Full text of "Psyche"
^R\ ES^^SMITHSONIAN INSTITUTION^NOlinillSNI NVINOSHiIl/\iS^S3 I d Vd 9 I-
Z * (/i z c/>
V < S S
> S >
illSNI_NVINOSHill^S^S3 I 9V9 9 n\l BRAR 1 ES^SMITHS0N1A^1^^NSTITUT10^
c ^ ^ c life ji; S 5 w -^^1'
5 e -' 5'
Z _J z
i\RIES SMITHSONIAN INSTITUTION NOIlfUliSNI NVINOSHilWS S3iyvaar
° ■ " 2 m i /53^x.n
CO
m
liliSNI NVINOSHillMS S3 I 9 Vd 9 H^U B R AR I ES^SMITHSONIAN ""iNSTITUTIOI
z CO z (o 2: V,
< y<<\vAS^;x S „< .cSa.- 2
z
o
CO
= CO
F §
5 XON^C^ 5 ^ ""v i
ARIES SMITHSONIAN INSTITUTION NOlifUliSNI NVINOSHJ.IWS*”s3 I a V« a I
CO ^ CO CO
O
iliSNl”NVINOSHillAIS S3ldV99n“’uBRARI ES^SMITHSONlAN”mST!TUTIOI
m '■ >5^' 2 m
ARIES SMITHSONIAN~INSTITUTION NOliniliSNI^NVlNOSHiiyMS S3ldVd9l
c/> in z in
Wi
^ >' S X^NOS^ >
illlSNI_NVIN0SHill^s'^S3 I d Vd 9 H^LI B RAR I Es'^SMITHSONIAN INSTITUTlOf
2 \ ^ ~ CO “
CO
O • ^ Q
ARIES SMITHSONIAN INSTITUTION NOlifUliSNI “NVINOSHilWS “S3 I a Va 8 1
Z f“ *3P r- —
CO — (/) CO
ARIES SMITHSONIAN INSTITUTION NOIlfUllSNI NVINOSHillNS S3IHVyai
(/> _ 2: T </> z in .
X
(/}
g V.n2«n;^ g | /' g ^
lillSNl NVINOSHilWs‘^S3 I a Vd 8 H^LI B RAR I Es”sMITHSONIAN^INSTITUTIO^
~ = <0 —
ARIES SMITHSONIAN INSTITUTION NOlifUliSNI NVINOSHiliAIS S3iavaai'
^ ^ m
nillSNI NVINOSHII^S S3 I dVH 8 n"u B RAR I ES^SMlTHS0NIAN“lNSTITUT10i
?AR I ES^SMITHSONIAN INSTITUTION NOIinillSNI^NV!NOSHlli^S^S3 I B VB 8 I
(/> ^ c/5
o
nillSNl“'NVINOSHiIlMS S3 I B VB 8 11 U B RAR I ES^SMITHS0NIAN"^INSTITUT10I
|VAs»^ m >5^' ^ xjvAs»i>^ m ^ m
? AR I ES SMITHS0NIAN”lNSTlTUT10N^N0linillSNl“NVIN0SHimS S3 I HV8 8 I
Z » CO ^ CO
S < A 2 S
A 2: -H 2
X
CO
o
__ 2:
> X^M^A5Vji>' — '>\W 5 ' ^ X^OS»^ >
^2 ^ Z S 2
niliSNl NVmOSHillMS S3IBV^an libraries smithsonian institutioi
^ ^ C/)
CO
o ■■ Q
Z ■' ' -j 2 _ ~
lARIES SMITHSONIAN INSTITUTION NOIlfliliSNI NVINOSHillAIS S3IBV88I
'.k
j:
»
i
PSYCHE
A Journal of Entomology
Volume 90
1983
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 M. D. Bowers
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. 89, nos. 3-4, for 1982, April 27, 1983
Vol. 90, nos. 1-2, for 1983, August 8, 1983
Vol. 90, no. 3, for 1983, December 9, 1983
ISSN 0033-2615
PSYCHE
A JOURNAL OF ENTOMOLOGY
founded in 1874 by the Cambridge Entomological Club
Vol. 90
1983
No.1-2
Studies on Upper Carboniferous insects: I. The Geraridae (Order Protorthop-
tera). Laurie Burnham I
Testing visual species recognition' in Precis (Lepidoptera: Nymphalidae) using
a cold-shock phenocopy. Arthur M. Shapiro 59
Defensive adaptations and natural enemies of a case-bearing beetle, Exema
canadensis (Coleoptera: Chrysomelidae). Richard B. Root and Frank J.
Messina 67
Studies on North American Carboniferous insects. 7. The structure and relation-
ships of £w6/c/;m.s r/oA7/W.v/ (Palaeodictyoptera). Frank M. Carpenter 81
The biology of Trichadenotecnum alexanderae Sommerman (Psocoptera:
Psocidae). 111. Analysis of mating behavior. B. W. Betz 97
Predatory behavior of bombardier beetles by a tabanid fly larva. Steven
Now ick i and Thomas Eisner 119
Prey selection by the neotropical spider, Alpaida tuonado, with notes on
web-site tenacity. Todd E. Shelly 123
Reproductive behavior of Claeoderes hivittata (Coleoptera: Brentidae).
Leslie K. Johnson 135
Polydomy in the slave-making ant, Harpagoxenus americanus (Emery)
(Hymenoptera: Formicidae). Maria Guadalupe Del Rio Pesado and
Thomas M. Alloway 151
Situation and location-specific factors in the compatibility in Rhytido-
ponera metallica {Uymenoplera: Formicidae: Ponerinae). Caryl P. Haskins
and Edna F. Haskins 163
Capture of bombardier beetles by ant-lion larvae. Jeffrey Conner and
Thomas Eisner 175
Reproductive plasticity in yellowjacket wasps: a polygynous, perennial
colony of Vespula maculifrons. Kenneth G. Ross and P. Kirk Visscher .... 179
CAMBRIDGE ENTOMOLOGICAL CLUB
Officers for 1982-1983
President
Vice-President
Secretary
Treasurer
Executive Committee
Frances Chew
Edward Armstrong
Margaret Thayer
Frank M. Carpenter
John Shetterly
Ronald McGinley
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
M. D. Bowers, Assistant Professor of Biology, Harvard University
Alfred F. Newton, Jr., Curatorial Associate in Entomology, Harvard
University
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: $1 1.00, 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 $10.00
each, and for full page half-tones, $12.00 each; smaller sizes in proportion. There is
ordinarily no additional charge for setting tables of less then six columns; for tables
of six or more columns the cost is $25 per page.
Psyche, vol. 89, no. 3-4, for 1982, was mailed April 27, 1983
The Lexington Press, Inc., Lexington, Massachusetts
PSYCHE
Vol. 90
1983
No. 1-2
STUDIES ON
UPPER CARBONIFEROUS INSECTS:
I. THE GERARIDAE (ORDER PROTORTHOPTERA)
By Laurie Burnham
Department of Entomology, Cornell University,
Ithaca, New York 14853;
and Museum of Comparative Zoology, Harvard University,
Cambridge, Massachusetts 02138*
Introduction
Despite the importance of the order Protorthoptera,' little is
known about its evolutionary history. While recent workers have
emphasized morphological and taxonomic diversity in the group
(Carpenter, 1971, 1977; Wootton, 1981), no one has undertaken
serious revisionary study at the family level. As a consequence, our
understanding of relationships within the order, as well as relation-
ships of the Protorthoptera to other Paleozoic insects, is rudi-
mentary at best. Clearly, revisionary studies on this group are badly
needed.
We know that the Protorthoptera first appear in the fossil record
at the base of the Upper Carboniferous (Namurian Stage) and
apparently flourished for 80 million years before becoming extinct
at the end of the Permian. We also know that they were remarkably
'It was one of the dominant orders of the Paleozoic (exceeding all other insects
both in number of species and in number of individuals), and is considered by many
to be ancestral to the Endopterygota (the group to which 90% of all living insects
belong).
■"Present address.
Manuscript received hy the editor March 5, 1983.
2
Psyche
[Vol. 90
diverse morphologically, and that diversity in the group {sensu lato)
far exceeded that of any other Paleozoic order (Carpenter, 1977).
Structural modifications normally associated with more recent
insects, including brightly patterned wings, raptorial fore legs, and
thoracic extensions of various kinds, are found throughout the
group.
Despite this fascinating array of characters. Carboniferous
Protorthoptera have generated little interest among systematists.
This is due, in part, to problems common to all paleoentomological
study: an overall lack of material (rarely is a species based on a large
series of specimens); preservational quality that ranges from excel-
lent to very poor; and a taxonomy that is highly subjective.
In addition, there are problems unique to the Protorthoptera
which make their study particularly difficult. First, they are neop-
terous, and as a consequence, are most frequently found with their
wings superimposed on one another. Interpretation of the venation
under these circumstances is not only difficult, but prone to error.
Second, the vast majority of Carboniferous Protorthoptera are
known from only two localities: Commentry in France, and Mazon
Creek in the United States, and were studied primarily by turn-of-
the-century workers.- As new material has become available for
study in recent years, the accuracy of much of this earlier work has
been questioned.
Finally, there is the problem of variation. Distinguishing species-
level differences from intraspecific variation in fossils that have such
a limited array of characters is not easily resolved. How, for
instance, does one recognize sexual dimorphism in a fossil species
when there are no genitalia or secondary sexual characters to serve
as guidelines? It is not surprising under the circumstances that
paleotaxonomy rests largely on subjective reasoning. But this,
unfortunately, has its pitfalls.
Previous work on the Geraridae is a case in point. Anton Hand-
lirsch, responsible for most of the earlier work on the family, de-
scribed a new species of gerarid for every specimen he examined.
-The Permian Protorthoptera are generally much better known than their Carbo-
niferous relatives. This is attributable to the fact that there are more than eight major
Permian deposits (including two in the U.S.) at which Protorthoptera have been
found. Furthermore, intensive studies on these insects have attracted the attention of
such well-known recent workers as Carpenter, Kukalova, and Sharov.
millions of years
1983]
Burnham — Geraridae
3
Fig. I. Geological Time Table of the Carboniferous. Note the different ages of
the Commentry and Mazon Creek localities.
basing his taxonomic decisions on small differences in wing vena-
tion (Handlirsch, 1906b, 1911, 1922). This approach, while render-
ing the decision making easier, is nevertheless open to criticism.
Studies on intraspecific variation in some Permian Protorthoptera
have shown, for instance, that two fore wings belonging to the same
specimen will exhibit noticeable differences in venation (Carpenter,
1966). From these findings we can conclude that intraspecific varia-
tion in the Protorthoptera was high, and that the variation Hand-
lirsch saw was no greater than that seen in a single specimen.
Further evidence that suggests Handlirsch was unrealistic in his
representation of species diversity comes from the low probability of
finding only one individual per species (for all species collected) in a
random sample of living insects. Similarly, we cannot reasonably
expect to find only one specimen per species in a paleontological
sample, particularly when fossilization was catastrophic (and hence
random) as is true for the Mazon Creek biota.
Nevertheless, in spite of these drawbacks to the study of fossil
insects, the field can be immensely rewarding. This is because it
provides us with concrete evidence (in the form of fossils) of what
early insect life was like. Without such proof, we would be guided
4
Psyche
[Vol. 90
only by our imagination, and having this proof provides a base on
which insect phylogeny and early insect evolution can be recon-
structed.
Clearly, revisionary studies on Paleozoic insects are important.
Fortunately, recent revisional work by Carpenter has greatly in-
creased our knowledge of certain Paleozoic orders (i.e., the Paleodic-
tyoptera, Megasecoptera, Diaphanopterodea, Protodonata, and
Caloneurodea), but much work on the Protorthoptera remains.
This revision of the family Geraridae is, at least, a beginning and is
intended to be the first in a series of family-level studies on Upper
Carboniferous Protorthoptera.
Selection of the Geraridae as a starting point was influenced,
ultimately, by two factors: 1) it is typical of many of the families in
the Protorthoptera, having last been studied in the early part of this
century (despite the discovery since then of new material assignable
to the family); and 2) the Geraridae are morphologically most unus-
ual insects. They were large (up to 75 mm in body length), and had
as their single most distinctive attribute, a prothorax that was elon-
gate, flask-shaped, and adorned with long, numerous spines. These
spines gave them the appearance of walking pincushions, and pre-
sumably provided some defense against vertebrate predators.
The systematic importance of the family plus the impact this work
has on current classifications of the Protorthoptera will be discussed
in subsequent pages. The remaining sections of this paper cover (in
the following order): the systematics of the Geraridae; paleoecologi-
cal differences between the two localities at which gerarids have
been found (Mazon Creek and Commentry); and the significance of
this study relative to phylogenetic relationships within the Pro-
torthoptera.
Systematics
Materials and Methods
The fossils examined for this study occur mostly as impressions
(imprints in a sedimentary matrix devoid of organic matter) but
some occur as compressions (in these, organic matter is present, but
usually coalified). Both types of fossils were prepared by degage-
ment,^ i.e., an uncovering of the fossil by removal of the overlying
rock matrix. This is generally done using a fine pneumatic drill and
From the French verb degager meaning to disengage, extricate, or get clear.
1983]
Burnham — Geraridae
5
compressed air gun. The technique is particularly effective at
revealing regions of an insect’s body (wing tips, legs, etc.) that are
found beneath the bedding plane. Following degagement, specimens
were studied under a Wild M-5 stereo microscope and photo-
graphed with a Zeiss 4" by 5" format camera.
Drawings were made of each fossil by tracing a general outline
from a photographic enlargement. Verification of detail was made by
referring back to the specimens and examining them frequently
under the microscope. The most complete reconstructions (e.g., fig. 2
of G. danielsi) were possible for those species that consist of a large
series of specimens. This is because one fossil rarely displays all
characters equally well, and, therefore, the larger the number of speci-
mens, the greater the likelihood of multiple character preservation.
Type specimens, including the holotypes, for all taxa considered
in this revisionary study were borrowed and examined using the
above methods. Pre-existing taxa were synonymized whenever pos-
sible, a decision based on the assumption that (for reasons cited
earlier) intraspecific variation in the Protorthoptera is great. Char-
acters of greatest taxonomic importance were venation and body
size and shape, particularly with respect to the prothorax. In situa-
tions where clearcut characters were lacking, as is true for several of
the Mazon Creek gerarids, I relied solely on size as a criterion for
specific assignment. While this may result in the recognition of some
dubious species, it seems preferable to relegating certain specimens
to incertae sedis status.
Since wing venation is such an im.portant taxonomic tool both in
paleoentomological and extant systematic study, it is surprising that
until recently no standardized wing terminology has been adopted.
This is particularly unfortunate for the Protorthoptera, 80% of
which have been described on the basis of wings alone. Inroads
have recently been made into this problem primarily by the efforts
of Carpenter in the United States and Wootton in Great Britain.
Both have stressed (Carpenter, 1966; Wootton, 1979, 1981) the
importance of a standardized venational nomenclature and Woot-
ton (1979) has proposed a terminology modified slightly from the one
used previously by Lameere (1922) and Martynov (1924, 1938).
Wootton proposes that the following nine major longitudinal
veins be recognized: Costa (C); Subcosta (SC); Radius (R); Radial
Sector (RS); Anterior Media (MA); Posterior Media (MP); Ante-
rior Cubitus (CUA); Posterior Cubitus (CUP), and Anals. In light
6
Psyche
[Vol. 90
of the historical basis for the nomenclature (used extensively in the
paleoentomological literature) and its conservatism (it may be used
to homologize the wing venation of all insects) I enthusiastically
concur with Wootton’s recommendations and will employ his sys-
tem here and in future systematic work.
A total of 58 specimens were made available for study through the
loans of various institutions and individuals. These are listed here
with their abbreviations:
Field Museum of Natural History (FMNH), Chicago, Illinois,
U.S.A. (This includes specimens collected by Jerry Herdina and
subsequently acquired by the Field Museum).
Institut de Paleontologie, Museum National d’Histoire Naturelle
(IP), Paris, France.
Museum of Comparative Zoology (MCZ), Cambridge, Massa-
chusetts, U.S.A.
United States National Museum (USNM), Washington, D.C.,
U.S.A.
Yale Peabody Museum (YPM), New Haven, Connecticut, U.S.A.
Daniel Damrow, of Mosinee, Wisconsin. Private collection.
(DMRW) (Includes specimens previously in the collection of Walter
Dabasinskas).
David Douglass, of Yachats, Oregon. Private collection. (DGLS)
Francis and Terri Wolff, of Port Charlotte, Florida. Private col-
lection (Wolff).
Order Protorthoptera Handlirsch
Family Geraridae Scudder, 1885
[Nom. correct. Handlirsch, 1906a (ex Gerarina Scudder, 1885)]
Gerarina Scudder 1885:342. Type: Gerarus Scudder.
Geraridae Handlirsch 1906a:146, 1906b:701, 1911:312, 1920:151.
Sthenaropodidae Handlirsch 1906a:141, 1919:37, 1920:150; Sharov 1968:19. Type:
Sthenarupoda Brongniart. nkw synonymy.
Genopterygidae Richardson 1956:41. Type: Genopteryx Scudder. new synonymy.
Description
Fore and hind wings similar in size and shape, but markedly
different in venation.
Fore wing: length 35-55 mm, and apparently not coriaceous; cos-
tal area broad in basal region, SC simple, terminating in C; R
parallel to SC, terminating at wing apex; RS originating from base
1983]
Burnham — Geraridae
1
of R near midpoint of wing; M either anastomosing with RS for a
short distance or connecting to it by a cross-vein; CUA strongly
developed, arising from base of M; CUP forked, arising independ-
ently from wing base.
Hind wing; length 30-48 mm; costal area not as broad as in fore
wing; SC simple, terminating in C; R parallel to SC, terminating at
wing apex; RS pectinate, arising from R near wing base; M forked,
arising near base of RS; CUA and CUP simple and parallel to one
another; CUA arising from base of RS near M, CUP arising inde-
pendently from wing base; anal area unusually reduced; cross veins
abundant in both fore and hind wings.
Body; prothorax elongate, flask-shaped, and distinctively spi-
nose; abdomen cylindrical; antennae filamentous; head small and
probably mobile; legs cursorial, tarsi five-segmented.
Diagnosis
In many ways the Geraridae are typical Orthopteroidea, having
mandibulate mouthparts, hypognathous heads, and filamentous
antennae. But they differ from other orthopteroids in two important
characters: their well-developed prothorax which is armed with
spines (the latter reach a length of 10 mm in G. danielsi), and their
distinctive fore and hind wing venation. While gerarids can be read-
ily recognized on the basis of the prothorax alone, wing venation is
generally a better diagnostic character. Particularly distinctive are
the RS-M veins in the fore wing, and the R-RS veins in the hind
wing. In the fore wing RS is reduced and M is expanded with 5 to 6
branches. The apical branch of M either anastomoses with RS for a
short distance or is connected to it by a short cross vein. In the hind
wing, the opposite is true: M is greatly reduced and RS expanded
into 5 to 6 branches.
It is worth noting that the anal fan in the hind wing, if present,
was very small (see fig. 17). This suggests that in gerarids the fore
and hind wings may have functioned equally well in flight, unlike in
extant Orthoptera, which rely primarily on expanded hind wings for
flight propulsion. The abdomen is essentially unknown for the fam-
ily, but was probably shorter than the wings, a claim based on the
comparison of wing length to legs, thorax, and head. No cerci are
preserved, but because the Geraridae are orthopteroid, it may be
assumed that they were present.
8
Psyche
[Vol. 90
Fig. 2. Gerarus clanielsi, composite drawing, based primarily on specimens
FMNH PE 5276, 31973, 32027, 32029; and USNM 31973.
1983]
Burnham — Geraridae
9
Remarks
The family Geraridae was first established by Scudder (1885) for
several fossil insects from Mazon Creek noted for their slender
bodies, which tapered “greatly anteriorly” (Scudder, 1885:344), and
for the distinctive branching pattern of RS in the hind wing.
Scudder placed the family in the order Paleodictyoptera, section
Neuropteroidea, where it remained until 1906 when Handlirsch
erected the order Protorthoptera and transferred the Geraridae to
it.^^ Most of the subsequent work on the family was carried out by
Handlirsch who added a total of two new genera and nine new
species to it (Handlirsch, 1906a, 1906b, 1911, 1920).
This revision is the first systematic work carried out on the family
since then, and rectifies many of the taxonomic errors made by these
earlier workers. To a large extent, the mistakes made by Scudder
and Handlirsch may be attributed to the limited availability of
material at their disposal, and the preservation of most gerarids with
all four wings lying over one another. Nevertheless, their errors were
of grave consequence. To begin with, neither worker apparently
recognized the extent to which intraspecific variation occurs in the
family, and therefore each named only monotypic species. But,
more importantly, owing to the difficulties of wing overlap, neither
Scudder nor Handlirsch correctly interpreted the wing venation of
Gerarus\ both managed to interpret the venation of one wing (the
hind wing) and then assumed that fore and hind wings were identi-
cal, although neither actually saw the fore wing.
The advantage of having more material at my disposal made it
possible for me to overcome the problems that faced these workers.
Certain well-preserved specimens (especially FMNH-PE 5276,
31973, 32027; IP 5, 23) were instrumental in demonstrating the
complete venational differences between fore and hind wings. A
comparison of figs. 6a and 6b shows how strikingly different the
fore wing actually is from the hind wing. This, in itself, was quite a
revelation. But it was only later, when searching through the litera-
ture looking for venational similarities with other groups, that the
■^Prior to this, all Carboniferous insects were included in the one order Paleo-
dictyoptera in accordance with Scudder’s beliefs that ordinal differentiation had not
taken place 'n the Insecta as early as the Carboniferous. We know, of course, that this
was incorrect; a total of 1 1 orders are now recognized from that Period (Carpenter,
1977; Wootton, 1981).
10
Psyche
[Vol. 90
full significance of the discovery emerged. It became immediately
apparent, based on fore wing characters, that the type genus (Sthe-
naropoda) for the family Sthenaropodidae^ is inseparable from
Gerarus. The consequences of this are twofold: 1) it extends the
geographic range of the Geraridae from North America to Europe,
strongly suggesting that the family was once large and successful;
and 2) the synonymy casts serious doubts on current classifications
of Paleozoic orthopteroids such as those proposed by Sharov (1968)
and Rasnitsyn (1980). The implications of this are addressed in the
discussion section at the end of this paper.
Geological range: Carboniferous — Westphalian D to Stephan-
ian. Occurrence: Mazon Creek, Illinois, U.S.A.; Commentry,
France. Type genus: Gerarus.
Synonymies
The families Genopterygidae and Sthenaropodidae are synony-
mized here with the Geraridae, since 1 find no unique characters by
which to recognize them as independent taxa. All major veins and
body characters are in complete agreement with the definition of the
Geraridae. Although the Genopterygidae are described from the
hind wing alone, and this synonymy may therefore change as addi-
tional material is found, the venational similarities between Genop-
teryx and Gerarus are striking (see fig. 7). This, in my mind, is
sufficient reason at this time to synonymize these families. The syn-
onymy of the Sthenaropodidae with the Geraridae is based not only
on the venation of both wings, but also on the prothorax (complete
with spines) and body size. The two families are so similar in charac-
ter that synonymy at the species level could almost be justified were
it not for their separation both geologically and geographically.
^The Sthenaropodidae, like the Geraridae, were the focus of taxonomic work
largely at the turn of the century. Brongniart first described Sthenaropoc/a (based on
S. fischeri) in 1885 and placed it with a series of other Carboniferous Protorthoptera
in the family Paleoacridiodea. Eight years later he synonymized Sthenaropoda with
Oedischia (now recognized as belonging to the Orthoptera), believing their differ-
ences too slight to warrant generic separation. In 1906 Handlirsch restored the genus
Sthenaropoda and placed it in its own family. His decision was later defended by
both Lameere (1917) and Sharov (1968), who felt that the oedischiids, by virtue of
their saltatorial legs, were true Orthoptera, and that the sthenaropodids, which
lacked well-developed jumping legs, were clearly members of the Protorthoptera.
The ramifications of this are discussed in the concluding pages of this paper.
Fig. 3. Handlirsch’s reconstructions of two species of Gerarus. a. C7. danielsi\ b.
G. collari.s (^lon^icullis). (From Handlirsch, 1920:152,153).
12
Psyche
[Vol. 90
Genus Gerarus
Gerarus Scudder 1885:344; Handlirsch 1906a: 147, I906b:702, 1911: 313, 1919:38.
Type: Gerarus vetus Scudder (original designation).
Sihenaropuda Brongniart 1885:59; Handlirsch 1906a: 148, 1906b: 704, 1919:30. Type:
St henaropoda fischeri (or\g\n-d\ designation), nkw synony my.'’
Genopteryx Scudder 1885:327; Handlirsch 1906a: 148, 1906b:704, 1919:30. Type:
Genopteryx constricta (original designation). Nt:w synonymy.^
Archaeacridites Meunier 1909a:39. 1909c: 145; Handlirsch 1919:39. Type: Archaea-
cridites hruesi (original designation), nhw s^ non'i my.
Rossites Richardson 1956:44. Type: Rossites inopinus (original designation).
M W SYNONYMY.
Description
Fore wing: membranous, larger than hind wing, rounded at
apex; SC long, weakly turning anteriorly to fuse with C at point
three-fourths to two-thirds the length of the wing; R parallel to SC,
terminating in C slightly anteriorad to wing apex. Fore wings differ
from hind wings in the following veins: in the fore wing, RS
branches from R in the basal third of the wing, and bifurcates two
or three times. M four- or five-branched, either connecting to RS by
a cross vein or fusing with it; CUA coalesces with M for short
distance at wing base and may be weakly branched; CUP simple,
elbowed towards CUA; network of anal veins present.
Hind wing: RS has three to six distinct pectinate branches and
does not fuse with M; M multiply branched, arising from RS; CUA
^Some doubt exists concerning the date of publication of this paper with respect to
Scudder’s 1885 article, but 1 have concluded for the following reasons that Scudder
had priority of publication: 1 ) Although we do not know the month of publication
for Brongniart’s paper, we do know that Scudder’s was published early in April,
1885. Unfortunately, attempts to obtain the exact date of publication for Brongni-
art’s article from the Museum d’Histoire Naturelle de Rouen and the Societe des
Amis des Sciences Naturelles de Rouen have met with no response. 2) Citations of
these two papers (e.g., Handlirsch, 1906a, 1922) have consistently listed Scudder’s
paper before Brongniart’s. 3) Scudder’s 1885 account of Gerarus includes a full
description, figures, and designation of a type species {G. vetus), whereas Brongni-
art’s paper only mentions Sthenaropuda and gives no formal description.
''Gerarus and Genopteryx were named and described by Scudder in the same paper
(1885). In accordance with the I.C.Z.N. procedures, and as the first reviser, I have
treated Genopteryx as the junior synonym of Gerarus, the better known and
larger genus.
1983]
Burnham — Geraridae
13
simple or with one bifurcation; CUP simple and parallel to CUA;
anal area slightly expanded, but unusually reduced for the Orthop-
teroidea. See fig. 17.
Diagnosis
Gerarus may be distinguished from the other genera in the
Gerandae {Nacekomia, Progenenfonium, Genentonmm and Gerarulus)
by size (members of this genus are large, fore wing is 40 mm to 55
mm in length); and the nature of the RS and M veins in the fore
wing. In Gerarus RS branches two or three times; in Progenentonium,
it branches at least four times. M in Gerarus is four- or five-branched,
and either coalesces with RS for a short distance or is connected to
it by a well-developed cross vein. In contrast, M in Naeekomia is
distinct from RS, and in Genentonium, M is only three-branched
and these branches are distinctly parallel to one another. Other
characters such as the shape of the thorax and number of prothoracic
spines may ultimately prove important in distinguishing these
genera from one another, but as yet, we lack the well-preserved
specimens necessary for separating all four genera in the family on
the basis of such additional characters.
Remarks
Handlirsch (1911:313) characterized Gerarus by its prothorax,
described as “a broad base, either provided with tubercles or
smooth, but in every case, produced into a long neck-like part bear-
ing the head.” While he was correct about the nature of the “neck,”
he was incorrect in his assessment of the “tubercles,” which were
presumably present in all adult gerarids as fully produced spines,
not tubercles. He was also slightly inaccurate in describing the pro-
thorax as “a broad base.” This study has shown the width of the
prothorax to vary from 5 mm to 13 mm depending on the species. A
better description for the genus is one based on wing venation.
Geological range: Upper Carboniferous — Westphalian D to
Stephanian. Occurrence: Mazon Creek, Illinois, U.S.A.; Com-
mentry, France. Type species: Gerarus vetus.
Synonymies
As indicated in previous pages, clarification of the venation of
both fore and hind wings has led to several important synonymies.
A comparison of Sthenaropoda with Gerarus reveals that the vena-
tional differences lie largely in the number of branches of M and
14
Psyche
[Vol. 90
this, in my opinion, does not justify distinction above the species
level. Similarly, Archaeacridites, lacking distinct venational charac-
ters, cannot be separated from Gerarus.
I am also synonymizing two genera from the order Caloneurodea
with Gerarus: Genopteryx and Rossites, for which Richardson
(1956) erected the family Genopterygidae. Genopteryx, originally
described by Scudder (1885), and placed in the family Homotheti-
dae, was transferred subsequently to the Geraridae by Handlirsch
(1906a). Richardson (1956:41) removed Genopteryx from the Gera-
ridae and placed it in the order Caloneurodea. He did so on the
basis of its “heavy cross veins and the close straight parallel CUA
and CUP” these being the “two characters regarded by Carpenter
(1943) as prescribing inclusion in the order Caloneurodea.” Richard-
son then states that "'Rossites has delicate cross veins and its CUA
deviates from strict parallelism with CUP, yet the venation is nearly
identical with that of Genopteryx,''" and for that reason placed the
two genera in the same family. While these genera do seem to belong
together, 1 see no reason to include them in the Caloneurodea. One
result of the present study was the discovery that CUA and CUP are
typically parallel in the hind wing of the gerarids, and that place-
ment and number of the cross veins is variable. Therefore, with the
disappearance of the supposed diagnostic venational characters that
Richardson used to justify their inclusion in the Caloneurodea and
the discovery of synapomorphies by which they may be linked to
Gerarus, 1 feel that there is every reason to include these species in
the Geraridae. It is interesting to note, however, that the parallel
positions of CUA and CUP, characteristic of this family, may ulti-
mately indicate a closer relationship with the Caloneurodea than
previously recognized.
Gerarus vetus
Figures 4 and 5
Gerarus vetus Scudder 1885:344, 1890:308; Handlirsch 1906a: 147, 1906b:702,
1919:30.
Description
Fore wing: length 45-55 mm, width 13 mm; RS two- to
three-branched, fusing with M for short distance at point where M
elbows towards RS; M three-branched; CUA simple, CUP forked.
1983]
Burnham — Geraridae
15
Fig. 4. Gerarus veiu.s, a. composite drawing of the fore wing, based on DGLS I
and USNM 38136. b. composite drawing of the hind wing, based on DGLS I and
USNM 38136.
Hind wing: length 42-50 mm, width 11-12 mm; RS at least
four-branched; M forked; CUA and CUP not known.
Body: prothorax much smaller in this species than in G. danielsi.
Width about 5 mm at its widest point, length 10-15 mm. Largest
measurable spine 7 mm. Unfortunately, the arrangement and number
of spines in this species is uncertain, but nine are expected in
keeping with the genus. Head small, 4-5 mm in length. Coxae
possibly enlarged; tibiae and femora long and slender.
Diagnosis
This species is distinguished from G. danielsi on the basis of its
long and slender appearance, its diminutive prothorax and narrow
wings. Unfortunately, the venation is not sufficiently preserved in
any of the specimens assigned to this species to be useful as a diag-
nostic character. Although body length is intermediate between that
of G. danielsi and G. collaris, this species is clearly more slender
than the other species in the genus. Compare fig. 5 with figs. 9 and
13.
Geological range: Westphalian D. Occurrence: Mazon
Creek, Illinois, U.S.A.
Fig. 5. Gerarus vet us, photograph of specimen FMNH PE 32024. Note the
narrow prothorax, and head, on which a compound eye is visible. Length of wing, 52
mm. e - eye; s = spine.
Psyche
[Vol. 90
1983]
Burnham — Geraridae
17
Holotype: USNM 38136. Specimen examined. This specimen
consists of a prothorax, including spines, and the hindwings which
overlap one another. Unfortunately, details of the venation are
incomplete.
New' material
FMNH PE 32022. Obverse and reverse halves. The prothorax
and five spines are preserved in this specimen but unfortunately, due
to overlap, venational details are obscured.
FMNH PE 32024. Obverse and reverse halves. The head is well
preserved, and complete with a compound eye, clypeus, one mandi-
ble, and a spine base. Parts of all three legs are also preserved. The
enlarged prothoracic femur is probably an artifact of preservation,
the result of lateral compression during burial.
DOES 1. Obverse and reverse halves. This is a somewhat dis-
torted specimen with the prothorax pushed into the mesothorax.
Fore and hind wings on the right side are slightly splayed apart and
reveal most of the venation of the hind wing.
Wolff 301. Obverse and reverse halves. Only a fragmentary spec-
imen with a poorly preserved prothorax and spine bases.
DMRW 2 (Dabasinskas 2). Obverse and reverse halves. An
excellent specimen showing posterior part of the head, complete
prothorax (although the spines are broken), mesothorax, meta-
thorax, and wings, which unfortunately overlap. Its long and
slender appearance and slim prothorax place it in this species.
Gerarus danielsi
Figures 2, 3, 6, 7, 8, 9, 10 and 1 1
Gerarus danielsi HandVnsch 1906a: 147, I906b;703, 1919:30.
Gerarus longus Handlirsch 1906a:147, 1906b:702, 1919:30. new synonymy.
Gerarus angusius Handlirsch 1906a:148, 1906b:703, 1919:30. new synonymy.
Gerarus laius Handlirsch 191 1:313, 1919:30. new synonymy.
Gerarus reductus Handlirsch 191 1:314, 1919:30. new synonymy.
Genopieryx constricta Scudder 1885:327; Handlirsch 1906a:148, 1906b:704, 1919:30.
NEW SYNONYMY.
Rossites inopinus Richardson 1956:44. new synonymy.
Description
Fore wing: length 53-55 mm, width 17-19 mm; SC unbranched,
parallel to C, connecting to latter by multiple cross veins; R simple,
parallel to SC, terminating at wing apex; RS pectinate with 2 to 3
18
Psyche
[Vol. 90
branches, originating from R in basal third of the wing; CUP
simple, originating from very base of R, connecting to RS by a short
cross vein; M four-branched, fusing for approximately 9 mm at its
base with CUA; CUA also four-branched; CUP simple, arising
independently of CUA at the wing base. CUP elbows towards CUA,
connecting to it by short cross vein. Anal veins slender and
bifurcating. Well-developed reticulation present in anal area.
Hind wing: length 40-48 mm, width 14-16 mm; SC and R same
as in fore wing; RS pectinate with five branches, arising from R near
wing base; M arising from near base of RS and deeply cleft with two
or more terminal bifurcations; CUA and CUP parallel and
independent at wing base. Anal area not well preserved, only
slightly expanded, and with reticulated venation.
Prothorax: distinctly large and swollen posteriorly. There are
nine prominent spines symmetrically arranged in the swollen region
(see fig. 8). Width at widest point 10-13 mm, length 20-22 mm;
spines 7-10 mm in length. One, possibly two, vertical spines extend
from the anterior of prothorax, posterior to head.
Body: large, ranging from 70 mm to 75 mm (tip of wing to
anterior tip of prothorax). Legs long and thin.
Diagnosis
This species may be distinguished from the other species in the
genus by the large prothorax and well-developed spines (longer in
this species than in any other); and the overall body size which is
distinctly larger than that of either G. vetus or G. collaris from
Mazon Creek. Although venational characters do vary intraspecifi-
cally, it should be noted that in the fore wing M connects to RS by a
small cross vein, and that the anterior branch of CUP connects to
CUA also by a small cross vein. This contrasts with the other species
in the genus in which one finds an actual anastomosis of these veins.
Geological range: Westphalian D. Occurrence: Mazon Creek,
Illinois, U.S.A.
Holotype: USNM 35574. Specimen examined. Reverse half
only.^ The bulbous region and spines of the prothorax are well
preserved in this specimen. Only the costal margins of the fore wings
are present, but venation of the hind wings is clear, except in the
anal area. All evidence of the head and anterior region of the pro-
thorax has been lost.
'<The obverse half was originally in the Daniels collection, the location of which is
not known (see Carpenter, 1965, for further details on this collection).
1983]
Burnham — Geraridae
19
Fig. 6. Gerarus tlanielsi, a. composite drawing of the fore wing, based primarily
on FMNH PE 5276, 31973; and DMRW 1. b. composite drawing of the hind wing,
based primarily on USNM 35574, FMNH PE 32031, and MCZ 222.
sc
Fig. 7. Gerarus danielsi, originally Genupteryx eunsiricta. Drawing of hind wing,
based on USNM 38148. Compare this with the hind wing in Fig. 6b.
20
Psyche
[Vol. 90
Fig. 8. Gerarus clanielsi, photograph of the prothorax, PE 32029. Total length of
prothorax, not including spines, 21 mm, s = spine.
Synonymies
Gerarus latus YPM 33. Specimen examined. Obverse and reverse
halves. This species is synonymized here with G. danielsi by virtue of
its size (hind wing as preserved is 45 mm long, but is short several
millimeters at its apex) and the shape of its prothorax. The latter,
despite some distortion, clearly has spines of the same length and
pattern as G. danielsi. A single spine base is present at the anterior
1983]
Burnham — Geraridae
21
Fig. 9. Gerarus cianielsi, photograph of the holotype, USNM 35574. Length ot
fore wing 42 mm, as preserved, fw = fore wing; hw = hind wing; s = spine.
22
Psyche
[Vol. 90
tip of the prothorax. Fore and hind wings overlap. Only the costal
area of the fore wing is visible, but most of the hind wing venation is
visible under close scrutiny.
Gerarus reductus YPM 35. Specimen examined. Obverse and
reverse halves. This is an unusual specimen in that two wings are
preserved in the same concretion, but nothing suggests that they
necessarily belong to the same specimen — or even to the same spe-
cies. The specimen is badly fractured into four distinct pieces, and
the two wings appear to be in different bedding planes. Handlirsch
(1911) described the two wings as fore and hind wings of the same
species, but expressed reservations concerning their generic assign-
ment. I am convinced that these wings do not belong to the same
specimen but believe instead that they are both hind wings belong-
ing to different species. The specimen that Handlirsch (191 1:315, fig.
20) considered to be a fore wing is here designated the lectotype of
G. reductus and it is herein synonymized with G. danielsi. The
specimen that he interpreted as the hind wing of G. reductus
(191 1:314, fig. 19) undoubtedly is a hind wing, but not sufficiently
well preserved to warrant family determination and it is placed here
in Protorthoptera incertae sedis.
Gerarus longus USNM 38822. Specimen examined. Obverse and
reverse halves. Fore and hind wings overlap but the venation is very
similar to that of the holotype of G. danielsi: RS is pectinate, with
five branches, and M is deeply cleft. Also the prothorax, although
badly preserved, does have spines, two of which are visible on the
left side. This, plus the size of the specimen (fore wing measures
approximately 55 mm, hind wing 44 mm) warrants synonymy of G.
longus with G. danielsi.
Gerarus angustus USNM 38811. Specimen examined. Obverse
half only. This is a poor specimen: all four wings overlap, and the
fossil has been weathered so the venation is only barely visible.
Nevertheless, in my opinion, the overall size of the specimen (fore
wing length is 55-57 mm) and the swollen prothorax justify its
synonymy with G. danielsi. Certainly it displays no unique charac-
ters by which it may be distinguished as a separate species.
Genopteryx constricta USNM 38148. Specimen examined. This
species was originally assigned to the Geraridae by Scudder (1885),
but later transferred by Richardson (1956) along with Rossites
inopinus (see below) to the Caloneurodea. Having examined both
1983]
Burnham — Geraridae
23
type specimens, I find no characters by which to separate either
genus from Gerarus. Because there are no significant differences in
venation between R. inopinus, G. constricta and G. danielsi (com-
pare fig. 6 with fig. 7), synonymy at this point seems justified.
Rossites inopinus FMNH PE 3304. Specimen examined. Obverse
and reverse halves. Only the basal half of the hind wing is preserved,
but it shows CUA and CUP very clearly. Length of the wing as
preserved measures 29 mm; actual length is estimated as 40 mm.
New material
FMNH PE 5276. Obverse and reverse halves. This specimen is,
without doubt, the most spectacular of all specimens examined for
this study. Both halves are excellent, and the obverse half gives a
particularly good three-dimensional effect (see fig. 11). The latter
also shows the entire prothorax and part of the head. The base of
the vertical spine at the anterior end of the prothorax may be seen in
the reverse half. Parts of all three legs are visible in the specimen and
unequivocally demonstrate the gracile nature of the femora.
FMNH PE 31973. Obverse and reverse halves. An almost perfect
specimen of a single G. danielsi fore wing. The apex of the wing is
missing, but the anal area is remarkably well preserved in this
specimen.
FMNH PE 31988. Obverse and reverse halves. This is a poor
specimen: fore and hind wings overlap, and are only partially pres-
ent. However, venation and size both place it in G. danielsi.
FMNH PE 32023. Obverse and reverse halves. This is not a well-
preserved specimen, but venation and size both conform to the
species description.
FMNH PE 32027. Obverse half. The prothorax, pterothorax, and
basal areas of the right hind wing and left fore wing are evident. The
prothorax bears the characteristic arrangement of nine spines and
also has a tiny lateral spine projecting from its anterior left side.
Although smaller than the other specimens in this species (width of
fore wing is 12 mm) it is included in G. danielsi because it is, in all
other respects, identical to the holotype.
FMNH PE 32029. Obverse and reverse halves. This specimen,
which has an impressive array of spines on the prothorax, and a
vertical spine at its anterior tip, is magnificent. Hind wings are pre-
served, but overlap.
24
Psyche
[Vol. 90
Fig. 10. Gerarus danielsi, photograph of FMNH PE 5276, obverse half. Length
of fore wing 50 mm, as preserved, s = spine; fw = fore wing; hw = hind wing.
1983]
Burnham — Geraridae
25
Fig. 11. Gerarus Janielsi, stereophotograph of FMNH PE 5276, reverse half.
Total length = 77 mm. Photograph by F. M. Carpenter.
FMNH PE 32031. Obverse and reverse halves. This is an excel-
lent specimen that shows the venation of the hind wing, and an
outline of the prothorax, complete with spines.
DMRW 1 (Dabasinskas 1). Obverse and reverse halves. This is a
beautifully preserved specimen, showing almost the entire fore wing,
and two-thirds of the hind wing. It differs from the holotype in the
nature of M, which has only three major branches, but is otherwise
consistent with G. danielsi. Recognition of DMRW 1 as a new
species merely on the basis of M, given that nothing is known of the
body, does not seem warranted at this time.
26
Psyche
[Vol. 90
Wolff 491. This specimen consists of head, thorax, and the basal
area of two wings, but nothing can be made of the venation. The
prothorax is large and bears at least seven spine bases. The head is
preserved at a slight angle to the prothorax. Labrum is visible, as are
one antenna and both eyes.
MCZ 222. Reverse half. Costal margin of the fore wings and most
of the hind wings preserved. This insect is small for the species (hind
wing measures 45 mm long, 14 mm wide) but the venation is indis-
tinguishable from that of the holotype.
Gerarus collaris
Figures 12 and 13
Gerarus collaris Handlirsch 1911:314, 1919:30.
