaceus, 1
Pogonomyrmex, 25
Polistes gallicus in Massachusetts (Hy-
menoptera: Vespidae), 169
Polymorphism and division of labor in
Orectognathus, 23 1
Prionyx, 1
Reticulitermes, 75
Rhytidoponera impressa, 89, 109

393

Silberglied, Robert E., dedication to,
197

Slave-making ant, 331

Sound production by males of Phidip-
pus, 199

Statary Behavior in nomadic colonies of
army ants: the effect of overfeeding,
151

Stridulation in green lacewings, 375

Tachysphex, 1

The emigration behavior of two species
of the genus Pheidole (Formicidae:
Myrmicinae), 135

The ontogeny of Lyssomanes viridis
(Walckenaer) (Araneae: Salticidae) on
Magnolia grandiflora, 127
Territoriality, nest dispersion, and com-
munity structure in ants, 265
Trail communication in Orectognathus,
231

Tremulation in green lacewings, 375
Trophallaxis in Harpagoxenus, 331
Walker types of Hydropsyche, 259
Wasps, 1

394

CAMBRIDGE ENTOMOLOGICAL CLUB

BACK VOLUMES OF PSYCHE

Requests for information about back volumes of Psyche should
be sent directly to the editor.

F. M. Carpenter
Editorial Office, Psyche
16 Divinity Avenue
Cambridge, Mass. 02138

FOR SALE

Reprints of articles by W. M. Wheeler

c/>

UIISNI^NVINOSHIMS S3 ! H VB 8 ll^ll B R AR I ES SMITHSONIAN^ INST i TUT 1 01
r- v 2 r- 2 r~

m v -V??'' z: m

ARIES SMITHSONIAN INSTITUTION^ NOlinillSN I NVIN0SH1IWS S3l8Vyai

2 \ C/> 2 CO

I
C/)

o
z ,

> s ' w > *' 2

2 co **v z oo

UIJ.SNI_NVINOSHJ.INS S3UJVU9I1 LIBRARIES SMITHSONIAN INSTITUTIOI

Z \ £ 2 ^

X/ —

AS£i/ o o Xi'VosvV^x _ o

^ Z „j Z

ARIES SMITHSONIAN INSTITUTION NOIlfUllSNI NVINOSHIIWS S3 I 8V88I
2: t“ 2 r- 2

^5ov7x 0 . m O Z . o

nilJLSNl NVINOSHimS S3 I avyairLIBRAR I ES ^SMITHSON IAN INSTITUTE
z ^ £ z to 2 .v

< E - .vW vv

f i 5

I AR I ES^SMITHSONIAn" INSTITUTION^NOIJ-nillSNI NVINOSHJIWS^SS IBVaa

00 5 ___ < n — 00

ItJ 7k CJ CJ

nj.ij.sm nvinqshjjns saiavaan libraries Smithsonian institute

r <, 2 r- z ^

ca x S ^ ®

Wr'^y m ^ m N^/vostoS?' ■ '/. /' '' ^

!AR I ES</>SMITHS0NIAN~INSTITUTI0NtnN0linilJSNI~NVIN0SHlHMS S3 I BVdai

^2 ^ . 2 00 2 GO

- 1 ! jftv 1

y 1" 2

OO '*** Z W

MwmnOLi 1 iiaic* oi 1 viwvi a n i » r> r% a n i c o puitupamiam IMCTITI IT! D

niiism nvinoshiiws saiavaan libraries Smithsonian institutio

SMITHSONIAN INSTITUTION NOlinillSN!

nvinoshiiws S3 lava a i

LIBRARIES SMITHSONIAN INSTITUTIO

rn ~~

TAR I ES SMITHSONIAN INSTITUTION NOIXfUliSNI NVINOSH1IWS S3iaVaai

RAR I ES^SMITHSONIAN^ INSTITUTION ^NOlinillSN I

in m

.niusNi NViNosHims S3 1 y vd a n libraries

NVINOSHIIWS S3ldVd8l
co

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "Psyche"

See other formats