Gerarus longicollis Handlirsch 1911:315, 1919:30. new synonymy.
Description
Fore wing: length 45-50 mm, width not known. Venation of fore
wing obscured in all specimens.
Hind wing: length 40-45 mm, width 10-12 mm; RS apparently
five-branched, M deeply forked; CUA and CUP parallel.
Prothorax: small, 1 1 mm in length and narrow (approximately 6
mm wide). Posterior, or bulbous region, 7-8 mm long. Broken
spines are present on all G. collaris specimens examined, but no
more than six can be seen on any one specimen.
Diagnosis
This is the smallest of the Gerarus species. Unfortunately, the
venation in all known specimens is not clear enough to serve as a
species level character. G. collaris is, therefore, best recognized by
its prothorax, which tapers gradually from the anterior to the poste-
rior end, and is much narrower and shorter than in other species of
Gerarus. The distinctive nature of the prothorax, and its usefulness
as a species-specific character, can be seen by comparing figs. 5, 1 1
and 13.
Geological range: Westphalian D. Occurrence: Mazon Creek,
Illinois, U.S.A.
Holotype: Gerarus collaris YPM 34. Specimen examined.
Obverse half. Fore and hind wings overlap in this specimen, only the
costal margin of the fore wing is well preserved. The prothorax is
intact and several spine bases are visible, but the overall preserva-
tion is mediocre.
1983]
Burnham — Geraridae
27
Fig. 12. Gerarus col/aris, drawing of hind wing, based on holotype YPM 34.
Synonymy
G. longicollis YPM 36. Specimen examined. Obverse half. A
fragmentary specimen. Wings are poorly preserved, and only RS
and CUA/CUP in the hind wing are visible. This species is syn-
onymized here with G. collaris on the basis of its prothorax, which
is small and tapers gradually from the posterior to the anterior end,
as it does in all known members of G. collaris.
New Material
USNM 38835. Obverse half. Only the base of the wings and the
prothorax are preserved in this small specimen. The prothorax,
which bears at least six spine bases, is clearly narrow and elongate,
and is the reason for including this specimen in G. collaris.
Gerarus fischeri
Figures 14, 15, 16 and 17
Sihenaropoda fischeri ^xongmdLXi 1885:59; Handlirsch 1906a:142, 1919:38.
Oedischia fischeri Brongniart 1894:559.
Sihenaropoda lerichei Lameere 1917:178. new synonymy.
Sihenaropoda agnusi Lameere 1917:178. new synonymy.
Description
Fore wing: length 40-50 mm, width 14-15 mm; SC parallel to C
turning upward to fuse with it at point that is two-thirds length of
wing; R parallel to C, terminating at wing apex; both SC and R
connecting to C and SC respectively by numerous sigmoidal cross
veins; RS diverging from R at midpoint of wing and branching
twice; M originating at base of R, anastomosing with RS basally
before branching off and forking once; CUP forked, originating
28
Psyche
[Vol. 90
Fig. 13. Gerarus coUaris, photograph of holotype, YPM 34. Length of fore wing
48 mm. fw = fore wing; hw = hind wing.
1983]
Burnham — Geraridae
29
a
Fig. 14. Gerarus fischeri, a. drawing of fore wing, based on specimens IP 4, 5, 7,
and 23. b. drawing of hind wing based on specimens IP 5, 7, and 10.
separately from CUA at wing base; anterior branch of CUP fusing
with CUA for approximately 9 mm before weakly breaking away;
multiple veins and well-developed reticulation present in anal area.
Hind wing; length 39-47 mm, width 13-14 mm; SC and R same as
in fore wing; RS parallel to R and pectinate, with number of
branches varying from four to five; spacing of these branches rela-
tive to one another also variable (in one specimen IP 2 the first
branch of RS is connected to the main stem of RS by a strengthened
cross vein, forming a small oval in the middle of the wing); M deeply
cleft with one or two branches; CUA and CUP parallel. The anal
area of hind wing not known for this species, but appears to be
slightly expanded, judging by overall wing shape. See fig. 17.
30
Psyche
[Vol. 90
Diagnosis
G.fischeri is remarkably similar to G. danielsi in many respects:
prothorax, size, venation (compare fore and hind wings of each in
figs. 6 and 14). The only obvious difference lies in the nature of
CUA and CUP in the fore wing, \nfischeri CUA forks only once (in
danielsi it has many small branches) and the anterior branch of
CUP fuses with CUA for a distance of 9 mm. In danielsi the two are
connected by a small cross vein.
Remarks
G. fischeri was first described by Brongniart for a series of
orthopteroid insects recovered from the Commentry Coal Basin.
The series is remarkable not only because it contains a large number
of individuals, but because most of these individuals are exception-
ally well preserved. Under these circumstances it is somewhat odd
that affinities between the Commentry species and the Mazon Creek
species went unrecognized for so long. Many of the Commentry
specimens (especially IP 5, IP 7, and IP 23) have most of the body,
including the prothorax, preserved and demonstrate the same
arrangement of spine bases seen in the Mazon Creek material.
Moreover, venation of the fore and hind wings in these specimens is
unequivocally clear. Handlirsch might have recognized the similari-
ties between Sthenaropoda and Gerarus had he examined the
Commentry material himself, but this is debatable since the fore
wing for Gerarus was unknown at the time. The similarities
between G.fischeri and G. danielsi, given above in the diagnosis, are
extraordinary. While separation of the two species on siich minor
morphological differences might be subject to debate, I have chosen
to recognize the two species as distinct from one another on geogra-
phical and geological grounds. G. danielsi comes from Mazon
Creek in North America (Westphalian in age) and G.fischeri from
Commentry in France (Stephanian in age).
Geological range: Stephanian. Occurrence: Commentry,
France.
Holotype: Gerarus fischeri. IP 5. Specimen examined. Obverse
half only. This is probably the most spectacular of all the Commen-
try gerarids and of great taxonomic significance because the wings
are splayed apart and venation of both fore and hind wings is read-
ily visible. The insect is preserved dorso-laterally and the three legs
19831 Burnham — Geraridae 31
Fig. 15. Gerarus fischeri, photograph of holotype, IP 5. Total length = 75 mm.
s = spine; h = head; a = antenna, f = femur; t = tibia; ts = tarsomere; fw = fore wing;
hw = hind wing.
on the right side are preserved, as are the thorax and head. Spine
bases are present on the bulbous region of the prothorax, although
the spines themselves have broken off (see fig. 15).
Synonymies
S. lerichei. Holotype. IP 23. Specimen examined. Obverse half.
S. agnusi. Holotype. IP 19/21. Specimen examined. Obverse and
reverse halves.
I am synonymizing these species with G. fischeri as there are no
obvious specific level differences by which they may be recognized.
The specimen of 5". lerichei is a well-preserved, dorsal compres-
sion of almost the entire insect. Because the wings are separated, it is
possible to interpret the venation of both fore and hind wings, and
especially that of the fore wing. The venation, the prothorax (includ-
ing spine bases) and the size of this insect are perfectly compatible
with G.fisheri.
The specimen of S. agnusi is a single fore wing, superbly pre-
served. Although the apex of the wing is missing, the basal area.
32
Psyche
[Vol. 90
Fig. 16. Gerarus fischeri, photograph of IP 7. Length of fore wing = 50 mm.
f ■ femur; fw ■ fore wing; hw - hind wing.
including cross-veins, is extraordinarily well-preserved, as are all
major veins. I cannot find sufficient differences between this speci-
men and the others already included in S. fischeri to warrant sepa-
rate species status.
Several species previously included in Sthenaropoda are trans-
ferred here to family, genus indet. These are S. elegantissima Meu-
nier and Sthenaropoda minor Handlirsch, the types of which 1 have
examined, and I do not believe are similar enough to Gerarus to
warrant inclusion in the family.
New Material
IP 7. Reverse half. This specimen, although fragmentary, does
show venation of fore and hind wings. The specimen is preserved
dorso-laterally; three femora on the left side are visible, but the rest
of the body including the thorax and head is missing. See fig. 16.
IP 8. Obverse half. This insect has both fore wings, a prothorax,
complete with spine bases, a head bearing moniliform antennae, and
1983]
Burnham — Geraridae
33
parts of all six legs present. The hind femora are not enlarged and
demonstrate unequivocally their cursorial nature.
IP 6. Reverse half. This specimen is not particularly well pre-
served due to apparent post-burial distortion of the insect. The
pro- and mesothoracic legs on the left side are detached from the
body, and the antennae, although present, are detached from the
head. Fore and hind wings on the left side overlap, but the venation
of the fore wing is preserved, and nothing of the hind wing. The
prothorax is largely intact and shows the spine bases.
IP 4. Reverse half. The fore and hind wings on the left side are
separated, and the venation of the left fore wing is clear. Unfortu-
nately, little can be seen of the remaining three wings.
IP 2. Obverse half. A single well-preserved hind wing. Anal area is
missing but may be folded under the wing. This wing differs from
most other gerarid hind wings because the first branch of RS con-
nects to the main stem of RS by a strengthened cross vein, forming a
small triangle in the center of the wing (see fig. 17c).
IP 3. Reverse half. This is a partially preserved insect and shows
most of the right hind wing but only a fraction of the other three
wings. It is interesting, however, for one feature: the right hind wing
shows an anastomosis of the first branch of RS with the main stem
of RS as seen in IP 2. Because the anastomosis in this specimen is
smaller than in IP 2, and present in only one of the hind wings, it
may be assumed that it is a form of intraspecific variation, and not
significant at a higher level.
IP 9. Obverse half. Although the venation is virtually obscured,
this specimen is important because the insect has been compressed
laterally and all six legs are spread apart. The fore legs are only
partially preserved, but the meso- and metathoracic legs on both
sides are magnificent. This is the only specimen from Commentry in
which one can count tarsal segments. There are five tarsomeres, and
a pair of tarsal claws. The prothorax and its spine bases are also
present in the fossil.
IP 10. Reverse half. This is a single hind wing and well preserved
except at the apex and in the anal area, which is folded over.
IP 1 1. Reverse half. The specimen is a single hind wing, and so
poorly preserved that the specimen is almost useless.
34
Psyche
[Vol. 90
Fig. 17. Gerarus fischeri, a. photograph of fore wing, specimen IP 19. Length 42
mm, as preserved, b. photograph of hind wing, specimen IP 10. Length 40 mm, as
preserved, c. photograph of hind wing, specimen IP 2. Length 40 mm, as preserved.
Note the small triangle formed by the anastomosis of one branch of RS with the main
stem of RS.
1983]
Burnham — Geraridae
35
Gerarus hruesi
Figures 1 8 and 19
Archaeacridites hruesi Meunier I909a:39.
Sihenaropoda hruesi Handlirsch 1919:39.
Description
Fore wing: length 45 mm (as preserved, estimated as 48 mm),
width 15 mm; SC terminating in apical third of wing, at C; R
parallel to C, connecting to it apically by several cross veins; RS
branching twice, each branch forking once distally; M expanded,
with five main branches; CUA four-branched, fusing with M at its
base; CUP elbowed towards CUA, connecting to the latter by a
strong cross vein; anal veins present; well-developed reticulation
present in area basal to CUA.
Hind wing: unknown.
Diagnosis
This species is based on a single, but almost perfectly preserved,
fore wing from Commentry. Meunier originally described bruesi
and assigned it to the genus Archaeacridites because he felt that this
species was in some way ancestral to the extant Acrididae (order
Orthoptera). While the relationships of the Protorthoptera (including
the Geraridae) to the true Orthoptera have yet to be resolved, 1 do
believe that synonymy of Archaeacridites with Gerarus is warranted.
1 have studied the holotype, and can find no characters to justify
separate generic status for this species. However, I do think that
species separation is warranted on the basis of CUA which connects
to M only by a cross vein and does not anastomose with it as in G.
fischeri. The nature of CUA in G. bruesi is much more reminiscent
sc
Fig. 18. Gerarus hruesi, drawing of fore wing, based on holotype no. IP 20.
36
Psyche
[Vol. 90
of G. danielsi, where CUA also connects with M by a short cross
vein. M, however, is more expanded (more branched) in G. hruesi
than in G. danielsi.
Geological range: Stephanian. Occurrence: Commentry, France.
Holotype: Gerarus hruesi. IP 20. Specimen examined. This
specimen is a single fore wing only, but beautifully preserved. All
veins except those at the very apex of the wing are clear and can be
interpreted without difficulty (see fig. 19).
Genus Genentunium
Genentonium Scudder 1885:329; Handlirsch 1906a: 144, 1906b:700.
Description
Fore wing: SC and R parallel to C; RS branched, originating
from R in basal third of wing; M distinctive with 3 to 4 branches, all
parallel; CUA parallel to first branch of M; CUP elbowed towards
CUA; anal area with several veins.
Hind wing: SC and R parallel to C; RS multi-branched, arising
from R near wing base; M, CUA, CUP, and anal veins not known.
Diagnosis
This genus may be distinguished from the other genera in the
family on the basis of M, which in the fore wing has the unique
branching pattern described above, and the strong topography of
the major longitudinal veins displayed by the two species assigned
here to this genus.
Geological range: Westphalian D. Occurrence: Mazon Creek,
Illinois, U.S.A. Type species: Genentonium validum Scudder.
Genentonium validum
Figures 20 and 21
Geneniomuni validum Scudder 1885:329; Handlirsch 1906a; 145, 1906b;700, 1919;40.
Genentonium Cockerell 1917:81. nhw synonymy.
Description.
Fore wing: length 45 mm (estimated), width 14 mm; SC parallel to
C, connecting to it by a series of cross veins; costal margin narrow;
R parallel to C; RS at least two-branched, originating from R in
basal third of wing; M three-branched, and distinctive for the genus;
CUA parallel to first branch of M; CUP elbowed towards CUA;
anal area with several fine longitudinal veins.
1983]
Burnham — Geraridae
37
Fig. 19. Gerarus hruesi, photograph of holotype no. IP 20. Length of fore wing
45 mm, as preserved.
Fig. 20. Genentonmni validunu drawings based on holotype no. USNM 38135. a.
fore wing. b. hind wing.
38
Psyche
[Vol. 90
Fig. 2 \ .Geneniumuni valiJuni, photographs of holotype no. USNM 38135. a. fore
wing. Length 42 mm, as preserved, b. hind wing. Length 40 mm, as preserved.
Hind wing: length 40 mm, width 14 mm; SC and R as in fore
wing; RS four-branched, originating from R near wing base.
Remarks
Originally described by Scudder as a member of the Homothetidae
(Neuroptera), this species was subsequently transferred to the family
Oedischiidae by Handlirsch (1906b) on the basis of M, which
anastomoses with RS in the fore wing. Of course Handlirsch had no
idea that this character is also found throughout the Geraridae. My
inclusion of Genentomum in the Geraridae is based on the study of
all major veins and these are completely consistent for the family.
Geological range: Westphalian D. Occurrence: Mazon Creek,
Illinois, U.S.A.
Holotype: USNM 38135. Specimen examined. Obverse and
reverse halves. Both fore and hind wings are preserved, although the
apex of the fore wing extended beyond the edge of the concretion
1983]
Burnham — Geraridae
39
and was therefore lost. Base and anal area of the hind wing also
missing. Wings are separated and almost at right angles to one
another.
Synonymies
Genentomwn carri USNM 65023. Obverse half. This insect is an
impression of the wings only. Although the wings do overlap, the
fore wing on the right side and the hind wing on the left side are
visible.
I am synonymizing Genentomum carri with Genentomwn validum.
Cockerell (1917) described this specimen as a new species on the
basis of R which he figured with a single anterior branch. After close
examination of the holotype, I find that he was incorrect in his
interpretation of this vein as a branch of R. Either it is a very weak
cross vein, or a wrinkle in the wing membrane. Branching pattern of
RS in both fore and hind wings is similar to G. validum, and M is
virtually identical in both.
The following genera and species are described from single
specimens (most from single wings). Their descriptions are therefore
somewhat approximate. There seems to be no justification for
removing any of these taxa from the Geraridae at this point,
although the discovery of more nearly complete specimens may
provide characters that will alter this arrangement.
Genus Progenentomum
Progenentumum Handlirsch 1906a: 145, I906b:70I, 1919:40.
Description
Fore wing: SC terminates in C at point two-thirds distance from
wing base to apex; R parallel to anterior margin of wing, fusing with
margin just before wing apex; RS pectinate with several branches;
M more branched than RS and elbowed distally, touching RS at
that point; branches of RS and M close to one another and parallel;
numerous cross veins present; CUA, CUP, and anal region not
known.
Hind wing: unknown.
Diagnosis
Progenentomum is close to Gerarus but separated from it by RS,
which has at least four branches in the fore wing. Compare with
Genentomum.
40
Psyche
[Vol. 90
Geological range: Westphalian D. Occurrence: Mazon Creek,
Illinois, U.S.A. Type species: Progenentomum carbonis.
Progenentomum carbonis
Figures 22 and 23
Progenentomum carbonis Handlirsch 1906a:145, 1906b:70l, 1919:40.
Description
Fore wing: length 30 mm preserved (estimated as 50 mm), width
15 mm; RS pectinate with four branches; distal branch of M elbows
up to touch RS. A distinct cross vein connects branches two and
three of M, and probably acted as a brace vein.
Diagnosis
This species differs from all others in the family in having a linear
series of punctations between R and RS. These may or may not
have been pigmented, but because this specimen is only an impres-
sion, no organic material remains.
Geological range: Westphalian D. Occurrence: Mazon Creek,
Illinois, U.S.A.
Holotype: USNM 35580. Specimen examined. Obverse half.
Genus Nacekomia
Nacekomia Richardson 1956:33.
Description
Fore wing: SC terminates in C two-thirds from wing base; R
parallel to SC; RS two-branched; M four-branched; CUA and CUP
simple.
Hind wing: unknown.
Diagnosis
Nacekomia differs from Gerarus only in the nature of M, which is
separate from RS, and not connected to it except by several very
small cross veins. This, in my opinion, warrants separate generic
status but not separate family status.
Remarks
This monotypic genus was originally included in the family
Cacurgidae (order Protorthoptera), but is here transferred to the
Geraridae on the basis of its fore wing venation. While one cannot
1983]
Burnham — Geraridae
41
SC
Fig. 22. Progeneniumuni carhonis, drawing of fore wing based on holotype no.
USNM 35580.
Fig. 23. Progenentomum carhonis, photograph of fore wing, based on holotype
no. USNM 35580. Length 30 mm, as preserved.
be certain that this genus belongs in the Geraridae until a more
nearly complete specimen is found, the fore wing is so similar to that
of Gerarus that I do not hesitate to include it in the family. 1 cer-
tainly can see no justification for the inclusion of Nacekomia in the
Cacurgidae, where it was placed by Richardson (1956). The vena-
tion of Nacekomia differs considerably from that of Cacurgus. In
the latter, R is branched, RS simple, M is reduced, CUA simple, and
CUP many branched. In the former, R is simple, RS branched, M
has many branches, and CUA and CUP are simple.
Geological range: Westphalian D. Occurrence: Mazon Creek,
Illinois, U.S.A. Type species: Nacekomia rossae.
42
Psyche
[Vol. 90
Nacekomia rossae
Figures 24 and 25
Nacekomia rossae Richardson 1956:34.
Description
Fore wing: length 43 mm, but apex is missing; SC terminates in C,
two-thirds distance from wing base; R parallel to SC, terminating at
wing apex; RS two-branched, diverging from R at midpoint of
wing; M with four well-developed branches; CUA strongly convex
and fused at base with M, nature of CUP uncertain.
Hind wing: unknown.
Diagnosis
It is impossible to designate specific characters when the genus is
based on a single specimen, but in all probability the nature of RS
(with only two branches) may be important.
Geological range: Westphalian D. Occurrence: Mazon Creek,
Illinois, U.S.A.
Holotype: Nacekomia rossae FMNH PE 791. Specimen exam-
ined. Obverse half.
Genus Gerarulus
Gerarulus Handlirsch 1911:318, 1919:30.
Description
Fore wing: SC parallel to C, terminating on it; R simple, RS
branched; M with multiple branches, anastomosing briefly with RS;
CUA simple, CUP elbows towards CUA.
Hind wing: SC and R same as fore wing; RS pectinate; M simple;
CUA and CUP independent from one another and parallel.
sc
Fig. 24. Nacekomia rossae, drawing of fore wing based on holotype no. FMNH
PE 791.
1983]
Burnham — Geraridae
43
Fig. 25. Nacekomia russae, photograph of fore wing of holotype no. FMNH PE
791 . Length 43 mm.
Diagnosis
Although the prothorax is unknown, venation of fore and hind
wings is typical for the family. I have retained this as a distinct genus
only on the basis of the diminutive size of its one species, and this
may change as more material is found.
Geological range: Westphalian D. Occurrence: Mazon Creek,
Illinois, U.S.A. Type species: Gerarulus radialis.
Gerarulus radialis
Figures 26 and 27
Gerarulus radialis Handlirsch 1911:316, 1919:30.
Description
Fore wing: length 25 mm as preserved (estimated as 35 mm),
width 1 1 mm; SC parallel to C, terminating on it; R simple, RS
branched (at least two or three times); M anastomosing briefly with
RS, appearing to be four-branched; CUA simple, CUP elbowed
towards CUA; two anal veins visible, each forking once.
Hind wing: length 21 mm as preserved (estimated as 30 mm),
width 10 mm; SC and R simple and parallel to C; RS pectinate,
although the number of branches is unknown; CUA and CUP inde-
pendent from one another and parallel; anal area slightly enlarged;
abdomen, although indistinctly preserved, appears to be rather
slender; prothorax missing.
44
Psyche
[Vol. 90
Fig. 26. Gerarulus racUalis, drawings based on holotype no. YPM 37. a. fore
wing. b. hind wing.
Diagnosis
This is the smallest of all the gerarids and its size seems, at pres-
ent, to be the most distinguishing feature of this species.
Geological range: Westphalian D. Occurrence: Mazon Creek,
Illinois, U.S.A.
Holotype: YPM 37. Specimen examined. Obverse and reverse
halves. This specimen is somewhat unusual in that both fore and
hind wings are stretched out on one side of the body and do not
overlap at all. The abdomen appears to be rather slender. The pro-
thorax is missing.
Genus Anepitedius
Anepiteciius Handlirsch 1911:318, 1919:30.
Description
Owing to the poor state of preservation of the type specimen, it is
impossible to describe diagnostic characters for this genus.
1983]
Burnham — Geraridae
45
Remarks
It is clear that this genus belongs in the family Geraridae: the
prothorax is distinctively shaped, and the limited venational charac-
ters are in accordance with the family. Unfortunately, since this is a
monotypic genus and based on a single, poorly preserved specimen,
it is impossible to assess its relationship to other taxa in the family.
Geological range: Westphalian D. Occurrence: Mazon Creek,
Illinois, U.S.A. Type species: Anepitedius giraffa.
Anepitedius giraffa
Anepiiedius giraffa HancUirsch 1911:318, 1919:30.
Description
Fore wing: length 40 mm, width 10 mm; M converges with RS,
connecting to it by a short cross vein before diverging.
Hind wing: apical half of wing missing and only costal margin
visible.
Remarks
This species deviates from the other species in the family by hav-
ing a combination of narrow wings and a robust prothorax. Unfor-
Fig. 27. Gerarulus raclialis, photograph of holotype no. YPM 37. Length of fore
wing 25 mm, as preserved.
46
Psyche
[Vol. 90
tunately, the poor preservation of the one specimen known
pjecludes a more detailed species description.
Geological range: Westphalian D. Occurrence: Mazon Creek,
Illinois, U.S.A.
Holotype: YPM 38. Specimen examined. Obverse and reverse
halves.
Depositional Environment
For most of this century, Mazon Creek and Commentry have
reigned as the two major localities of Upper Carboniferous insects (a
third locality has just been added to this list; see Burnham, 1981).
Both have provided large numbers of superbly preserved specimens
and have contributed greatly to our understanding of early insect
evolution. Because the family under consideration is represented at
both places, a comparison of their geological history is warranted.
Roughly three hundred million years ago, in the Stephanian
Stage of the Upper Carboniferous, the Commentry Coal Basin was
a shallow lake — 9.6 km long, 3.2 km wide, and encircled by moun-
tains (Fayol, 1887; Stevenson, 1909). Two principal streams de-
scended from the surrounding mountains into the lake, where deltaic
swamps formed from the deposition of fine-grained sediments.
These were, in many respects, typical coal swamps, characterized by
Cordaites, and, in lesser numbers, other coal swamp flora such as
Lepidodendron and Stigmaria. Fossilization at the site was almost
instantaneous, the result of flooding that deposited massive amounts
of sediment in the lake and bordering swamp. The remarkable pres-
ervation of the Commentry insects would not have been .possible
without their immediate burial under these catastrophic conditions,
and it is assumed that they were buried with minimal post-mortem
transportation.
The first fossils at Commentry were discovered in the mid-
nineteenth century as a result of extensive coal exploration in the
central region of France, and were made accessible to collectors
only because of intense mining activity in the area. Once the coal
supply began to diminish, about 1915, the mines were closed down
and filled in, and further fossil collecting prohibited. For an histori-
cal account of the Commentry collections and a review of the litera-
ture on Commentry insects, see Carpenter (1943).
1983]
Burnham — Geraridae
47
Mazon Creek, in contrast, was once part of a major delta on the
edge of the Illinois Basin Sea. Periodic floodings in this area
resulted in the burial and preservation of a wide diversity of organ-
isms, both plant and animal. Two assemblages are recognized: the
Essex fauna (mostly marine organisms); and the Braidwood assem-
blage (freshwater to brackish flora and fauna). According to
Richardson (1956:1 1-12) the Braidwood fossils “represent the fauna
that lived on an aggrading plain, [just] above sea-level” and con-
sisted of “more than 200 species of small animals, including insects,
arachnids, mussels, and amphibians.” Over 140 species of insects
have so far been described from this locality (Richardson, pers.
comm.) and many of the specimens show exceptionally fine detail.
Unlike the Commentry fossils, which are preserved in shale,
Mazon Creek fossils are found primarily in iron carbonate or side-
rite concretions. These concretions (also called nodules) form due to
decay of the organism contained within them, but will do so only
under the right conditions (iron-rich sediments, high pH, rapid bur-
ial). They are characteristic of certain Upper Carboniferous coal-
bearing strata and have been recorded from localities in the United
States, France, England, and Germany. Nodules are shaped roughly
according to the dimensions of the organism they contain and can
be split along the bedding plane to reveal their fossilized contents.
Preservation is generally good, although appendages (particularly
legs) are frequently lost due to insufficient chemical reaction in the
extremities. For a more detailed account of concretion formation
see Woodland and Stenstrom (1979).
The Mazon Creek biota has been known since the middle of the
nineteenth century (Nitecki, 1979), but their initial discovery (unlike
Commentry) was due to the erosion of fossil-bearing strata by
stream action rather than by mining exploration. Concretions
washed out by the stream (Mazon Creek) accumulated along its
banks, and were found there by local collectors. Eventually the area
became the focus of extensive mining exploration and several pit
mines were dug in an effort to obtain coal. This was enormously
beneficial to paleontologists because it exposed great numbers of
concretions that then became available for study. Although most of
the mining has now ceased, at least one mine remains open (pit
eleven) from which fossils are still being collected, primarily by an
48
Psyche
[Vol. 90
avid corps of amateur collectors, many of whom have made their
finds available for scientific study.
The presence of the Geraridae at both Mazon Creek and at
Commentry may seem somewhat surprising,*^ The Mazon Creek
locality is roughly 5 to 10 million years older than that at Commen-
try and the two formed under quite different circumstances. How
can the presence of Gerarus at both be explained?
To find an answer, one must look at land mass movements during
the Carboniferous, and at their influence on climatic patterns and
continental distributions (see fig. 28). The collision of the continents
Gondwana and Laurasia during this period had two major conse-
quences. These were 1) the formation of the Allegheny Mountain
range in North America; and 2) the alignment of eastern North
America and western Europe so that they were contiguous at zero
latititude. The significance of these events for the family Geraridae
is twofold. One, the separate land masses were fused into a single
continent, and two, their new position along the equator resulted in
the formation of extensive coal swamps throughout North America
and Europe. These events made dispersal of insects from one region
to the other relatively easy. Although the creation of the Allegheny
Mountain chain may have acted as a barrier to dispersal for some
insects (and other animal and plant species), this was probably not
so for those that were strong fliers. It is likely, therefore, that the
Geraridae were able to cross the barrier, and in so doing, passed
from one coal swamp habitat to another. It is assumed, being
orthopteroids, that they were herbivores, and probably restricted in
their feeding habits to plants found in these swamps. It is not sur-
prising, then, that they should have been so widespread and success-
ful during the Upper Carboniferous. For the same reason, it is not
surprising that they died out by the end of the Carboniferous when
climatic changes led to the drying up of the great coal swamps and
the concomitant extinction of the coal swamp fauna and flora.
“^Three other genera common to both these localities have previously been
reported. They are Honialoncura (Carpenter, 1964) and Spilaptera (Carpenter and
Richardson, 1971) in the order Paleodictyoptera, and Mischoptera (Carpenter and
Richardson, 1968) in the order Megasecoptera.
1983]
Burnham — Geraridae
49
Fig. 28. Geographic map of the Upper Carboniferous. Stippled areas represent coal swamps. Note location of Mazon
Creek relative to Commentry. (After Bambach, et al, 1980).
50
Psyche
[Vol. 90
Discussion
Recognition of the Protorthoptera has had a varied and unsettled
history. Despite the unquestioned importance of the order in the
evolution of the higher Insecta (as ancestors to extant Orthoptera
and possibly to the Holometabola) there is as yet little agreement
about affinities within the group. Our understanding of relation-
ships within the order is still rudimentary. This is well documented
by the present study in which the families Sthenaropodidae and
Geraridae (previously assigned to two distinct orders) are synony-
mized. Many attempts have been made to separate the Paleozoic
Orthopteroidea into more “natural lineages,” but it is currently pro-
posed (Carpenter, 1966) that recognition of one order Protorthop-
tera {sensu lato) is preferable until a better understanding of the
group’s true phylogeny emerges. While this forces acknowledgement
of the Protorthoptera as a “taxonomic wastebasket” and the group
“as thus constituted is almost certainly polyphyletic” (Carpenter,
1966), adoption of a presumably phylogenetic classification would,
at this time, only misrepresent the actual evolutionary relationships
of these insects. 1 believe that previous work on the Protorthoptera
(particularly the Sthenaropodidae) is a good example of how such
misrepresentation can occur as the result of inadequate study of a
given fossil group.
The Protorthoptera were first recognized in 1906 when Hand-
lirsch split the Paleozoic orthopteroids into three orders: the Protor-
thoptera, Protoblattodea, and Protorthoptera vel Protoblattodea
(for species that seemed to merge the characteristics of the first two).
In 1938 the Soviet scholar Martynov made an alternative sugges-
tion: that the fossil Orthopteroidea be divided into the two orders
Protorthoptera and Paraplecoptera according to whether they pos-
sessed saltatorial legs, as in the Protorthoptera, or cursorial ones, as
in the Paraplecoptera. The Geraridae were placed at this time in the
Paraplecoptera, and Martynov considered them, on the basis of size
and cursorial legs, to be typical representatives of that order. Sharov
(1960, 1962) originally suggested that these orders be reorganized
into the Protorthoptera, Paraplecoptera, and Protoblattodea, but
later (1968) expressed agreement with Carpenter that the Paraple-
coptera and Protoblattodea cannot be recognized as distinct orders.
1983]
Burnham — Geraridae
51
Convinced that the Protorthoptera should reflect direct relation-
ship to the Orthoptera, Sharov narrowed the order to include only
a single family, the Sthenaropodidae, synonymized here with the
Geraridae. Unfortunately, not having had the opportunity to
examine the Commentry types, he erroneously believed that the
family consisted entirely of saltatorial forms and used this to justify
its placement in the Protorthoptera. The various families previously
assigned to the Protorthoptera were placed in the Paraplecoptera
(containing the Geraridae), and the true Orthoptera. More complete
accounts of these various classifications are given by Carpenter
(1966, 1977) and Sharov ( 1968).
The most recent classification of Orthopteroidea was proposed by
Rasnitsyn (1980) in his work entitled The Historical Development
of the Insecta. Here he distributes the Paleozoic orthopteroids
among ten separate orders and proposes a new order Gerarida for
which Geraridae is the type family. The order Gerarida includes six
Mid to Late Upper Carboniferous families (previously assigned to
the Protoblattodea and Paraplecoptera), which Rasnitsyn consid-
ered related to one another on the basis of their elongate protho-
races and free, highly mobile heads. He includes in this order the
Eucaenidae, Spanioderidae, Dieconeuridae, Ischnoneuridae, Cne-
midolestidae, and Geraridae.
Rasnitsyn ( 1980: 165) admits that recognition of the Gerarida and
its division into these families is “extremely provisional owing to
insufficient study of its members.” Inasmuch as this revision of the
Geraridae has shown the degree to which detailed study of a particu-
lar taxon can affect higher levels of paleoentomological classifica-
tion, it would seem premature to accept Rasnitsyn’s ordinal
classification at this time. In my opinion, it is preferable to continue
to recognize the Protorthoptera sensu lato until we have valid syn-
apomorphies by which the true monophyletic groups in the Protor-
thoptera can be recognized.
The relationship of the Geraridae to other Carboniferous Protor-
thoptera must consequently remain unresolved. Nevertheless, there
are several interesting possibilities to consider. The first of these is
that Rasnitsyn may be correct in grouping together those families
with elongate prothoracic segments. It is perfectly possible that they
52
Psyche
[Vol. 90
represent a monophyletic offshoot of the Insecta that left no
descendants."^
Other possible relationships, however, may be construed on the
basis of venational characters, particularly the nature of M in the
fore wing. This vein is distinctive in that it either anastomoses with
RS for a short distance, or is connected to it by a cross vein. Because
a similar trend is seen in other groups of Protorthoptera, it may
suggest common descent. Families, aside from the Geraridae,
known to possess this anastomosis between M and RS are the
Streptocladidae, Oedischiidae, Nugioneuridae, and Tococladidae.
While many of these families (particularly the Streptocladidae) have
a much more complex venation than the Geraridae, it may be that
they represent an earlier stage in the evolution of the group — one
that led eventually to the saltatorial forms represented by the oedi-
schiids. Because the oedischiids were clearly saltatorial as far back
as the Carboniferous, it is reasonable to speculate that the gerarids
fall into a proto-saltatorial complex of Upper Carboniferous Pro-
torthoptera and may represent a line of evolution quite distinct from
that of the cursorial orthopteroids living today.
Summary
The family Geraridae, previously thought restricted to North
America, and known only from Mazon Creek, was apparently a
widespread and fairly successful group in the Upper Carboniferous.
Careful examination of Commentry Protorthoptera has resulted in
the synonymy of Sthenaropoda with Gerarus from Mazon Creek
and illuminates the problems inherent in the classifications pro-
posed by several recent authors. Recognition of the family Sthena-
ropodidae as the sole family of the order Protorthoptera and the
Geraridae as members of the order Paraplecoptera or Gerarida is no
longer tenable.
While further study is required to determine whether the Gerari-
dae are more closely related to the Mazon Creek families considered
"’Rasnitsyn is not the first to propose that the elongate prothorax is a
synapomorphic character. Others, especially Handlirsch, have already suggested that
the Geraridae are close relatives of the Spanioderidae on this basis. An argument
against this relationship, however, is the fact that they have distinctly different
patterns of venation. (In the Spanioderidae CUA is multiply branched and R
branches only once. Neither character is true for the Geraridae.)
1983]
Burnham — Geraridae
53
by Rasnitsyn as belonging to the “Gerarida,” or to the oedischiid
complex of true Orthoptera, at least monophyly for the family is
now established. What remains is the task of clarifying the relation-
ships of these other families; not only in terms of their relationships
to each other, but to their extant descendants as well.
TABLE I. Past and Present Classifications of the Geraridae.
Classification proposed Classification proposed
by previous workers by Burnham in this article
Order Paraplecoptera
Family Geraridae
Gerarus
vet us
danielsi
longicollis
longus
angustus
talus
rectucius
coUaris
niazonus
Genopieryx
constricta
Gerarulus
radialis
Anepitedius
girqffa
Order Protorthoptera
Family Sthenaropodidae
Sthenarupoda
fischeri
elegantissima
hruesi
minor
a gnu si
lerichei
Order Orthoptera
Family Oedischiidae
Genentomuni
validuni
carri
Prugenentomum
carhonis
Order Protorthoptera
Family Geraridae (= Genopterygidae,
Sthenaropodidae)
Gerarus (— Sthenaropoda, Rossites,
Genopieryx, A rchaeacridiies)
veius
danielsi (= talus, reducius,
tongus, angusius, consirieius,
inopinus)
cottaris (= tongicottis)
fischeri (= terichei, agnusi)
hruesi
Genenioniuni
vatiduni (=carri)
Progeneniomum
carhonis
Nacekomia
rossae
Gera rut us
radiatis
A nepiiedius
girqffa
54
Psyche
[Vol. 90
Acknowledgements
Many persons have contributed in varying ways to this paper. I
would like to begin by thanking Professor Frank M. Carpenter for
his superb advice and guidance. 1 thank him for his demanding
sense of scholarship, his never failing encouragement of my re-
search, and his incredible ability to understand even my most pecul-
iar idiosyncrasies. In honor of his 80th birthday, and in recognition
of his role as friend, mentor, and adviser, I dedicate this paper to
him.
Robert Mawhinney is a special person and Harvard colleague
who deserves thanks for his keen, theoretical mind and inspirational
nature, both of which he has so willingly shared with me. I thank
him for the spirit with which he has gauged my progress and has
offered many helpful, insightful, and nonambiguous suggestions.
Professor James Wilkinson of the History and Literature De-
partment at Harvard has shown strong interest in the Geraridae and
provided encouragement and editorial assistance whenever needed.
His enthusiasm for paleoentomology and his support of my work
have meant a great deal to me.
Dr. E. S. Richardson, Jr., of the Field Museum was particularly
helpful in sending unidentified material to me for examination;
without this material this revision would have had little significance.
His warm and enthusiastic support of my research is gratefully
appreciated. His premature death is a great loss to many of us.
Professors W. L. Brown, J. L. Cisne, and G. C. Eickwort of
Cornell University read an earlier draft of this manuscript and
offered many excellent suggestions for its improvement.
Beverly Strassmann, Dr. Naomi Pierce, and Professors Ruth Tem-
ple and Geneva Sayre are colleagues and friends who provided en-
couragement and support at the right times. They are appreciated in
many ways.
Dean John Wootton of Cornell University offered assistance and
advice when they were most needed. I am grateful to him for this.
K. M. Horton provided invaluable typing and editing skills. Her
standards of excellence have greatly improved this manuscript and I
am grateful for her support.
The staff at the Institut de Paleontologie in Paris were warm,
helpful, and hospitable during my visit there. In particular, I would
like to thank Dr. J. C. Fischer (Directeur), Mme. Yvette Gayrard,
1983]
Burnham — Geraridae
55
Dr. Daniel Heyler, and Dr. Agnes Lauriat-Rage for their wonderful
assistance.
The following individuals very kindly supplied the specimens on
which this study depended: Mr. F. J. Collier of the United States
National Museum; Mr. David Douglass of Yachats, Oregon; Mr.
Daniel Damrow of Mosineee, Wisconsin; Mme. Y. Gayrard of the
Institut de Paleontologie; Mrs. J. S. Lawless of the Yale Peabody
Museum; and the late Dr. E. S. Richardson of the Field Museum of
Natural History.
Partial financial support of this research is gratefully acknowl-
edged to National Science Foundation Grant no. DEB 82-05398 (F.
M. Carpenter, Principal Investigator), the Society of the Sigma Xi,
Cornell Chapter, and to Sigma Xi, the Scientific Research Society,
for a Grant-in-Aid of Research.
Literature Cited
Bambach, R. K., C. R. Scotese and A. M. Ziegler.
1980. Before Pangea: The geographies of the Paleozoic world. Amer. Sci.
68:26-38.
Brongniart, C.
1885. Les insectes fossiles des terrains primaires. Bull. Soc. Amis Sci. Nat.
Rouen, pp. 50-68.
1894. Recherches pour servir a I’histoire des insectes fossiles des temps pri-
maires. These Fac. Sci. Paris, no. 821: 1-494.
Burnham, L.
1981. Fossil insects from Montceau-les-Mines (France): A preliminary report.
Bull. Soc. Hist. Nat. Autun no. 100: 5-12.
Carpenter, F. M.
1943. Studies on Carboniferous insects from Commentry, France: Pt. 1. Intro-
duction and families Protagriidae, Meganeuridae and Campylopteridae.
Bull. Geol. Soc. Amer. 54: 527-554.
1964. Studies on North American Carboniferous insects. 3. A spilapterid from
the vicinity of Mazon Creek, Illinois (Palaeodictyoptera). Psyche 71(3):
117-124.
1965. Studies on North American Carboniferous insects. 4. The genera Met-
ropatur, Euhleptus, Hapaloptera and Hadentumum. Psyche 72(2):
175-190.
1966. The Lower Permian insects of Kansas. Pt. 11. Psyche 73: 46-88.
1971. Adaptations among Paleozoic insects. Proc. N. Amer. Paleo. Conv.,
Chicago, 1969, vol. I, pp. 1236-1251.
1977. Geological history and evolution of the insects. Proc. 15th Int. Congr.
Ent., Washington, D.C., 1976, pp. 63-70.
56
Psyche
[Vol. 90
Carfentj r, F. M. and E. S. Rkhardson, Jr.
1968. Megasecopterous nymphs in Pennsylvanian concretions from Illinois.
Psyche 75(4): 295 309.
1971. Additional insects in Pennsylvanian concretions from Illinois. Psyche
78(4): 267-295.
COCKHRHLL, T. D. A.
1917. Fossil insects. Ann. Ent. Soc. Amcr. 10(1): 1-22.
Favol, H.
1887. Etudes sur le terrain houiller de Commentry. Premiere partie: Litholo-
gic et stratigraphic. Bull. Soc. Ind. Min. XV(3 & 4). Synopsis [in] Bull.
Geol. Soc. France 16(3) ( 1889).
Handlirscii, a.
1906a. Die fossilen Insekten und die Phylogenie der rezenten Formen. Ein
Handbuch fur Palaeontologen und Zoologen. Leipz., Engelmann. 1430
pp.
1906b. Revision of American paleozoic insects. Proc. U.S.N.M. 29: 661-820.
191 1. New paleozoic insects from the vicinity of Mazon Creek, Illinois. Amer.
J. Sci. 31: 297-326.
1919. Revision der palaeozoischen Insekten. Denkschr. Akad. Wiss. Wien
Math. Naturw. 96: 1-82 (51 1 592).
1920. Geschichte, Literatur, Technik, Palaeontologie, Phylogenie, Syste-
matik. Handb. Ent. 3: 1 18-306.
1922. Insecta palaeozoica [in] Fossilium Catalogus. 1. Animalia, pars 16, C.
Diener, ed.. pp. 1 230.
Lamhhrh, a.
1917. Revision sommaire des insectes fossiles du Stephanien de Commentry.
Bull. Mus. Paris 23: 141 200.
1922. Sur la nervation alaire des insectes. Bull. Class. Sci. Belgium, 1922:
138 149(transl. Psyche 30: 123 132, 1930).
Martynov. A. V.
1924. Sur rinterpretation de la nervation et de la tracheation des ailes des
Odonates et des Agnathes. Rev. Russd’Ent. 18: 145-174 (transl. Psyche 37:
245-280, 1930).
1938. Etudes sur I’histoire geologique et de phylogenie des ordres des
insectes (Pterygota). Trav. Inst. Paleont. Acad. Sci. URSS 7: 1-150.
Meunter, F.
1909a. Nouveaux insectes du Stephanien de Commentry. Bull. Mus. Hist. Nat.
Paris. 15: 37-49.
1909b. Nouveaux paleodictyopteres du Stephanien de Commentry. Ann. Soc.
Sci. Bruxelles. 33: 139-140.
1909c. Nouvelles recherches sur les insectes du Terrain Houiller de Commen-
try. Ann. Paleont. 4: 125-152.
Nite:cki, M. H.
1979. Mazon Creek fauna and flora: a hundred years of investigation [in]
Mazon Creek Fossils, M. Nitecki, ed.. Academic Press, pp. 1-13.
1983]
Burnham — Geraridae
57
Rasmtsvn, a. P.
1980. The historical development of the insects. Trans. Paleont. Inst. 175:
1-268.
Richardson, E. S., Jr.
1956. Pennsylvanian invertebrates from the Mazon Creek area, Illinois.
(Insects). Fieldiana Geol. 12: 15-56.
Sri DDKR, S. H.
1885. Palaeodictyoptera: on the affinities and classification of Paleozoic
Hexapoda. Mem. Boston Soc. Nat. Hist. 3: 319-351.
1890. The fossil insects of North America, vol. 1. The pretertiary insects.
Sharov, A. G.
1960. On the system of the orthopterous insects. Int. Congr. Ent., Wien, 1960,
vol. 1, pp. 295-296.
1962. The order Protorthoptera [in] Osnovy paleontologii, B. Rohdendorf,
ed., Moscow, pp. 145-146.
1968. Phylogeny of orthopteroid insects. Tr. Paleontol. Inst. Akad. Nauk
SSSR 1 18: 1-218. (In Russian. English transl. 1971, Israel Program Sci.
Transl. Ltd., 251 pp.)
Stevenson, J. J.
1909. The coal basin of Commentry in central France. Ann. N. Y. Acad. Sci.
19(2): 161-204.
Woodland, B. G. and R. C. Stenstrom
1979. The occurrence and origin of siderite concretions in the Francis Creek
Shale (Pennsylvanian) of northeastern Illinois [in] Mazon Creek Fossils,
M. Nitecki, ed.. Academic Press, pp. 69-104.
Wootton, R. j.
1979. Function, homology and terminology in insect wings. Syst. Ent. 4:
81-93.
1981. Palaeozoic insects. Ann. Rev. Ent. 26: 319-344.
TESTING VISUAL SPECIES RECOGNITION IN PRECIS
(LEPIDOPTERA; NYMPHALIDAE) USING A COLD-SHOCK
PHENOCOPY
By Arthur M. Shapiro'
Department of Zoology, University of California
Davis, California 95616
There have been many studies of the role of color and pattern in
mating and species recognition in butterflies. For example. Crane
(1955) manipulated the bold color pattern of Heliconius spp. (Heli-
coniidae), affecting mating success; Burns (1966) claimed on the
basis of spermatophore counts that differential attractiveness of
female morphs helped to maintain a mimetic polymorphism in Papi-
lio glaucus L. (Papilionidae); and Silberglied, Aiello, and Lamas
(1980) found that modifying the pattern of Anartia (Nympha-
lidae) affected mating success but not survivorship.
Recently Hafernik (1983) demonstrated that the conspicuous pale
dorsal forewing band serves as a visual species-recognition charac-
ter, contributing to reproductive isolation between the partly sym-
patric buckeye butterflies Precis {=Junonia) coenia Hubner and P.
nigrosuffusa (Barnes & McDunnough). In hybridization experi-
ments these entities are quite compatible genetically and develop-
mentally; Hafernik concluded that differences between them “are
probably not associated with major genomic reorganization, but are
rather the result of allelic differences at a few loci,” including pre-
sumably those that control the presence or absence of the forewing
band.
The experiments done by Hafernik to test the hypothesis of visual
reproductive isolation were modeled on the work of Scott (1972),
involving presentation of reared virgin females to wild patrolling
males afield. There were four sets of experiments (i) actual combina-
tions of nigrosuffusa and coenia; (ii) coenia painted to resemble
nigrosuffusa; (iii) “wing transplants” (wings of one type glued onto
the wings of a living animal of the other; (iv) paper models. All of
these tended to indicate that coenia males discriminate against
^Manuscript received hy the editor January 6, 1983
59
60
Psyche
[Vol. 90
bandless females, and that species-specific pheromones need not be
invoked to account for reproductive isolation. None of Hafernik’s
females actually mated, but Scott (1972) showed that darkening the
wings of male coenia does not lower their courting success with
conspecifics, as it does when females are darkened.
Despite the consistency of these results, there are possibly con-
founded variables whenever one tests using entire genomes (as in i
above, in which pheromonal and subtle behavioral cues cannot be
controlled for), or altered phenotypes (as in ii and iii, where the
“similarity” to the other species is questionable, and wing loading
and odor may be altered by glues or paints). However, another test
is available, not exploited by Hafernik: pure coenia genomes can be
induced to produce nigrosuffusa-WkQ phenotypes, which may then
be presented to coenia males afield. This situation arises from the
sensitivity of coenia to temperature shocks applied shortly after
pupation.
A named aberration of coenia, “schracleri, ” figured in color by
Comstock (1927, plate 43), resembles nigrosuffusa in lacking the
band. Other characters, including the hindwing ocelli, are in the
^ coenia rather than the nigrosuffusa state. Schrader’s specimen was
reared, but similar individuals do occur in nature. One shown in fig.
If has the hindwing ocelli and distal pattern obsolescent; others are
normal for these pattern elements. The actual frequency of bandless
buckeyes is unknown. 1 have taken two at the same locality in eleven
years, during which time I must have seen hundreds of thousands of
individuals. No clear-cut genetic basis for bandlessness has been
established, but the same phenotypes are readily inducible by
subjecting wild California pupae to sustained low temperatures. Fig.
2 shows chilled individuals from three different families. The
extensive variation in individual response to treatment is character-
istic of such experiments. The involvement of the ocelli and distal
pattern is partly controllable by age of the pupa at onset of chilling,
but even very precise timing can only reduce, not eliminate, the
variation. Such indeterminacy was characterized as early as 1913 in
Pictet’s “law of melanization and albinization of parts,” which is a
statement of the partial independence of different pattern-deter-
mining processes during wing development.
Several broods of pure coenia from Solano County, California
were reared and subjected to a potent cold-shock treatment (3 weeks
1983]
Shapiro — Species recognition
61
Fig. I. Phenotypes of wild-collected Precis. A,C, male, B,D, female P. nigro.suf-
fusa from Arizona, USA and Sinaloa, Mexico. E, normal female P. coenia, Solano
Co., California. F. "\schraJeri'\ Suisun Marsh, Solano Co., CA. viii.28. 1978.
62
Psyche
[Vol. 90
at 3°C beginning 8 hr after pupation). The usual spectrum of
phenotypic response was observed. About one-third of the animals
which eclosed were seriously crippled and unusable for mating tests.
The remainder — 46 females in toto — were classified into three more
or less arbitrary phenotypic categories: (i) essentially unaltered (fig.
2e), (ii) bandless but with ocelli unaltered (figs. 2a-d), and (iii)
bandless and with ocelli obsolete (fig. 20- These were used in
experiments modeled on Scott’s and Hafernik’s, carried out on a
total of 9 days at Suisun City, Solano County, and Rancho
Cordova, Sacramento County, in urban vacant lots and annual
grassland from late September to early November 1982. Wild male
coenia were common throughout this period.
My methodology differed from Hafernik’s in a few points. Virgin
females were held, unfed and unflown, in the dark at 3°C for 3 to 1 1
days before use. This treatment did not diminish their attractiveness
relative to Hafernik’s females. They were transported in a cooler in
the dark to the study sites and allowed to warm in the sun (air
temperatures 14-24° C). After a test they were usually recaptured,
rechilled for at least 15 min, and re-used. A few were lost, and about
one-fourth mated successfully and were not re-used. As in Hafer-
nik’s work, only releases in which the male at least investigated the
female were scored. Females were considered to have elicited a
courtship if the male either attempted to copulate or remained
oriented toward the female for at least 20 sec. The durations of
about a third of the courtships were recorded.
The percentages courted were overall higher than seen by Hafer-
nik. For Point Richmond, California female coenia X male coenia
at Point Richmond, Hafernik had 64% courtship. When female
nigrosujfusa from Texas were used, this dropped to 10%. My corre-
sponding figures (table 1) are 74% and (pooled classes ii and iii)
49%. The difference remains highly significant, however, and the
discrepancy in frequency may be due to differences in weather
conditions or to the torpidity of my females. For timed courtships,
bandless females elicited less persistence than banded ones, but the
difference was not statistically significant. Most of the actual
copulations were essentially instantaneous, regardless of phenotype.
This experiment does not rule out pheromones in Precis court-
ship, but as in previous work indicates that visual cues are impor-
1983]
Shapiro — Species recognition
63
Fig. 2. Phenotypes of female P. coeniu from northern California, induced by
chilling the young pupa. A— D, grade ii (bandless, ocelli unaltered). E, grade i
(essentially unaltered). F, grade iii (bandless, ocelli obsolete).
64
Psyche
[Vol. 90
tant and possibly adequate to account for reproductive isolation. (It
is conceivable that pheromones could be physiologically coupled to
phenotypes such that the most phenotypically deviant females
would also be pheromonally abnormal.)
The control of pattern in Precis has been studied by Nijhout
(1980a, b), who has shown that all of the wing pigments in P. coenia
are melanins and that the ocelli are determined by well-defined foci
in the early pupal wing. His work does not permit a causal analysis
of how pupal chilling phenocopies the normal phenotype of nigro-
suffusa, though the phenocopy “straw” has been linked functionally
to its genocopy in Drosophila (Seybold, Meltzer, and Mitchell,
1975). In Shapiro’s (1981) model of the evolution of phenotypic
plasticity, a genetic basis for bandlessness could be established by
selection of modifiers bringing the latent ''schraderr response to the
surface under normal developmental temperatures. The derivative
character of bandlessness is shown clearly by its variable penetrance
(especially in females) in pure nigrosuffusa populations. But how
did it become virtually fixed? Discrimination by male coenia against
bandless females, even genotypicaly normal ones with intact wings,
should lead to selection against any bandless allele, however ori-
ginated. Under the conventional model for enhancement of pre-
zygotic reproductive isolating mechanisms in secondary sympatry,
one could rationalize bandlessness as a device protecting the gene
pool of nigrosuffusa. This, however, presupposes a disadvantage to
Table 1. Success of cold-shocked female Precis coenia in attracting courtships by
wild males in field tests in northern California.
Type of female
Number of
99
Total
releases
Number of
$9 mated
Number of
courtships
(i) essentially
-
unaltered
phenotype
18
42
6
31
(ii) bandless, ocelli
unaltered
20
55
5
27
(iii) bandless, ocelli
obsolete
8
17
0
8
Totals
46
1 14
1 1
66
'-test for courtship proportions, (i) \’.v. (ii + iii): z
= 2.7845 (significant at 0.01)
1983]
Shapiro — Species recognition
65
hybridization which outweighs the discrimination against bandless
females, and no such disadvantage has been found. Bandlessness
may be quite incidental to hybridization, but that still leaves the
question of why it persists.
Acknowledgements
This research was supported by the Department of Zoology, UC
Davis, to which 1 express my thanks. This paper is dedicated to the
memory of Robert E. Silberglied, who would no doubt have had
great fun with shock phenotypes in behavioral studies.
References
Burns, J. M.
1966. Preferential mating \’.v. mimicry. Science 153: 551-553.
COMSTOC K, J. A.
1927. Butterflies of California. Publ. by author, Los Angeles. 335 pp.
Crane, J.
1955. Imaginal behavior of a Trinidad butterfly, Heliconius erato hydara
Hewitson, with special reference to the social use of color. Zoologica 40:
167-196.
Hafernik, j. E., Jr.
1983. Phenetics and ecology of hybridization in buckeye butterflies (Lepi-
doptera: Nymphalidae). Univ. Calif. Pubs, in Ent., 96: 1-109.
Nijhout, H. F.
1980a. Pattern formation on Lepidopteran wings: determination of an eyespot.
Devel. Biol. 80:267-274. 1980b. Ontogeny of the color pattern on the
wings of Precis coenia (Lepidoptera: Nymphalidae). Devel. Biol.
80: 275-288.
Pic tet, A.
1913. Recherches experimentaies sur les mecanismes du melanisme et de
I’albinisme chez les Lepidopteres. Mem. Soc. Physique et Hist. Nat. de
Geneve 27: 1 1 1-278.
Scott, J. A.
1972. Comparative mating and dispersal systems in butterflies. Unpubl. Ph.D.
thesis, Univ. Calif. (Berkeley).
Seybold, W. D., P. S. Meltzer and H. K. Mitc hell.
1975. Phenol oxidase activation in Drosophila: a cascade of reactions. Biochem.
Genet. 13:85-108.
Shapiro, A. M.
1981. Phenotypic plasticity in temperate and subarctic Nymphalis antiopa
(Nymphalidae): evidence for adaptive canalization. J. Lepid. Soc. 35:
124-131.
Silberglied, R. E., A. Aiello, and G. Lamas.
1980. Neotropical butterflies of the genus Anartia: systematics, life histories,
and general biology (Lepidoptera: Nymphalidae). Psyche 86:219 260.
DEFENSIVE ADAPTATIONS AND NATURAL ENEMIES
OF A CASE-BEARING BEETLE,
EXEMA C/lA/4D£A5^/N(COLEOPTERA: CHRYSOMELIDAE)
By Richard B. Root and Frank J. Messina*
Section of Ecology and Systematics,
Corson Laboratory, Cornell University
Ithaca, New York 14853, U.S.A.
Introduction
The larval habit of constructing and carrying a portable case has
evolved many times in the Holometabola. It is a widespread trait
of the Trichoptera and Lepidoptera (e.g. the Coleophoridae and
Psychidae). Among the Coleoptera, casebearing is found in four
related subfamilies of the Chrysomelidae, the so-called camptoso-
mates: Clytrinae, Cryptocephalinae, Chlamisinae, and Lamproso-
matinae (Boving and Craighead 1931). The larval case of many
insects is thought to function primarily in defense by providing
armor or camouflage (Otto and Svensson 1980). Here we describe
the uses of the case and other defenses in a chlamisine beetle, Exema
canadensis Pierce, and speculate briefly on the evolution and conse-
quences of the case-bearing habit.
The genus Exema Lacordaire contains nine species in North
America (Karren 1966). All of the species appear to be univoltine
and to feed on a fairly restricted range of herbaceous or shrubby
genera in the Asteraceae (Jenks 1940; Karren 1966, 1972). In central
New York E. canadensis is commonly found on goldenrods {Soli-
dago spp.) and asters {Aster spp.). Its life cycle was summarized by
Messina and Root (1980). Le Sage (1982) recently described the
immature stages.
Methods
We observed the life history and natural enemies of E. canadensis
during 1979 and 1980 at Whipple Farm, 8 km N.E. of Ithaca, New
*Present address; Boyce Thompson Institute for Plant Research, Tower Road,
Ithaca, New York 14853, U.S.A.
Manuscript received by the editor January 24, 1983.
67
68
Psyche
[Vol. 90
York. Field-collected larvae and pupae were reared to measure the
incidence of parasitoids. Voucher specimens of E. canadensis and its
enemies were placed in the Cornell University Insect Collection (Lot
no. 1068).
The morphology of E. canadensis was examined with the scan-
ning electron microscope. Larval cases and adults were air-dried
and mounted on metal stubs with double-sided tape. Larvae and
pupae were dehydrated in an ethanol series and critical-point dried
with COt before mounting. All specimens were sputter-coated with
gold-palladium (ca. 200A). Measurements of larvae and cases were
made with an ocular micrometer.
We did experiments on the protective functions of the case by
exposing beetles to three predaceous insects: Podisus maculiventris
(Say) (Pentatomidae), Nahicula suhcoleoptrata (Kirby) (Nabidae),
and Hippodamia glacialis (F.) (Coccinellidae). For these experi-
ments we carefully extracted 4th-instar larvae from their cases; this
procedure did not appear to harm the larvae. “Exposed” and
untreated (“encased”) larvae were placed in petri dishes containing
moist filter paper and a few goldenrod leaves. In trials using P.
maculiventris, two 5th-instar nymphs were taken from a vigorous
lab culture and added to dishes containing three exposed and three
encased E. canadensis larvae (a choice situation). We recorded the
number of each prey type that were consumed by the stink bugs
after 6 and 24 h. In trials using A. suhcoleoptrata and El. glacialis,
field-collected adults were starved for 24, 48, or 72 h before being
added to dishes containing either exposed or encased E. canadensis
larvae (a no-choice situation). Each dish held three predators and
five prey. We recorded prey consumption hourly for up to 5 h.
Results
The larval case and adaptations associated with its use
In chlamisine beetles the female parent provides the initial larval
case in the form of an egg case or “scatoshell” (Hinton 1981). The
female deposits a single yellow egg that is attached to the plant on a
smooth, yellowish stalk (Figs. 1-3) that appears to be continuous
with the egg chorion. The attachment is shaped into the contours of
the leaf or stem surface (Fig. 3), suggesting that the base of the stalk
is extruded in a plastic state. The female then systematically sur-
rounds the egg with strips of green fecal material. She starts around
1983]
Root & Messina — Exenia canadensis
69
Figs. 14: Cases of E. canadensis. F Egg case with egg stalk attached to stem. 2,
View from top of case before cap has been added (egg was removed). 3, Close-up of
egg stalk and base. 4, Case of 1st instar larva. Arrow indicates juncture between
original egg case and larval additions. Scale bars = 500 (Fig. 4), 200 (Figs. 1-2), or
100 (Fig. 3) /um.
70
Psyche
[Vol. 90
the stalk and periodically twists the egg with her hind legs as she
builds up the sides until the egg is enclosed in a cuplike case with a
flared ridge (Figs. 1-2). A flat top is then added to seal the egg in the
case. The flexible egg stalk often remains twisted beneath the case
(Fig. 3). The entire deposition process takes 20-30 min in the labor-
atory. The egg case turns a drab brown color as it dries.
The larva emerges by chewing through the flat top of the case. It
then flips the case over, presumably after severing the connection
with the egg stalk (Karren 1972). Inside the inverted egg case, the
larva assumes the characteristic folded posture of the camptoso-
mates with the mouth and anus both adjacent to the single case
opening.
The larva begins to feed and gradually enlarges the case by adding
its own fecal material to the rim around the opening. The juncture
between the contributions of the mother and the larva remains dis-
tinct (Figs. 4-5), and the original egg case eventually appears as a
small nipple projecting from the tail of the larval case. A larva
passes through four stadia, always molting within the enlarging case
(Le Sage 1982). Case length is a moderately good predictor of larval
instar, as determined by the width of the head-capsule (Table 1).
Larvae of E. canadensis possess several morphological features
that are probably related to the case-bearing habit. The legs are
unusually long; each coxa is movable and so elongated that it
exceeds the length of the femur (Fig. 6). The legs can extend later-
ally beyond the rim of the case when the larva is walking. If dis-
turbed, the larva retracts its legs and pulls the case down so that the
rim is appressed to the foliage (Wallace 1970). The strongly recurved
tarsal claws (Fig. 7) may facilitate this maneuver by providing a
firmer grip on the substrate. The larval cuticle, which is normally
covered, is sclerotized in only a few areas (Le Sage 1982). Setae
(usually tricoid sensillae) are sparse, but spiny or rounded protuber-
ances are scattered over much of the surface. These protuberances
serve may to increase traction between the larval cuticle and the
case. The larval spiracles are uniforous and annular (Fig. 8). The
requirements for spiracular closure and moisture retention may be
reduced in a case-bearer; Karren (1964) reports that artificially
exposed Exema larvae are highly vulnerable to desiccation.
The prepupa seals the case rim to a leaf or stem with a layer of
frass. It then reorients itself so that the posterior end is against the
1983]
Root & Messina — Exema canadensis
71
Table I: Head capsule widths and case lengths (in mm) ol the immature stages of
E. canailensis.
Egg
'
11
111
IV
Head width
x(SE)
0.29(±.01)
0.37(±.02)
0.50(±.02)
0.67(±.02)
N
7
7
7
9
Case length'
x(SE)
1.05(±.05)
1.62(±.38)
2.69(±.33)
3.55(±.32)
4.34(±.26)
Range
0.96 1.12
1.04 2.32
2.16 3.36
2.80 4.08
3.84 4.80
N
12
23
18
32
45
'As measured from case opening to tip of original egg case.
substrate and the anterior end faces the nipple at the tail. Fully
sclerotized adults cut a circular cap in the tail of the case with their
mandibles; this cap is pushed off as the beetles emerge. The barrel-
shaped, tuberculate adults (Figs. 9 10) can be easily mistaken for
caterpillar frass by humans (Jenks 1940; Karren 1964; and our per-
sonal experiences). It may be that vertebrate predators overlook
them in the same way. The adults exhibit the widespread chryso-
melid trait of quickly withdrawing the legs and dropping off the
substrate when they are disturbed. This escape mechanism is elabo-
rated in Exenia\ the deep sternal grooves (Fig. 10) allow the adult to
retract its appendages so completely that the falling beetle bounces
and rolls off the foliage. The compact adults also slide deeply into
the litter beneath the plant and often come to rest in a deep recess
where they are extremely difficult to find.
Natural enemies
No predators were seen to attack the larvae of E. canadensis
during the many hours that we and our associates, E. W. Evans and
J. A. Gowan, have spent observing the goldenrod fauna in the field.
The three species of predaceous insects used in our experiments,
however, were frequently observed to kill the larvae of other chry-
somelid species that are associated with Solidago in central New
York (Evans 1982; Messina 1982). In the laboratory, exposed E,
canadensis larvae were readily captured and eaten by these preda-
tors (Table 2). In contrast, few encased larvae were consumed even
though the confined space in the petri dishes must have increased
72
Psyche
[Vol. 90
Figs. 5-8: Case and larval morphology of E. canadensis. 5, Case of 3rd instar larva.
6, 3rd instar larva, lateral view. 7, Tarsal claw. 8, 2nd abdominal spiracle. Arrow
indicates juncture between original egg case and larval additions. Scale bars = 500
(Figs. 5-6) or 10 (Figs. 7 8) yum.
the frequency of encounter between predator and prey far above the
usual conditions in nature. The coccinellid, H. glacialis, never suc-
ceeded in capturing an encased larva and nine of the ten encased
prey consumed by the pentatomid, P. maculiventris, were taken
only after all of the exposed larvae in the dish had been eaten. The
rate that exposed larvae were consumed by N. subcoleptrata and H.
glacialis was increased by starvation (Fig. 1 1).
The protective function of the larval case is further illustrated by
its influence on predator behavior. The predator appeared to
approach in response to prey movement with the outcome that
attacks were launched, without apparent discrimination, on both
exposed and encased larvae. Attacks on exposed larvae were
quickly and invariably successful. Upon encountering an encased
larva, the predators with sucking mouthparts {N. subcoleoptrata
and P. maculiventris) touched the case with their forelegs and
extended their beaks. They were never able to penetrate the case
1983]
Root & Messina — Exenia canadensis
73
Figs. 9-10: Adult E. c anadensis. 9, Dorsal view. 10, Ventral view. Scale bars = 500
/urn.
even though they made repeated probes. In those instances when
these hemipterans did consume encased prey, they fed through the
case opening on the few occasions when a larva had been knocked
on its side. This is an unlikely event in nature because dislodged
larvae fall from the plant. The chewing predator, H. glacialis,
attacked the encased larvae by attempting to insert the mandibles
under the rim of the case; we never observed success in this
endeavor.
The case is an ineffective barrier to certain adapted paraSitoids.
Larvae at both field sites were parasitized by a Tetrastichus sp.
(Eulophidae); this was possibly T. chlamytis Ashmead, a species
that is only known to attack chlamisine beetles (Burks 1979). Rates
of parasitism ranged from 16 to 42% (Table 3). We obtained an
average of 8.6 Tetrastichus adults/ infested host (range, 5-14 wasps;
n = 37 hosts). Parasitoids emerged from larvae that were collected in
the field as both early (I-Il) and late (III-IV) instars. The cuticle of a
parasitized larva turns from white to black and the host dies shortly
before the time it would normally pupate. The wasps usually
emerged from the case opening, but a small exit hole was observed
74
Psyche
[Vol. 90
Table 2: Consumption of exposed and encased larvae of E. canadensis by three
arthropod predators in laboratory arenas.
% available prey consumed
Predator
Exposed
Encased
N'
P^
Podisus niaculiventris
69
19
54
.001
nymphs
Nahicula subcoleoptrata
70
7
30
.001
adults
Hippodamia glacialis
87
0
30
.001
'Number of each prey type offered.
-Chi-square test, where expected values assume equal consumption of each prey type.
in the case of a larva that had cemented the opening to the substrate
before it died. We could not determine if Tetrastichus females ovi-
posit through the case wall or under the rim. In the field, however, we
commonly observed Tetrastichus adults that remained perched on
the side of a larval case for prolonged periods. Perhaps these wasps
were waiting for the larva to move and thus expose a vulnerable
spot for oviposition.
^ N. subcoleoptrata H. glacialis
Hours after release
Predator starvation ;• 24,° 48, ° 72 h
Fig. 1 1: Consumption of exposed E. canadensis larvae by Nahicula subcoleoptrata
and Hippodamia glacialis in the laboratory. Predators were starved for 24-72 h
before release. Ten prey were offered per trial.
1983]
Root & Messina — Exenia canadensis
15
Six of 22 pupae that were collected on August 21, 1979, were
parasitized by Spilochaleis albifrons (Walsh) (Chalcidae), a species
that has previously been reported from Exenia dispar Lacordaire
(Burks 1979). One chalcid emerged from each host. No S. albifrons
were found in E. canadensis that were collected as larvae. If this
chalcid attacks only pupae, it must be able to oviposit through the
case wall because the case rim is sealed to the substrate by the
prepupae. In this regard, it is interesting to note that S. albifrons has
been taken from a broad range of unrelated case-bearing and leaf-
mining insects. Moreover, at least four of the seven species in the
side group, to which S. albifrons belongs, parasitize case-bearing
chlamisines and coleophorids (Burks 1979). These observations
suggest that the evolution of specializations in Spilochaleis is more
closely linked to the abilities required to penetrate materials that
cover the host than it is to factors that are more narrowly associated
with the host’s taxonomic affinity.
Larval mermithid nematodes were found in dissections of a few
field collected E. canadensis larvae. These parasites are functionally
similar to parasitoids, killing the hosts as they exit the body follow-
ing development (Nickle 1974).
Larvae of an erythraeid mite, Leptus sp., were found attached at
several locations on beetles. In a survey done in late August 1979,
67% of the 43 adult beetles sampled bore at least one mite. There
was an average of 1.7 mites on the infested beetles and as many as
five mites were found on a single host. Nothing is known about the
influence of these mites on the beetles.
Table 3: Per cent of E. canadensis larvae parasitized by a Tetrastichus sp. in 1979
and 1980. (Sample sizes in parentheses.)
Site
Collection date
Brooktondale 1979
3 July
29 July
21 August
21 August'
17(107)
16(57)
38(21)
27(22)^
Whipple Farm 1980
28 June
1 July
4 July
16 July
23(61)
42(19)
40(55)
31(13)
'These cases contained pupae; the larvae had cemented the case rim to the substrate
prior to collection.
-An additional 27% of the pupae were parasitized by Spilochaleis albifrons.
76
Psyche
[Vol. 90
Discussion
The defensive adaptations of the immature and adult stages of E.
canadensis are quite different even though they occur in the same
microhabitats, overlap in their seasonal occurrences, and encounter
similar predators. The cases that cover the eggs and larvae appear to
deter most, if not all, of the several invertebrate predators that
forage on goldenrods (see Messina 1982 for a list). Wallace (1970)
has found that the case of another chlamisine, Neochlamisus gibbo-
sus (F.) (= Anthrochlamys plicata F.), protects the larvae from
imported fire ants. The defenses of adult chlamisines require further
investigation. Nevertheless, it seems obvious that a variety of escape
mechanisms are derived from the adults’ morphology. As a conse-
quence of their hard, compact body form, adults are well-armored
against the initial thrusts of predators and they are more likely to
tumble into a refuge after dropping from the foliage. Furthermore,
because of their resemblance to caterpillar frass, adults may be over-
looked by many predators that rely on vision (Jenks 1940).
Many chrysomelids are chemically defended against predators
(e.g. Meinwald et al. 1977; Howard et al. 1982). Adults in the camp-
tosomate group, however, lack the defense glands found in most
chrysomelid subfamilies (Deroe and Pasteels 1982). This suggests
that chlamisine adults must rely primarily on the mechanical and
behavioral defenses discussed above.
The major enemies of E. canadensis are the parasitoids, S. albi-
frons and T. chlamytis. Specialized parasitoids have been highly
successful in overcoming most of the defenses (e.g. reflex bleeding,
fecal shields, glandular secretions) employed by chrysomelid larvae
to deter predators (Eisner et al. 1967; Wallace 1970; Matsuda and
Sugawara 1980).
Several characteristics of E. canadensis can be grouped into an
adaptive syndrome that is associated with the case-bearing habit.
This coordinated set of traits includes the bowed posture, long legs,
and other morphological adaptations that accommodate the larvae
to life within the confinement of a case. In addition, casebearing
probably influences other aspects of the natural history. For
instance, the time and case-building material that the female must
invest in each egg may result in a lowered reproductive rate. We
observed that 30 females laid an average of only 1.2 eggs per day
over a six-day period; Karren (1972) reports similar oviposition
1983]
Root & Messina — Exenia canadensis
11
rates. This low output is reflected in the females’ reproductive mor-
phology. In dissections we found that females of E. canadensis have
only four or five ovarioles per ovary and that each ovary never
contains more than one fully mature oocyte. Camptosomate beetles,
in general, have relatively few ovarioles per ovary (Robertson 1961 ;
Suzuki 1974; Mann and Singh 1979). Beetles may compensate for
gradual egg production by ovipositing over an extended period. In
central New York, overwintered females begin egg-laying in early
May and continue until mid-July.
The low fecundity of E. canadensis may be related to its normally
low and relatively stable population size. Over a three-year period,
the population densities of five other chrysomelid species that feed
on goldenrod fluctuated by at least an order of magnitude (Messina
and Root 1980). During this same period the population of E. cana-
densis varied less than twofold. Furthermore, during the course of
our long-term investigations on the goldenrod fauna at several local-
ities in central New York, we have yet to observe a host plant that
was significantly depleted by E. canadensis. Karren (1964) has also
noted the stable densities of Exenia populations. Le Sage (1982),
however, reported that during 1980-81, populations of E. canaden-
sis increased greatly over a large area in southern Canada.
The evolutionary steps that produced the case-bearing habit are
unclear. Since the larval case is added to the egg case, it can be
argued that the defense originated with the female’s habit of cover-
ing the eggs with fecal material (this may be mixed with secretions
from the anal gland; Hinton 1981). This initial step is exhibited by
other chrysomelids, e.g. the eumolpine, Chrysochus auratus (Fabri-
cius). The extant species of camptosomate beetles differ in their
manner of oviposition and egg case deposition. Some clytrine bee-
tles lay eggs in clusters (a typical trait of non-camptosomate chry-
somelids), with each egg connected to the substrate by a separate
stalk (Hinton 1981). A cryptocephaline, Pachybrachis bivittatus
(Say), apparently does not connect the egg to the substrate at all.
Instead, the female covers the egg with fecal material while holding
it with her hind legs, and then simply drops the egg to the ground
(Lawson 1976). Further comparative data on the details of egg-case
provisioning are needed to trace further the evolution of the case-
bearing habit and the often enigmatic phylogeny of the camptoso-
mate line (Mann and Crowson 1981).
78
Psyche
[Vol. 90
Acknowledgements
We thank E. W. Evans and M. K. Hausmann for technical
assistance. N. F. Johnson and B. M. O’Connor identified the parasi-
toids and mites respectively. The research was supported by NSF
grant DEB77-25120.
Summary
Morphological and behavioral defenses of Exema canadensis are
illustrated with scanning electron micrographs. In laboratory exper-
iments, the fecal case was shown to protect larvae from three pre-
daceous insects (a nabid, a pentatomid, and a coccinellid) that occur
in the same microhabitats with E. canadensis. Exposed larvae were
readily consumed by predators. The case did not deter parasitoids;
larvae were heavily parasitized by a eulophid, Tetrastichus sp., and
pupae were attacked by a chalcid, Spilochalcis albifrons. Other
enemies include mermithid nematodes and erythraeid mites. The
adaptive syndrome associated with the case-bearing habit and its
possible evolution are discussed.
Literature Cited
BOving, a. G. and F. C. Craighead
1931. An illustrated synopsis of the principal larval forms of the order Coleop-
tera. Entomol. Amer. 11: 1-351.
Burks, B. D.
1979. Chalcididae, pp. 860-873 and Eulophidae, pp. 967-1021. In K. V.
Krombein, P. D. Hurd, Jr., O. R. Smith, and B. D. Burks (eds.). Catalog
of Hymenoptera in America north of Mexico. Washington, D.C.,
Smithsonian Institution Press.
Deroe, C. and J. M. Pasteels
1982. Distribution of adult defense glands in chrysomelids (Coleoptera: Chry-
somelidae) and its significance in the evolution of defense mechanisms
within the family. J. Chem. Ecol. 8: 67-82.
Eisner, T., E. Van Tassell, and J. E. Carrel
1967. Defensive use of a “fecal shield” by a beetle larva. Science 158:
1471-1473.
Evans, E. W.
1982. Feeding specialization in predatory insects: hunting and attack behavior
of two stinkbug species (Hemiptera: Pentatomidae). Amer. Midi. Nat.
108: 96-104.
Hinton, H. E.
1981. Biology of insect eggs, vol. II. Oxford, Pergamon Presss. 778 pp.
1983]
Root & Messina — Exenia canadensis
79
Howard, D. F., M. S. Blum, T. H. Jonks, and D. W. Phillips
1982. Defensive adaptations of eggs and adults of Gastrophysa cyanea (Coleop-
tera: Chrysomelidae). J. Chem. Ecol. 8: 453-462.
Jenks, G. E.
1940. Dwarfs that live in their hats. Nature Mag., pp. 337-340.
Karren, J. B.
1964. Protective coloration and form in the North American genus Exema
(Chrysomelidae, Coleoptera). Proc. North Central Branch, Entomol.
Soc. Amer. 19: 77-79.
1966. A revision of the genus Exema of America, north of Mexico (Chrysome-
lidae, Coleoptera). Univ. Kansas Sci. Bull. 46: 647-695.
1972. A revision of the subfamily Chlamisinae of America north of Mexico
(Coleoptera: Chrysomelidae). Univ. Kansas Sci. Bull. 49: 875e988.
Lawson, F. A.
1976. Egg and larval case formation by Paehyhraehis hivinatus. Ann. Ento-
mol. Soc. Amer. 69: 942-944.
Le Sage, L.
1982. The immature stages of Exema eanadensis Pierce (Coleoptera: Chry-
somelidae). Coleop. Bull. 36: 318-327.
Mann, J. S. and R. A. Crow.son
1981. The systematic positions of Orsudaene Latr. and Syneia Lac. (Coleop-
tera, Chrysomelidae), in relation to characters of larvae, internal anat-
omy and tarsal vestiture. J. Nat. Hist. 15: 727-749.
M ANN, J. S. andJ. B. Singh
1979. Ovariole number in the family Chrysomelidae (Coleoptera: Polyphaga)
from northern India. J. Entomol. Res. 3: 217-222.
MATStIDA, K. AND F. SUGAWARA
1980. Defensive secretion of chrysomelid larvae Chrysomela viginiipunetata
eosteUa (Marseul), C. populi L., and Gastrolina depressa Baly (Coleop-
tera: Chrysomelidae). Appl. Entomol. Zool. 15: 316-320.
Meinwald, j., T. H. Jones, T. Eisner, and K. Hicks
1977. New methylcyclopentanoid terpenes from the larval defensive secretion
of a chrysomelid beetle {Plagiodera versieolora). Proc. Nat. Acad. Sci.
U.S.A. 74: 2189-2193.
Messina, F. J.
1982. Comparative biology of the goldenrod leaf beetles, Trirhahda virgaia Le
Conte and T. borealis Blake (Coleoptera: Chrysomelidae). Coelop. Bull.
36: 255-269.
Messina, F. J. and R. B. Root
1980. Association between leaf beetles and meadow goldenrods. (Solidago
spp.) in central New York. Ann. Entomol. Soc. Amer. 73: 641-646.
Nickle, W. R.
1974. Nematode infections, pp. 327-376. In G. E. Cantwell (ed.). Insect dis-
eases, vol. II. New York, Marcel Dekker.
Otto, C. and B. S. Svensson
1980. The significance of case material selection for the survival of caddis
larvae. J. Anim. Ecol. 49: 855-865.
80
Psyche
[Vol. 90
Robertson, J. G.
1961. Ovariole numbers in Coleoptera. Can. J. Zool. 39: 245-263.
SnzoKi, K.
1974. Ovariole number in the family Chrysomelidae. J. Coll. Lib. Arts,
Toyama Univ. 7: 53-70.
Wallace, J. B.
1970. The defensive function of a case on a chrysomelid larva. J. Georgia
Entomol. Soc. 5: 19-24.
STUDIES ON NORTH AMERICAN CARBONIFEROUS
INSECTS. 7. THE STRUCTURE AND RELATIONSHIPS OF
EUBLEPTUS DANIELS/ (PALAEODICTYOPTEKA)*
By Frank M. Carpenter
Museum of Comparative Zoology
Harvard University, Cambridge, Mass. 02138
Eubleptus danielsi was described by Handlirsch in 1906 from a
single, poorly preserved specimen in a concretion from the Francis
Creek Shale, Illinois (Middle Pennsylvanian). The systematic posi-
tion of the insect has been controversial. It was placed by Hand-
lirsch in a new family, Eubleptidae, in the order Palaeodictyoptera.
However, Martynov, in 1938, expressed doubt about its assignment
to that order, and in 1952 Laurentiaux transferred it to a new order,
Eubleptidodea, which Rohdendorf accepted in the Osnovy Paleon-
tologii in 1962. Neither Laurentiaux nor Rohdendorf gave a diagno-
sis of the new order, although vague reference was made to the
presence of large eyes and to the absence of pronotal lobes. From
my study of the reverse half of the holotype (all that is now known) 1
came to the tentative conclusion (1965) that the insect was a member
of the Palaeodictyoptera, probably related to the family Spilap-
teridae.
During the past decade, many additional specimens of Eubleptus
have been found in a strip-mine pit on the Will-Kankakee County
line, Illinois, mostly by private collectors. These new specimens,
some of which are exceptionally well preserved, have been loaned to
me for study. The purpose of this paper is to present the results of
my examination of these specimens and to discuss the relationships
of the insect, as it is now known.
I am grateful to Mr. Frederick J. Collier of the Department of
Paleobiology, National Museum of Natural History, Washington,
for the loan of the holotype of Eubleptus danielsi; and to Mrs. J. S.
Lawless of the Peabody Museum of Natural History, Yale Univer-
sity, for the loan of the holotype of Athymodictya parva, a synonym
♦Partial financial support of this research is gratefully acknowledged to the National
Science Foundation, Grant No. DEB 82 05398, F.M. Carpenter, principal investi-
gator.
81
82
Psyche
[Vol. 90
of danielsi. 1 am especially grateful to the following private collec-
tors for the opportunity of studying their specimens: Mr. Paul Harris,
now of Mountain Home, Arkansas; Helen and Ted Piecko, Chicago;
Mr. and Mrs. Francis Wolff, now of Port Charlotte, Florida; Mr.
Daniel Damrow, Mosinee, Wisconsin; Mr. Raymond Bandringa,
Willow Brook, Illinois; Mr. Joseph Pohl, Belgium, Wisconsin; and
Mr. Richard Rock, Crest Hill, Illinois. As will become apparent from
the account below, our present extensive knowledge of Eubleptus has
resulted mainly from their fossil collecting and their cooperation in
making the specimens available for study.
1 am deeply indebted to the late Dr. Eugene S. Richardson, Jr.,
formerly of the Department of Geology, Field Museum of Natural
History, for his unfailing cooperation and his assistance over the
past fifteen years in the course of our investigations on the insects in
the concretions from the Francis Creek Shale.*
Order Palaeodictyoptera
Family Eubleptidae Handlirsch
Eubleptidae Handlirsch, 1906a, p. 679; 1906b, p. 1 1 1.2
Eubleptidae Laurentiaux, 1953, p. 423.
Eubleptidae, Carpenter, 1965, p. 178.
Small species, with slender, pointed wings. Fore wing; SC extend-
ing nearly to wing apex, terminating on the costal margin; RS dichot-
omously forked, with 4 (rarely 5) terminal branches; M forking just
basad of the origin of RS; MA with a long fork; MP with 3 terminal
branches; CUA with a short fork; CUP with 3 (rarely 2) terminal
branches; 3 short anal veins present; relatively few cross veins,
unbranched, and forming a distinct pattern; archedictyon absent.
Hind wing: similar to the fore wing in venation but slightly broader
near or before mid-wing, the hind margin strongly curved. Body:
moderately slender; antennae very long and thin; head apparently
'Shortly before his death in January, 1983, Dr. Richardson and I completed a joint
paper on the Archaeognatha (Insecta) in the concretions. This will be published in
the next issue of Psyche.
2The family, genus, and species were described and designated as new in both of
Handlirsch’s 1906 publications; the 1906a article obviously has priority, since ijiany
of its pages are cited by number in the 1906b work.
T\gmt\.Eublep,usdanielsi. Photograph of specimen PH 1 5 (reverse) in the Paul Harris collection. Length of fore wing, 13 mm.
84
Psyche
[Vol. 90
small in dorso-ventral view, but eyes prominent and protuding;
beak well developed; prothorax short, with small and weak pronotal
lobes; mesothorax and metathorax subequal; abdominal segments
apparently with small lateral lobes; cerci very long; female with
short, curved ovipositor.
The family is known only from the Francis Creek Shale.
Genus Euhleptus Handlirsch
Euhlepius Handlirsch, 1906a, p. 681; 1906b, p. 111.
Athynwciictya Handlirsch, 191 1, p. 298. nhw synonymy.
Fork of MA at nearly the same level as the first fork of RS; first
fork of MP well before mid-wing, its posterior branch forked near
the wing margin.
Type-species: Eubleptus danielsi Handlirsch; by monotypy.
Euhleptus danielsi Handlirsch
Figures 1-8
Euhleptus danielsi H-dxxdWx^cK I906a:681; I906b:112; 1920:137. Rohdendorf, 1962:
54. Carpenter, 1965: 180.
Athyniodictya parva Handlirsch, 191 1:298. new synonymy.
Fore wing: length 13-14 mm; maximum width, 3. 5-3. 8 mm; hind
wing: length 13-14 mm; maximum width, 4-4.3 mm; length of
antennae (complete), 1 1 mm. The venational pattern is shown in
figure 2. Only slight individual variations seem to occur: RS usually
with four terminal branches, but a fifth, short branch may be pres-
ent; CUP usually with three branches, though the shortest one may
be absent. Head about 3 mm wide across the eyes, and about 1.5
mm long as seen from above (i.e., not including the beak, which is 3
mm long). Pronotum about 1 mm long and 2.5 to 3 mm wide,
including the small pronotal lobes; meso- and metathoracic seg-
ments apparently subequal, although the compression of the body
has probably altered the true proportions of both segments; the
abdomen is about 13 mm long and 2 mm wide at mid-length.
Holotype: no. 38731, U.S. National Museum of Natural History,
Washington (L.E. Daniels, collector). This is a poorly preserved
specimen, showing the proximal three-fourths of a fore wing and
very little of the hind wings and body. Handlirsch described the
species from both obverse and reverse halves, but when I examined
1983]
Carpenter — Euhleptus
85
Figure 2. Euhleptus danielsi. Venational diagram of fore and hind wings. SC,
subcosta; R, radius; RS, radial sector; MA, anterior media; MP, posterior media;
CUA, anterior cubitus; CUP, posterior cubitus; A1 and A2, anals. Drawing based
mainly on specimen PH 15 Paul Harris collection, with some details from specimens
PE32046 and PE32045.
the specimen in 1965 only the reverse half could be found."* Having
now examined many additional specimens, I am convinced that
both Handlirsch and I incorrectly interpreted several of the vaguely
indicated structures in the type. Most of the cross veins that I de-
scribed and figured are obviously wrinkles in the wing membrane; in
the well-preserved specimens discussed below the cross veins are as
strongly developed as the longitudinal veins. Also, the structures
that 1 considered to be pronotal lobes are, in part, the large eyes to
which Handlirsch referred. The pronotal lobes are indeed present
but they are small.
Handlirsch’s Athymodietya parva, described in 1911 from a sin-
gle, poorly preserved specimen (YPM 18ab) in the Peabody Museum
^According to the records of the National Museum, counterparts of some of the
Daniels specimens were kept by Mr. Daniels after Handlirsch had studied them;
their present location is unknown.
86
Psyche
[Vol. 90
at Yale University, is without question a synonym of cianielsi. Hand-
lirsch mentioned a fine archedictyon on the wings and he placed the
insect in the family Dictyoneuridae, but I can find no suggestion of
it in the fossil. The matrix of that particular concretion is unusually
granular and I surmise that Handlirsch interpreted the granulation
as an archedictyon. If the fossil is moistened with alcohol, the char-
acteristic cross veins of Euhleptus are discernible. The venational
pattern, even as shown in Handlirsch’s drawing, is identical with
that of danielsi, although his figure incorrectly depicts some of the
veins with pectinate instead of dichotomous branching. The type of
parva is about the size of that of danielsi, the fore wing being 13
mm. long, with a maximum width of 4 mm.
Specimens of Eubleptus danielsi Studied
I have been able to examine seventeen specimens of danielsi dur-
ing this investigation. For convenience of reference, I include here
an annotated list of these:'*
1. National Museum of Natural History, Washington, No. 38731
(reverse half only). Mazon Creek. Holotype of danielsi. Poorly pre-
served, showing about three-fourths of a fore wing, but virtually
nothing of the body and hind wing. Fore wing, as preserved, 13 mm.
long.
2. Peabody Museum of Natural History, Yale University, No.
18. Mazon Creek. Holotype of Athymodictya parva. Poorly pre-
served, showing proximal portions of fore and hind wings, as well as
pronotum, pterothorax, and parts of abdomen.
3. Paul Harris collection, no. PH 15. Pit Eleven. Excellent pres-
ervation of entire insect, except end of abdomen; the best specimen
known. Especially good are the wings (which include the color
markings), the pronotum, and the head, which shows the antennae,
and eyes, and the location of the beak.
^There are apparently only two exposures of the Francis Creek Shale at which
specimens of danielsi have been found: Mazon Creek, the bed of the stream about 4
miles west and a mile north of Coal City; and Pit Eleven, a strip mine of the Peabody
Coal Co., in Will and Kankakee Counties, Illinois.
1983]
Carpenter — Eu hie plus
87
Figure 3. Euhleptus cianietsi. Photograph of specimen PH 15 (obverse), Paul Harris
collection. Dorsal view of head and thorax; a, antenna; h, head; e, eye; p, pronotum;
ms, mesonotum; mt, metanotum; cv, small cavity in the matrix of the concretion,
several millimeters deep and partially filled with kaolinite; cavity originally occupied
by the beak. Width of mesonotum, 3 mm.
88
Psyche
[Vol. 90
Figure 4. Euhleptus danielsi. Photograph of specimen 236, Daniel Damrow collec-
tion; dorsal view of thorax, but frontal view of head: b, beak; e, eye. The three
thoracic segments are partially covered by kaolinite. Maximum width of fore wing in
photograph, 3.5 mm.
4. Field Museum, No. PE32046 (J. Herdina collection, no.
H424). Pit Eleven. Very good preservation of all wings, especially of
basal parts; also, thorax and abdomen, including proximal part of
cerci. Head crushed.
5. Field Museum, No. PE32045 (J. Herdina collection, no.
H540). Pit Eleven. Good preservation of basal parts of all four
wings; body very poorly preserved.
6. Francis and Terri Wolff collection. No. 229. Pit Eleven. Good
preservation of basal portions of all wings; poor preservation of
body, but good view of head from above; one fore leg present.
1983]
Carpenter — Euhleptus
89
Figure 5. Euhleptus danielsi. Photograph of head of same specimen shown in
figure 4, with greater magnification and different illumination. Scale line is 1 mm
long.
7. Francis and Terri Wolff collection. No. 233. Pit Eleven. Good
preservation of basal two-thirds of wings; most of body not
preserved.
8. Raymond Bandringa collection. No. 66-PBSM-l l-(3). Pit
Eleven. Good preservation of basal parts of wings, part of antennae,
eyes, and thorax.
9. Field Museum collection. No. PE22016 (from Dwayne Stone
collection). Pit Eleven. Excellent preservation of whole insect in
lateral view; wings overlapped but venation clear; shows abdomen,
including cerci; head, including eyes; beak, in side view.
90
Psyche
[Vol. 90
10. Daniel Damrow collection. No. 236. Pit Eleven. Good
preservation of basal part of wings; thorax strongly compressed;
entire head well preserved in front view, showing beak, with excel-
lent preservation.
1 1. Joseph Pohl collection. No. MPH8. Pit Eleven. Good preser-
vation of basal parts of all four wings, with general features of body.
12. Helen and Ted Piecko collection. No. 402. Pit Eleven. Fair
preservation of basal parts of all wings, but body not clear.
13. Helen and Ted Piecko collection. No. 422. Pit Eleven. Good
preservation of most of all four wings and parts of thorax and
abdomen.
14. Helen and Ted Piecko collection. No. 432. Pit Eleven. Fair
preservation of body and of basal portions of all wings.
15. Helen and Ted Piecko collection. No. 436. Pit Eleven. Poor
preservation of wings and body.
16. Richard Rock collection, no. 729. Pit Eleven. Good preserva-
tion of body and of basal parts of fore and hind wings.
17. Richard Rock collection. No. 817. Pit Eleven. Good preserva-
tion of basal parts of wings, poorly preserved body.
A composite drawing of Eubleptus danielsi is given in figure 8.
The general habitus of the insect, as drawn, is based on the Paul
Harris specimen, PH 15 (see figures 3 and 4), but details from other
fossils have been added, as follows (the numbers refer to the speci-
mens in the above lists): head, PE22016, Wolff 229, Damrow 236;
beak, PE22016, Damrow 236; pronotum, YPM 18; mesothorax and
metathorax, PE32046,YPM 18; fore leg, Wolff 229; abdomen, PE32046,
PE22016; ovipositor, PE22016; cerci, PE32046,PE22016; wings,
PE32046,PE32045, Wolff 229, HTP 422. All structures shown in the
composite drawing are present in one or more of the fossils studied.^
Discussion of the Structure of Eubleptus danielsi.
Head: The head of danielsi was obviously hypognathous. In the
specimens preserved in dorsal view (i.e., PH 15, figure 2) there is a
distinct hole in the matrix, at about the center of the insect’s head,
marking the point at which the beak penetrated the matrix; and in
^Handlirsch’s restoration of Eubleptus, based on the unique type (1920), bears little
resemblance to the insect in this composite drawing.
1983]
Carpenter — Euhleptus
91
Figure 6. Euhleptus cianielsi. Photograph of specimen YPMI8 (holotype of Athy-
mudictya parva Handlirsch. Dorsal view (reverse); pr, pronotal lobes, maximum
width of left fore wing in photograph, 4 mm.
the one specimen preserved in lateral view (PE22016, figures 4 and
5) the head is clearly hypognathous. The antennae are extraordinar-
ily long and thin (PH 15, Wolff 229; Bandringa specimen 66-
PBSM); for most of its length it is .04 mm in diameter and the
segments are about .1 mm long. The antennae of PH 15 include
about 1 10 segments and are almost certainly complete. The beak, as
preserved in lateral view in PE22016 is 3 mm long and slender;
several stylets project from its end. In specimen Damrow 236, the
beak is 2.8 mm long and as seen in front view (figure 4) is triangular
in shape, relatively broad basally, and bears long striae, as has been
noted in other species of Palaeodictyoptera (Kukalova, 1970). The
eyes are large and bulging, as shown in PH 15, Wolff 229, Bandringa
specimen 66-PBSM, and especially in PE22016, in which the eye, in
lateral view, is preserved in strong relief.
Thorax. The prothorax is very small and, as Handlirsch showed
in his drawing of parva, bears small lateral lobes about 1 mm wide
(YPM18); the folded and twisted condition of the lobes in some
specimens suggests that they were thin and weak. The legs ae known
92
Psyche
[Vol. 90
Figure 7. Eubleptus danielsi. Photograph of specimen PE22016. Lateral view: b, beak; c, cerci; o, ovipositor. Length of
ovipositor as preserved, 10 mm.
1983]
Carpenter — Eubleptus
93
only from a single fore leg in Wolff 229; the preserved part, appar-
ently consisting of the femur, tibia, and tarsus, has a total length of
3.5 mm; the tarsus appears to have five subequal segments. The
wing venation, as previously noted, shows only a slight amount of
variation among the 17 specimens examined. The shape of the wings
is more diverse, but that has undoubtedly been determined to some
extent by the process of preservation and the amount of movement
of the sediment in which the specimens were entombed. The degree
of variation in wing shape in the specimens of danielsi seems to be
comparable with that reported by Kukalova-Peck (1971) for the
Permian Dunbaria faseiipennis. The wing markings, consisting of
four triangular spots along the anterior margins of both wings, are
similar in both specimens in which they are preserved (PH 15 and
PE32046).
Abdomen. The segmentation of the abdomen is nearly homo-
nomous, except for the 9th and 10th segments, which are slightly
smaller than the others. The lateral margins of the tergites are
extended posteriorly only slightly (YPM 18), about as in the Spilap-
teridae. The ovipositor, preserved only in PE22016, is strongly
curved and only 2.5 mm long, not extending beyond the end of the
abdomen. The cerci (PE22016, and PE32036) are preserved to a
maximum length of 10 mm, but since they end at the edge of the
concretion, that is almost certainly not their full length. Segmenta-
tion of the cerci is clear at intervals; the segments are .3 mm long
(beyond the basal segments) and .3 mm wide, and covered with
short hairs. The largest piece of a cercus includes about 34 segments.
Relationships of Eubleptus
Study of the new specimens of Eubleptus danielsi provides no
evidence to justify the recognition of the order Eubleptidodea. On the
contrary, all the evidence supports Handlirsch’s assignment of the
family Eubleptidae to the Palaeodictyoptera. Furthermore, both the
wing venation and the newly acquired knowledge of the body struc-
ture of Eubleptus show a close relationship to the family Spilapter-
idae of the Palaeodictyoptera. The wings of Eubleptus have the
same general shape as those of the spilapterids, the hind wings
being slightly broader than the fore wings. The only significant
difference between the venational patterns of the two families is the
reduction of CUA in the family Eubleptidae: it has only a small
94
Psyche
[Vol. 90
Figure 8. Eubleptus danielsi. Reconstruction based mainly on specimen PH 15, in
the Paul Harris collection, with some details from specimens PE32046,PE32245,
PE22016, YPM18, USNM35576,FTWolff 229 and 233, HTP 422, and Bandringa
66-PBSM. All structures shown are preserved in at least one of these fossils. Length
of fore wing, 13 mm.
1983]
Carpenter — Euhleptus
95
terminal fork, whereas in the spilapterids CUA has several long
branches. This difference serves to justify the separation of Eublep-
tus into its own family, but does not have any significance at the
ordinal level. The body structure of Eubleptus turns out to be very
similar to that of the spilapterids. The pronotal lobes are small in
both, the beaks are relatively small and of similar shape in both, the
legs (so far as they are known) are short in both, and the ovipositors
are similarly formed. Eubleptus danielsi is the smallest known spe-
cies in the Palaeodictyoptera, but it is not much smaller than the
Permian Dunbaria faseiipennis of the family Spilapteridae.
Literature Cited
Carpenter, F.M.
1965. Studies on North American Carboniferous insects. 4. The genera Met-
rupator, Eubleptus, Hapaloptera and Hadentomum. Psyche, 72: 175-
190.
Handlirscii, Anton
1906a. Revision of American Palaeozoic insects. Proc. U.S. Nat. Mus., 29
( 1 44 1):66 1-820.
1906b. Die fossilen Insekten und die Phylogenie der rezenten Formen. II.
Abschnitt: Palaeozoische Insekten, p. 53-393. Engelmann, Leipzig.
I91 1. New Paleozoic insects from the vicinity of Mazon Creek, Illinois. Amer.
Journ. Sci. (4) 21:297-326,353-377.
1920. Palaontologie, in Handbuch der Entomologie (ed. C. Schroder).
3:117-306.
Kukalova-Peck, J.
1971. The structure of Dunbaria (Palaeodictyoptera). Psyche, 78: 296-305.
Laurentiaux, D.
1953. Classe des insectes. In Traite de Paleontologie (ed. Piveteau), p. 397-527.
Martynov, A.V.
1938. £tudes sur I’histoire geologique et de phylogenie des ordres des
insectes. Trav. Inst. Paleont., Akad. Nauk USSR, 7 (4): 1-148.
Roiidendorf, B.B.
1962. Osnovy Paleontologii: Tracheata, Mandibulata. Akad. Nauk USSR., p.
1-374.
■ -a., “:^:'
®li' .
• ,,V. . ~
^ -j\^
v*. .V,v ’- *-'"^^ ■ ' ■■
r ■ 'j
-1
iV^iivi'-:';
y«vi?^ i>»£k«hw., ^
>.• ;.fi'
iji.Wvf S-
‘\. ■:• -ir-
^memM
^ *' **^ ‘ '^ *■ *“■' i "■
I
t^:-iff^P ’ ..ji,
/ " 'f 1^' :jBf ;ci
.
. *^v.i%5.. i' '-'..a''*
.C’A-' As' ' ^'■•^' ■'
' .•■ •" * i.-. v-
<^4?; '
. V "
^■>- i-^ '-V-#
j
- ^ - r ‘
■ <■“ .-
ir - '• ■ " '■■ ' ’
, *i 1L. >w *
1 r :, »j" ./•
THE BIOLOGY OF TRICHADENOTECNUM ALEXANDERAE
SOMMERMAN(PSOCOPTERA: PSOCIDAE).
III. ANALYSIS OF MATING BEHAVIOR
By B. W. Betz'
Introduction
Several authors have described mating behavior in species of Pso-
coptera (Pearman 1928, Sommerman 1943a, 1943b, 1944, 1956,
Badonnel 1951, Thornton and Broadhead 1954, Klier 1956, Mock-
ford 1957, 1977, Broadhead 1961, Eertmoed 1966). Only one or at
most a few matings in a species were observed. This paper presents a
comprehensive analysis of pre- through post-copulatory behavior in
Trichadenotecnum alexanderae Sommerman. Evidence is presented
for a sex-attractant pheromone, produced only by females that were
receptive to mating.
Trichadenotecnum alexanderae is a relatively common psocid in
eastern United States (Betz 1983a). The species inhabits trees and
rock outcroppings providing its principal food source, pleurococ-
cine algae. Betz (1983a) found that T. alexanderae is capable of
facultative thelytoky. Formerly, the species was confused morpho-
logically with three other species, all obligatorily thelytokous, which
have been identified and described as T. castum Betz, T. merum
Betz, and T. innuptum Betz (Betz 1983a).
This paper is part of a series (cf. Betz 1983b, c, d) detailing the life
history of T. alexanderae.
Materials and Methods
Cultures of T. alexanderae were obtained from three populations
in Illinois: at Moraine View State Park (McLean County), along the
Sangamon River at Lake of the Woods (Champaign County), and
along the Salt Fork River at Champaign County Forest Preserve
District — Homer Lake (Champaign County).
Specimens were collected from tree trunks with an aspirator and
kept with pieces of bark in cotton-stoppered test tubes. Cultures
were transported to the laboratory over ice-water in a cooler.
MOOO North Lake Shore Drive, Chicago, Illinois, 6061 1.
Manuscript received by the editor August 16. 1982.
97
98
Psyche
[Vol. 90
Laboratory cultures were kept in cotton-stoppered test tubes.
Each tube was supplied ad libitum with food in the form of pleuro-
coccine algae on bark. Culture tubes were stored in closed, glass
desiccator jars over a saturated potassium chloride (KCl) solution to
maintain a relative humidity of 80 ±5%. The temperature regimen
for rearing was 23.3° : 1 8.0° C light:dark, and the photoperiod was 1 5
h light:9 h dark. Illumination was 4300 lumens/ m^, supplied by
incandescent and fluorescent lamps.
Because the other species of the T. alexanderae complex are oblig-
atorily thelytokous and often occur sympatrically with the biparental
species, I began a laboratory culture of T. alexanderae from each
locality with females mated in the laboratory to assure the identity
of the culture as the biparental species. Several breeding pairs
were used to begin a culture, an attempt to represent the genetic
diversity of the original sample from the field population. 1 also
examined the morphology of the original breeding pairs to verify
that they were T. alexanderae. 1 used bark obtained only from the
original field locality in cultures; bark was examined for eggs before
it was placed in a culture.
Mating behavior was studied in adults from Lake Dawson, Lake
of the Woods, and Salt Fork cultures. Adults were isolated as late
stage nymphs and reared in shell vials (four dram size). Each vial
was supplied with a flat piece of bark which lessened the interference
of the substrate on mating behavior. Females were 2-3 days old and
males were 2-5 days old when brought together, the times when they
were the most receptive to mating (Betz 1983c). Isolated specimens
were brought together by the following method. The cotton stopper
on each vial was removed, the open ends of pairs of vials were
apposed, and the vials were tilted carefully until the piece of bark in
the vial containing a male contacted the bark in the vial containing a
female. The open ends of pairs of vials were kept together and the
vials were not moved during observation of the insects. The method
I used to bring together isolated specimens did not appear to disturb
the insects greatly, and thus probably provided accurate observa-
tions of courtship behavior.
Results
Precopulatory Behavior
The behavior of male and female T. alexanderae was somewhat
variable among the successful matings (N = 99) and the unsuccessful
1983]
Betz — Biology of Trichadenotecnum
99
attempts (N = 45) I observed. Most precopulatory behavior fol-
lowed the patterns outlined in Figure 1.
When bark bearing an isolated, receptive female was brought
together with bark bearing a sexually active male, the male always
ran onto the female’s bark. A male displayed a higher level of activ-
ity under these conditions than if his piece of bark was brought
together with bark bearing a nymph, a female of T. alexanderae in
an unreceptive state, or a piece of bark without an insect (N = 7). In
five of the mating encounters I observed, the male flew onto the
female’s bark before the two pieces of bark were touching.
This higher level of activity in males occurred even if females were
placed out of the males’ sight. Almost immediately after a male ran
onto the bark of a female, he began a search over the substrate.
Sometimes a female remained motionless during this search, even
though she might have been active prior to the introduction of a
male. The manner in which males elicited this reaction of females
remains unknown, although the reaction may have resulted from
the slight disturbance caused by the introduction of pieces of bark
into the females’ vials.
A male searched in the direction of a female, often stopping
momentarily to flick his antennae and adjust his course.
When a male approached within about 1 cm of a receptive female
he began a quick, sideways gait while moving toward her, even
though she may have remained hidden from the male’s view. The
sideways gait lasted about 1-2 seconds. Occasionally a male ap-
proached a receptive female, or courted her, without the sideways
gait (N= 12). Females always fled from these encounters. When a
female fled, a male remained in the vicinity of the encounter and
spun completely around one or more times flicking his antennae.
Then a male usually ran off in the general direction of a female’s
flight. Unless unsuccessful courtship occurred many times (usually
the result of a male not performing the sideways gait), a female
would always acquiesce at the next courting.
After performing the sideways gait, a male ran up to a female’s
side, about midway along her length; a male approached a female
almost perpendicularly from her side. If features on the substrate
made a male’s approach difficult, his contact with a female was as
perpendicular to her as the substrate permitted. When a male ran up
to a female, he touched her briefly (less than a second) with both his
antennae. A male’s antennae usually struck a female’s thorax or
head and the distal end of her forewing because his antennae were
100
Psyche
[Vol. 90
Figure 1. Schematic diagram depicting precopulatory behavior in Trichadenotecnum alexanJerae. Typical behavior is indicated by bold
face arrows. Female behavior coinciding with male behavior is indicated by dashed arrows.
1983]
Betz — Biology of Trichaclenotecnuni
101
usually held about 90° apart and at about 45° off the substrate at
rest and also during the search for a female. Directly after contact-
ing her, a male backed away slightly, then rapidly fanned his wings
over his body. The wings were fanned at such a rapid rate and at
such a small angle (never more than 90°) that they became blurred.
As a male continued to fan his wings, he began to move anteriorly
along a female’s side, while still remaining perpendicular to her. A
male continued this motion until he was almost facing a female.
When this occurred, a male stopped fanning, dropped his wings
slightly, turned about 180°, and backed underneath a female
between her legs. The genitalia were apposed in this manner.
Occasionally a male stopped fanning when he was only laterally
apposed to a female’s head, then turned about 90°, and tried to back
under her. Of the 14 mounting attempts I observed progressing
in this way, only two of them led directly to copulation. Of the failed
attempts, eight were unsuccessful because males were blocked from
backing in by females’ legs; in the remaining four attempts, males
turned around farther than necessary and kept moving backward
beside the females rather than beneath them.
When a female remained hidden from a male’s view, he some-
times approached her directly from the front (N = 3). In two encoun-
ters, the female assumed the receptive posture (see below) as the
male fanned his wings. However, both of these males were unable to
locate the female after they turned around and began to move
backwards. One of the males approached the female on her left,
turned counterclockwise about 240°, and finally stopped after mov-
ing 8 mm away from her. He then tried to mount her end-to-end
(i.e., facing away from her), but this failed. The pair remained
motionless in the end-to-end position for about 30 seconds with
their genitalia nearly touching, then the male courted the female
from her side and was able to orient himself correctly.
The male of the third encounter courted frontally, but the female
turned her body slightly instead of assuming the receptive posture,
and the male was unable to move far enough backward for their
genitalia to come together. The courtship did not lead to copulation.
Most (73.7%) courting attempts were successful on the first try
(Table 1, A). When an attempted mounting failed, a male and
female always remained motionless for about 10-30 seconds. After
this period, if a male and female remained within about 1 cm after
102
Psyche
[Vol. 90
Table 1. Precopulatory behavior in TrichaJenoiecnum alexanderae
% of Total
Number of courting attempts
1. One
73.7
by a male (including the one
2. Two
15.2
leading to copulation)
3. Three
10.1
4. Four
0.0
5. Five
1.0
N=99^ 100.0
B. Length of time between the x
introduction of a male s.d.
and the beginning of mating range
(i.e., genitalic contact) N
C. Stage in courtship when a female
assumed the receptive posture
1. During the approach of a male
2. When touched by a male’s antennae
3. Just before a male backed underneath
N=96
1.1 minutes
1.3
0. 1-9.0 minutes
62
15.6
83.3
1.1
100.0
^The number of mating pairs for which the states of this behavioral character
were recorded.
they separated, a male courted again by approaching a female on
her side, touching her with his antennae, and fanning his wings.
Once either sex had fled from an unsuccessful courtship, males
always began further courtship with a sideways gait.
I observed the behavior of females that had fled from a failed
courtship (N = 4). Each female eventually stopped moving, and at
this time I observed each female flexing her valvulae dorsoventrally
in a pairwise manner, and making about ten contractions of her
abdomen.
The time required by males to establish genitalic contact after
they were introduced to females varied among mating encounters
(Table 1, B). This period was usually less than 1 minute if the first
courtship was successful. A male ran directly to a female in some
encounters, and the time between introduction and the beginning of
copulation was usually less than 30 seconds. Some males (N = 5)
were slower to find females because each remained within a small
1983]
Betz — Biology of Trichaclenotecnum
103
area on a female’s bark. Even though a female was not nearby, three
males exhibited sideways gaiting, wing fanning, and backward move-
ment, while two others only displayed a higher level of activity. All
five males were active, and each found a female about 10 minutes
after introduction to the female’s bark.
A female had to raise the anterior part of her body for a male to
be able to fit beneath her. Females always assumed a characteristic
appearance for this purpose that I here term the receptive posture.
In the receptive posture, the fore- and midlegs were moderately
extended, the hindlegs were slightly extended, and the antennae were
swept back along a female’s body (Fig. 2b). Most (83.3%) females
assumed the receptive posture when males touched them with their
antennae (Table 1 , C). Some ( 1 5.6%) females assumed the receptive
posture when males performed the sideways gait. One female waited
until a male was backing underneath her.
Females assuming the receptive posture early in courtship (i.e.,
before antennal contact was made by a male) elicited less wing
fanning from males. Males exhibited all of the actions involved in
courtship, but performed them more rapidly. On the other hand, the
female not assuming the receptive posture until a male began to
move beneath her did not appear to inhibit the male from courting
normally.
After a courtship failed, a male often fanned his wings around a
female for a longer period of time during the next one, regardless of
whether a female assumed the receptive posture when a male was
approaching or courting (N=10). These prolonged courtships
always led to copulation.
A male had to crouch slightly just prior to moving beneath a
female, even though she had assumed the receptive posture. This
position is shown in Figure 2a. Males remaining in a standing posi-
tion were blocked from moving past the coxae of the females’ legs
(N = 2). Furthermore, at this time a male’s abdomen became slightly
arched along its length, raising the posterior end.
A male’s wings were kept extended over his body as he backed
underneath a female, and she rested her fore- and midlegs on him.
When the genitalia of a pair were apposed, the head of a female was
positioned between and slightly caudad of a male’s raised hindwings.
As a male moved under a female, the shelf of his epiproct, which
normally rested in a posterodorsal position, struck a female’s sterna
104
Psyche
[Vol. 90
a
b
Figure 2. Typical precopulatory behavior in Trichadenotecnuni alexanderae.
a. A male in position to move beneath a female. b. The receptive posture of a
female.
and was brought anteriorly. This flattened the shelf against the
male’s abdominal terga, and caused his paraprocts to extend slightly
posteriorly. When fully beneath a female, a male’s epiproctal shelf
was brought into apposition with the basal arms of a female’s sub- i
genital plate and the sterna of her posterior abdomen.
I studied how males orient themselves with females to produce a
successful mounting. Individual females (N = 18) that had been
freshly killed (by pressure from a forceps) from a Lake of the Woods
culture were placed in a standing position on the substrate. Males
always courted and mounted the females without losing orientation
(N = 5). When I reoriented a female while a male was turning
around after touching her with his antennae, he did not reorient to
mount successfully (N = 3). If a female was moved as a male
approached (that is, prior to antennal contact), a male was always
able to orient and mount in the proper direction (N = 10). Hence,
antennal contact by males appeared to be important for a successful \
mounting.
When I placed a teneral male with a receptive female (N = 1), he
ran to her, paused briefly at her side, but then did not exhibit any \
other courting behavior (e.g., antennal contact, wing fanning, etc.).
Instead, he repeatedly climbed over her for about 10 seconds until '
she fled. The male made no attempt to mount, and the female did not ,
assume the receptive posture. ^
1983]
Betz — Biology of Trichaclenotecnum
105
Copula lory Behavior
When a male was fully beneath a female, the posterior end of his
abdomen probed for hers. The valvulae of a female dropped
ventrally somewhat, and moved until contact was made with a male.
The genitalia of male and female T. alexanderae interlocked strongly
together during copulation.
When their genitalia became locked, a male lifted a female off the
substrate by extending his legs, which were still crouched from back-
ing beneath her.
A normal copulatory position for a pair of T. alexanderae is
shown in Figure 3. The hind legs of most (67.8%) males were
extended slightly more than the other pairs of legs, causing a male’s
head to be lowered, and raising and slightly arching his abdomen
(Table 2, A). Males greatly extending their hindlegs usually also had
greatly arched abdomens; males extending all pairs of legs about
I 1
1.0 mm
Figure 3. Trichadenotecnum alexanderae in copulation.
106
Psyche
[Vol. 90
equally had no abdominal arch. Because the degree of abdominal
arching and the extension of the legs compensated for each other,
the overall position of a male, and thus the relative positions of a
male and female, generally differed little among the matings 1
observed. A male’s abdomen became slightly more arched and his
head dropped lower as copulation progressed. Some males lowered
their heads so much during copulation that their maxillary palps
touched the substrate.
After a female was raised off the substrate, a male’s abdomen
contracted more or less rhythmically. A female contracted her
abdomen in synchrony with a male. Her paraprocts were periodi-
cally flexed medially during copulation.
At the beginning of copulation, the wing pairs of a male were
usually extended high over his body, forming a small angle between
them (Table 2, B). As copulation progressed, the angle formed by
the wing pairs increased; the angle of the wings at the end of copula-
tion was about 60° greater (Table 2, C). 1 found 45.7% of all males
had both wing pairs locked at the nodus (Table 2, D). Wing pairs
remaining free were in a position as though they were locked, and
thus did not interfere with copulation.
Table 2 (E and F) shows that a male’s antennae generally remained
in a normal resting position during copulation.
A female was lifted off the substrate until the angle formed by
her body and the substrate was about 3 1°-60° (Table 2, G). Females
whose bodies inclined more than 60° (7.4%) had been pushed up
into this position when males mounted beneath them.
A female placed her fore- and midlegs on a male when he moved
beneath her. Table 2 (H and I) shows the distribution of the place-
ment of a female’s fore- and midlegs, respectively. Most (41.3%)
forelegs were placed on males’ hindwings and most (50.0%) midlegs
were placed on males’ anterior abdominal pleura. However, as
reflected in Table 2, the first two pairs of legs were positioned in
many other ways.
Table 2 (J) shows the distribution of the placement of a female’s
hindlegs. Most (93.3%) females kept their hindlegs on the substrate.
The rigidity of the hindlegs indicated that they were supporting
some of a female’s weight.
The forewing tips rested lightly on the substrate in 84.8% of the
females (Table 2, K). This contact did not appear to support much
of a female’s weight.
1983]
Betz — Biology of Trichadenotecnuni
107
Table 2. Copulatory behavior in TrichaJenoiecnum alexamierae
% of Total
Position of a male
A. Extent of the arch of
the abdomen
B. Angle formed by the forewings
at the beginning of mating
C. Angle formed by the forewings
at the end of mating
D. Number of wing pairs locked
at the nodus
E. Angle formed by the antennae
F. Angle formed by the antennae
and the bark substrate
1 . Not arched
21.9
2. Slightly arched
67.8
3. Greatly arched
10.3
N = 87
100.0
1. 0°-30°
52.3
2. 3l°-60°
38.5
3. 6l°-90°
6.1
4. 9I°-I20°
3.1
N=65
100.0
1. 0°-30°
5.6
2. 3l°-60°
II. 1
3. 6l°-90°
37.0
4. 9I°-120°
40.7
5. I2I°-150°
0.0
o
O
oo
o
3.7
7. I8I°-2I0°
1.9
N=54
100.0
1. None
35.7
2. One
18.6
3. Both
45.7
N=70
100.0
1. 0°-30°
5.2
2. 31°-60°
32.8
3. 61°-90°
56.9
4. 91°-120°
3.4
5. 121°-150°
0.0
6. 151°-180°
1.7
N=58
100.0
1. 0°-30°
40.5
2. 31°-60°
59.5
N=74
100.0
108
Psyche
[Vol. 90
Table 2. (continued)
% of Total
Position of a female
G. Angle formed by the body of a
female and the bark substrate
H. Placement of a foreleg on a male
I. Placement of a midleg on a male
J. Placement of a hindleg on a male
1.
0°-30°
1.90
2.
3P-60°
90.7
3.
o
O
O'
1
o
SO
7.4
N=54
100.0
1.
Forewing
5.6
2.
Base of forewing
29.1
3.
Hindwing
41.3
4.
Base of hindwing
3.1
5.
Metanotum
8.7
6.
Anterior abdominal
terga
6.6
7.
Anterior abdominal
pleura
4.6
8.
Hind tibia
1.0
N=196^
100.0
1.
Hindwing
3.7
2.
Hindwing base
3.1
3.
Metanotum
17.7
4.
Anterior abdominal
terga
6.3
5.
Anterior abdominal
pleura
50.0
6.
Anterior abdominal
sterna
0.5
7.
Hind tibia
13.5
8.
Hind femur
0.5
9.
Bark substrate
4.7
N=192^
100.0
1.
Anterior abdominal
pleura
1.0
2.
Anterior abdominal
sterna
0.5
3.
Hind tibia
5.2
4.
Bark substrate
93.3
N=194^
100.0
1983]
Betz — Biology of Trichadenotecnum
109
Table 2. (continued)
% of Total
Position of a female
K. Position of the forewing tips
1.
On the bark substrate 84.8
2.
In the air
15.2
N=92
100.0
L. Angle formed by the antennae
1.
0°-30°
25.0
2.
o
O'
o
0
51.7
3.
61°-90°
18.3
4.
9I°-120°
5.0
N=60
100.0
M. Angle formed by the antennae and
the bark substrate
1.
0°-30°
88.3
2.
O
k
o
11.7
N=60
100.0
General Features
N. Relative positions of a male
and female
1.
2.
Both in-line
Female skewed
55.3
right or left
44.7
N=47
100.0
O. Duration of copulation
X
14.3 minutes
s.d.
2.2
range
8.3-19.9 minutes
N
99
^Recorded for both legs.
In contrast to the position of a male’s antennae, a female’s anten-
nae were swept back along her body when she assumed the receptive
posture. Usually the antennae were held close to her sides, and the
angle formed by the antennae was between 31° -60° (Table 2, L).
Most (88.3%) antennal pairs were slightly raised from the substrate
(Table 2, M).
Copulation did not always proceed uneventfully. The number of
positions among the pairs reflects the incidence of minor problems
that were encountered during mating. The conformations occurring
110
Psyche
[Vol. 90
with lower frequency were usually the result of some difficulty dur-
ing copulation. The relative positioning of a male and female (Table
2, N) generally indicates whether a problem occurred during copu-
lation. If the bodies of a male and female were in-line, copulation
usually had proceeded normally. A female was skewed left or right
in 44.7% of the matings. Some of the complications that occurred
during mating are discussed below.
Sometimes a male lifted a female far off the substrate, so her head
was almost over his. Other males pushed females backward during
mounting, then locked genitalia, causing females to be inclined
almost vertically and males to have greatly arched abdomens.
When events such as these occurred, the positioning of a pair, espe-
cially a female’s legs, changed to maintain her balance. For exam-
ple, the hindlegs of a female typically rested on the substrate.
However, when a male lifted a female relatively far off the substrate,
copulation proceeded with more stability when her hindlegs were
placed on a male’s abdomen. When a male’s abdominal contractions
increased to an amplitude that caused his hindwings to strike a
female on the head, she placed one (N = 8) or both (N = 3) forelegs
about midway up on a male’s fore- or hindwings. This response
lessened the force of a male’s contractions. Sometimes a mating pair
fell on their sides (N = 3), but their genitalia remained locked. These
pairs never regained a normal copulatory position, yet they did not
break off copulation because of this problem.
Except for the abdominal contractions and adjustments for stabil-
ity, other movements by a mating pair were uncommon. Occasion-
ally the maxillary palps of a male or a female moved or pulsed
rapidly.
1 observed courtship and mating on a vertical bark substrate in
the laboratory (N = 3). When a male courted a female, she oriented
herself so her head faced downward, then she assumed the receptive
posture. A male lifted a female off a vertical substrate during copu-
lation. A female was positioned dorsally and posteriorly on a male,
similar to mating on a horizontal surface, but in a vertical orienta-
tion a female balanced directly above a male. On a vertical substrate
the fore- and midlegs appeared to have supported more of a female’s
weight.
The duration of copulation varied considerably (Table 2, O),
although it was never less than eight or more than 20 minutes.
1983]
Betz — Biology of Trichadenotecnum
HI
Postcopulatory Behavior
The final stage of copulation was indicated by a slowing of the
rate of abdominal contractions and by contractions of a slightly
more spasmodic nature. Males suddenly became active and broke
off copulation (Table 3, A) by quickly running forward, dragging
along the females for about 1 second until their genitalia unlocked.
Of the 60 pairs I observed for this behavioral character, only two
females (3.3%) appeared to break off copulation. In each of these
matings, the female became active during copulation and tried to
dismount laterally, but her genitalia were locked with the male’s and
this caused her to fall on her side. One female successfully dis-
mounted, thus terminating copulation. The other female failed to
dismount, and instead tried to assume a normal mating position
three times, but because the genitalia were locked she was kept
off-balance. Copulation continued with the female supported tenu-
ously off a side of the male.
A spermatophore was passed in all copulations, including the one
broken off by a female.
Table 3 (B) shows the reactions of males after copulation was
broken off. Most (53.5%) males ran forward about 1 cm, then
remained motionless for at least 5 minutes. Some of these males had
their wings parted slightly and held laterally along their bodies, but
most males brought the wings back to a normal resting position.
Some (28.2%) males were highly active after copulation, and ran
over the substrate without stopping for over 5 minutes. Other
(12.7%) males broke off copulation and remained almost at the
place where copulation occurred. A few (5.6%) males courted the
females they had just mated, but this always caused the females to
flee.
The reactions of females after copulation (Table 3, C) were
somewhat different than those of males. Most (60.0%) females
remained in the area where copulation occurred; almost all of these
females spun around about 90°, some females spun around about
180°. If a female ran off, she usually ran in the direction of a male
because both faced in the same direction during mating. Females
that ran, even for 1 cm, never spun around more than 90°.
Unless disturbed by another insect, once females stopped walking
after mating they rarely moved until the contents of the spermato-
phore that they held had been transferred. Even the antennae did
112
Psyche
[Vol. 90
Table 3. Postcopulatory behavior in Trichadenoiecnuni alexanJerae
A. Identity of the sex
breaking off copulation
B. Reaction of a male
after copulation
was broken off
C. Reaction of a female
after copulation was
broken off
% of Total
1. Male
96.7
2. Female
3.3
N=60
100.0
1. Ran off ( > 1 cm)
28.2
2. Ran about 1 cm.
then remained still
53.5
3. Stayed in the area
where mating
occurred
12.7
4. Tried to court the
female again
5.6
N=71
100.0
I . Ran off ( > 1 cm)
8.6
2. Ran about 1 cm,
then remained still
31.4
3. Stayed in the area
where mating
occurred
60.0
N=70
100.0
not change position during this time. Only one of three postcopula-
tory females 1 observed at length changed her location once, about 2
minutes after copulation, but did not move after this.
Initially, a spermatophore had an appearance of a whitish, semi-
opaque, hemispherical droplet, protruding between the terminalia
of a female. A female manipulated her terminalia so its contents
passed her genital opening. When it was first visible on a female, a
spermatophore seemed adhesive and somewhat fluid in shape,
allowing it to be manipulated on a female’s terminalia. During the
transfer of its contents, a spermatophore covering seemed to lose
adhesiveness and harden, allowing a female to dispose of it easily.
1983]
Betz — Biology of Trichaclenotecnum
113
A female manipulated a spermatophore between her paraprocts
and valvulae. These movements were accomplished by a rhythmical,
medial flexion of the terminalia. Contractions were spaced about
1-2 seconds apart, and each contraction lasted about 1-2 seconds.
The epiproct was less active in this respect; it was flexed once about
every 30 seconds. Epiproctal flexion probably forced the contents of
a spermatophore into a female’s genital area. The abdominal con-
tractions were pronounced for about the first 4-5 minutes after
copulation, then slowly decreased in rate and intensity.
The females I observed (N = 3) required about 10-25 minutes to
transfer the contents of a spermatophore and discard it. A spermato-
phore either fell from a female’s genital area when she ran (N = 2),
or a female dragged the posterior end of her abdomen along the
substrate for about 1 mm to discard it (N = 1 ). A female sometimes
intermittently flexed her terminalia after a spermatophore had been
discarded.
Evidence for a Sex-attractant Pheromone
The following observations present evidence indicating that the
attraction of males to females of T. alexanderae was mediated by a
pheromone. All females discussed here were unmated, receptive
females from cultures of all three field localities unless noted
otherwise.
Males introduced to bark bearing receptive females often became
more active than when they were introduced to pieces of bark
which had no exposure to receptive females. Some males became so
active they flew across the gap between the pieces of bark before I
could join these together. A male was usually able to find a receptive
female, even though she may have remained hidden from his view.
Additionally, males would court females of T. castum and T.
merum, two obligatorily thelytokous species of the T. alexanderae
species complex, if vials containing these females had previously
contained receptive females of T. alexanderae (Betz 1983a).
Females of T. alexanderae which had just mated ceased rapidly to
be a source of attraction to males, although males occasionally tried
to court females engaged in mating. Teneral females, or females in
the stage of oviposition, failed to attract males (Betz 1983c).
In one mating encounter involving a male and female from a Lake
of the Woods culture, when the male was introduced into a vial
114
Psyche
[Vol. 90
containing the female, she deposited a clear droplet from her genital
area on the bark. The droplet was absorbed rapidly. The male was
highly active and quickly found the female, who was still in the area
where she deposited the droplet, and mated with her. In another
encounter, the female deposited a droplet about 3 minutes after a
failed courtship attempt. This pair mated eventually. Another
female intermittently dragged the tip of her abdomen over the sub-
strate after the male was introduced into her vial. In two encounters,
each involving a male and female from a Lake of the Woods culture,
two from a Lake Dawson culture, and one from a Salt Fork culture,
the male persistently courted a particular place on the female’s sub-
strate, even though she was not nearby. Eventually, after about 10
minutes, males stopped courting these areas.
I have not observed females of T. castum or T. merum depositing
any type of droplet in the above manner, or observed males courting
places on a substrate bearing females of T. castum or T. merum.
In the orientation experiment, it was important to use freshly
killed females because after about 5 minutes they lost attractiveness
to males. Anesthetizing females with ether (N = 3) or carbon dioxide
(from dry ice) (N = 3) caused an immediate loss of interest by males.
From the evidence cited above it appears the females of T. alex-
anderae produce a pheromone that attracts males. It appeared to be
highly volatile; a loss of mating receptivity in a female was almost
immediately evident, as indicated by the lack of attractiveness to
males.
The area around a female in which the pheromone was effective in
attracting males was rather small, having a radius of about 1 cm. 1
determined this by placing individual receptive females (N = 10) in
uncovered petri dishes (standard size), then introducing sexually
active males.
Discussion
Mating behavior in T. alexanderae followed a pattern outlined by
Pearman (1928) for “winged Psocids.” This courtship pattern, which
has since been categorized (Badonnel 1951) and further documented
(Klier 1956), is the one found in most species of Psocoptera that
have been studied. This pattern differs from those in other species in
two details: males do not run over the dorsum of females prior to
mounting, and the duration of copulation is relatively long.
1983]
Betz — Biology of Trichadenotecnum
115
A receptive posturing by females has been noted in several other
species of Psocoptera (Pearman 1928, Sommerman 1943a, 1956,
Schneider 1955, Broadhead 1961). Apparently only females of T.
alexanderae have been observed assuming the receptive posture
before males began backing underneath them.
The receptive posture appeared necessary only to permit males to
fit beneath females during mating. Lifting the anterior end of a
female’s body did not communicate a female’s orientation to a male.
Males courting one of the freshly-killed females always moved ante-
riorly along her body, even though she was not in the receptive
posture. However, the contact made by a male’s antennae after the
sideways gait probably was important in discovering how a female
was oriented because a male was unable to adjust his course to find
a reoriented female after antennal contact was made. Also, males
that had difficulty moving beneath females began the next courtship
with wing fanning, but males not contacting females recourted with
a sideways gait and antennal contact. A differential concentration of
pheromone along a female’s body may have informed a male of her
orientation. A perpendicular approach to a female allowed maxi-
mum extension of a male’s antennae along her body, thus perhaps
facilitating a determination of her relative position.
1 found that the receptive posture was assumed by some females
of T. alexanderae when males approaching to court began a side-
ways gait. Largely auditory cues, rather than visual ones, were proba-
bly given by a male to signal his approach, thereby eliciting the
receptive posture in a female. The sideways gait may cause stridula-
tion of a male’s Pearman’s organs because a male approaching a
female in this way never caused her to flee, even though he may have
remained hidden from her view during the sideways gait.
The purpose of the females’ genital movements and abdominal
contractions after failed courtship is unknown. These motions were
only observed in females involved in some phase of the mating
process. Perhaps this action released more pheromone to attract
males again.
The role of the droplets (apparently containing pheromone)
which were deposited by females is also uncertain. This behavior
would assist a male in locating a female only if she remained in the
area where a droplet was deposited. Because the pheromone appears
to be highly volatile, to have any effect on males a female probably
must deposit many of these droplets during her receptive period.
116
Psyche
[Vol. 90
The epiproctal shelf of a male played an important role during
mating in T. alexanderae: the shelf and a female’s hindlegs on the
substrate supported almost all of a female’s weight. A male’s epi-
proctal shelf and the basal arms of the subgenital plate of a female
are structures apparently functioning to distribute her weight
because both structures are well-sclerotized and have large surface
areas.
The support given by the epiproctal shelf and the interlocking
genitalia apparently increased the lateral stability of a mating pair.
This can be adduced in the following observations. A mating pair
had greater stability during copulation than when a male had
backed fully beneath a female but before their genitalia had inter-
locked. Also, it was difficult for a mating pair to fall over to one side
when problems in stability occurred during copulation, although
once they fell regaining a normal mating position was impossible.
The stability supplied by the contact between the epiproctal shelf-
subgenital plate and posterior abdominal sterna, the positioning of
a female’s legs, and the interlocking genitalia probably also expe-
dited the lifting of a female by a male, although the reason for the
necessity of lifting a female is unknown.
Acknowledgements
I thank Dr. E. L. Mockford of Illinois State University, whose
beneficial discussions about Psocoptera and review of the manu-
script were greatly appreciated. 1. N. Holod and D. D. Pierce assisted
in the production and the typing of the manuscript, respectively.
Summary
Pre- through postcopulatory behavior in Trichadenotecnum alex-
anderae Sommerman is here quantified and discussed. Mating
behavior follows a pattern described for many other species of Pso-
coptera, in which a male approaches a female, fans his wings over
his head, and backs underneath her without running over her dor-
sum. Additional behavioral actions, including possible stridulation
and antennal contact of a female by a male and a female assuming a
receptive posture prior to mounting by a male, are believed to
promote copulation. Evidence is presented for a sex-attractant
pheromone produced by receptive females.
1983]
Betz — Biology of Trichacienotecnum
117
Literature Cited
Badonnel, a. 1951. Ordre des Psocopteres. Pages 1301-1340 in Grasse, P. P.,
ed., Traite de Zoologie, Vol. 10, Fasc. 2. Masson et Cie., Paris.
Betz, B. W. 1983a. Systematics of the Trichadenoiecnum alexanJerae species
complex (Psocoptera: Psocidae) based on an investigation of reproductive
modes and morphology. Can. Entomol. (in press).
1983b. The biology of Trichadenoiecnum alexanderae Sommerman (Pso-
coptera: Psocidae). 1. Habitat, life stages and events. Entomological News
(in press).
1983c. The biology of Trichadenoiecnum alexanderae Sommerman (Pso-
coptera: Psocidae). II. Duration of biparental and thelytokous reproductive abili-
ties. J. Kansas Entomol. Soc. (in press).
1983d. The biology of Trichadenoiecnum alexanderae Sommerman ( Pso-
coptera: Psocidae). IV. Mechanism of genitalic coupling. J. Kansas Entomol.
Soc. (in press).
Broadhead, E. 1961. The biology of Psoquilla marginepunciaia (Hagen) (Corro-
dentia, Trogiidae). Trans. Soc. Brit. Ent. 14: 223-236.
Eertmoed, G. E. 1966. The life history of Peripsocus quadrif asciai us (Psocopiexa:
Peripsocidae). J. Kansas Entomol. Soc. 39: 54-65.
Klier, E. 1956. Zur Konstruktionsmorphologie des mannlichen Geschlechtsap-
parates der Psocopteren. Zool. Jb. (Anatomic) 75: 207-286.
Mockford, E. L. 1957. Life history studies on some Florida insects of the genus
Archipsocus (Psocoptera). Bull. Fla. State Mus., Biol. Sci. 1: 253-274.
1977. Asiopsocus sonorensis (Psocoptera: Asiopsocidae): A new record,
augmented description, and notes on reproductive biology. Southwestern Nat.
22:21-29.
Pearman, j. V. 1928. Biological observations on British Psocoptera. HI. Sex
behavior. IV. Miscellaneous. Entomologist’s Mon. Mag. 64: 263-268.
Schneider, H. 1955. Vergleichende Untersuchungen uber Parthenogenese und
Entwicklungsrhythmen bei einheimischen Psocopteren. Biol. Zentr. 74: 273-310.
Sommerman, K. M. 1943a. Bionomics of Eciopsocus pumilis (Banks) (Corroden-
tia, Caeciliidae). Psyche 50: 53-64.
1943b. Bionomics of «/v6/7/.v (Aaron) (Corrodentia, Caeciliidae).
Can. Entomol. 75:99-105.
1944. Bionomics of (Walsh) (Corrodentia, Psocidae).
Ann. Entomol. Soc. Amer. 37: 359-364.
1956. Two new species of Rhyopsocus (Psocoptera) from the U.S.A.,
with notes on the bionomics of one household species. J. Wash. Acad. Sci. 46:
145-149.
Thornton, I. W. B., and E. Broadhead. 1954. The British species of Elipsocus
Hagen (Corrodentia, Mesopsocidae). J. Soc. Br. Ent. 5: 47-64.
PREDATORY CAPTURE OF BOMBARDIER BEETLES
BY A TABANID FLY LARVA*
By Stephen Nowicki and Thomas Eisner
Section of Neurobiology and Behavior
Cornell University, Ithaca, NY 14853
While collecting bombardier beetles {Brachinus spp.) on the eve-
ning of August 27, 1982, by a pond near Portal, Cochise County,
Arizona, a group of us, including Rodger Jackman of the British
Broadcasting Corporation and Maria Eisner, came upon an unusual
phenomenon. Thousands of young adults of the spadefoot toad,
Scaphiopus multiplicatus, were active beside the pond on that
night, having just emerged from the water after metamorphosis. On
close observation we noted a number of these toads that were dead
or dying and in various stages of partial submergence in the mud.
Each had been grasped from beneath by a larva of the horsefly
Tabanus punctifer, a mud-dwelling predator, which had seized it with
its hooked mouthparts, had pulled it partly into the substrate, and
was embibing its body fluids. Details of this first known occurrence
of predation by a fly larva on an adult amphibian will be published
elsewhere. Our purpose here is to call attention to another extraor-
dinary ability of this larva: the capture of bombardier beetles.
We transported several of the larvae to our Cornell laboratories
and established them individually in mud-filled enclosures, where
they quickly buried themselves, leaving only their mouthparts
exposed at the surface (Fig. lA). We maintained the larvae on
young spadefoot toads, which they captured as they had in the field,
and also on insects, which judging from published accounts on
tabanid larvae (Webb and Wells, 1924; Oldroyd, 1964; Burger,
1977), must be a principal staple of their diet. They proved capable
even of capturing large crickets ( Teleogryllus oceanicus), which they
hooked by a leg, drew partly into the mud, and then held for hours
while sucking out their body contents.
*Paper No. 73 of the series Defense Mechanisms of Arthropods. Paper No. 72 is
T. Eisner and J. Meinwald, Psyche 89, 357-367, 1983.
Manuscript received by the editor January 6, 1983
119
120
Psyche
[Vol. 90
Figure 1. (A) Mouthparts (arrow) of a buried T. punciifer larva, projecting just
above the surface of the mud, as the animal lies in wait of prey. (B) Bombardier
beetle, caught by a submerged larva and partly drawn into the mud, in the process of
being eaten (the larvae can also pull beetles into drier mud than here shown). (C) Full
grown T. punctifer larva beside an average-sized bombardier beetle. Reference bars =
5 mm.
Bombardier beetles are doubtless among the most invulnerable of
insects. The quinone-containing spray that they eject from their
abdominal defensive glands when attacked is hot (100°C) and is
aimed accurately toward the predator by rotation of the abdominal
tip (Eisner, 1958; Aneshansley et al., 1969). A number of broadly
insectivorous predators have been shown to be repelled by the
spray, including ants, spiders, preying mantids, and toads (Eisner,
1958; Eisner and Dean, 1976; Dean, 1980).
We staged encounters between bombardier beetles and T. punc-
tifer larvae by releasing the beetles singly onto the mud in the larval
1983]
Nuw'icki & Eisner — Bombardier beetles
121
enclosures. The beetles all stemmed from the larval collecting site,
where they were taken at the very places on the edge of the pond
where the larvae were also abundant. In three encounters we were
fortunate to witness the beginnings of the attack. The events pro-
ceeded quickly and were the same in each case. No sooner had a
beetle brought one of its tarsi to rest upon the mouthparts of the
larva, than it was grasped by that tarsus and caused to spray. There
were sometimes several discharges, audible at times and visible as
misty puffs, but the larva, which had withdrawn below the surface
the moment it hooked on to the beetle’s leg, was already out of
reach. The beetle struggled as it was gradually pulled into the mud,
but the larva never released its hold. Partly submerged, the beetle
eventually died (Fig. 1 B), and when retrieved next day was found to
be largely eaten out. Five additional encounters that were not wit-
nessed from the outset were equally fatal to the beetles. We assume
that the death of the beetles was hastened by the salivary toxins that
tabanid larvae are said to inject into their insect prey (Schmidt,
1982).
Given the ecological co-occurrence of T. punetifer larvae and
bombardier beetles, we feel that encounters between the two must
inevitably occur also in nature, with the same outcome as in the
laboratory. Moreover, the larvae must have access also to diverse
other insects that discharge noxious secretions, includings ants, tiger
beetles, and additional Carabidae. Species of Chlaenius, for exam-
ple, whose odor was unmistakably suggestive of the phenolic output
that characterizes other beetles of the genus (Eisner et ai, 1963),
scurried about together with Braehinus at our collecting site at
night. Against such chemically protected insects, the predatory tac-
tic of lurking just beneath the surface, and of withdrawing into the
mud for total cover the moment a victim is seized and caused to
activate its defenses, doubtless serves the larvae well. Other mud-
dwelling tabanid larvae of similar opportunistic feeding habits might
equally profit from the tactic.
Acknowledgements
Study supported by Grant AI02908 from NIH; we thank Drs.
John F. Burger and Rodolfo Ruibal for helpful information and for
identifying the larva and toad respectively, and Maria Eisner for
technical help.
122
Psyche
[Vol. 90
References Cited
Aneshansley, D. J., T. Eisner, J. M. Widom, and B. Widom
1969. Biochemistry at 100° C: Explosive secretory discharge of bombardier
beetles (firflfc/7/>7w.s). Science 165:61-63.
Burger, J. F.
1977. The biosystematics of immature Arizona Tabanidae (Diptera). Trans.
Amer. Ent. Soc. 103:145-258.
Dean, J.
1980. Encounters between bombardier beetles and two species of toads (Bufo
americanus, B. nmrinus)\ Speed of prey-capture does not determine
prey-capture success. J. Comp. Physiol. 735:41-50.
Eisner, T.
1958. The protective role of the spray mechanism of the bombardier beetle,
Brachynus hallistarius Lee. J. Insect Physiol. 2: 215-220.
Eisner, T. and J. Dean
1976. Ploy and counterploy in predator-prey interactions: orb-weaving spiders
versus bombardier beetles. Proc. Nat. Acad. Sci. USA 73:1365-1367.
Eisner, T., J. J. Hurst, and J. Meinwald
1963. Defense mechanisms of arthropods. XI. The structure, function, and
phenolic secretions of the glands of a chordeumoid millipede and a
carabid beetle. Psyche 70:94-1 16.
Oldroyd, H.
1964. The Natural History of Flies. Norton: New York.
Schmidt, J. O.
1982. Biochemistry of insect venoms. Ann. Rev. Entom. 27:339-368.
Webb, J. L. and R. W. Wells
1924. Horse-flies: Biologies and relation to western agriculture. U. S. Dept.
Agr. Bull. 1218.
PREY SELECTION BY THE NEOTROPICAL SPIDER
ALPAIDA TV ON A BO
WITH NOTES ON WEB-SITE TENACITY'
By Todd E. Shelly
Department of Biology
University of California
Los Angeles, CA 90024
Introduction
Prey selection by web-building spiders includes 2 principle com-
ponents. First, webs may catch a nonrandom sample of the avail-
able prey (passive selection). Among items caught in the web, the
spider may then feed on preferred prey but reject unsuitable prey
(active selection). As evident from a recent review (Riechert and
Luczak 1982), quantitative field measurements of either component
are relatively rare and particularly so for tropical species.
Here I compare the web contents of Alpaida tuonabo (Chamber-
lin and Ivie) with sticky trap samples of available prey. Field work
was conducted at one site over a relatively short period of time thus
reducing potential complications arising from habitat and seasonal
differences in prey availability. As Olive (1980) and Uetz et al.
(1978) found, however, prey availability may vary over short verti-
cal distances, and to examine this possibility potential prey were
sampled at several different heights.
In addition, a second comparison was made between captured
items being eaten and those left unattacked and uneaten. Since prey
ignored during the day may have been consumed at night with the
web, uneaten prey did not necessarily represent rejected prey. This
comparison, however, does quantify the probability of immediate
attack upon different types and sizes of captured prey.
'While Araneus is the accepted generic designation, this species is not closely related
to other members of this genus and should perhaps be placed in the genus Aplaida
(H. Levi pers. comm.).
Manuscript received by the editor December 20, 1982.
123
124
Psyche
[Vol. 90
Materials and Methods
The study was conducted between July 23 and August 25, 1980,
on Barro Colorado Island (BCI), Panama. This time period falls
near the middle of a rainy season which annually extends from late
April to mid-December (Croat 1978). The island is covered by a
lowland tropical moist forest (Holdridge et al. 1971). Alpaida tuo-
nabo was most abundant on the island’s central plateau, and all
work was conducted there.
Very little is known about the biology of A. tuonabo. A descrip-
tion of the female has been published (Chamberlin and Ivie 1936),
but males have not yet been described (H. Levi, pers. comm.).
Females are relatively small; the mean wet weight and body length
of 8 adult females were 0.023 g (SD 0.005) and 5.6 mm (SD 0.94),
respectively. Females appeared to construct and tend webs during
the day and consume them at night. In 4 nights of searching, I never
saw a female or an intact web. On BCI A. tuonabo is abundant only
in the mid to late wet season (July to December) and is rarely found
during the rest of the year (Lubin 1978).
Flying insects were sampled at 10 different sites. At each site I
implanted a 2.7 m PVC pole (diameter 25 mm) by driving 0.30 m —
0.45 m of its length into the ground. Wooden rods (length 30 mm;
diameter 5 mm) were then fastened to the pole at 0.3 m intervals
(from 0.3 m to 2.1 m above ground). Fastened at one end, each rod
projected perpendicularly from the vertical pole and hence was
parallel to the ground’s surface. Insects were collected on tanglefoot
covered traps suspended from the wooden rods. Each trap was
a 15 cm by 23 cm rectangle of 3 mm thick transparent plastic coated
on both sides with tanglefoot. Insects were sampled during the day
only on August 7-9. Each day the traps were set between 0800
hrs-0900 hrs, taken down between 1600 hrs-1700 hrs, and stored
overnight in closed boxes. Aside from Diptera and Hymenoptera,
all trapped insects were identified to order. Flies were categorized as
either nematocerous or non-nematocerous, and hymenopterans
were subdivided into bees and wasps, parasitoids, and winged ants.
All trapped insects were measured to the nearest 0.1 mm using a
dissecting microscope equipped with a disc micrometer.
Each day of the study 1 walked through different areas of the
forest (between 0900-1630 hrs) and examined every web encoun-
tered. All caught items were collected and labelled as either eaten or
1983]
Shelly — Alpaicia tuutiaho
125
Diptera
nematocerous
non-nematocerous
Hymenoptera
ants
parasitoids
Coleoptera
Frequency
Figure I. Vertical distributions of the major prey categories. Each value repres-
ents the total number of individuals captured on 10 sticky traps suspended at a
particular height. See text for details of sampling method.
uneaten. Uneaten prey were also examined for evidence of wrap-
ping. For each web thus sampled, the height of the spider was also
recorded. Collected prey were later assigned to the appropriate prey
category and measured to the nearest 0.1 mm.
Prey selectivity was quantified using Ivlev’s (1961) index of elec-
tivity. Electivity (E) is calculated as follows: E = (ri — pi)/(ri + pi)
where rj is the proportion of the predator’s diet represented by prey
type (or size class) i, and pi is the proportion of the available prey
represented by prey type (or size class) i. Values of E range from
— 1.0 (complete avoidance) to +1.0 (complete preference). In this
study electivity values with absolute values less than 0.40 were not
considered to differ significantly from zero. In addition, two sets of
electivity values were calculated. For web selectivity (Ew) ri is the
proportion of the web contents (both eaten and uneaten items)
represented by prey type i, and pi is the proportion of available prey
(as measured by the sticky traps) represented by prey type i. For
spider selectivity (Eg) ri is the proportion of the spider’s observed
diet (the eaten prey) represented by prey type i, and pi is the propor-
tion of the web contents (both eaten and uneaten items) represented
by prey type i.
Results
Alpaicia tuonabo females generally constructed webs in relatively
open sections of the forest or at the edges of tree-fall gaps. Most web
126
Psyche
[Vol. 90
sites were shaded, and only rarely was a web placed in an area that
received direct sunlight. Various web support structures were util-
ized, including leaf tips, herbaceous stems, woody vines and
branches, and palm fronds. The circular webs averaged 21.6 cm in
diameter and 350 cm^ in catching area (n = 8).
Individuals do not appear to remain at a particular web-site for
more than 1-2 days. On August 3 I marked the location of 20
occupied webs. These sites were then revisited daily for 7 days, and
the presence or absence of the spider and the web was recorded. In
terms of the number of spiders remaining at their initial site, the
results obtained were as follows: Day 1 — 12; Day 2 — 3; Days 3 and
4 — 2; Days 5 and 6 — 1 ; Day 7 — 0. In no instance was a spider absent
but the web present; spider and web were always both present or
both absent. In addition, in examining a 2 m-3 m radius about each
vacated web-site, I never observed the presence of a newly con-
structed web.
Five prey categories comprised 89.0% of the total sample, and
vertical abundance patterns were examined for these groups only.
Beetles, parasitoid Hymenoptera, nematocerous and non-
nematocerous Diptera all exhibited a similar trend in vertical abun-
dance (Figure 1). That is, the greatest numbers of individuals were
collected at the two lowest sampling heights (0.3 m and 0.6 m).
While similar numbers of parasitoid Hymenoptera were captured at
the two lowest sampling heights, nearly twice as many beetles, nema-
tocerous and non-nematocerous Diptera were captured at 0.3 m
than 0.6 m. Ants were captured in relatively constant numbers over
all sampling heights.
Although the numbers of trapped individuals varied greatly with
height for 4 prey categories, each major category comprised a rela-
tively constant proportion of the total sample at each height (Figure
2). Similarly, within each category size frequency distributions did
not vary with height in any obvious manner (Figure 3). Thus, while
the abundance of flying insects varied with height, the taxonomic
and size composition of this fauna did not.
The vertical distribution of A. tuonabo did not closely match that
observed for available prey (Figure 4). Alpaida tuonabo preferred
web-sites between 0.6 m-1.2 m, and approximately 60% of the spi-
ders measured were within this range. Thus, while traps nearest the
ground caught the greatest numbers of flying insects, only 18% of A.
tuonabo were found below 0.6 m.
1983]
Shelly — Alpaicia tuonaho
127
Diptera
• nematocerous
o non-nematocerous
2.1r
0.9-
<D
zn
0.5-
40 ' 80 ' 120
Hymenoptera
• ants
o parasitoids
Coleoptera
40
80 120
Number
Figure 2. Relative abundances of major prey categories over all heights sampled.
Each value represents a proportion of the total number of individuals captured on 10
sticky traps suspended at a particular height. See text for details of sampling method.
3;
Frequency
Figure 3. Size frequency distributions for the major prey categories over the 7
heights sampled. Within a category each value represents the proportion of individu-
als captured at a particular height that fell within a particular 1 mm interval. The
symbols used for the various size classes are: 0 — 1 mm (•), 1 — 2 mm (O), 2 —
3 mm (X), and >3 mm (A).
128
Psyche
[Vol. 90
A total of 446 insects representing 6 orders were taken from 320
webs of A. tuonabo. Approximately 95% of these insects belonged
to those 5 prey categories which were most abundant in the sticky
trap samples. Consequently, analysis of both web and spider selec-
tivities will focus only upon these groups. In addition, since the
composition of the flying insect fauna did not much vary with
height, both the data regarding prey availability and diet were com-
bined over all heights.
Web selectivity values did not differ greatly from zero for beetles,
nematocerous Diptera, or parasitoid Hymenoptera (Table 1). Non-
nematocerous Diptera, however, comprised a small proportion of
the web contents relative to their proportion on the traps. Con-
versely, ants represented a large proportion of the web contents
relative to their proportion on the traps.
Aside from nematocerous Diptera, A. tuonabo were observed to
consume prey types in proportions roughly equal to their propor-
tion in the web (Table 2). Spider selectivity values for beetles, ants,
non-nematocerous Diptera, and parasitoid Hymenoptera were all
less than 0.20 (absolute value). In contrast, the Es value for nema-
tocerous Diptera was large and negative.
Figure 4. Vertical distribution of A. tuonaho and available prey. Heights of
hub-resting spiders were measured to the nearest cm and then placed into 0.3 m
intervals. Values for prey represent the total number of insects captured on 10 sticky
traps suspended at a particular height. See text for details of sampling method.
1983]
Shelly — Alpaicia tuonaho
129
Table I. Web selectivity (Ew) values for prey types collected from webs of A.
tuonaho.
Prey type
Collected from webs
(eaten and uneaten)
no. rj
Captured on traps
no. Pi
Ew
Beetles
56
12.5
320
19.2
-0.21
Nematocerous Diptera
164
36.8
337
20.2
+0.29
Non-nematocerous Diptera
34
7.6
453
27.1
-0.56
Ants
128
28.7
1 19
7.1
+0.60
Parasitoid Hymenoptera
42
9.4
264
15.8
-0.25
Others
22^
4.8
175^^
10.4
♦Others include: butterflies (6), bees and wasps (10), leafhoppers (4), thrips (2)
♦♦Others include: butterflies (2), bees and wasps (2), leafhoppers (80), thrips (27),
Hemiptera (8), Orthoptera (5), Collembola (3), Zoraptera (4), Plecoptera (3),
Isoptera (21), Psocoptera (20)
As the Es values imply, the majority (87%) of uneaten prey were
nematocerous Diptera. Most of these, in turn, did not appear to
have been wrapped. Many, in fact, were observed struggling in web
while stuck by a single wing. Similarly, most uneaten non-
nematocerous Diptera and parasitoid Hymenoptera were appar-
ently unwrapped. In contrast, 9 of the 12 uneaten ants had clearly
been attacked and wrapped.
Only 2 groups, nematocerous Diptera and ants, were found in
webs in sufficient numbers to allow meaningful calculation of web
selectivity values for different size classes. Nematocerans less than 1
mm were relatively more abundant in webs than on the traps, while
the opposite was true for those between 1 mm-2 mm (Table 3a).
Web selectivity values, however, did not differ greatly from zero for
either size class. Ants in webs were rather uniformly distributed
among 8 size classes (Table 3b). The majority (76.0%) of ants on the
sticky traps, however, were less than 3 mm long. Consequently, web
selectivity values for the 1 mm-2 mm and 2 mm-3 mm size classes
were large and negative, while those for larger classes were all large
and positive.
Only ants were eaten in sufficient numbers to allow meaningful
calculation of spider selectivity values for different size classes. Yet,
since nearly all (90.6%) of the ants taken from webs were being
eaten, these selectivity values provide little new information. Among
the remaining groups, only nematocerous Diptera had large enough
130
Psyche
[Vol. 90
Table 2. Spider selectivity (E^) values for prey types collected from webs of A.
tuonaho.
Prey type
Collected from webs
(eaten only)
no. rj
Collected from webs
(eaten and uneaten)
no. Pi
Es
Beetles
52
18.8
56
12.5
+0.20
Nematocerous Diptera
32
1 1.6
164
36.8
-0.52
Non-nematocerous Diptera
28
10.1
34
7.6
+0.14
Ants
116
42.0
128
28.7
+0.19
Parasitoid Hymenoptera
32
11.6
42
9.4
+0.10
Others
16^
5.8
22**
4.8
♦Others include: butterflies (6), bees and wasps (10)
♦♦Others include: butterflies (6), bees and wasps (10, leafhoppers (4), thrips (2)
numbers of eaten (32) and uneaten (132) individuals to permit com-
parison. Mean body lengths for eaten (x = 1.6 mm; SD = 1.8) and
uneaten (x = 0.8 mm; SD = 0.29) nematocerans were significantly
different (t = 4.86; p <.00 1).
Discussion
The present findings highlight 2 features of the predatory behav-
ior of A. tuonabo. First, the webs captured and the spiders con-
sumed nonrandom samples of the available prey. Nonrandom web
captures have been recorded for other spiders (e.g. Uetz and Biere
1980; Brown 1981; Turnbull 1960) and most likely reflect differing
abilities for web avoidance or escape among different prey. While
no avoidance was observed, 1 did see several large flies (Asilidae and
Tabanidae) strike webs but then successfully escape. Among insects
successfully restrained by the web, the spider may attack, ignore, or
reject different types and/or sizes of prey. Numerous studies (e.g.
Robinson and Robinson 1970, 1973; Riechert and Tracy 1975;
Turnbull 1960) note rejected prey, but few studies (Uetz and Biere
1980) quantify attack vs. ignore probabilities for different prey.
Here, the tendency of A. tuonabo to ignore nematocerans probably
does not reflect avoidance but rather the inability of these small,
weak-flying insects to escape or damage the web. Thus, A. tuonabo
may have ignored these weak prey only to consume them with their
web in the evening. Interestingly, the mean body length of nema-
tocerans being consumed was nearly twice that of nematocerans
1983]
Shelly — Alpaicia tuonaho
131
Table 3. Web selectivity ( values for size classes of nematocerous Diptera and
ants collected from webs of A. tuonaho.
a. Nematocerous Diptera
Size (mm)
Collected from webs
(eaten and uneaten)
no. rj
Captured on traps
no. Pi
Ew
O-I
1 18
71.9
138
40.7
+0.28
1-2
40
24.4
163
48.1
-0.33
2-3
4
2.4
31
9.1
-0.58
>3
2
1.2
7
2.1
-0.27
b. Ants
Collected from webs
Captured on traps
Size (mm)
(eaten and uneaten)
no.
n
no.
Pi
Ew
0-1
0
0.0
0
0.0
1-2
17
13.3
43
36.7
-0.47
2-3
II
8.6
46
39.3
-0.64
3-4
21
16.4
7
6.0
+0.46
4-5
16
12.5
3
2.6
+0.65
5-6
10
7.8
3
2.6
+0.50
6-7
24
18.7
7
6.0
+0.51
7-8
13
10.2
2
1.7
+0.71
>8
16
12.5
6
5.1
+0.42
caught in the web but ignored. Spider selectivity for larger prey has
also recently been demonstrated for Micrathena gracilis (Uetz and
Biere 1980).
Second, A. tuonabo did not construct their webs at heights where
total prey abundance was greatest. Since the taxonomic and size
composition of the flying insect fauna varied only slightly with
height, A. tuonabo was apparently not responding to the vertical
distribution of a particular type or size of prey. Several factors
potentially affect web height in A. tuonabo. First, although females
use various support structures, the number of suitable “web spaces”
may be limited (Lubin pers. comm.). Also, other species of similar
size (e.g. Pronous tuberculifer, Edricus crassicaudus, and Leucauge
sp.) construct webs closer to the ground (Lubin 1978; Shelly pers.
obs.). Thus, the higher webs of A. tuonabo may reflect a behavioral
means to lessen interspecific competition for food. In addition.
132
Psyche
[Vol. 90
increased web height may reduce risks of predation from ground- or
litter-dwelling predators.
Acknowledgements
I thank H. Levi, M. Robinson, Y. Lubin, and particularly C.
Olive for helpful comments on an earlier draft. H. Levi identified
the species. D. Weinberger kindly helped process the sticky trap
samples. The Smithsonian Tropical Research Institute provided
logistic support.
IjTI RAH RI ClTHD
Brown, K.
1981. Foraging ecology and niche partitioning in orb-weaving spiders. Oeco-
logia 50: 380-385.
ChAmhkri.in, R. V. .and W. Ivn .
1936. New spiders from Mexico and Panama. Bull. Univ. Utah 27: 1-103.
Croat, T. B.
1978. Flora of Barro Colorado Island. Stanford Univ. Press, Stanford.
Holdridgh, L. R., W. C. Gri NKh, W. H. Hathhwav, T. Liang, and J. A. Tost, Jr.
1971. Forest environments in tropical life zones: a pilot study. Pergamon Press,
San Francisco.
IVLHV, V. S.
1961. Experimental ecology of the feeding of fishes. Yale Univ. Press, New
Haven.
Ft BIN, Y. D.
1978. Seasonal abundance and diversity of web-building spiders in relation to
habitat structure on Barro Colorado Island, Panama. J. Arachnol. 6:
31-51.
Olivh, C. W.
1980. Foraging specializations in orb-weaving spiders. Ecology 61: 1 133-1 144.
RiK( HHRT, S. E. AND C. R. TrA(A .
1975. Thermal balance and prey availability: bases for a model relating web-
site characteristics to spider reproductive success. Ecology 56: 265-284.
RiTGIII RT, S. E. AND J. Lu( ZAK.
1982. Spider foraging: behavioral responses to prey. In Spider communica-
tion. eds. P. N. Witt and J. S. Rovner. Princeton Univ. Press, Princeton,
NJ.
Rohinson, M. H. and B. Robinson.
1970. Prey caught by a sample population of the spider Argiope argentata
(Araneae: Araneidae) in Panama: a year’s census data. Zool. J. Linn.
Soc. 49: 345-357.
Robinson, M. H. and B. Robinson.
1973. Ecology and behavior of the giant wood spider Nephila maculata
(Fabricius) in New Guinea. Smithsonian Contrib. Zool. No. 149.
1983]
Shelly — Alpaicia monaho
133
Turnbi'll, a. L.
I960. The prey of the spider Linyphia triangularis (Clerck) (Araneae,
Unyphiidae). Can. J. Zool. 38: 859-873.
Uktz, G. W., a. D. Johnson, and D. W. Schhmskk.
1978. Web placement, web structure, and prey capture in orb-weaving spiders.
Bull. Br. Arachnol. Soc. 4: 141-148.
Uktz, G. W. and J. M. Bihrk.
1980. Prey of Micrathena gracilis (Walckenaer) (Araneae: Araneidae) in com-
parison with artificial webs and other trapping devices. Bull. Br. Arach-
nol. Soc. 5: 101-107.
REPRODUCTIVE BEHAVIOR OF
CLAEODERES BIVITTATA
(COLEOPTERA: BRENTIDAE)*
By Leslie K. Johnson
Department of Zoology
The University of Iowa
Iowa City, Iowa 52242
Promiscuous aggregations of adult brentid weevils often occur on
host trees, where females gather to oviposit (Meads 1976; Johnson
1982). In such a circumstance, in which a male can potentially
inseminate many females, intense competition by males for females
typically occurs (cf. Thornhill 1976; Alexander and Borgia 1979;
Fincke 1982). In addition, members of the family Brentidae show
considerable intraspecific variation in adult size (Sharp 1895; Soares
1970; Damoiseau 1967). From the numerous studies that show that
larger body size enhances competitive aggressive success (e.g. John-
son and Hubbell 1974; Hamilton et al. 1976; Heinrich and Bartho-
lomew 1979), it might be predicted that larger male brentids would
enjoy greater mating success in breeding aggregations, and —
provided that male size is an important competitive characteristic —
that variation in male mating success would be commensurate with
variation in male body size. I tested these predictions on an aggrega-
tion of C/aeoderes bivittata Kirsch. (Coleoptera: Brentidae) in
which the adults varied more than ten-fold in body weight.
The results of the present study support the idea that body size is
an important trait. Males of nearly equal size engaged in a ritualized
contest which appeared to permit sensitive assessment of relative
size, and larger males enjoyed greater success in fights over females.
However, small ( 1 1-22 mm) as well as large males (31-39 mm long)
were disproportionately represented in mating. Small males had
greater than expected success partly because they at times took
shelter under, rather than guarded, their females, emerging for cop-
ulation when a larger rival was not present.
* Manuscript received hy the editor December 12, 1982
135
136
Psyche
[Vol. 90
Materials and Methods
Claeocleres hivittata adults were studied on a dying tree of Qua-
rarihea asterolepis (Bombacaceae) on Barro Colorado Island (9° 09'
N, 79°51' W) in the wet season of 1980.
On June 9 all weevils from ground level to 2 m on the standing
tree were collected, placed in a bag, sexed, measured in length to the
nearest mm, and replaced. On June 13 all weevils up to a height of 2
m were collected and brought to the laboratory, where they were
sexed, measured, cleaned of most mites with masking tape, and
weighed to the nearest tenth mg on a Mettler H35AR balance. On
June 14 these weevils were replaced on the trunk. On six dates
between June 28 and July 14 the behavior of individually marked
weevils of different sizes was described into a portable tape recorder,
for a total of 13 hours. Rectangular and trapezoidal arenas about
1 / 3 m" were drawn on the sides of the trunk between buttresses. On
a given date the trunk was circled clockwise. The reproductive and
competitive behavior that was centered around all male-female pairs
in an arena was recorded, until none of the pairs originally in the
arena remained. Durations of acts were timed with a stopwatch.
Weevil density on the trunk slowly dwindled from 27-36/ m- on
June 28 to 15 or fewer/ m- on July 14. A few weevils were collected
in alcohol for identification and dissection.
Description of Weevil Activities
Oviposition
Before drilling, a female walks slowly over the smooth trunk,
touching the substrate with her antennae. When a favorable site is
found the female chews for 30-60 min until her rostrum is buried to
the depth of the antennal insertion. Periodically she withdraws her
snout, lifts her head, and expels sawdust from her jaws.
To oviposit, a female turns around and locates the drilled hole by
tapping with rear end and hind legs. She then everts her telescoped
sclerites, bringing the ovipositor to the hole, and remains still for 70
sec to 3 min. The hole drilled is the right size for one egg.
After oviposition the female rocks by bending and straightening
her forelegs 12 times per min for 3.5-12 min, repeatedly moving the
tip of her abdomen between the hole and positions further back. As
the female rocks out, a bristled tergite is everted, to which bits of
1983]
Johnson — Claeoderes hivinaia
137
sawdust and other debris adhere. As she rocks in, the material
appears to be added to the hole.
A female may drill and oviposit three times in succession (Fig. 1).
Female Aggression
Aggression is instigated by females before they drill and by
females that have just completed oviposition. The aggression is usu-
ally directed against drilling females. The encounter may involve
only an intention movement, or the instigator may push, poke, or
swat a drilling female with her snout, or pry her out of her hole by
sticking the snout under her abdomen and lifting. A fleeing female
may be pursued several cm. Reciprocated aggression may result in a
fight lasting 6 min or more in which the combatants kick, face one
another and swivel their heads and forebodies, or thrust the snout
under the other and lift suddenly; females of the same length may
also stack themselves head-to-tail and sweep their snouts over the
tip of their opponent’s abdomen. Fights end when a female leaves
or is flipped from the tree.
Guarding
A guarding male stays with a female as she antennates the trunk,
drills, or oviposits, keeping his rostrum or his body over her (Fig. 2).
He responds aggressively if a rival male draws near, and he may also
threaten a female if she approaches his female too closely, by facing
her, advancing on her, or chasing her with a yawing movement of
the head.
Mating
A male mates with the female he is guarding one to several times
during drilling, and is especially likely to do so Just before the female
pulls her beak out of the wood to oviposit (the onset of oviposition
occurs less than a minute after the termination of copulation in
about 80% of the cases (see Fig. 1)). A few seconds before mating a
male accelerates his movements, antennates the female, and then
mounts, sometimes trying the female’s head. Copulation lasts about
a minute.
Rejection
A female not ready to drill or oviposit will walk away from males
that approach. A drilling female can thwart mating attempts by
walking her hind end in a circle around the pivot point of her snout
in wood, or by withdrawing her snout entirely and walking away.
Guard
138
[Vol. 90
Psyche
c
i
ir>
c :s
^ e
^ 0>
, >
<N O
Tf XJ
<
i: 3
c
i I
o i:
c td
c aj
lu -a
t; u
C3 w
■3 ^
C3 CO
^ -o
II ^
‘"S c
2 o o
.“• (u CL -r:
o 3 >
O (U o o
- II t
0^ c
> o
t/2
.. o O
(U T3 '
CJ OJ ID CO
= 11“
(U '-o u
^ .5 II ■•=
•s .2 O c
00 C LT
1 1 1 ^
CO
CJ ^
^ =5 s
“ V- CL 3
_ O o a>
•2 § II o
C D g o
^ "S
CO "O C
E c ii .SP
(U CO Cl ’-3
w- CL CL 3
E £ - t:
c o ^
c
iCi' ti o
<N L» -
c § t; g
W CL CO ^
u >> JJ -r
I -5 E °
C 3 ^ O
t <L> , c/3
E -3 (U -
^ ^ 3 2^
m , --D c
(N "O o
-
.2 ■= o
u. • 2
flQ ^ c/
. “ :l "
-
E ciil
^ u CO :
L» C L
c/i C «■
d
o
_«
3
CL
O
CJ
*o O*-
1983]
Johnson — Claeoderes hivittata
139
Fig. 2. A male, with enlarged jaws for nipping rivals, guards a female Claeoderes
hivittata as she prepares to drill an oviposition hole. A guarding male typically
places all or part of his body (particularly the rostrum) over the female. A small male
in the presence of a large male, however, may insert himself partly under the drilling
female.
Male Aggression
Males fight for access to females. Initially they may intermingle,
jerk, or lash their antennae. Attacks involve nipping, kicking, pok-
ing with the snout, or putting the snout under the rival’s body and
Jerking upwards 3 times/ sec. A male may also interfere \yith copula-
tion by thrusting his snout between a mating pair and pushing. An
attacked male may flee, or reciprocate in kind.
Two males of approximately the same length may engage in a
more stylized contest in which they align themselves, side by side,
1-10 mm apart, facing opposite directions. On the side of the rival a
male taps his antenna and hind leg 4-5 times/ sec, and when the
opponent does likewise, the males fence leg against antenna at either
end. The “appendage-fencing” contests observed in this study lasted
3 sec to 9 min.
Results
Size Variation
Male weevils in the June 13 sample (n = 67) ranged from 12-38
mm in length and from 19-334 mg wet weight, i.e., the biggest male
was 3 times as long and 17 times as heavy as the smallest. Males 1 1
140
Psyche
[Vol. 90
and 39 mm long were found subsequently. The females (n = 81)
ranged from 12-29 mm and 19-247 mg with the biggest female 2Vi
times as long and 13 times as heavy as the smallest. At all lengths
females were heavier than males, and they increased in weight faster
with length than did the males (Fig. 3).
A frequency histogram of the lengths of males (n = 101) and
females (n = 128) measured June 9 is shown in Fig. 4. Mean male
length ±S.D. was 25.91 ±7.21; mean female length ±S.D. was 20.97
±4.35.
Five females were dissected. Each had two ovarioles and 3 or 4
large, yolked eggs. The length of the largest yolked egg increased
monotonically with female length, from a 1.3 mm egg m a 13 mm
female to a 2. 1 mm egg in a 29 mm female.
Size ami Aggressive Success
In aggressive encounters between females the female that fled was
deemed the loser. The winners by this criterion were larger in 14 of
14 contests involving weevils of unequal length (p = .0001, sign test).
Even if four additional encounters involving females of equal length
were conservatively counted as victories for the smaller weevil, the
winners were still significantly more likely to be the larger (p =
.0154).
In male encounters the winner was considered to be the male that
remained by the female. Here again the larger weevil was signifi-
cantly more likely to win (p < .005, sign test). Defending males (the
ones originally with the female) were not significantly more likely to
win encounters than intruding males (p .18).
The relative size of the rivals was also a factor in the occurrence of
the appendage-fencing contest. An analysis of the differences in
length between the rivals in five encounters in which the contest
occurred and sixteen encounters in which it did not, showed that
rivals using the contest were significantly more similar in length (p =
.002, Mann-Whitney U test). The mean ±S.D. difference in length
for rivals using the contest was 1.8 ±2.0 mm; for rivals not using it,
8. 1 ±7.4 mm.
One effect an intruding male may have, whether or not he wins
the female, is to shorten the duration of the defending male’s copu-
lation. Uninterrupted copulations lasted a mean ±S.D. of 82.4
±48.7 sec, with 2/3 of the copulations lasting between 40 and 90 sec.
Copulations interrupted by rivals, however, lasted 31.0 ±15.1 sec
(p = .026, Mann-Whitney U test).
I9K3] Johnson — Claeoderes hivittata 141
Fig. 3. Log,o length (mm) vs. log,o wet weight (mg) of male (•) and female (O)
ClaeuJeres hivinaia. For males, log, q( weight, mg) = -1.53 + 2.54 log, q (length, mm).
For females, log,^ (weight, mg) = -1 .78 + 2.83 log,o (length, mm).
142
Psyche
[Vol. 90
Size and Mating
Individual weevils were compatible with mating partners of many
sizes. In 52 different pairings, females mated with males as much as
10 mm shorter than themselves, and males with females as much as
16 mm shorter. Despite this, mating was size-assortative overall.
The Pearson product moment correlation for male and female
length was r = .323 (p = .021) for the 52 different pairings, and r =
.398 (p = .002) if multiple matings of a pair were included.
Females, however, tended to reject males smaller than themselves
when such males attempted to mate. In 57% of the cases of rejection
(4 out of 7) the female was larger, whereas in only 37% of the cases
of mating (22 out of 60), was the female larger. When lengths of
males rejected and accepted for mating were examined, it was found
that rejected males were shorter (p < .05, 1 -tailed, Mann-Whitney U
test).
Given the more frequent rejection of small males, and the greater
success of larger males in aggressive encounters over females, it was
expected that males found mating would tend to be larger than
males simply present in the aggregation. Whereas females that
mated were larger than unattended drilling females (p < .02, 2-
tailed, Mann-Whitney U test), males that mated were not signifi-
cantly larger than guarding males, males in a random sample, or
males that were alone (Table 1). Instead, a frequency histogram of
mating males showed a bimodality in the size of males that mated
compared to an unimodal distribution of males in the breeding
aggregation (Fig. 4). There appeared to be a dearth of medium-sized
mating males. Indeed, a chi-square test on the 52 different pairings
found that mating males were significantly more likely to be large
(^ 31 mm) or small (^ 22 mm) than would be expected if they
mated in proportion to their abundance in the random sample (x^ =
4.87, 1 df, p = .027).
Extra opportunities for small males to mate could arise if guard-
ing males drove away small rivals less frequently than they did rivals
more their size. With this in mind, I compared the 7 cases in which
two males co-occurred at a drilling female for 3 min or more with
the 19 cases in which one male drove off the other within the first
minute. In 6 of the 7 cases of co-occurrence, one male was small
(^ 22 mm) and the other large (^ 31 mm). In the remaining case
both males were medium-sized. In the 7 cases of co-occurrence the
1983]
Johnson — Claeocleres hivittata
143
Table I. Lengths (mm) of ClaeoJerc.s hivinata individuals in different categories.
The two means marked with an asterisk are the only two compared within a sex that
are significantly different.
Females
Males
n
X
S.D.
n
X
S.D.
99 drilling alone
1 1
19.73*
2.45
(5(5 without partners
14
25.21
7.30
Guarded, drilling
99
49
20.08
3.67
Guarding 55
49
24.55
6.76
Random sample of
99
128
20.97
4.35
Random sample of
(5(5
101
25.91
7.21
99 that mated
52
22.21*
4.20
55 that mated
52
24.77
8.09
mean ±S.D. size difference between the males was 13.6 ±8. 5mm. In
the 19 cases of intolerance, the mean S.D. size difference was only
5.8 ±5.7 mm. The males in the cases of co-occurrence were, in fact,
significantly more disparate in size (p < .02, 2-tailed, Mann-
Whitney U test).
The joint attendance of a drilling female by the two medium-sized
males was short-lived (4 min). The small and large male combina-
tions, on the other hand, were more persistent (x = 19.8 ±10.5 min).
Stability was achieved in part because the small male kept a “low
profile.” The small males were unaggressive, even if poked, and 5/6
of them spent most of their time partway under the drilling female.
Usually it was the rostrum that was tucked under the female, but
two individuals crawled under the female at right angles to her long
axis and centered themselves beneath her. Postures in which a male
placed part of himself under the female were exhibited only by small
males in the presence of a large male guard.
Opportunities to mate did arise for 5 of the 6 small males, despite
the existence of the larger guards. Three of the small males mated
while the large male was fighting off a large intruder. One small
male mated while the large male stood with his snout resting on the
female’s head. Another small male waited until the onset of oviposi-
tion, when the large male left. He then interrupted the post-
oviposition rocking behavior of the female in order to copulate. The
small male that did not mate was driven off by the large male guard,
who was aroused from quiescence by a 38 mm intruder who nipped
him and mated with his female.
144
Psyche
[Vol. 90
2 « t:
w T3 •-
^ 2 2
c iS
^ g.
^ a
w
■5 =
00 ■“
c ^
OJ c3
“ 3
-O
J= o> iS
CJ
^ -C
^ .E
o -
~0 <U
c c
03 O
■c E
S • —
s ^
^ c
(U
T3 3
3 O'
3 .-r
>
0> U
3 "E
o ^
(U
on
o “
g
3 eo
O' 3
OJ T3
U. !E
•o
. C
Tt ■■"
. 00
00 c
ul 3
P ^
o —
c r
•— (U
jj .— •
E ^
os _L
O) m
3
3 ca
j= O
t/5
-a
"O (U
■ > "g
"I
(U 3
E ^
— (U
1983]
Johnson — Claeoderes hivinata
145
Despite their mating successes, the small males were not the equal
of the large ones. The copulations of the small males appeared to
require the absence or inattention of the larger attendant. In con-
trast, the large males mated when they chose. All 6 large males, for
instance, were the last to mate before the female oviposited, after
which they left.
Discussion
In Claeoderes hivittata, there is a great deal of variation in adult
size, presumably due in part to variable growth conditions expe-
rienced by the larvae (Kleine 1933; Haedo Rossi 1961 ; Galford 1974;
Peters and Barbosa 1977). As with another brentid, Brentus ancho-
rago (Johnson 1982), the larger individuals have several reproduc-
tive advantages. In both species, larger females can clear from the
region in which they just oviposited, more of their drilling rivals
(thus possibly reducing later crowding of their larvae), and are less
likely to be ousted from their chosen drilling site. Larger females
also lay larger eggs, an initial advantage which in other beetles has
been shown to significantly affect final adult size (Palmer 1983).
Larger males do more mating than average, and assortatively mate
with larger females who have the reproductive advantage of larger
eggs and greater competitive success at oviposition sites.
There is, however, a principal difference between B. anchorago
and C hivittata. In B. anchorago, the bigger the male, the more he
mates (Johnson 1982). In C. hivittata, the middle-sized males mate
the least. In B. anehorago, males are highly intolerant of other males
at a drilling female. In C. hivittata, a female can sometimes have
two attendants if one is large and one is small.
The circumstances that permit the co-occurrence of a large and
small male at one female need further investigation. In a proximal
sense, small males may be less easily perceived than larger ones.
Certainly, the small males appeared to assist this process by making
themselves less conspicuous. They frequently tucked their snout
under the female, along with the antennae which in other encounters
permit male-male recognition. The small males were not seen to
advertise their presence by initiating acts of aggression. Similar
unprovocative tactics were noted in the smallest males of the wood-
boring weevil, Rhinostomus harhirostris, at females guarded by
large males (Eberhard 1980). Then too in the ultimate sense, it may
146
Psyche
[Vol. 90
not be worth the energy expenditure for a large male to keep small,
persistent males from the vicinity of the female. Despite matings by
small males, large males may enjoy most of the paternity.
A true answer to the question of the relative reproductive success
of large and small males awaits determination of the mode of sperm
competition in C hivittata. Whatever the mode, the relative repro-
ductive success of a small male is probably less than the relative
number of matings he achieves. If there is sperm mixing, the small
males (which in the six cases observed here averaged 200 mg lighter
than the males with which they co-occurred), probably transfer less
sperm per copulation than the large ones. In two species of helico-
niine butterflies, for example, smaller males transfer smaller sper-
matophores (Boggs 1981). If there is sperm precedence, we would
expect large males, with the advantage of weight and strength in
aggressive encounters, to copulate at will when the probability of
fertilizing the egg is the highest. Small males, mating when they
could, might or might not transfer sperm at the opportune time.
The mode of sperm competition is unknown in C. hivittata;
however, sperm displacement has been found thus far to be the rule
in Coleoptera (Walker 1980). If sperm displacement does occur in
C. hivittata, the last male to mate before oviposition would have
the advantage in paternity. That last male advantage occurs in C.
hivittata is suggested by the fact that copulation immediately pre-
cedes oviposition, and that when the female ceases to explore the
trunk and drill, the male ceases to guard her.
I would argue, then, that small males of C. hivittata do not enjoy
nearly as much reproductive success as their proportion of the copu-
lations would suggest, and that there has not been intense- selection
for large males to assiduously expend energy excluding them from
drilling females they are guarding. For small males, however, there
must at times be an advantage to lingering near a female guarded by
a larger rival, for otherwise one would predict that small males
would avoid such females. If there is complete or partial sperm
mixing in C. hivittata, there exists a possibility, however small, that
a given copulation by any male at any time will result in fertiliza-
tion. Even if sperm displacement is complete, there remains the
possibility that the larger rival, distracted by competitors or a more
attractive female, will not return before oviposition begins, leaving
the way open for the small male to copulate last. Similarly, a small
male that mates just after oviposition might still fertilize the next
I9S3]
Johnson — Claeoderes hivinaia
147
egg to ripen if by chance the female went unmated during her next
drilling.
The above arguments do not provide an ultimate explanation for
why small males of C. /?/v/7/^7/^7 enjoy greater mating success than
small males of B. anchorago. Comparative studies are planned for
these two species, which have similar breeding ecologies. Sperm
competition and methods of detecting rivals will be explored, and
the behavior and reproductive input of small, medium, and large
males of both species will be compared. Possibly the system in C.
bivittata represents an early stage of the development of dual male
strategies, and may be a step on the evolutionary road to male
dimorphism (Eberhard 1980). If so, elucidation of the differences
between C. bivittata and B. achorago could help our understanding
of the selective environments favoring dimorphic male behavior and
structure.
Summary
Adults of Claeoderes bivittata aggregated on a Quararibea tree
in Panama. Males ranging in length from 1 1-39 mm guarded and
mated with females 12-29 mm long as they bored holes in the wood
for their eggs. Fights often ensued as females tried to pry other
females from their drilling sites; larger females more often won.
Males fought males for access to females; larger males won signifi-
cantly more often. Disputes involving males of similar size could be
settled by a contest in which the two males stood closely parallel
head-to-tail, while an antenna lashed a hind leg at either end. Such
an appendage-lashing contest may permit rivals to assess one
another’s relative size.
Although individuals differing by at least 16 mm in length could
couple, significant size-assortative mating was observed (r = .4).
Due to the greater aggressive success of larger males and the fact
that males rejected by females were smaller than males they
accepted for mating, it was expected that mating males would be
above average in size. Instead, mating males were significantly more
likely to be large 31 mm) or small 22 mm). The dispropor-
tionate mating of small males may be explained in part by the
tendency of smaller males to wait partly sheltered under a drilling
female, emerging for copulation when larger males are not guarding
the female.
148
Psyche
[Vol. 90
Acknowledgments
I am grateful to Lenny Freed, Jim Palmer, Bob Silberglied, and
Barbara Stay for statistical and biological input, and to Charles
O’Brien for identifying C. hivittata. I also thank Hank Howe, who
found the tree with weevils, and Steve Hubbell, who lent a hand
weighing them. The Smithsonian Tropical Research Institute pro-
vided laboratory space and facilities.
Ijil RATl Rh Cm-;D
Au xandi R. R. D. AND G. Borgia.
1979. On the origin and basis of the male-female phenomenon. In: M. Blum
and N. Blum (eds.). Sexual Selection ami Repruduetive Competition in
Insects. Aeademic Press, N.Y.
Boggs, C. L.
1981. Seleetion pressures affecting male nutrient investment at mating in heli-
coniine butterflies. Evolution 35: 931-940.
DAMOlSl At , R.
1967. Monographic des Coleopteres Brentidae du Continent Africain. Mus.
Royal de L’Afrique Cent., Tervuren, Belg. Ann. Ser. IN-8°-ser. Zool.
160: 1 507.
EhI RIIARD, W. G.
1980. Horned beetles. Scientific American 242 (3); 166-182.
Ein( Kh, O. M.
1982. Lifetime mating success in a natural population of the damselfly, Enal-
laf^nia hu^’eni (Walsh) (Odonata: Coenagrionidae). Behavioral Ecology
and Sociobiology 10: 293-302.
GaI I ORD, J. R.
1974. Some physiological effects of temperature on artificially reared oak bor-
ers. ,1. Econ. Ent. 67: 709 710.
Hai do Rossi, J. A.
1961 . Brent idos Argentinos ( Brent hidae, Coleoptera). Opera Lilloana 6: 1-317.
Hamiiton, W. j. Ill, R. E. Buskirk and W. H. Buskirk.
1976. Social organization of the Namib Desert tenebrionid beetle Onymaeris
ruyatipennis. The Canadian Entomologist 108: 305-316.
Hi nirk II, B., AND G. A. Bartiiolomkw.
1979. Roles of endothermy and size in inter- and intraspecific competition for
elephant dung in an African dung beetle, Searahaeus laevistriatus. Phys-
iological Zoology 52: 484-496.
.loiINSON, 1.. K.
1982. Sexual selection in a brentid weevil. Evolution 36: 251-262.
.loiINSON, L. K. AND S. P. HuBMELL.
1974. Aggression and competition among stingless bees: field studies. Ecology
55: 120 127.
1983]
Johnson — C/aeocIeres hivittata
149
Kl.HINh, R.
1933. Weitere biologische Mitteilungen iiber Brenthiden. Ent. Rdsch. 50:
25 28, 49 50.
Mhads, M. J.
1976. Some observations on Lasiorhynchus harhicornis (Brentidae; Coleop-
tera). New Zealand Entomologist 6: 171 176.
PaI MKR, J. O.
1983. Seasonal body size variation in the milkweed leaf beetle, Lahidomera
(7/wVo///.v (Kirby). Unpublished manuscript.
Pktkrs, T. M. and P. Barbosa.
1977. Influence of population density on size, fecundity and development rate
of insects in culture. Ann. Rev. Entom. 22: 431-450.
Sharp, D.
1895. Brenthidae. Biologia Centrali-Americana 4 (6): 1-80.
SOARKS, B. A. M.
1970. Contribui(;ao ao estudo das especies do genero Acratus Lacordaire.
(Coleoptera, Brentidae). Studia Entomologica 13: 193-252.
Tiiornhiu,, R.
1976. Reproductive behavior of the lovebug, Plecia nearciica (Diptera:
Bibionidae). Annals of the Entomological Society of America 69:
843-847.
Walkkr, W. F.
1980. Sperm utilization strategies in nonsocial insects. American Naturalist
115: 780-799.
POLYDOMY IN THE SLAVE-MAKING ANT,
HARPAGOXENVS AMERICANVS (EMERY)
(HYMENOPTERA: EORMICIDAE)*
Bv
Maria Gi adaluph Del Rio Pesado and Thomas M. Alloway
Erindale College
University of Toronto
Mississauga, Ontario L5L 1C6
Canada
Introduction
Slavery in ants is a form of social parasitism in which parasitic
“slave-making” species exploit the labor of workers derived from
host-species colonies. The slave makers raid host-species nests,
where they capture all or part of the brood. Subsequently, workers
maturing from the captured brood form a social attachment to the
slave makers and perform all the usual worker-ant functions in the
parasites’ colony (see review in Buschinger et al. 1980).
Harpagoxenus aniericanus (Emery) is an obligatory slave maker
living in eastern North America, where it forms mixed colonies with
members of certain Leptothorax species (see Alloway 1979). Two
kinds of H. aniericanus nests are found: “primary nests” containing
a single slave-maker queen and slaves with or without slave-maker
workers, and “secondary nests” consisting of slave-maker workers
and slaves without a slave-maker queen (Creighton 1927; Sturtevant
1927; Buschinger & Alloway 1977). Primary nests are apparently
established when a parasite queen successfully invades a host-
species nest (Wesson 1939), but the origin of secondary nests is
questionable. The problem is compounded by the fact that secon-
dary nests are usually more numerous than primary nests and fre-
quently produce slave-maker females (workers and / or queens) from
'This research was supported by a scholarship from CONACYT [Consejo Nacional
de Ciencia y Tecnologia, Mexico] to the first author and by a grant from the Natural
Sciences and Engineering Research Council (Canada) to the second author. The
authors would like to thank Victor Chudin for his assistance in collecting the data
and Robin Stuart for his constructive comments on the manuscript.
Manuscript received hy the editor February 19, 1983.
151
152
Psyche
[Vol. 90
their broods (Wesson 1939; Buschinger & Alloway, 1977). In the
related European slave maker, H. sublaevis, “ergatoid queens”
(individuals with fully functional ovaries and a spermatheca, but
with a more or less worker-like external morphology) are common
and function as the usual female reproductives (Buschinger, 1978).
However, Buschinger and Alloway (1977) found that, while many
H. americanus workers have functional ovaries and lay eggs that
can mature to produce males, they rarely, if ever, possess a spermat-
eca. Thus, “ergatoid queens” are absent or very rare in H. ameri-
canus. Moreover, these authors thought that thelytoky was unlikely
in this species.
Wesson (1939) observed the formation of secondary nests in the
laboratory. Slave-maker colonies which had conducted several
ordinary slave raids sometimes concluded the final raid of the sea-
son by splitting into two components. In these cases, a few slave-
maker workers and slaves remained indefinitely in the raided nest
with part of the captured brood. Wesson suggested that this might
be a sufficiently common late-season activity to account for the
frequent occurrence of secondary nests. However, even if secondary
nests are formed in this manner, there are still two possibilities for
the relationship between secondary and primary nests and for the
origin of the female slave-maker brood in secondary nests. Follow-
ing their formation, secondary nests might become autonomous
entities functionally separate from their parental primary nests. In
this case, if we exclude thelytoky, the female slave-maker brood in
secondary nests would have to be derived exclusively from brood
carried over from the primary nest when the secondary nest was
initially occupied (Buschinger & Alloway 1977). Alternatively, the
primary nest and one or more secondary nests might comprise a
single multiple-nest (polydomus) colony (Sturtevant 1927). Interac-
tions between the nests of such polydomous colonies would be pro-
tracted, and the slave-maker queen would continue to supply female
brood for all nests in her colony. The latter possibility is supported
by the fact that polydomy of this type has recently been demon-
strated in two of the host species of H. americanus, Leptothorax
ambiguus Emery and L. longispinosus Roger (Alloway et al.
1982). The objective of the present study was to examine these two
possibilities by collecting and mapping H. americanus and host-
species nests in nature, reconstructing their spatial relationships in
the laboratory, and observing the interactions among them.
1983]
Pesach & Alloway — Harpagoxenus aniericanus
153
Material and Methods
The ants were collected on the Erindale Campus of the University
of Toronto in Mississauga, Ontario, during the spring and summer
of 1980 and 1981. Since our purpose was to determine whether
colonies of H. americanus sometimes occupy more than one nest,
we looked for areas where two H. americanus nests occurred within
2 m of each other. Whenever such a place was located, we layed out
a 2 m by 2 m quadrant centering on the two nests and then collected,
numbered and mapped the location of every H. americanus nest and
every nest of its host species (L. ambiguus and L. longispinosus) in
the quadrant. In some cases, adjacent quadrants were combined to
permit the collection of a larger group of slave-maker nests.
In the laboratory, we removed the ants and their brood from their
natural nests and established them in artificial nests of the type
described by Alloway (1979). For censusing, the artificial nests were
placed in petri dishes (diameter = 14.5 cm; height = 1 .5 cm) contain-
ing a water bottle and food (Bhatkar & Whitcomb 1970). Then the
ants were transported to an unairconditioned, naturally lighted
room. On the floor of this room, quadrants were layed out with
masking tape; and the field maps were used to locate the position
occupied by each nest. A thick layer of petroleum Jelly on the mask-
ing tape formed a barrier which confined the ants to their respective
quadrants. A water bottle and food were placed near each nest. In
this way, it was possible to set up the artificial nests so that we
duplicated the spatial arrangement of the natural nests.
In addition to the quadrants collected from the field, we set up
one control quadrant to study behavioral interactions between two
H. americanus nests which had not been collected near one another
in nature. The sides of this control quadrant were 100 cm long, and
the two nests were placed 80 cm apart.
During the course of our observations, some of the ants were
marked so that they could be individually identified. Each mark
consisted of a very small dot of colored nail polish applied to the
dorsal surface of the gaster with the tip of a minuten pin embedded
in the end of a wooden stick. Ants remained marked for periods of 1
day to 1 month.
Observations were made 8 h a day, 5 days a week between 10 June
and 27 August 1980 and between 7 May and 30 August 1981. Five
quadrants were collected and observed during 1980; and 14 quad-
rants were collected and observed during 1981.
154
Psyche
[Vol. 90
Results
There was a total of 19 quadrants. However, quadrant 1 was
merged with quadrant 2 and quadrant 9 with quadrant 10 when an
additional H. americanus nest was found in close proximity to a
group of other H. americanus nests, but outside the original quad-
rant boundary. Altogether, the quadrants contained 49 H. ameri-
canus nests, 57 L. ambiguus nests, and 59 L. longispinosus nests (see
Table 1).
Our most common observation was “fusion” of all the H. ameri-
canus nests in a quadrant. By “fusion”, we mean that eventually all
the ants from two or more H. americanus nests peacefully moved
into a single nest after exchanging adult nest-mates and brood
among the different nests for varying lengths of time (Figure 1). This
exchange was carried out exclusively by slaves. The ability of nests
to fuse shows that there is no behavioral barrier to interactions and
exchange of nestmates among nests and thus indicates either that all
the ants are members of the same polydomous colony or that H.
americanus is a unicolonial species with no behavioral barriers
between its nests. Fusion of all the slave-maker nests was observed
in quadrants (1 + 2), 5, 6, (9 + 10), 11, 12, 13, 16, 17, and 18, in
which there was never more than one slave-maker queen.
However, we did not always observe fusion among H. americanus
nests:
a. In the control quadrant where two H. americanus nests
from different collection sites were arbitrarily set up near each
other, the ants showed no tendency to fuse; and the slave makers
from one nest successfully raided the other slave-maker nest.
b. Quadrants 3 and 8 each contained two H. americanus
queens living in different nests with slaves and a brood. In quad-
rant 3, there was little behavioral interaction between the ants in
the two nests. However, in quadrant 8, after the brood matured,
the H. americanus workers in one nest raided the other slave-
maker nest.
c. In quadrant 4, there was little contact between the ants in
the two H. americanus nests, but the small amount of contact
observed was hostile.
d. In quadrant 14, slaves and slave makers in two nests fused
and then fought with the slaves and slave makers in a third nest.
1983]
Pesach & AUoway — Harpagoxenus americanus
155
Figure 1 - Set-up date, duration of nestmate
exchange until date of fusion.
NO. OF
QUAD.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
f
1
\
f
k
, ' 1
J
\
1
T
1
1
k
T
1980
1981
10 20 30 10 20 30 10 20 30 10 20 30
MAY JUNE JULY AUGUST
I
\ set-up date
I nestmate exchange until fusion.
156
Psyche
[Vol. 90
e. In quadrant 19, both the slaves and the slave makers from
different nests fought whenever they met; and the slave makers in
one nest mounted an incomplete raid against the other slave-
maker nest.
These observations indicate that H. americanus is not a unicolo-
nial species. Aggressive behavioral barriers preventing the exchange
of nestmates and nest fusion exist between some H. americanus
nests. This fact strengthens the conclusion that exchange of nest-
mates and brood and nest fusion, when they do occur, are indicative
of the existence of a polydomous colony.
Somewhat peculiar partial fusions were observed in quadrants 7
and 15. In both quadrants, the slaves peacefully moved into a single
nest. In quadrant 7, the slave makers from the two nests fought. In
quadrant 15 where there were 5 H. americanus nests, some of the
slaves attacked slave makers which had been living in other nests
prior to the fusion. These partial fusions may represent situations in
which lack of contact between nests had begun to produce auton-
omy between nests.
Once the raiding season was over, we observed the formation of
secondary nests in quadrants 1 and 6. In both quadrants, some of
the ants which had been occupying a single nest moved into a
second nest. In both cases, exchange of nestmates and brood con-
tinued for two weeks, when observations ended.
In 7 quadrants, we were unable to find an H. americanus queen,
despite our efforts to collect each nest completely and to search
beyond quadrant boundaries for additional slave-maker nests belong-
ing to these nest groups (see Table 1). However, in each of these
cases, all the maturing H. americanus adults were males, a fact
which indicates that these particular nests had not been receiving
female brood from a primary nest and is consistent with the suppo-
sition (Buschinger & Alloway 1977) that thelytoky does not occur in
H. americanus.
The total number of adults of various species in all nests studied is
summarized in Table 1. The total number of H. americanus workers
was 1 15, with the average slave-maker nest containing about 2 H.
americanus workers. The largest number of slave makers in a single
nest at the time of the original census was 13; and the largest number
of nests in a single apparently polydomus H. americanus colony was
6 in quadrant (9 + 10). Altogether, this colony contained 19 H.
1983]
Pesacio & A I Iowa y — Harpagoxenus aniericanus
157
aniericanus workers, 54 L. longispinosus slaves and 53 L. ambiguus
slaves. The average distance between H. aniericanus nests in nest
groups apparently comprising a single colony was 43.5 cm, with a
range of 1 1 to 159 cm. The average distance between H. aniericanus
nests among which there were aggressive interactions was 61.36 cm,
with a range of 19 to 180 cm.
In the H. aniericanus nests, L. longispinosus slaves outnumbered
L. ambiguus slaves by a ratio of almost 4:1, the total number of
slaves being 803 (79.5%) for L. longispinosus and 207 (20.5%) for L.
ambiguus. All the H. americanus colonies contained L. longispino-
sus slaves, and 9 colonies contained slaves of both species. However,
none of the H. americanus colonies used in this study had only L.
ambiguus slaves, although such colonies are occasionally found in
the Toronto region (Alloway unpublished data). Nevertheless, 7 of
the quadrants studied contained no nests of free-living L. longispino-
sus, and one of the quadrants in which there were L. ambiguus
slaves contained no nests of free-living L. ambiguus (see Table 1).
Unenslaved nests of L. longispinosus were on average somewhat
more populous than the unenslaved nests of L. ambiguus, the mean
number of workers per nest being 25.6 for L. longispinosus and 16.8
for L. ambiguus.
Discussion
Our observations indicate that many Harpagoxenus americanus
colonies are polydomus. This conclusion is based primarily on
observations of peaceful interactions and of nest fusion among nests
collected close together in nature, contrasted with observations that
ants from different H. americanus nests do not always interact
peacefully. The fighting and raiding observed indicate that H. ameri-
canus does not possess a unicolonial population structure. Thus,
peaceful exchange of nestmates and nest fusions, when they occur,
signify the existence of polydomous colonies. However, polydomy
in H. americanus is not obligatory. New colonies are monodomous,
becoming polydomous as they grow. Finally, our observations of
partial fusions suggest that nests in polydomus colonies may gradu-
ally become autonomous, perhaps due to cessation of regular con-
tact between nests. Under these circumstances, new queenless
“secondary colonies,” similar to those envisaged by Wesson (1939),
could be formed.
Table 1. Total number of nests and individuals by species in each quadrant.
158
Psyche
[Vol. 90
sisau lie
*0 16)01
CO
-
LO
LA
O
LA
00
LA
o>
free living
s)sau
.'1
O
O
CO
O
<=
2
L a
D+
70
CM
o
o
66
o
o
107
261
Of
CM
CO
o
O
o
CM
=
o
O)
o>
s)sa
1
U
O
-
CO
LA
CO
LA
CO
CO
CM
CO
L 1
D+
o
53
63
112
116
166
291
125
CM
00
0+
o
-
CM
CM
CM
o
-
sisau
B H
CM
CO
CM
CM
CM
CO
CM
CM
CO
CM
slaves
<0
XH
S2
23
O
O
O
o
O
O
s
53
Of
o
o
o
o
=
=
o
Tin
CH
44
44
24
22
131
67
39
37
54
04
o
=
O
O
O
O
O
o
o
H a
XH
in
00
CO
CO
CO
00
CM
O)
O
-
-
CM
-
o
-
O
CM
o
O
Quad
No
-
CM
CO
LA
CO
r-
OO
Oi
1983]
PesacU) & AUoway — Harpagoxenus americanus
159
2
-
CO
o
s
CO
CO
CM
o
164
OJ
fO
r*-
LO
=
=
=
57
136
29
oo
100
57
o
958
00
CM
eo
=
o
o
=
59
=
o
=
o
oo
69
76
204
o
o
o
o
222
1509
-
in
=
=
O
o
o
o
-
30
CO
C>J
PO
PO
LO
CO
CO
CM
CM
49
PO
o
63
LO
42
=
eo
«=
°
207
=
=
=
-
o
o
O
-
64
37
22
52
84
36
70
803
o
=
o
o
O
O
o
o
O
=
-
PO
CO
LO
2
-
115
O
=
o
O
CM
PO
iO
CO
r-
eo
O)
TOT
arpagoxenus americanus. L 1 = Leptothorax longispinosus.
L a = Leptothorax ambiguus. $ = queen. $ = worker.
160
Psyche
[Vol. 90
Nevertheless, the fact that ants which had been living in several
different nests in nature so frequently moved into a single artificial
nest in the laboratory is somewhat problematic. Under our labora-
tory conditions, polydomy seldom persisted, possibly because our
artificial nests were somewhat more spacious than the acorn nests
which these ants inhabit in nature. If the ants live in more than one
acorn because no single acorn is large enough for the whole colony,
then giving the colony a larger artificial nest might produce nest
fusion. However, many factors other than space may be involved in
producing and maintaining polydomy in these ants (see discussion
in Del Rio Pesado, 1983).
Our observations of nest divisions in two quadrants further sup-
ports the polydomy hypothesis. However, the nest fusions which we
saw did not closely resemble those described by Wesson (1939). In
only one of our colonies did raiding parties tend to remain in target
nests; and even these raiders returned home after 1 to 3 days. What
we observed was that ants which had been occupying one nest came
to occupy two nests after the “raiding season” was over.
Several previous investigators have noted that many H. ameri-
canus nests are queenless (Buschinger & Alloway 1977; Creighton
1927; Sturtevant 1927; Wesson 1939). The usual conjecture has been
that most of these queenless nests are “branches” located near
queenright nests. Our data confirm this supposition by showing that
many queenless nests are parts of queenright polydomus colonies.
However, there were 7 quadrants in which we could not find a nest
containing an H. atnehcanus queen. Since these queenless nests
produced only male slave-maker brood, it is unlikely that they
represent components of a queen-right polydomous colony; and the
males produced in these nests are probably the offspring of H.
aniehcanus workers (Buschinger and Alloway 1977). Some of these
isolated nests may be remnants of colonies whose queen has died,
while others may be products of long-distance raids from which the
raiders failed to return. The presence in some H. americanus nests of
slaves belonging to a species for which there were no free-living
nests in the same quadrant suggests that H. americanus raids may
occur over distances of several meters; and far-ranging raiders may
sometimes fail to return to their base (Creighton 1927).
In our study area, L. longispinosus slaves outnumbered L. ambi-
guus slaves by a ratio of almost 4: 1 . This finding is typical through-
out southern Ontario and the adjacent parts of New York state,
1983]
Pesacio & Alloway — Harpagoxenus americanus
161
despite the fact that L. ambiguus colonies are generally more
abundant than L. longispinosus colonies (Alloway et al. 1982).
Two factors probably account for the prevalence of L. longispinosus
slaves in H. americanus nests. First, H. americanus seems to mani-
fest an ecological preference for rather cool, shady places, a habitat
preference which closely matches that of L. longispinosus. Second,
at our study site, we found that L. longispinosus nests were on
average ore populous than L. ambiguus nests. Thus, a raid against
a nest of L. longispinosus might net more worker pupae than a raid
against a nest of L. ambiguus.
Summary
Field maps were made while collecting nests of the slave-making
ant, Harpagoxenus americanus, and two of its host species, Lepto-
thorax ambiguus and L. longispinosus. The ants were then trans-
ferred to artificial nests arranged to reconstruct the natural spatial
relationships among nests. Ants from adjacent slave-maker nests
often exchanged nestmates and brood for a period of time before
moving into a single nest; and ants which had been living in a single
nest in the laboratory sometimes moved into two nests. However, in
other instances, ants from adjacent nests fought. These observations
were interpreted as indicating that colonies of H. americanus some-
times occupy more than one nest (facultative polydomy). Nest popu-
lation data were also presented and discussed.
References
Alloway, T. M.
1979. Raiding behaviour of two species of slave-making ants, Harpagoxenus
americanus (Emery) and Leptothorax duloticus Wesson. Animal Behav-
iour 27: 202-210.
Alloway, T. M., A. Buschinger, M. Talbot, R. Stuart, & C. Thomas.
1982. Polygyny and polydomy in three North American species of the ant
genus Leptothorax Mayer (Hymenoptera: Formicidae). Psyche, 89:
249-290.
Bhatkar, a. & W. H. Whitcomb.
1970. Artificial diet for rearing species of ants. Fla. Entomologist 53: 217-232.
Buschinger, A.
1978. Genetically induced origin of alate females in the slave-making ant, Har-
pagoxenus sublaevis (Nyl.) (Hym., form). Insectes Sociaux 25: 163-172.
Buschinger, A., & T. M. Alloway.
1977. Population structure and polymorphism in the slave-making ant Harpa-
goxenus americanus (Emery). Psyche 83: 233-242.
162
Psyche
[Vol. 90
Buschinger, a., W. Ehrhardt, and U. Winter.
1980. The organization of slave-raids in dulotic ants: A comparative study
(Hymenoptera: Formicidea). Zeitschrift fur Tierpsychologie 53: 245-264.
Creighton, W. S.
1927. The slave raids of Harpagoxenus americanus. Psyche, 36: 48-80.
Del Rio Pesado, M. G.
1983. Polydomy in the slave-making ant Harpagoxenus americanus (Emery)
(Hymenoptera: Formicidae). M.Sc. Thesis, University of Toronto.
Sturtevant, a. H.
1927. The social parasitism of the ant Harpagoxenus americanus. Psyche 34:
1-9.
Wesson, L. G., Jr.
1939. Contribution to the natural history of Harpagoxenus americanus Emery
(Hymenoptera: Formicidae). Transactions of the American Entomolog-
ical Society 65: 97-122.
Wheeler, W. M.
1910. Ants: Their Structure, Development, and Behavior. Columbia Univer-
sity Press, N.Y. & London, 663 pp.
Wilson, E. O.
1971. The Insect Societies. Belknap Press of Harvard University Press, Cam-
bridge, 548 pp.
SITUATION AND LOCATION-SPECIFIC FACTORS
IN THE COMPATIBILITY RESPONSE IN
RHYTIDOPONERA MET A LUC A
(HYMENOPTERA: FORMICIDAE: PONERINAE)*
By Caryl P. Haskins and Edna F. Haskins
Haskins Laboratories, Inc.
New Haven, Connecticut 06510
Introduction
Hangartner, Reichson, and Wilson (1970) reported some years
ago that individual communities of harvester ants of the genus Pogo-
nomyrmex are able to distinguish the scent of their own nesting
material from that of other conspecific colonies. Holldobler and
Wilson (1977) were able to show that the African weaver ants,
Oecophylla longinoda, mark and advertise individual community
territories by means of colony-specific pheromones deposited in the
rectal fluids. And Traniello (1980) has recently demonstrated that,
in the typically densely packed aggregations of colonies of Lasius
neoniger, persistent trunk trails are maintained which arise from
recruitment trails marked, again, with hindgut material. Here we
describe what we believe to be nest-area marking with hindgut
material in the primitive Ponerine ant Rhytidoponera metallica.
Experiments and results
The tests reported here were a continuation of a series carried on
for some years, and earlier reported in part (Haskins and Haskins,
1979). Material and methods were essentially as described there, and
need only be briefly reviewed. The specific population used in this
work was collected as a single, rather small colony taken at Mont-
ville, in the Blackall Range of northern Queensland, Australia, on
December 23, 1963. It was maintained as a closed inbreeding unit in
the laboratory until the fall of 1979, at which time it had greatly
increased in numbers, was active and vigorous, and contained
* Manuscript received by the editor February 24, 1983
163
164
Psyche
[Vol. 90
numerous “worker”, female and male brood.' Other things being
equal, it might have been expected to have attained considerable
genetic homogeneity, since new generations of “workers” and young
queens were fathered exclusively by males reared within the colony.
On November 4, 1979 this population was divided into two
roughly equal halves and placed in separate arenas standing side
by side on the same laboratory bench. All conditions were kept
constant for the two moieties, designated A and B, except that they
were maintained on differing diets, comprising crickets and dilute
honey water for A and mealworm larvae and dilute sugar water for
B. Two years later, on November 7, 1981, a series of compatibility
tests were run between pairs of individuals taken one each from the
two halves and allowed to encounter one another in fingerbowls, as
described earlier. These demonstrated only very limited incompati-
bility, as reported earlier (Haskins and Haskins, 1979), and sug-
gested that diet, though possibly a measurable influence, was almost
certainly not a critical factor in mediating compatibility as charac-
terized in this test procedure.
Individual pair-tests after isolation on the same diets
On November 1 1, 1981 a further separation of the population was
made by dividing Moiety B into two, designated B' and B'\ and
continuing to maintain both on the identical diets of mealworms
and sugar water, and continuing with no worker interchange or
communication between them. They were held in this manner for a
further year. Then, on November 15, 1982, fifty pair-tests were run
between Moieties B' and B". In all but two of these pairs, full
compatibility was exhibited in the fingerbowl trials. The same tests
run the next day, November 16, between members of one of the pair
of moieties maintained on the same diets {B' and B"), and the first
moiety. A, still maintained on crickets and sugar water, showed
' In /?. metallica reproduction occurs exclusively through fertilized ergatogynes which
may make up from 5% to as much as 15% of the colony population and are morpho-
logically indistinguishable from unfertilized sister workers. Thus reproduction is
continuous and self-sustaining. Colonies are thus characteristically highly polygy-
nous, and may persist nearly indefinitely under laboratory conditions. “True”
females, fully winged and otherwise morphologically typical, can also be produced
(and frequently were in the present population) but they seem to be without repro-
ductive function, soon dealating themselves, functioning briefly as workers, and
dying in a short time.
1983]
Haskins & Haskins — Rhyticloponera mefai/ica
165
results generally confirmatory of those reported earlier, though with
somewhat higher levels of aggression than the year before. Thus, of
fifty-one pairs tested, 37 showed full compatibility, 8 exhibited
“startle” reactions, in 2 cases there was momentary seizure with
immediate release, and in 4 cases there was violent attack. Thus
noncompatibility between the members of Moieties B' and B"
maintained for a year separately on the same diets, was virtually
negligible, while that of moiety A and the other pair, maintained on
different diets, was generally confirmatory of earlier findings: signif-
icantly higher but still, after two years of separation, not nearly
comparable with reactions toward the members of another, widely
separated population taken near Sutherland, N.S.W., as earlier
reported. From all these tests it might have been concluded that, as
indicated by pair compatibility encounters, genetic factors were sig-
nificant but were overlain by a measureable diet factor. In fact, the
situation now appears more complex.
Tests with whole nests
On June 27, 1982, a single nest, housing 70-100 workers of the
second moiety {B') (nests consisted of earth-filled Lubbock-type
glass “sandwiches” stacked) was transferred to the arena of the first
moiety {A). The introduced nest was placed as far away as possible
from the stacked A nests in the arena. Arenas used throughout were
fabricated from 5/8 cm. thick transparent polyster sheets glued
together to form lidless boxes of dimensions 59.5 cm. X 44.5 cm. X
18.5 cm. covered with screening set in wooden frames, and lined
with white paper.
The reaction was immediate, violent, and virtually universal.
Massed workers from A entered the introduced B' nest in force, show-
ing unequivocal hostility, seized and dragged out almost the entire B
population, ultimately killing a large fraction of them. The struggle
went on for two days, and resulted in the apparent total occupation
of the B' nest by A workers. Subsequently, this nest was fully incor-
porated into the A colony. Thus the reaction in this experiment was
in dramatic contrast to the very limited aggression shown in the
pair-tests.
It remained to determine whether similar behavior would occur
between moieties B' and B'\ which had been maintained on the
same diets and, as described, had exhibited nearly complete compat-
ibility in the pair-tests.
166
Psyche
[Vol. 90
At 2:03 p.m. on November 15, 1982 a nest of moiety B' was
transferred to the arena of moiety B" immediately following the
B'-B" pair-tests described above (workers of B' and B" which had
been used in the pair-tests were not returned to their respective
arenas until after the nest-transfer experiment was complete).
Again, in the most conspicuous contrast to the experience in B'-B"'
pair-tests, but in the same pattern as the reaction when the nest of B'
was introduced to A, immediate mass hostility was exhibited
between the two fragments. Eight minutes after introduction it had
become general, with many interlocked pairs. By 6:55 p.m. pairs
“clinched” and stinging were still present within the introduced B'
nest, and disturbed young males present in that nest were emerging
prematurely. This condition persisted until the following day, by
which time it appeared that occupation of the B' colony by
members of the B" moiety had been completed, and things settled
down, leaving many dead workers in the arena.
It therefore became clear that previous dietary history was not a
dominant factor in mediating the mass hostility so conspicuous
between A and B on the one hand and B' and F' on the other. It
remained to test whether it was in fact the presence of the “foreign”
nest with its soil that triggered the mass incompatibility or simply
the introduction of many alien workers at one time near the home
“site” — a “mass” effect of numbers on the one hand or the possible
influence of a familiar site for the test, rather than fingerbowls, on
the other. To check this, at 8:00 a.m. on November 18, 1982 ten
workers of B' were introduced together into the B" arena, being
placed close to the entrances of the B'' stack of nests. Reactions were
completely compatible until 8:25, when two of the introduced
workers were seen being dragged about. This continued for the next
five minutes, when one was released, the other being freed by 8:30.
There was then entire quiet and apparent compatibility until 3:15
p.m., with no further aggression except that a single worker (living
and uninjured) was being dragged about the arena at 12:00 noon of
the following day. The remaining nine were apparently “adopted”.
Simultaneously with this experiment, the reciprocal transfer was
carried out. (10 workers of B" introduced into the B' arena, near the
entrances to the B' nests). The experiment was begun at 8:10 a.m.
Here also there was complete compatibility, except for two workers
seen dragged out of a nest entrance at 3:15 p.m., as observations
1983]
Haskins & Haskins — Rhyticioponera nieta/lica
167
Tabu; 1
A. 10 WORKERS OF GROUP B' INTRODUCED CLOSE TO THE NEST ENTRANCE OF GROUP B"
8:05
Introduced at 8:00 a.m.
Totally amicable reception
8:30
2 $ $ being dragged
8:10
" " "
8:35
15"
8:20
" " "
8:40
All quiet. No dragging seen.
8:25
2 $ $ dragged (after
8:55
All quiet. No dragging seen.
introduction of A)
12:00M
1 5 seen still being dragged
2:00 pm
3: 15 pm
All quiet.
All quiet.
B. 10 WORKERS OF GROUP B" INTRODUCED CLOSE TO NEST ENTRANCE OF GROUP B'
Introduced at 8:05 a.m.
Totally amicable reception
8:10
"
"
"
8:30
All quiet. Most $ $ inside nest
8:20
"
”
"
8:35
"
8:25
8:40
8:55
I2:00M
2:00 pm
3: 15 pm
2 5 § seen dragged out of
nest entrance by 2 $ § each,
released unharmed.
C. 10 WORKERS OF GROUP A INTRODUCED CLOSE TO NEST ENTRANCE OF GROUP B"
Introduced at 8:15 a.m.
8:20 No attacks whatever. Slight
suspicion once or twice. Two
or three $ § bit briefly at
nesting material.
8:23 1 $ being dragged about
8:25 2 $ $ "
8:35 1 § "
8:40 All quiet. No further dragging
seen.
8:55 " " " " " "
3:15 pm " " " " " "
168
Psyche
[Vol. 90
were closed. They were shortly released unharmed. The contrast
with the B'-B" and B"-B' nest introductions could hardly have been
more vivid.
The same experiment was also carried out between the A and
moieties. At 8:15 a.m. on November 18, 10 workers of A were
introduced into the B" arena, again close to the nest entraces of B'\
Five minutes later there seemed complete compatibility. At 8:23 one
worker was seen being dragged about, and at 8:25 two were being so
treated. By 8:35 only one such pair was seen, and nothing further
developed through the cessation of observations at 9:15 p.m. The
results of all three of these experiments are summarized in Table I.
One further confirmation of these results was required. Only three
days had elapsed between the confrontation of nest B ' with that of
B" in the B" arena, (when the B' colony was apparently occupied by
the B" moiety) and the test introduction of ten B' workers into the
B" arena. If (as seemed likely) the introduced B' nest had been
occupied by B" workers, could not the passive reception of the new
B' workers be attributed either to the presence of other B' workers in
the arena or, alternatively (or in addition) might not B" workers
have become somewhat adapted to B' odors, modifying their
reaction? Though the introduced B' nest in the B" arena was
removed after the “nest experiment” and before the new experiment
with the ten B' workers, since but three days had elapsed between
experiments, both factors might well have been involved.
To check this, a longer time interval was allowed to intervene
before the 10-worker test was repeated. On February 15, 1983, 92
days after the preceding tests (all colonies having been left undis-
turbed in the meantime) 10 workers of B" were again introduced to
the B' arena, close to the stacked nests of B'. Introduction was made
at 3:45 p.m. At 4:10 two workers were “clinched” and mutually
stinging near a nest entrance. Five minutes later activity at the nest
entrance was much diminished, and the stinging pair was not seen.
At 4:12, and again at 4:30 p.m., general activity was much dimin-
ished but two workers presumed “alien” were being dragged about
the arena. At 4:35 p.m. no further hostility had developed, but one
or two males had emerged from a B' nest. At 5:00 p.m. the arena was
entirely quiet, with only two workers outside the nests. An hour
later the situation was similarly quiet, but one “alien” worker was
being dragged about the arena and two freshly killed workers were
1983]
Haskins & Haskins — Rhytidoponera nietallica
169
in a corner. Except for these three, no further attacks were wit-
nessed. The other seven workers appeared to have been “adopted”.
It is possible that the attacked workers were in fact egglaying indi-
viduals, which may have stimulated the hostile attacks, as found by
Holldobler (in litt.) for Novomessor in similar situations.
Simultaneously the reciprocal introduction was performed. Ten
workers of B' were introduced into the B" arena in similar fashion,
at 3:45 p.m. Here the reaction was even more passive. Observations
made at five minute intervals until 5:00 p.m. revealed no conflict
whatever. At 6:00 p.m. the same observation was repeated and at
8:00 a.m. the following day the situation remained the same. (Table
2.)
Thus these later tests seemed entirely to confirm the earlier ones:
the introduction of a “mass” of ten workers simultaneously pro-
voked reactions not essentially different from those observed in the
pair-tests on the one hand, and, on the other, in conspicuous con-
trast to the situation when whole nests were introduced. This was
true with moieties which had been maintained since isolation both
on the same and on differing diets.
Discussion
Experiments testing compatibilities between workers from three
moieties of an originally single nest population of Rhytidoponera
metallica after mutual isolation for a period of two years under
conditions identical except for diet on the one hand, and for another
year between halves of one of these moieties isolated and main-
tained under entirely identical conditions (including diet) led to
some interesting conclusions. Pair-tests in fingerbowls indicated
that some incompatibility, with accompanying suspicion or aggres-
sion, could occur between individuals from isolated moieties main-
tained on identical diets for a year, but it was infrequent. Both the
frequency and vigor of aggression were somewhat greater when the
tests were made between workers drawn from moieties isolated on
differing diets but under otherwise identical environmental condi-
tions. Thus it seemed possible that previous dietary history could
have a minor role in mediating compatibility, but not an impor-
tant— much less a decisive — one. Similar tests using ten-worker
samples introduced between the moieties in all combinations yielded
results essentially the same as the pair-tests, indicating that “mass
170
Psyche
[Vol. 90
Table 2
FINAL RECIPROCAL TESTS OF TEN WORKERS BETWEEN COLONY
FRAGMENTS B' AND B"
February 15-16, 1983:92-93 days after first reciprocal tests 11/15/82
February 15\
Workers BH into BU
3:45 p.m. 10 workers introduced from BO
Many workers clustered inside nest entrance, but no hostility, until
4:10 p.m. Two workers “clinched” and stinging near nest entrance.
4:15 p.m.
Activity much diminished at nest entrance. The “clinched” pair not
seen.
4:25 p.m.
4:30 p.m.
Generally quiet but two “alien” workers seen being dragged in arena.
Generally very quiet, but the two “alien” workers still being dragged
in arena.
4:35 p.m.
Some activity around nest entrance, and one or two males emerging.
No hostility observed.
5:00 p.m. Entirely quiet in arena with only two workers out. Some activity
6:00 p.m.
about nest entrance. No conflict.
One “alien” worker seen being dragged by two others. Otherwise all
normal and quiet.
February 16:
8:00 a.m.
Arena quiet with one or two males emerging from nest.
However, 1 dead worker (presumably alien) being dragged about
arena, and two freshly killed workers in corner. These three
presumably BD aliens.
Thus the general picture was one of no general arousal (as before) but ultimate
individual hostility to three out of ten workers, with eventual killing. Entirely
confirmatory of earlier results.
February 15:
Workers Bit into BD:
3:45 p.m. 10 workers introduced from B.
All introduced workers immediately disappeared into BU nests,
without causing any sign of disturbance.
4:20 p.m. Only 5 workers seen outside nests. No conflict and no signs of
disturbance.
4:25 p.m. All very quiet in arena. Only 2 workers out. No conflict.
4:30 p.m. All entirely quiet. 1 worker only seen in arena. No conflict.
4:35 p.m. Completely quiet. One worker seen in arena. No conflict.
5:00 p.m. Completely quiet. One worker seen in arena. No conflict.
6:00 p.m. All completely quiet. I worker seen in arena. No conflict.
1983]
Haskins & Haskins — Rhytidoponera metallica
171
Tabu-; 2 (continukd)
FINAL RECIPROCAL TESTS OF TEN WORKERS BETWEEN COLONY
FRAGMENTS B' AND B"
February 16:
8:00 a.m. Arena entirely quiet. Only 2 workers seen in arena. No hostility, and
no “alien” bodies found.
Thus, throughout this run, there was no hostility of any kind between host and
introduced individual workers. It should be noted that BCl was markedly less
numerous and strong than BU, and while BU contained considerable regenerating
brood, none was found in BO.
These test, therefore, were confirmatory of the earlier ones run on November 15,
1982. Like them, they emphasize the important role played by site nest marking, as
opposed to individual odor characteristics — an interesting convergence to the Tra-
niello findings (Naturwissenschaften 67, S. 361 (1980).
effects” were not demonstrable and almost certainly not signifi-
cantly involved.
In sharp contrast, the introduction of long-occupied earth-con-
taining Lubbock nests of one moiety into the arena of another,
whether the moieties had been maintained on identical or non-
identical diets, was very different, resulting in vigorous mass attacks
and the invasion and occupation of the introduced nest.
This dramatic contrast suggests that, as in the cases of Pogono-
myrmex, Oecophylla, and Lasius, colony-specific nest-site marking
with gut contents (perhaps containing colony-specific pheromones)
is important and regularly employed even in so primitive an ant, and
one with so diffuse and vagile a colony structure, as Rhytidoponera
metallica. This conclusion is reinforced by the extensive (though
apparently random) marking of the substrate with fecal droplets
that we have found general in arenas containing long-occupied
metallica nests, a typical example of which is illustrated in Figure I.
It strongly supports the recent findings of Holldobler (unpublished
ms.) that in the Ponerine ants Paltothyreus tarsatus, a species of
Leptogenys and in two species of Hypoponera fecal droplets depos-
ited at the nest entrances can serve as orientation cues in homing,
while in the last genus colony-specific preferences for these markings
could be demonstrated.
172
Psyche
[Vol. 90
Figure 1. Random marking with fecal droplets of territory surrounding nest in
Rhytidoponera metallica (Straight edge corresponds to margin of Lubbock nest.)
Summary and Conclusions
The following conclusions seem probable from the present work:
(1) As suggested in a previous paper (Haskins and Haskins, 1979)
“recognition” between the members of fragments of a single popula-
tion separated for a year or more appears to remain on the whole
stable through several “generations” of workers which have not
1983]
Haskins & Haskins — Rhytidoponera me tallica
173
been in direct contact during their ontogeny, when those workers
are pair-tested in fingerbowls on an individual basis. This compati-
bility is not universal, however. Incompatibility was observed in a
few cases even between workers of two halves of a population
separated for a year or more but maintained under identical envir-
onmental conditions, including diet, whether tested in pairs or in
groups of ten. When the diet had consistently differed markedly
throughout the period of separation, the numbers of workers exhib-
iting incompatibility appeared somewhat increased, but was still a
minor proportion. It is possible that such individuals eliciting attack
were in fact laying workers, as found by Holldobler in Novomessor.
(2) When earth-containing Lubbock nests occupied by one fraction
of the divided population throughout the periods of separation were
introduced into the arena of another, the situation was dramatically
altered. Mass hositility and mass raiding of the introduced nest by
the recipient moiety regularly followed, regardless of whether the
preceding dietary history was the same or different. We conclude
that, as reported by other investigators in a number of higher ant
genera {Pogonomyrmex, Oecophylla, Lasius) and in the Ponerine
genus Hypoponera) colony-specific nest site marking is important
also in Rhytidoponera metallica, despite its relative primitiveness
and the typical diffuseness and vagility of its colonies. Typical ran-
dom markings of the floors of arenas about earth-containing Lub-
bock nests long occupied by colonies of metallica, as illustrated,
indicate that, as with at least some higher ants, and in several Pone-
rine genera including Paltothyreus, Leptogenys and Hypoponera,
fecal contents are the characteristic marking “vehicle”, perhaps
including, as in the higher ants, colony-specific pheromones. If this
is true of R, metallica, as suggested in the experiments reported, it
becomes interesting to consider the factors involved in mediating
this specific reaction between two halves of a single population
separated for less than two years and maintained on identical diets
and in identical arenas placed side by side on the same laboratory
bench during that period. No evidence has been found of trail mark-
ing, or indeed of trail laying, in R. metallica.
Acknowledgments
We would like to express particular appreciation to Professor
Bert Holldobler of Harvard University for his invaluable assistance
and suggestions in this program.
174
Psyche
[Vol. 90
Literature Cited
Hangartner, W., J. M. Reichson, and E. O. Wilson
1970. Orientation to nest material by the ant Pogonomyrmex haciius (La-
reille). Animal Behaviour 18: 331-334.
HOlldobler, B., and E. O. Wilson
1977. Colony-specific territorial pheromone in the African weaver ant Oeco-
phylla longinoda (Latreille). Proceedings of the National Academy of
Sciences 74: 2072-2075.
Traniello, j. F. a.
1980. Colony specificity in the trail pheromone of an ant. Naturwissen-
schaften 67S: 361-362.
Haskins, C. P. and E. F. Haskins
1979. Worker compatibilities within and between populations of Rhytido-
ponera metallica. Psyche 86: 299-312.
CAPTURE OF BOMBARDIER BEETLES
BY ANT LION LARVAE'
By Jeffrey Conner and Thomas Eisner
Section of Neurobiology and Behavior
Cornell University, Ithaca, NY 14853
Ant lions (larvae of Myrmeleontidae) are well-known for their
unique method of prey capture (Wheeler, 1930). They construct a
conical pit in the sand and lie buried at the bottom with only their
sickle-shaped mandibles, or head and mandibles, exposed. When an
ambulatory arthropod falls into the pit it is seized and pierced by the
mandibles and sucked dry. Bombardier beetles, like other Carabi-
dae, are ground foragers and thus may be expected to fall into ant
lion pits. However, due to their singularly effective chemical defense,
some question remained whether they might be vulnerable to cap-
ture by ant lions. Bombardier beetles respond to attack by ejecting
an aimed spray of hot (100° C) repellent quinones from the tip of the
abdomen (Eisner, 1958; Aneshansley et ai, 1969). The spray is an
effective deterrent to a number of insectivores (Eisner, 1958; Eisner
and Dean, 1976). Several authors (Turner, 1915; Wheeler, 1930;
Lucas and Brockmann, 1981) have observed that ant lions may pull
their prey under the sand after grasping it. Lucas and Brockman
(1981) suggest that this behavior may protect ant lions from aggres-
sive prey. We here report that ant lions can capture bombardier
beetles providing the ant lions have pulled their head beneath the
sand by the time the beetles eject their spray.
Our observations were made at the Archbold Biological Station,
Lake Placid, Highlands County, Florida, where the ant lions {Myr-
meleon crudelis larvae) and bombardier beetles {Brachinus spp.)
were taken. Fifteen ant lions were placed in each of three metal
boxes (30 X 44 X 18 cm high) filled with sand to a depth of 8 cm.
After the ant lions had constructed pits, bombardier beetles were
released individually into the boxes and observed until they slid or
walked into a pit and were seized by an ant lion. Two things were
noted each time a beetle “fired” after being grasped: (1) whether the
'Paper No. 75 of the series Defense Mechanisms of Arthropods. Paper No. 74 is T.
Eisner and S. Camazine, Proc. Nat. Acad. Sci., in press.
Manuscript received by the editor February 25, 1983.
175
176
Psyche
[Vol. 90
ant lion’s head was above the sand or had already been withdrawn
below the surface, and (2) whether the beetle was retained in the ant
lion’s hold or released. Detection of firings posed no problem since
the discharges are accompanied by audible detonations (Eisner,
1958).
A total of 37 captures were witnessed. Five of these involved
beetles that were held only momentarily by the larvae and released
without being induced to discharge. Another three involved beetles
that also failed to discharge, although they were held persistently
and were eventually killed and eaten. The remaining 29 encounters
resulted in bombardier firings (Table 1). Eighteen of these ended
with the beetle escaping: single firings were involved in each case,
and the ant lion’s head was in all instances exposed when the firing
occurred. The beetles were released unharmed promptly after the
discharge. In the other 1 1 encounters in which firings occurred, the
ant lions had withdrawn the head beneath the sand by the time the
beetles fired, and although there were sometimes repeated dis-
charges, only one beetle secured its freedom. The other 10 were
killed and eaten. It is clear that with their heads submerged, the ant
lions are much less likely to be repelled by the spray.
One wonders why the larvae did not consistently withdraw into
the sand the moment they seized a beetle. We had noted that ant
lions commonly pull their victims into the sand, but usually only
when the prey is smaller than the predator itself. The beetles that we
tested were roughly of the size of the ant lions or even larger, sug-
gesting that the larvae may simply have lacked the strength to pull
themselves under while holding such prey. That large insects are
indeed commonly “feasted upon on the surface” had previously
been noted (MacLachlan, 1865).
In three instances when beetles fired at submerged ant lions, the
latter pulled away from the site of discharge by tunneling backward
just beneath the sand surface while keeping the beetle in tow. The
option of burrowing without loss of prey, in a substrate where bur-
rowing can potentially be quicker than the rate of diffusion of a
repellent chemical, could prove helpful to ant lions also in their
capture of chemically protected animals other than bombardier bee-
tles. Indeed, a substantial fraction of prey items ordinarily available
to ant lion larvae, including ants, carabid and staphylinid beetles,
and millipeds, possess dischargeable defensive glands. Interestingly,
1983] Conner & Eisner — Capture of honiharclier beetles 177
Table 1. Summary of the outcomes of all observed encounters between ant lions
and bombardier beetles in which the beetle “fired” defensive secretion. Beetles were
more likely to be killed if the ant lion had pulled itself under the sand by the time the
beetle fired [p <0.001, x2=24.8, 1 d.f., with a continuity correction used (Snedecor
and Cochran, 1967)].
No. Encounters
Position of Ant Lion
No. Firings/ Encounter
Fate of Beetle
18
head exposed
1
all escaped
11
head beneath sand
2.7 ±1.7
1 escaped
(range: 1-5)
10 eaten
one of the few other predators known to be able to capture bom-
bardier beetles is a tabanid larva that lies in wait while semisub-
merged in mud and feeds on the beetles by catching them by a leg
and dragging them into the substrate (Nowicki and Eisner, 1983).
Acknowledgements
We thank Dr. Lionel Stange for identifying the ant lions, John D.
Crawford and Stephen Nowicki for comments on the manuscript,
and the staff of the Archbold Biological Station for hospitality dur-
ing our stay.
References Cited
Aneshansley, D. J., T. Eisner, J. M. Widom, and B. Widom
1969. Biochemistry at 100°C: Explosive secretory discharge of bombardier
beetles (Brachinus). Science 165: 61-63.
Eisner, T.
1958. The protective role of the spray mechanism of the bombardier beetle,
Brachynus hatlistarius Lee. J. Insect Physiol. 2: 215-220.
Eisner, T. and J. Dean
1976. Ploy and counterploy in predator-prey interactions: orb-weaving spiders
versus bombardier beetles. Proc. Nat. Acad. Sci. USA 73: 1365-1367.
Lucas, J. R. and H. J. Brock man n
1981. Predatory interactions between ants and ant lions (Hymenoptera: For-
micidae and Neuroptera: Myrmeleontidae). J. Kans. Ent. Soc. 54:
228-232.
MacLachlan, R.
1865. Observations on the habits of the ant-lion (Myrmeleon formicarius).
Ent. Mon. Mag. 2: 73-75.
Nowicki, S. and T. Eisner
1983. Predatory capture of bombardier beetles by a tabanid fly larva. Psyche,
90: 119-122.
178
Psyche
[Vol. 90
Snedecor, G. W. and W. G. Cochran
1967. Statistical methods, 6th ed. Iowa State University, Ames, Iowa.
Turner, C. H.
1915. Notes on the behavior of the ant-lion with emphasis on the feeding
activities and letisimulation. Biol. Bull. 29: 277-307.
Wheeler, D. M.
1930. Demons of the Dust. Norton and Co., New York.
Wilson, D. S.
1974. Prey capture and competition in the ant lion. Biotropica 3: 187-193.
REPRODUCTIVE PLASTICITY IN YELLOWJACKET WASPS:
A POLYGYNOUS, PERENNIAL COLONY OF
VESPULA MACULIFRONS
By Kenneth G. Ross* and P. Kirk Visscher
Department of Entomology
Cornell University
Ithaca, NY 14853
Introduction
Social wasps in the family Vespidae are thought to have origi-
nated in the southeast Asian tropics (Richards, 1971; Spradbery,
1973a; but see Carpenter, 1982). Members of the subfamily Vespi-
nae presumably evolved monogyny and an annual colony cycle as
adaptations to cold winters in north temperate regions. Exceptions
to this characteristic social organization and colony ontogeny in
vespines have become increasingly apparent (Ross & Matthews,
1982). In climatically favorable areas of their natural range (Tissot
& Robinson, 1954; Duncan, 1939; Vuillaume et al., 1969; Akre et
al., 1980), and in areas newly colonized (Spradbery, 1973b; Perrott,
1975; Thomas, 1960) several species of Vespula (subgenus Paraves-
pula) facultatively form polygynous, perennial colonies. This capac-
ity demonstrates great plasticity in the behavioral ecology of
Paravespula species, and is intriguing in light of its implications for
theories concerning the evolution of eusociality in the Vespidae.
We here report the discovery of a polygynous, perennial nest of
Vespula maculifrons (Buysson) from the southeastern U.S. With
this discovery, all non-parasitic Nearctic representatives of the sub-
genus Paravespula have been shown to exhibit this atypical colony
ontogeny.
*Present address: Department of Entomology, University of Georgia, Athens,
GA 30602.
Manuscript received hy the editor February 25, 1983.
179
180
Psyche
[Vol. 90
Methods
Nest site and excavation
A large Vespula macu/ifrons colony was discovered on 25 Novem-
ber 1981, nesting in sandy soil in sand pine scrub habitat at the
Archbold Biological Station in Highlands County, Florida (27° 1 1'
N, 81° 21' W). Flight from the colony was observed on 1 February
1982 and 1 1 March 1982 and we excavated the colony on 20 March
1982.
We placed traps (similar to Fig. 193 in Edwards, 1980) on each of
the two entrances of the nest in the early morning and aroused the
nest by pounding the ground. Workers (and some males) flying
from the nest were caught in the traps. We discharged a carbon
dioxide fire extinguisher into one of the entrances, which chilled and
partially narcotized wasps remaining inside; we then excavated the
nest.
Because of inadequate narcosis of the wasps and because the nest
was intersected by several large roots, we were unable to remove the
nest intact. Many combs were broken into several pieces. The pieces
were placed in large polyethylene bags and kept frozen until
examined.
Analysis of nest contents
We traced each comb fragment onto a sheet of paper of uniform
weight, and recorded a visual estimate of the proportions of cells
containing capped brood, eggs, and uncapped brood. Of those cells
which contained eggs, we estimated the proportion which contained
more than one egg. Any large (queen-size) cells present were
counted individually in each of the above categories.
To determine the comb area of each fragment we cut out and
weighed each tracing (weight X cm^/g for the paper = area). We
counted the number of small (worker-size) cells on 12 representative
comb fragments totaling 3964 cells. We used the mean number of
cells/cm2 to estimate the number of cells in each of the above cell
content categories for each fragment. These per-fragment estimates
were then summed for the entire nest (205 comb fragments).
Sex ratio of colony
We estimated the sex ratio of the capped brood in small cells by
removing 33 pupae or pharate adults from each of 16 comb frag-
1983]
Ross & Visscher — Poly^^ynoids Vespula 181
c
c
<u
o>
d, ^
c c
3
ryi O
— CJ
O ^
'o
^ in
C =
0
Is
3. KJ
E “
S <2
■a <u
11
CQ -
1
I Si
__ 3
jy .1
sO o
■3-
O
O
3
o
'3-
O
rsi
o
rn —
r"
—
O'' ^
' '
o'
O'
o
■o
o
o
u-
_d
o
'O
■3
2:
r-
O'"
sO
iL)
c.
O'
O'
c.
“
3
O
3
>
no O
oo
VO
. — ,
?:i
"3-
'd
_3
rsi ^
<N
rsi
rsi
bb
c
'c
3
C
o
3
o
lU
E
3
in
>/o o
rsi
'O
"v
o
00
00
X
00
o
o
c
u.
OJ
r-'
o
u
1_
C-
(U
3
X)
i/5
E
3
Z
in
O'
00
r-
CiXj
C
00
00
o
no
Tj-
no
<>
CJ
O''
O''
>.
no X
r-
VO
O
D.
>/o
(N
no
VO
oo
E
O ^
o'
m
■3-
•3-
in
in
"iJ
<U
o
o
—
<u
3
3
E
3
O
CA
h-
182
Psyche
[Vol. 90
ments and determining their sex (n = 528). We counted and sexed all
pupae and pharate adults in large cells.
To estimate the sex ratio of the adult population of the nest we
weighed the two sub-populations of workers and males retrieved
from the nest interior and the entrance traps. We then weighed,
counted, and sexed 4 and 5 samples, respectively, of these popula-
tions. The mean number of wasps/ g and the mean sex ratios of these
samples were extrapolated to the population and sex ratio of the
colony adults as a whole.
Reproductive status of queens and workers
We dissected all queens in the nest and classified their reproduc-
tive status as follows: “Developed” ovaries had oocytes greater than
2 mm in length, “undeveloped” ovaries had oocytes less than 1 mm
in length. “Inseminated” queens had spermatozoa visible microscop-
ically in a squash mount of their spermathecae, while “not insemi-
nated” queens had none.
To estimate the reproductive status of workers in the colony, we
dissected a random sample of 100 workers and examined their ova-
ries. We classified them as “undeveloped” [ovarian index less than 1
(Cumber, 1949), oocytes not developed], “moderately developed”
(ovarian index 1-2, maximum oocyte length 0.9 mm), or “well devel-
oped” (ovarian index 3-12, oocyte length greater than 1.0 mm).
Results
Nest structure and brood composition
Estimated cell number and brood composition of the nest are
given in Table 1 (see also Fig. 1). From comb measurements the
volume of the nest cavity was estimated to be ca. 80 liters. The nest
structure was roughly ellipsoid, 71 cm by 45 cm. Two nest entrances
were in use at the time of discovery; flight activity from each was 95
± 14 and 15 ± 6 wasps/ min returning at midday on 1 1 March (here
and throughout this paper numbers in this form denote mean ± 1
SD, except as noted). The combs were present in at least 15 layers.
Total comb area was 15,652 cm^; the combs contained an estimated
100,120 cells.
Small (worker-size) cells comprised 99.3% of all cells in the nest
(4.0 ± 0.20 mm in diameter between parallel sides, range 3. 7-4. 7
mm, n = 44). Of the small cells, 16.7% contained eggs; 41.9% of
1983]
Ross & I'isscher — Poly^ynous Vespula
183
Figure 1. Comb fragments and adult inhabitants of perennial V. nwculifrons
colony. Combs positioned in left foreground contain large cells.
Figure 2. Supernumerary eggs and immatures in small cells of perennial V. niaculi-
frons colony. Note positions of eggs high on the cell walls.
184
Psyche
[Vol. 90
these contained supernumerary eggs (Fig. 2). In a sample of 119
small cells containing eggs, up to 5 eggs/ cell were present ( 1 .9 ± 0.89
eggs/ cell). Most of these eggs were positioned high on the cell walls.
Forty-one percent of the small cells contained no brood or eggs.
Most of these empty cells had reduced cell walls or were papered
over indicating disuse at the time of collection (Duncan, 1939).
The sex ratio of pupae and pharate adults in capped small cells is
presented in Table 2. Note that more than half of the capped brood
sampled were males.
We counted 744 large cells (5.8 ± 0.20 mm in diameter between
parallel sides, range 5.4-6. 1 mm, n = 22) in the nest (Table 1); these
were located on 8 comb fragments, 2 of which contained exclusively
large cells. Over one-third (34.5%) of the large cells were empty;
some had been papered over and the cell walls reduced. Of the
49.6% of large cells with eggs, 1 6.5% contained supernumerary eggs.
One small comb of 29 cells appeared newly constructed. The paper
was light and fragile, no meconia were present, cells on the perime-
ter were shallow, and each cell contained a single egg.
The capped large cells contained predominantly queen pupae and
pharate adults (Table 2), but a large percentage (30.4%) were males.
Adult inhabitants
An estimated total of 11,817 ± 210 (95% confidence interval)
adult wasps were collected (Table 2). Of these adults, 25.8 ± 2.9%
(95% Cl) were males. Of the remaining female adults, only 23 were
queens. An undetermined number of flying workers and males were
not captured as the nest was collected.
We dissected all 23 queens found within the colony. Fifteen of
these queens had undeveloped ovaries; only one of these was insem-
inated. Six queens were inseminated and possessed well developed
ovaries. Two queens were classified as senescent. One of these had
well developed ovaries but had no spermatozoa in her spermatheca;
the oviducts appeared degenerated, pigmented, and clogged. Very
little fat body was present. The other queen, found in the bottom
third of the nest cavity, had apparently been dead for some time.
Some abdominal sclerites had been punctured and the viscera were
desiccated. One wing was missing while the other was very frayed.
Both senescent queens and the six inseminated queens with devel-
oped ovaries had frayed wings and abdominal cuticular markings
characteristic of physogastric, aged queens (Spradbery, 1973a).
1983]
Ross & Visscher — Polygynous Vespula
185
Table 2 — Number of capped brood and captured adults of each caste in perennial
V. macuUfrons colony (percentages in parentheses). Values for adults and capped
brood in small cells are estimates, those for capped brood in large cells are counts.
Queens
Males
Workers
Total
Capped brood
large cells
small cells
49 (69.6)
21 (30.4)
10,454 (53.9)
8,941 (46.1)
70(100)
19,395 (100)
Adults
23 (0.2)
3,044 (25.8)
8,749 (74.0)
11,816(100)
We dissected 100 workers chosen at random. While 74% of the
workers exhibited no ovarian development, 12% had moderately
developed ovaries, and 14% had well developed ovaries.
Discussion
Although this is the first record of such a nest for V. macuUfrons,
the nest size and number of inhabitants are typical for polygynous,
perennial colonies of other Paravespula species (Ross & Matthews,
1982; Spradbery, 1973a; Edwards, 1980). In contrast, annual colo-
nies of V. macuUfrons from northern Georgia and western North
Carolina average 6,104 small cells and 2,551 large cells at their
greatest development (MacDonald & Matthews, 1981). The peren-
nial V. macuUfrons colony, while containing fewer large cells, had
more than 16 times as many small cells as average conspecific
annual nests. Perennial nests of V. germanica (F.) from Tasmania
and New Zealand are reported to contain up to 180 combs and four
million cells (Thomas, 1960; Spradbery, 1973b). The study colony
also contained almost twice as many adult workers as the most
populous conspecific annual colony studied by MacDonald &
Matthews (1981).
The prodigious size of perennial, polygynous Vespula colonies
does not result simply from the cumulative effects of two seasons of
growth. The pre-existing nest structure and worker force presum-
ably support a rate of production by each of the colony’s queens
early in the season attained only much later by annual colonies. In
addition, newly recruited queens in such a colony avoid the inherent
risks of haplometrotic colony founding, such as predation while
foraging and early colony failure (Archer, 1980).
186
Psyche
[Vol. 90
Several to 50 queens are typically present in perennial Vespula
colonies, although Spradbery (1973b) found up to 1000 in perennial
nests of V. germanica in Tasmania. On the other hand, Thomas
(1960) reported only a single queen in each of the perennial V.
germanica nests he studied from New Zealand. Six of the 23 queens
we found in the study colony were inseminated and possessed well
developed ovaries; these were probably functional queens. The large
number of worker brood we found corroborates the evidence for
several egg laying queens. An additional two queens were probably
former reproductives. The remaining queens could have emerged
recently, as queens at all stages of development were present in the
colony. Newly emerged queens in perennial colonies may mate in
the nest with sibs (Ross, 1983; R. E. Wagner, personal communica-
tion) or embark on mating flights and return to the parental nest as
newly recruited reproductives (Spradbery, 1973a, b). Thus, func-
tional queens in polygynous Vespula colonies are typically regarded
as being daughters of the original foundress (Spradbery, 1973b;
Edwards, 1980; but see Ross & Matthews, 1982).
The study colony contained fewer large cells than do average
annual V. maculifrons colonies, indicating that fewer than normal
queens had been reared during the first developmental season. The
presence of large numbers of male and queen brood and adults
indicates that the colony had been rearing reproductives throughout
the winter [as is typical for other perennial Vespula (Edwards,
1980)], and a newly initiated queen comb with eggs suggests that
queens would have continued to have been reared into the spring.
Presumably the number of queens produced over two seasons
would far exceed the productivity of an annual colony.
Over one-half of the capped brood sampled in small cells were
males. The occurrence of so many male brood in the spring, the
great number of supernumerary eggs, and the positions of eggs high
on the cell walls suggest the likelihood of laying workers (R. W.
Matthews, personal communication; Akre et al., 1982). Dissections
confirmed that at least 14% of the workers possessed well developed
ovaries and were probably ovipositing.
Greater than 30% of the brood being reared in large cells at the
time of colony collection were males. This represents a considerably
larger figure than has been previously reported for annual Para-
vespula colonies, in which large cells contain almost exclusively
1983]
Ross & Visscher — Polygynous Vespula
187
queen brood (MacDonald & Matthews, 1981; Spradbery, 1971;
MacDonald et al., 1974). Perhaps pressure to find empty cells
resulted in workers ovipositing in large cells.
The presence of many laying workers in large, diffuse nest struc-
tures is not unexpected if queens exert reproductive control via the
dissemination of volatile or trophic pheromones (Ikan et al., 1969;
Landolt et al., 1977). The percentages of laying workers and super-
numerary eggs we found were in close agreement to those reported
for a queenless nest of Vespa simillima Smith (Yamane, 1974). Lay-
ing workers in annual vespine colonies may be common during the
phase of colony decline, also suggesting diminished queen control
(R. W. Matthews, personal communication; Montagner, 1966; Akre
et al., 1982). These workers appear to occupy regions of the nest not
frequented by the queen (Edwards, 1980).
Vespine workers may indeed represent “hopeful reproductives”
(West Eberhard, 1978; Lin & Michener, 1972): while many workers
never lay eggs, a significant proportion of them do and all can be
regarded as having some probability of directly contributing genes
to subsequent generations. This factor has not been adequately con-
sidered by theories attempting to explain the origin of eusociality by
reference to a polarized view of reproductive castes (Hamilton,
1964a, b; Alexander, 1974; Spradbery, 1973a).
Our discovery of an overwintered, polygynous colony of V. macu-
lifrons completes the series of free-living North American species in
the subgenus Paravespula with this life history. No members of the
subgenus Vespula or the aerial-nesting genus Dolichovespula have
been reported to exhibit this atypical colony cycle (Akre & Reed,
1981). Perennial, polygynous colonies have been reported for V.
squamosa (Tissot & Robinson, 1954; Ross & Matthews, 1982),
whose affinities with other Vespula species are unclear (Akre et al.,
1980; see also Archer, 1981).
The ability to retain colony social cohesion through two develop-
mental seasons and to tolerate the existence of multiple functional
queens points to great ecological and behavioral plasticity in the
subgenus Paravespula. Members of this group differ from other
temperate vespines in additional biological attributes including: (1)
delay of reproductive production until fall or early winter and con-
sequent increased duration of colony life span, (2) development of
populous colonies and large nests, (3) ability to successfully colonize
188
Psyche
[Vol. 90
new areas of the world when introduced by man, and (4) tendency
for workers to become scavengers on carrion or human food and
refuse in the late summer. The common possession of these derived
features supports a monophyletic origin of this group within the
Vespinae. The interaction of these same features accounts for the
greater public health importance of these species relative to other
vespines.
The vespid subfamilies Polistinae and Vespinae, comprised exclu-
sively of eusocial species, are thought to have evolved from a com-
mon social ancestor (Carpenter, 1982). The Polistinae are diverse in
their methods of colony founding and number of functional repro-
ductives (Iwata, 1976; Jeanne, 1980); their social behavior appears
loosely associated with a tropical or temperate existence. In con-
trast, all vespines are characteristically haplometrotic and mo-
nogynous, and form annual colonies (thought to be temperate
adaptations), regardless of their distribution (Iwata, 1976; van der
Vecht, 1957; Akre et al., 1980). Thus, the characteristic social organ-
ization of the Vespinae appears to be the expression of a common
ancestral trait, rather than an immediate response to ecological
conditions. This interpretation suggests that the Vespinae may have
originated in temperate regions rather than in the tropics (Carpen-
ter, 1982), as has been previously assumed.
The evolution of eusociality in vespids is thought to have
occurred by one of two general routes: (1) the subsocial or matri-
filial monogynous route in which prolonged maternal care provides
opportunities for social interaction between a foundress and her
offspring (Evans & West Eberhard, 1970; Spradbery, 1973a), or (2)
the polygynous or parasocial route in which nesting associations of
foundresses of the same generation lead to increasingly complex
levels of social organization (Lin & Michener, 1972; West Eberhard,
1978). Insofar as the occurrence of occasional polygyny in the Ves-
pinae bears on the social origins of this group, the recurrent ability
of colonies to tolerate multiple functional reproductives strengthens
an argument for the evolution of eusociality via the parasocial route
in the Polistinae + Vespinae. The occurrence of perennial, polygy-
nous colonies of Vespula may be viewed as a reversion to a more
primitive behavioral and physiological mode. Further investigations
of this phenomenon should aid in elucidating the environmental and
social contexts under which it occurs.
1983]
Ross & Visscher — Polygynous Vespula
189
Summary
We describe a polygynous, overwintering colony of Vespula
maculifrons from central Florida. The nest contained about 100,000
cells; many brood of all castes, at all developmental stages; over
8000 adult workers and 3000 adult males; and 23 adult queens, at
least six of which were functional egg-layers. Supernumerary eggs
were found in 7000 of the small cells, often placed high on the cell
walls. Of 100 workers dissected, 14 exhibited substantially devel-
oped ovaries and had probably been laying eggs.
With this report, all non-parasitic neartic species of the subgenus
Paravespula are known to occasionally exhibit this unusual life his-
tory, in contrast to the uniformly monogynous, annual species in the
subgenus Vespula. Paravespula also exhibit more plasticity in their
nesting and foraging habits. We discuss the ability of perennial
colonies to tolerate multiple queens and relate this ability to the
question of the evolution of eusociality in the Polistinae+Vespinae.
Acknowledgements
We thank Chester Winegarner for discovering the colony, moni-
toring its survival, and helping in excavation. James M. Carpenter
and George C. Eickwort reviewed the manuscript and made helpful
suggestions. We thank James Layne and Archbold Expeditions for
use of the facilities of the Archbold Biological Station, and for a
grant-in-aid to the junior author. Additional funding through NIH
grant 5 ROl All 601 1-03, Vespid Venom Collection, awarded to Dr.
Roger A. Morse.
References
Alexander, R. (1974) The evolution of social behavior. Annual Review of Ecol-
ogy and Systematics 5,325-81.
Akre, R. D. & Reed, H. C. (1981) A polygynous colony of Vespula pensylvanica
(Saussure) (Hymenoptera: Vespidae). Entomological News 92,27-31.
Akre, R. D., Greene, A., MacDonald, J. F., Landolt, P. J., & Davis, H. G. (1980)
Yellowjackets of America North of Mexico. U.S. Dept, of Agriculture, Agricul-
ture Handbook No. 552.
Akre, R. D., Reed, H. C., & Landolt, P. J. (1982) Nesting biology and behavior
of the blackjacket, Vespula consohrina (Hymenoptera: Vespidae). Journal of
the Kansas Entomological Society 55,373-405.
Archer, M. E. (1980) Population dynamics. In: Edwards, R. Social Wasps Their
Biology and Control. Rentokil Ltd., East Grinstead, Britain (pp 172-207).
190
Psyche
[Vol. 90
Archer, M. E. (1981) The Euro-Asian species of the rw/a group (Hymen-
optera: Vespidae), with descriptions of two new species and one new subspecies.
Kontyu 49,54-64.
Carpenter, J. M. (1982) The phylogenetic relationships and natural classifica-
tion of the Vespoidea (Hymenoptera). Systematic Entomology 7,1 1-38.
Cumber, R. A. (1949) The biology of bumblebees with special reference to pro-
duction of the worker caste. Transactions of the Royal Entomological Society
of London 100,1-45.
Duncan, C. D. (1939) A contribution to the biology of North American vespine
wasps. Stanford University Publications in the Biological Sciences 8,1-272.
Edwards, R. (1980) Social Wasps Their Biology and Control. Rentokil Ltd.,
East Grinstead, Britain.
Evans, H. E. & West Eberhard, M. J. (1970) The Wasps. University of Michi-
gan Press, Ann Arbor.
Hamilton, W. D. (1964a) The genetical theory of social behavior. 1. Journal of
Theoretical Biology 7,1-16.
Hamilton, W. D. (1964b) The genetical theory of social behavior. II. Journal of
Theoretical Biology 7,17-52.
Iran, R., Gottlieb, R., Bergmann, E. D., & Ishay, J. (1969) The pheromone of
the queen of the Oriental hornet, Vespa orientalis. Journal of Insect Physiology
15,1709-12.
IwATA, K. (1976) Evolution of Instinct Comparative Ethology of Hymenoptera.
Amerind, New Delhi.
Jeanne, R. L. (1980) Evolution of social behavior in the Vespidae. Annual
Review of Entomology 25,371-96.
Landolt, P. j., Akre, R. D., & Greene, A. (1977) Effects of colony division on
Vespula atropilosa (Sladen) (Hymenoptera: Vespidae). Journal of the Kansas
Entomological Society 50,135-47.
Lin, N. & Michener, C. D. (1972) Evolution of sociality in insects. Quarterly
Review of Biology 47,131-59.
MacDonald, J. F., Akre, R. D., & Hill, W. B. (1974) Comparative biology and
behavior of Vespula atropilosa and V. pensylvanica (Hymenoptera: Vespidae),
Melanderia 18,1-66.
MacDonald, J. F. & Matthews, R. W. (1981) Nesting biology of the eastern
yellowjacket, Vespula maculifrons (Hymenoptera: Vespidae). Journal of the
Kansas Entomological Society 54,433-57.
Montagner, H. (1966) Sur I’origine des males dans les societes de Guepes du
genre Vespa. Comptes Rendus des Hebdomadaires des Seances, Paris 263,
785-7.
Perrott, D. C. F. (1975) Factors affecting use of mirex-poisoned protein baits for
control of European wasp (Paravespula germanica) in New Zealand. New Zea-
land Journal of Zoology 2,491-508.
Richards, O. W. (1971) The biology of the social wasps (Hymenoptera: Vespi-
dae). Biological Reviews 46,483-528.
Ross, K. G. ( 1983) Laboratory studies of the mating biology of the eastern yellow-
jacket, Vespula maculifrons (Hymenoptera: Vespidae). Journal of the Kansas
Entomological Society, In Press,
1983]
Ross & Visscher — Po/ygynous Vespula
191
Ross, K. G. & Matthews, R. W. (1982) Two polygynous overwintered Vespula
squamosa colonies from the southeastern U.S. (Hymenoptera: Vespidae). Flor-
ida Entomologist 65,176-84.
Spradbery, J. P. (1971) Seasonal changes in the population structure of wasp
colonies. Journal of Animal Ecology 40,501-23.
Spradbery, J. P. (1973a) Wasps An Account of the Biology and Natural History
of Solitary and Social Wasps. University of Washington Press, Seattle.
Spradbery, J. P. (1973b) The European soc\a\ wasp, Paravespula germanica (F.)
(Hymenoptera: Vespidae) in Tasmania, Australia. Proceedings of the Vllth
International Congress of the International Union for the Study of Social
Insects, London 1973,375-80.
Thomas, C. R. (1960) The European wasp {Vespula germanica Fab.) in New
Zealand. New Zealand Department of Scientific and Industrial Research
Information Series 27,1-74.
Tissot, a. N. & Robinson, F. A. (1954) Some unusual insect nests. Florida
Entomologist 37,73-92.
VAN DER Vecht, J. (1957) The vespinae of the Indo-Malayan and Papuan areas
(Hymenoptera, Vespidae). Zoologische Verhandelingen 34,1-83.
VuiLLAiiME, M., ScHWANDER, J., & RoLAND, C. (1969) Note preliminaire sur
Texistence de colonies perennes et polygynes de Paravespula germanica.
Comptes Rendus des Hebdomadaires des Seances, Paris 269,2371-2.
West Eberhard, M. J. (1978) Polygyny and the evolution of social behavior in
wasps. Journal of the Kansas Entomological Society 51,832-56.
Yamane, Sk. (1974) Observations on an orphan nest of Vespa simillima Smith
(Hymenoptera, Vespidae). Kontyu 42,404-15.
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. 90 1983 No. 3
CONTENTS
Nest building behavior and development of the sunflower leafcutter bee:
Euniegachile (Sayapis) pugnata (Say) (Hymenoptera: Megachilidae).
D. R. Frolich and F. D. Parker 193
Anthicidae of the Greater Antilles and a new species from Venezuela (Coleop-
\evdi). Floyd G. Werner 211
Nest architecture and brood development times in the paper wasp, PoUstes
(Hymenoptera: Vespidae). J. F. Sfrassmann and
M. C. Ferreira Orgren 237
New species of the ant genus A/ rop/a.v (Hymenoptera: Formicidae: Ponerinae).
Robert B. Willey and William L. Brown, Jr 249
Larvae of wrack Coleoptera in the families Corylophidae. Rhizophagidae, and
Lathridiidae. Donald S. Chandler 287
The guest ant, Symmyrniica chamherlini, rediscovered near Salt Lake
City, Utah (Hymenoptera, Formicidae). Alfred Buschinger and
Andre Francoeur 297
Emigration raids by slave-making ants: a rapid transit system for colony relo-
cation. Fllen C. Kwait and Howard Topoff 307
Defense of bracken fern by arthopods attracted to auxiliary nectaries.
Matthew M. Douglas 313
Natural history of the workerless inquiline ant Pogonomyrmex colei (Hy-
menoptera: Formicidae). Steven W. Rissing 321
CAMBRIDGE ENTOMOLOGICAL CLUB
Officers for 1983-1984
Margaret Thayer
Donald S. Chandler
Scott E. Miller
Erank M. Carpenter
Andrew Spielman
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, Cornel! 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
M. D. Bowers, Assistant Professor of Biology, Harvard University
Alfred F. Newton, Jr., Curatorial Associate in Entomology, Harvard
University
E. O. Wilson, Baird Professor of Science, Harvard University
President
Vice-President
Secretary
Treasurer
Executive Committee
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: $1 1.00, 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 required 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 $10.00
each, and for full page half-tones, $12.00 each; smaller sizes in proportion. There is
ordinarily no additional charge for setting tables of less then six columns; for tables
of six or more columns the cost is $25 per page.
Psyche, vol. 90, no. 1-2, for 1983, was mailed August 8. 1983
The Lexington Press, Inc., Lexington, Massachusetts
PSYCHE
Vol. 90
1983
No. 3
NEST BUILDING BEHAVIOR AND DEVELOPMENT
OF THE SUNFLOWER LEAFCUTTER BEE:
EUMEGACHILE (SAYAPIS) PUGNATA (SAY)
(HYMENOPTERA: MEGACHILIDAE)
By D. R. Frolich* and F. D. Parker
Bee Biology & Systematics Laboratory
Agricultural Research
Science & Education Administration
USDA
Utah State University, UMC 53
Logan, Utah 84322
Introduction
Eumegachile (Sayapis) pugnata (Say), formerly Megachile (Say-
apis) pugnata Say (Mitchell 1981), is a large (13-18 mm) leafcutter
bee that is widely distributed throughout the United States and
southern Canada (Hurd 1979). Eumegachile pugnata nests in a wide
variety of situations including man-made borings in wood and is
easily trapped in the wild (Medler 1964, Krombein 1967, Parker &
Frohlich in prep.).
Since E. pugnata is oligolectic to flowers of the Compositae
(Tepedino & Frohlich 1982), attention has recently been directed
toward developing the bee as a pollinator of commercial sunflower.
Parker and Frohlich (1983) described its use in hybrid sunflower
pollination; Tepedino and Frohlich (1982) discussed mortality fac-
tors, pollen utilization and sex ratio; and Frohlich (1982) described
various aspects of its ecology. The purpose of this study was to
■Current address; University of Idaho, SW Idaho Research and Extension Center,
Parma, Idaho 83660.
Manuscript received by the editor March 29, 1983.
193
194
Psyche
[Vol. 90
elucidate the within-nest biology of E. pugnata, including develop-
ment, nesting and provisioning behaviors, and nest architecture.
Methods and Materials
Within-nest behaviors were observed from a wooden box
(lXlX3m) located in a green house (6X6X5m). Nests of 2 types were
fastened to cardboard sheets which were then mounted onto the
observation box. 1. Elderberry sticks that had been drilled (9mm
diameter) and planed lengthwise, were covered with a glass plate to
expose the boring; and 2. Glass tubes with plastic inserts were taped
to cigarette filters to facilitate handling (8mm diameter) (Fig. 1).
The end of the glass tube that served as the nest entrance was dipped
in black India ink and inserted into a cork ring to allow the bee
secure footing (Torchio 1972). Nests were darkened with paper slip
covers until cell construction began. Removal of slip covers after the
onset of nesting did not appear to affect females, though no females
nested in uncovered nests. A small swamp cooler mounted above
the wooden box maintained temperatures below 40° C in order to
avoid egg-larval mortality due to heat buildup.
Commercial Helianthus annuus L. and 3 garden variety compos-
ites (Cosmos, Bachelor’s Button, Callendula) were provided as
pollen and nectar sources in beds of approximately equal size.
Because of its usefulness in similar studies of other megachilids
(Parker & Tepedino 1982, Frohlich 1983) Oenothera hookeri
T. & G. was used as nest partition material. A tape recorder,
otoscope, and stopwatch facilitated within nest observations.
As nests were completed, most were removed and replaced. Com-
pleted nests were incubated at 30° C and used to study aspects of
larval development and behavior. The glass plates on the elderberry
sticks were removed prior to incubation and replaced with clear
plastic food wrap. The plastic inserts of the glass tube nests were
also removed and provisions containing eggs were cut away and
placed separately in BEEM® capsules, commonly used in electron
microscopy. As each egg eclosed, the emergent instar was marked
with a tiny spot of pink fluorescent Day-Glo® powder applied with a
watchmaker’s forceps. Disappearance of spots indicated molting
and new marks were made. Larvae were inspected several times a
day and various behaviors associated with each instar were observed
1983]
Frolich & Parker — Eumegachile
195
Figure 1. Glass covered stick and glass tube with plastic insert used for nests.
Figure 2. Schematic drawing showing construction of a partition. Whole leaf
pieces are added in sequence (starting with No. 1) and are sealed to the nest wall, each
subsequent piece partially covering the previous piece.
Figure 3. Egg in late embryogenesis, attached to provision.
Figure 4. Cocoon containing prepupal larva, showing incorporated fecal pellets.
with a dissecting microscope fitted with fiber optics lighting (to
reduce heat load). Larvae that died and examples of each instar
were preserved in picroformalin.
Results
Within-Nest Biology
Females began nesting in the greenhouse 4 June 1981, within 3
days after release. The following is a composite account, in temporal
sequence, from selection and preparation of a new nest to nest
closure. Each activity discussed was observed for several different
females.
Nest Selection — Preparation. Before beginning cell construc-
tion females investigated both types of potential nest substrates.
Sticks and glass tubes that were not covered (darkened) in some way
were either ignored or only casually inspected. Usually before pre-
196
Psyche
[Vol. 90
paring her nest a female would sit quietly at the back of the stick or
just inside the entrance for a few minutes to an hour. Once a choice
was made, extraneous pith particles were picked up with the man-
dibles and Jettisoned outside of the nest during flight. Females did
not make the nest walls completely smooth but cut away gross
irregularities with the mandibles and removed large pith particles.
As many as 24 pith removal trips were observed before nest initia-
tion. During this period of preparation females were especially sen-
sitive to any activity around the nest site. On several occasions
females abandoned nests when an observer approached the nest
entrance. In general nesting E. pugnata were very wary of intruders.
Partition Building. Basal and apical partitions of each cell were
constructed similarly and were composed of the same materials so
construction details of each will be considered together.
After preparing her nest site for cell construction the female left
the nest to retrieve a strip of O. hookeri leaf. The bee landed on the
plant, straddling the leaf, and quickly cut, while walking backwards,
a thin strip V2 to Ya, as long as her body, and returned to the nest.
After entering the nest with the unmodified leaf in her mandibles the
female masticated it into a shiny ball which was pressed into the
back wall, or along the floor where the cell was to be initiated. From
the leaf material a thin ring of moist chewed leaf was formed around
the inner circumference of the tunnel. Three to 6 trips were usually
required to complete the ring. The female then left and returned
with a large oval-shaped leaf piece that was carried beneath the
body by all 6 legs and the mandibles.
The mandible and front legs were used to spread and position the
leaf piece along a portion of the chewed ring thus closing a portion
of the circle (Fig. 2). The outer edge of the unmodified leaf confluent
with the ring was chewed into the ring and the 2 were sealed. The
female also used her head in an extremely fast jackhammer-like
motion to tamp the ring and leaf pieces together. The clypeus and
proximal outer surfaces of the mandibles appeared to be the point
of impact. Subsequent leaf pieces were brought in and fastened to
the ring in the same manner until the base of the cell was covered
(Fig. 2). Three or 4 oval-shaped leaf pieces were required to form the
base of the partition. After the leaf pieces were positioned more
masticated Oenothera strips were used to form a second ring in the
1983]
Frolich & Parker — Eumegachile
197
same position as the first thus further sealing the leaf pieces to the
walls.
Once the second ring was in place the female continued to add to
it by placing more masticated Oenothera on the inside of the ring
and chewing and spreading it toward the center with the mandibles
until a thin layer of moist leaf pulp covered the whole leaf pieces.
Next, moist soil particles (not mud) were collected and placed at the
base of the partition. These clods were cut into many tiny slivers
which were taken singly or in groups and pressed into the pulpy
partition with the mouthparts. These were then tamped in with the
head as before. Oenothera and soil particles were retrieved alter-
nately until the partition approached its ultimate size.
As the partition increased in thickness the periods of tamping
with the head grew longer. During the last half hour of partition
construction tamping often lasted as long as 5 minutes and became
combined with a grooming behavior. Before tamping the female
groomed the posterior portion of the abdomen with her hind legs
and collected a droplet of fluid that was passed to the middle legs
and then the front legs. The fore tarsi with the secretion were then
used to wipe down the face and antennae; especially the clypeal and
mandibular areas that came in contact with the partition during
tamping. Possibly the act of tamping or packing at this point not
only shaped and defined the partition but incorporated a secretion
as well.
After the last leaf pulp and soil were added the concave surface of
the partition was further modified. The female laid on her back and
groomed the posterior portion of the abdomen and again passed a
droplet of liquid to the middle and fore-legs. This time the secretion
was placed between the mandibles and chewed vigorously. The
female then chewed and licked the outer surface of the partition. As
this was finished, provisioning ensued. No threshold or rudiment of
an apical partition was laid down prior to provisioning.
Provisioning. The female first backed into the cell with a load of
pollen carried on the abdominal scopa. Deposition of the first
pollen load began about 3 mm in front of the basal partition and
was spread backwards with the feet in the kicking motion. The
pollen was removed first by the hind legs rubbing together toward
the middle of the sterna. Pollen remaining on the venter between the
198
Psyche
[Vol. 90
fore and mid-legs was scraped off initially by the mid-legs and then
the fore-legs. Both pairs of legs then transferred the pollen to the
hind legs where it was deposited by rubbing the legs together in a
“hand washing motion.” Pollen removal by the legs was aided by a
complementary telescoping motion of the abdomen and elevation of
sternal hairs. As the legs brushed pollen from the side, toward mid-
sternum and backwards, the abdomen contracted so that the tarsi
came in contact with the entire surface of the abdomen. The abdo-
men then elongated and the contraction-brushing motion began
again.
The first load of nectar was brought in on the second provisioning
trip. The female entered head first and picked up the pollen left on
the first trip with her mouthparts, mixing nectar and pollen into a
moist paste that she spread over the concavity in the basal partition.
She then went to the nest entrance, turned around outside on the
nest face, backed in, and kicked any pollen remaining from the first
deposition toward the partition. Before pollen deposition this time
the female arched her body into a ‘U’ shape, with head and abdomen
as its highest points. Front legs and hind legs were placed approxi-
mately halfway up opposite walls of the nest, while mid-legs rested
on the floor. The abdomen was arched and was backed into the
cavity of the basal partition. Pollen removal then proceeded as
before and the load fell into the concavity or onto the floor in front
of the partition. On subsequent trips the female entered head first,
swinging her head back and forth as she approached the provision,
picking up stray pollen with her mouthparts. The dry pollen from
the previous trip was then chewed and mixed with nectar to form a
paste which she molded into a loaf with her mandibles. Pollen was
then deposited atop the growing provision and the sequence was
repeated.
Prior to nectar regurgitation, the bee usually cleaned her face and
antennae, removing pollen with her front legs and passing it to her
hind legs, where it was deposited along the sides of the abdomen.
She also stopped just in front of the entrance and preened again
before embarking on the next foraging trip.
Once the pollen loaf was approximately its ultimate size the
female used the abdomen tip to plunge a shallow hole in the loaf
after each pollen deposition. This hole was then filled with nectar on
1983]
Frolich & Parker — Eumegachile
199
the next trip and masticated. Dry pollen was deposited on it and a
new hole was formed with the abdomen tip. This behavior con-
tinued until the provision was about V3 its ultimate size whereupon
the female tended to sprinkle pollen evenly over the entire surface.
Nectar was also deposited more uniformly and the whole surface
was chewed after each trip, incorporating pollen and nectar.
On the last few pollen trips the bee used her face to flatten the
vertical surface of the pollen loaf, using a motion similar to the
tamping during partition construction.
Oviposition and Cell Closure. Once the cell was provisioned the
female collected an unmodified Oenothera strip. She masticated it
into a moist ball and wiped down the floor in front of the provision,
picking up loose pollen. As when making the basal partition she
used the leaf pulp to form a ring around the inner circumference of
the tunnel close to the edge of the pollen loaf. Two or 3 leaf gather-
ing trips sufficed; the ring was the initiation of the apical partition.
The leaf pulp ring completed, the female made 3 or 4 more forag-
ing bouts each time returning with only nectar. On returning from
the first bout the bee plunged her mouthparts deeply into one side of
the face of the provision and continued to do so in an extremely fast
up and down fashion for several seconds. With the mandibles mov-
ing in a cutting fashion much of the provision was pushed to the side
opposite the female. After the next trip the other side of the pollen
loaf was worked in a similar fashion until the front half of the entire
provision had been thoroughly kneaded. At the end of the final
foraging bout the female regurgitated a large quantity of nectar onto
the middle of the provision face and plunged her mandibles in an
around its center until a small wet hillock was formed. The front
half of the provision was thoroughly wetted with nectar and
appeared much darker in color than the back half. This completed,
the female turned around at the entrance, backed in and oviposited.
As she backed into the cell, she inserted her ovipositor into the
upper half of the hillock, appearing to anchor to the provision. A
series of pumping motions forced the egg onto the hillock where it
appeared to sink into the nectar. When the egg was about halfway
extruded from the female the pumping motions ceased and she
pulled away, leaving the anterior portion of the egg free and at
about a 45° angle (Fig. 3). During oviposition the female remained
200
Psyche
[Vol. 90
fairly rigid with the exception of the abdominal pumping motion
and a slight rocking of the body. The head was cocked downward
somewhat and the antennae wiggled slightly. The whole process
lasted about 60 seconds.
Immediately after oviposition the female left the nest and
returned with leaf material. Most often this was a large oval-shaped
piece that was sealed to the leaf pulp ring. An occasional female
returned with Oenothera strips and added to the ring but most often
the entrance to the cell was immediately closed by adding the oval-
shaped pieces. Once the cell was closed, the apical partition was
constructed in the same manner as the basal partition.
In almost all nests at least 1 partition, not associated with a
provisioned cell, was constructed in the front of the nest to form a
vestibular and an intercalary cell. This partition was constructed in
the same manner as partitions defining provisioned cells, i.e., soil,
leaf pulp, and whole leaf pieces were incorporated. The nest plug
made to close the entrance was also constructed of the same mate-
rial as partitions but was considerably thicker. The behaviors
involved in plug construction were identical to those involved in
partition formation. In addition to size, the closing plug differed
from partitions in that it was often a series of partitions interspersed
with soil and leaf pulp placed one atop the other. The outside sur-
face of the plug was also different in that it contained much more
soil than partition surfaces. Often what appeared to be pure soil was
found on the outside surface of the plug, although leaf pulp was still
used as the binding matrix.
Usually E. pugnata built 1 cell a day, but occasionally some
females began provisioning a second cell. In the greenhouse E. pug-
nata provisioned cells in the morning when pollen was available and
built partitions and plugs in the afternoon and early evening hours.
Cell provisioning took 3.5 hours on the average. The number of
pollen-nectar trips per cell varied from 36-44. Nectar and pollen
deposition took roughly the same amount of time; nectar deposition
= 38.7 sec. (standard deviation, sd = 12.3), pollen deposition = 32.4
sec. (sd = 6.6). Foraging trips ranged from 2 min. 28 sec. to 9 min. 22
sec. and averaged 4 min. 59 sec. (sd = 1 min. 38 sec.). Plug and
partition construction took approximately the same amount of time
as provisioning so that a nest with 1 cell, 1 intercalary partition and
1983]
Frolich & Parker — Eumegachile
201
a plug took about 7 hours to complete. Approximately 15 Oeno-
thera, 15 soil, and 3-4 large oval leaf collecting trips were required
per partition. Plug construction required roughly twice those num-
bers. Collection of oval leaf pieces took longer than collection of
Oenothera strips (x = 1 min. 23 sec., sd = 49 sec. vs x = 40 sec., sd =
9.4 sec.) and soil collecting trips were shortest of all (x = 22.5 sec., sd
= 7.7 sec.).
In most cases females constructed nests in hollow sticks. How-
ever, when undrilled sticks, with shallow (5 mm) starter holes drilled
in the side, were placed in the greenhouse for use by another bee 2
E. pugnata widened the cavities and nested therein.
Development
Egg Hatching. The egg, which was attached to the provision by
its posterior Va, was opaque when deposited but gradually became
translucent as it developed. It measured \-\V2 mm wide anteriorly
and posteriorly, 3-4 mm in length, and was straight (Fig. 3).
Embryogenesis took an average of 5.1 days at 30° C (Table 1) and
some structures became grossly visible through the chorion approxi-
mately 1 day before eclosion.
Eclosion usually took from 10 to 12 hours and became evident with
the appearance of a clear fluid-filled area in the region of the poste-
rior attachment. At this time the dorsal vessel, spiracles and major
tracheal branches were visible. As the fluid increased in the poste-
rior pole the embryo exhibited undulating waves that passed from
anterior to posterior and perhaps aided in concentrating the fluid in
the posterior region. Thus, the chorion was stretched very tightly
over the head of the enclosed embryo. After fluid disappeared from
the posterior pole the embryo appeared to remain quiescent for a
short time. Fluid then began to collect at the anterior pole of the
egg, accompanied by undulating waves moving in the opposite
direction (posterior to anterior). As the chorion became tightly
stretched over the posterior embryo a longitudinal-lateral split in
the chorion became visible at the level of the spiracles. This rupture
divided the chorion into upper and lower halves. As the pressure
and peristaltic waves receded the lower half of the chorion slipped
from the larva and came to lay directly between it and the pollen
202
Psyche
[Vol. 90
mass under most of the body (except the head). The top half of the
chorion including that surrounding the head seemed to dissolve. If
the larva swallowed any portion of the chorion it was not evident.
As eclosion continued the larva came to lay directly on top of the
pollen mass, with all segments touching it, and began to feed.
Feeding Stages. The second stadium was short (Table 1) and the
second instar fed differently than the other instars. The larva
remained nearly motionless with the head in direct contact with
food in an area of the provision that was considerably higher in fluid
and nectar content than other areas. As the larva fed, a back and
forth motion of the head was apparent and it appeared to suck up
fluid like a small pump. The mouthparts were partially buried in the
provision but almost no movement was detectable in that area dur-
ing feeding.
The actual process of molting was not observed but larvae
marked with powder on the dorsal side of the body were noticed
lying on the old exuvium that bore the powder mark after a molt. It
appeared then that the entire old integument was sloughed off and
not dissolved away. The instars molted in the same manner so that
after molts the body was attached to old exuviae which in turn were
attached to the provision.
Ingestion of solid food, aided by the mandibles, began in the third
stadium. Subsequent instars fed in a similar manner but the last
instar consumed the bulk of the provision. As the larva fed, bi-
dentate mandibles shovelled food into the mouth and appeared to
be aided by a pumping motion of the head capsule. As the head
capsule retracted the mandibles pulled the food in and as the head
capsule extended the mandibles opened outwardly. Larvae tended
to feed in bouts of approximately 5 minutes, stopping to swallow
and pass food into the gut with a series of peristaltic waves between
feeding bouts. As the provision was consumed the larva began to
turn from white to yellow and the pollen-filled gut became visible.
The third instar began feeding in the place where the second instar
fed. The fourth instar fed in the same place, hollowing out a cavity
beneath itself. By the middle of the fourth stadium many larvae had
become detached from the provision but were much more mobile
and continued to feed. Regardless of position (attached, detached
with venter on floor, detached with dorsum on floor) the last 3
1983]
Frolich & Parker — Eumegachile
203
Table I. Life history and developmental times (days) for stages.
Stage
X
sd
range
n
Oviposition to Eclosion
5.1
1.0
4-7
12
Eclosion to Solid Food
1.2
.4
1-2
12
Solid Food to First Defecation
6.5
1.9
5-9
8
First Defecation to Cocoon Spinning
14.9
2.1
10-18
19
Cocoon Spinning to Complete Cocoon
3.7
.5
3-4
11
Oviposition to Complete Cocoon
26.6
1.5
24-28
7
instars appeared to feed in a similar manner: with the body of the
larva extended, the mouthparts were planted on the provision, then
the body closed into a ‘C’ shape and several mouthfuls of food were
taken in while contracting, forming a trough on the provision. The
body then extended and the process was repeated in the same
groove cut previously or adjacent to it so that the whole provision
was systematically consumed. Feeding continued into the last sta-
dium after the onset of defecation and lasted up to about 3 days
before cocoon spinning. The time from the first ingestion of solid
food (3rd instar) to first defecation (last instar) averaged 6.5 days
(Table 1). The 3rd stadium averaged 1.6 days.
Defecation. Defecation began a few hours after molting into the
last larval instar. The midgut in the early instars was a blind sac, not
continuous with the hindgut. At the molt to the last instar the gut
was connected and defecation was possible. The last instar was also
distinguishable from other instars by its longer body setae.
Most of the feeding and growth took place during the last sta-
dium. The average length of time from first defecation to the onset
of cocoon spinning was 14.9 days (Table 1).
Feces were small, squat, yellow cylinders and were deposited
away from the provision while feeding continued. As the provision
was nearly consumed the cell began to fill with pellets and the larva
smeared fresh feces on the walls instead of depositing them behind
it. Defecation continued for about 3 days after feeding ceased up to
204
Psyche
[Vol. 90
the time the cocoon was spun. Most of the pellets were incorporated
into the cocoon.
Cocoon Spinning. Before fecal pellets were spun together a web-
like matrix was laid down on the walls. The larva pressed its salivary
lips onto various points of the walls and partitions and deposited a
small droplet of material from which a short strand of silk was
pulled and anchored elsewhere. The apical partition was covered
with many more strands than the walls or basal partition. Most of
the strands were attached anteriorly to the apical partition which
was also that portion of the cell where most of the fecal pellets had
been deposited. The larva anchored a pellet by holding it with the
mandibles, depositing a small drop of material with the mouth,
pulling away and attaching the other end to another pellet, leaf hair
or portion of the wall. As the salivary component was daubed onto
various structures by the salivary lips, the labium appeared to be
split so that the silk was pulled through as if being threaded, and a
steady pressure was maintained. The fecal pellets were spread evenly
across the anterior portion of the cell and when all were anchored a
cavity lined by white threads covering the entire cell had been
formed. During this time the larva showed much mobility and agil-
ity, moving freely about the cell and turning completely around
several times as necessary.
Once the fecal pellets were spun together more tiny strands were
laid down within the cavity until a fairly dense network of threads
that would be the template for the cocoon was formed. The cocoon
was composed of one thin transparent and cellophane-like layer.
The larva deposited the layer in one of two ways. Either a single
thread was grasped with the mandibles and a clear liquid was
exuded as the head moved up and down the strand or the mandibles
separated 2 or more strands, depositing the liquid between them,
moving the head back and forth until the layer dried. A few fecal
pellets were incorporated into the matrix and flattened and spread
out. No recognizable nipple was formed anteriorly. Instead, an area
somewhat more transparent and of similar thickness to the rest of
the cocoon was formed (Fig. 4). The average time from initiation of
cocoon spinning to completed cocoon was 3.7 days (Table 1).
Pupation and Adult Emergence, On 27 July 1981, 10 overwinter-
ing larvae were placed in an incubator at 30° C in order to observe
1983]
Frolich & Parker — Eumegachile
205
pupation. Average time from incubation to pupation was 9.7 days
(sd = 4.3, range = 7-18, n = 9). The transformation from overwinter-
ing larva to adult took an average of 22.3 days (sd = 2.9, range =
20-27, n = 6), with males completing development prior to females.
Pigmentation changes were first observed in the eyes which turned
yellow in approximately 11 to 13 days. At 13 to 14 days both com-
pound eyes and ocelli had turned dark brown. Wing buds became
evident and turned yellow at 1 1 to 14 days. Mouthparts began to
darken at 15 days and had usually turned black within 16-161/2 days.
Coloration of general body regions started at 16 days and began
with patches of integument at the bases of hairs on the vertex, frons,
thoracic terga and abdominal sterna. Hairs quickly turned dark and
pigmentation spread to the remaining portion of the head and
thorax followed by the abdomen. Generally proximal portions of
appendages changed color first with distal portions of the legs
changing color last. From 16 to 20 days the body remained dull
black while wings darkened. A shiny appearance to the body and
hairs did not appear until just prior to ecdysis. Bees emerged from
cells shortly after wings had darkened and proboscides had been
retracted.
Discussion
The incorporation of glandular secretions into nest linings,
widespread in the Apoidea, is believed to have evolved as a mecha-
nism to protect larvae and provisions from dehydration and/ or the
microbial consequences of excessively humid environs. Batra (1972,
1980), Cane (1981), and Eickwort et al. (1981) have discussed the
inclusion of salivary and/or Dufour’s gland components into cell
linings and provisions of the ‘short tongued bees’ (Colletidae, Halic-
tidae, Andrenidae) and the Anthophoridae. While the phenomenon
has likely figured prominently in the evolution of these groups, little
or no attention has been directed toward similar behaviors in the
Megachilidae. Indeed, the evolution of the megachilidae has been
viewed in a different framework. Eickwort et al. (1981) see the bulk
of the megachilids (Megachilinae) as having evolved from a soil
dwelling ancestor that developed the ability to gather foreign mate-
rials (leaf pieces, mud, resin, etc.) to line cells as an alternative to
206
Psyche
[Vol. 90
glandular secretions in overcoming the constraints of humidity.
While Eickwort et al.’s (1981) hypothesis of nesting evolution in the
Megachilidae is interesting, more recent evidence points to the fact
that glandular secretions are important features of megachild nest-
ing biologies. Parker and Tepedino (1982) observed the application
of salivary secretions to bare walls of Osmia marginata Michener
nests. Frohlich (1983) observed the incorporation by Osmia bruneri
Cockerell of an abdominal secretion into the provision, and also
noted the application of a salivary secretion to partitions. Dianthi-
dium ulkei ulkei (Cresson) also incorporates an opaque viscous sub-
stance, originating from the abdomen, into resinous cell walls and
spreads the material over bare areas of the cell (Frohlich and
Parker, in prep.).
Since the twig nesting megachilids probably arose from soil nest-
ing megachilids (Eickwort et al. 1981) and since the Megachilidae is
distantly related to the other soil nesting families (Michener 1974)
we propose that the Megachilidae have retained the habit of using
glandular secretions to line cells. It seems likely that a waterproof
layer of some sort is necessary to maintain the humidity of the cell
within tolerable limits. Lining cells with leaves in soils that are
particularly moist would have little effect on reducing humidity and
controlling fungal growth nor would lining pre-existing cavities
such as twigs prevent dehydration. It is important, therefore, that
we thoroughly examine the behavioral, and more importantly chem-
ical, components of nesting in order to gain an understanding of the
role of nest architecture in evolution.
The paucity of information available on the nesting biologies of
other species of Eumegachile make it difficult to confirm (or, refute)
Mitchell’s (1981) recent revision of the old genus Megachile on the
basis of behavior or nest architecture. The available data does con-
firm the separation of Eumegachile from Megachile, since the nest
architectures of the two are radically different. Eumegachile pug-
nata nests are similar to, though somewhat more elaborate than,
known nests of other species in the subgenus. Krombein (1967)
reported that E. inimica inimica Cresson makes unlined cells with
partitions of agglutinated sand a little larger than the inner circum-
ference of the nest. Eumegachile inimica sayi Cresson also uses a
single leafcutting as a partition but covers it with leaf pulp, incorpo-
rating five pebbles (Krombein 1967). Eumegachile policaris Say lays
1983]
Frolich & Parker — Eumegachile
207
more than one egg per provision, makes partitions and plugs com-
posed of leaf pulp and two layers of small compressed leaflets with
no soil or pebbles, and constructs no vestibular or intercalary cells.
No other biologies in the subgenus Sayapis are known and no biol-
ogies or nest architectures are known in the other subgenera of
Eumegachile.
The manner in which E. pugnata constructs individual cells rend-
ers its adoption as a potential pollinator of commercial sunflower
somewhat problematical under certain circumstances. Eumegachile
pugnata construct cells that are separated from each other by parti-
tions and are not surrounded by a leaf envelope. This is unfortunate
because E. pugnata is susceptible to a chalkbrood fungus, Ascos-
phaera aggregata Skou, the treatment of which in other bees takes
advantage of nest architecture. The disease sometimes decimates
populations of the alfalfa leafcutting bee. Megachile rotundata
(Fabricius), which construct cells that are completely leaf lined.
During treatment, nests are opened, individual cells are separated as
discrete leaf lined units, treated, and stored (Parker and Torchio
1980). When M. rotundata emerge only egress from individual cells
is necessary and adults are not required to chew through cells con-
taining dead larvae with infectious spores. Since E. pugnata are
protected only by a thin cocoon and no leaf lined envelope, removal
from the nest would cause excessive mortality. This “loose cell”
management is also used to control various M. rotundata parasites.
In the case of E. pugnata parasites could emerge from individual
cells and reparasitize other cells without leaving the nest.
A second point that will have to be considered in commercial
pollination is the fact that E. pugnata incorporates a fair amount of
nectar into the provision. If growers are going to increase bee popu-
lations, sunflower cultivars that provide adequate nectar will have
to be available.
Finally, one trait that makes E. pugnata a good candidate for
sunflower pollination is its habit of provisioning cells early in the
day. Male fertile sunflower cultivars dehisce overnight and in the
early morning. Thus, the greatest amount of pollen is available
during the time that E. pugnata are provisioning and pollinating
flowers.
208
Psyche
[Vol. 90
Acknowledgments
Critical comments by G. C. Eickwort, V. J. Tepedino, and P. F.
Torchio substantially improved and benefited our manuscript. We
thank them for their conscientious efforts. Additionally, B. Pei-
tersen, C. Schmitz, D. Veirs and T. Griswold rendered invaluable
field assistance.
References Cited
Batra, S. W. T. 1972. Some properties of the nest-building secretions of Nomia,
Anthophora, Hylaeus and other bees. J. Kans. Entomol. Soc. 45: 208-218.
1980. Ecology, behavior, pheromones, parasites and management of
sympatric vernal bees Colletes inaequalis, c. thoracicus and C. validus. J. Kans.
Entomol. Soc. 53:509-538.
Cane, J. H. 1981. Dufour’s gland secretion in the cell linings of bees (Hymenop-
tera: Apoidea). J. Chem. Ecol. 7:403-410.
Eickwort, G. C., R. W. Matthews, and J. Carpenter. 1981. Observations on
the nesting behavior of Megachile rubi and M. texana with a discussion of the
significance of soil nesting in the evolution of megachilid bees (Hymenoptera:
Megachilidae). J. Kans. Entomol. Soc. 54:557-570.
Frohlich, D. R. 1982. The biology and ecology of Eumegachile (Sayapis) pug-
nata (Say) (Hymenoptera: Megachilidae), a pollinator of commercial sun-
flower. M.S. Thesis. Utah State University, Logan, Utah. 1 10 pp.
1983. On the nesting biology of Osmia (Chenosmia) bruneri Cockerell
(Hymenoptera: Megachilidae). J. Kans. Entomol. Soc. 56: 123-'130.
Hurd, P. D., Jr. 1979. Superfamily Apoidea. Pp. 1741-2209 in K. V. Krombein,
P. D. Hurd, Jr., D. R. Smith, and B. D. Burks, eds. Catalog of Hymenoptera in
America North of Mexico. Smithsonian Institution Press, Washington.
Krombein, K. V. 1967. Trapnesting wasps and bees: life histories, nests, and
associates. Smithsonian Press, Washington, 570 pp.
Medler, j. T. 1964. A note on Megachile (Sayapis) pugnata pugnata Say in
trap-nests in Wisconsin (Hymenoptera: Megachilidae). Can. Entomol. 96:
918-921.
Michener, C. D. 1974. The social behavior of the bees. Harvard University
Press, Cambridge, Massachusetts. 404 pp.
Mitchell, T. B. 1981. A generic revision of the megachiline bees of the Western
Hemisphere. Cont. Entomol. Dep. N. Carolina State University, Raleigh. 95
PP-
Parker, F. D. and D. R. Frohlich. 1983. Hybrid sunflower pollination by a
manageable composite specialist: The sunflower leafcutter bee (Hymenoptera:
Megachilidae). Environ. Entomol. 12: 576-581.
Parker, F. D. and V. J. Tepedino. 1982. The behavior of female Osmia margi-
nata Michener in the nest (Hymenoptera: Megachilidae). Pan-Pac. Entomol. In
press.
1983]
Frolich & Parker — Eumegachile
209
Parker, F. D. and P. F. Torchio. 1980. Management of wild bees. pp. 144-160
In: Beekeeping in the United States. USDA Agriculture Handbook 335. 193 pp.
Tepedino, V. J. AND D. R. Frohlich. 1982. Mortality factors, pollen utiliza-
tion, and sex ratio in Megachile pugnata Say (Hymenoptera: Megachilidae), a
candidate for commercial sunflower pollination. J. N. Y. Entomol. Soc.
90:269-274.
Torchio, P. F. 1972. Sapyga pumila Cresson, a parasite of Megachile rotundata
(F.) (Hymenoptera: Sapygidae; Megachilidae). I: Biology and description of
immature stages. Melanderia 10:1-22.
■J ■
1 ,
,
@1
■ ■ ■■ ■: ■ ■■ ■ '■ ■■^•! ■< .-
Vj-
^ Vr' -^■•*- - V, ‘ ’"-
I? ;v.. .r,, ■■ vr- 'V-‘‘'^ -w:«:Tr.>'
s«i« 1H j,;- rijgi-j
f* , Tte-^ r- , ^ Stf-.t_ .^ . - f ' ''■■■ ' ■» ■ ;
St
M;’
' i' ' ■ i “■
■ ^,M"v’’^‘'^j| ^ r:..^
■,., ||»»^ r, ;•: j K* ■!*_;■ I Ml ^ *S .
•.'*1
ANTHICIDAE OF THE GREATER ANTILLES,
AND A NEW SPECIES FROM VENEZUELA
(COLEOPTERA)'
By Floyd G. Werner
Department of Entomology
University of Arizona
Tucson 85721
Thirteen of the 29 species that are known or reported from the
Greater Antilles appear to be endemic. Five (Anthicus darlingtoni,
hispaniolae, macgillavryi, soledad and subtilis) make up the subtilis-
group, which does not seem to have near relatives on the mainland.
Three others stand quite isolated in their genera: Acanthinus schwarzi
in an almost exclusively neotropical genus, Anthicus blackwelderi
and russoi in a world-wide genus that contains many diverse ele-
ments. A. blackwelderi is counted among the endemic species
because it has different color patterns on the islands that it is known
to inhabit; the form of the internal sac of the male genitalia is very
different from that of possible relatives on the mainland. A. russoi is
probably not properly placed in Anthicus, and is unlike any anthicid
known to me in several details; Menozzi’s (1930) evidence that it is a
myrmecophile with a native ant makes local origin seem logical.
The 5 other endemic species are similar to mainland New World
species. Mecynotarsus hispaniolae and jamaicanus belong to the
elegans-group, which has species from Florida to Central America.
Notoxus bipunctatus and jamaicus have been assigned to the
monodon-group (Chandler 1978), which ranges from Canada to
northern South America. Finally, Anthicus antilleorum seems to
have originated in the Greater Antilles and spread to the Virgin
Islands and Bahama Islands; its nearest relatives are found around
the southern Caribbean.
Within the 13 endemic species, there is inter-island variation in
color pattern in 3: Anthicus antilleorum, blackwelderi and soledad:
in each instance the Jamaican population is different from that of
'Arizona Agricultural Experiment Station, Journal paper No. 3662.
Manuscript received by the editor August 30, 1982.
211
212
Psyche
[Vol. 90
the adjacent island of Cuba (and Hispaniola in antilleorum and
blackwelderi).
Two of the other 16 species listed are based on records that can-
not be verified: Acanthinus ebeninus on an old specimen with a
“Cuba” label, and Amblyderus sp. on some specimens from Puerto
Rico that cannot now be located. Ten are shared with continental
areas of the New World: Acanthinus angusticollis, concinnus, quin-
quemaculatus and scitulus, Anthicus pallidus, Sapintus similis and
teapensis, Thicanus texanus, and Vacusus holoxanthus and vicinus.
These may have reached the Greater Antilles without human help,
but Vacusus holoxanthus is found mainly from Chile to Bolivia, and
Acanthinus scitulus seems not to have been present in the lowland
localities that were extensively collected in the 1930’s, so is probably
of recent introduction. Finally, 4 species of Anthicus are of Old
World origin: floralis and formicarius, which are almost cosmopoli-
tan; tobias, which is expanding its range in several parts of the
world; and crinitus.
Two large genera, the world-wide Tomoderus and the New World
Ischyropalpus, are conspicuous by their absence. The latter genus,
at least, should have been collected if it was present; mainland spe-
cies are often abundant on blossoms. That the fauna has not been
completely sampled is indicated by the addition of a species of
Mecynotarsus from Hispaniola through the recent collecting of J.
and S. Klapperich.
Acknowledgements
A critical part of the material reported here was collected by
Philip J. Darlington, Jr. in the 1930’s. I wish to thank Dr. Darling-
ton and subsequent curators of the M. C. Z. for permission to retain
these specimens until they could be studied in comparison with
continental faunas. The second large lot came from the collecting of
Richard E. Blackwelder and Edward A. Chapin, and was made
available by the U. S. National Museum; it was collected in the same
period. The largest recently collected lot originated in the collecting
of J. and S. Klapperich in the Dominican Republic, and was made
available by Dr. M. Branucci of the Naturhistorisches Museum in
Basel.
1983]
Werner — Anthicidae
213
Key to Greater Antilles Anthicidae
1. Prothorax with an anterior horn that extends over the
head 25
— Prothorax without a horn 2
2. Sides of mesosternum curved outward to form a broad plate,
with a variably developed fringe of setae along its
edges 15
— Mesosternum with sides diagonal and nearly or quite
straight, without fringe setae 3
3. First visible abdominal sternum with a transverse, pubescence-
lined invagination behind each hind coxa. Elytral pubescence
double, the under layer more appressed, diagonal 24
— First visible abdominal sternum without invaginations. Elytral
pubescence usually single, double in Anthicus pallidus 4
4. Elytral pubescence double, undercoat more diagonal. Pale,
somewhat flattened, elytra with dark brown midband and
suture, markings usually isolating a pale zone in basal and
apical fourth of each elytron; head truncate. 2.25-2.35 mm.
Hispaniola, Puerto Rico Anthicus pallidus Say
— Elytral pubescence simple 5
5. Vertex of head somewhat produced, edge nearly straight
from middle to weak temporal angles (Fig. 11). Uniform
pale brown, somewhat shiny, elytra sometimes with a weak
median cloud. Ca. 2.6 mm. Hispaniola, Puerto Rico
Thicanus texanus (LaFerte)
— Base of head from truncate to evenly rounded 6
6. Base of head truncate, temporal angles narrowly rounded
7
— Base of head rounded, temporal angles broadly rounded or
not evident 10
7. Head microreticulate between punctures. Rufescent to
brown, elytra usually brown except across base. Elytral
pubescence very short and inconspicuous. 2. 9-3. 2 mm.
Jamaica Anthicus formicarius (Goeze)
— Head smooth and shiny between punctures 8
8. Elytral setae sparse and as long as width of a femur, subde-
cumbent. Rufescent, shiny, elytra with dark markings that
214
Psyche
[Vol. 90
usually isolate a common pale spot in apical third, 2, 5-3. 2
mm. Puerto Rico, Virgin Islands
Anthicus crinitus LaFerte
— Elytral pubescence shorter than width of a femur; dark
elytral markings not enclosing a common pale spot in apical
area 9
9. Elytral pubescence short and even, the erect tactile setae
extending well above the decumbent setae. Prosternum with
uniformly distributed punctures and pubescence in front of
coxae. Elytra pale at base and usually in an obliquely oval
spot in apical third of each. 2.0-2. 3 mm. Jamaica, Cuba,
Hispaniola, Virgin Islands and Bahama Islands; elytra
usually lacking posterior pale spots in Jamaican popula-
tion Anthicus antilleorum, sp. n.
— Elytral pubescence longer and less decumbent, the tactile
setae barely evident among the setae. Prosternum with front
half of portion in front of coxae smooth, back half bearing
some coarse punctures and setae. Uniformly dark (Jamaica)
or elytra pale across base and at apex, the posterior marking
rounded in front (Cuba, probably Hispaniola). 2.27-2.55
mm Anthicus blackwelderi, sp. n.
10(6) Rufescent or paler, with pale appendages; elytra usually
with suture and whole apical half black except for a round,
very pale spot on each in apical third. 2. 6-3.0 mm. $ tegmen
with apex knob-like, lacking lateral tufts of setae. Jamaica,
Virgin Islands Anthicus tobias Marseul
— Elytra usually with a complete or interrupted dark midband
and an oblique pale subapical band, never with a round pale
spot on each in apical third. $ tegmen pointed, with a tuft of
setae on each side. Anthicus subtilis-group 11
11. $ front tibiae excavated in apical 2/5. Elytral midband often
complete. 2.11-2.24 mm. Cuba
Anthicus macgillavryi Buck
— S front tibiae simple 12
12. $ tegmen gradually tapered to apex, slender. Elytral mark-
ings dark, all connected along suture, including a dark zone
across base. Ca. 2.5 mm. Hispaniola
Anthicus hispaniolae, sp. n.
— $ tegmen not evenly tapered to apex 13
1983]
Werner — Anthicidae
215
13. $ tegmen very bluntly truncate at apex except for a small
median point. Elytral midband complete in Cuban specimens
seen, interrupted at suture in Jamaican specimens. 2.22-2.53
mm. Cuba, Jamaica Anthicus soledad, sp. n.
— (5 tegmen with sides slightly constricted beyond middle ....
14
14. Antennae unusually long and slender. Elytral midband
reduced to a pale brown triangle with point toward suture,
on each side. 2.47-2.76 mm. Hispaniola
Anthicus subtilis LaFerte
— Antennae not so slender. Elytral midband interrupted at
suture, but mark truncate toward suture on each side. Elytra
slightly inflated. 2.02-2.42 mm. Hispaniola
Anthicus darlingtoni, sp. n.
15(2) Pronotum with a pair of small bumps near anterior edge of
disc. Fringe setae of mesosternum closely appressed to
mesepisterna. Rufous, elytra black or brown with basal
fourth rufous in a well-demarcated zone; appearing glabrous
and subopaque. 2. 6-3. 2 mm. Jamaica, Puerto Rico, Virgin
Islands Anthicus floralis (L.)
— Pronotum without such bumps. Fringe setae of meso-
sternum at least slightly raised from surface of mesepi-
sterna 16
16. Sides of prothorax not constricted, almost evenly tapered
from widest part, near front, to basal impressed line 17
— Sides of prothorax at least slightly constricted anterior to
basal impressed line 18
17. Shiny, only erect tactile setae very obvious; luteous to
rufous, elytra with apex and an interrupted submedian band
dark. 2. 3-2. 8 mm. Jamaica, Cuba, Hispaniola, Puerto Rico,
Virgin Islands Vacusus vicinus (LaFerte)
— Shiny but with surface partly obscured by appressed pubes-
cence; tactile setae short and inconspicuous. Moderately
slender, entirely tannish. 1. 8-2.0 mm. Jamaica
Vacusus holoxanthus (Fairmaire & Germain)
18(16) Pubescence fine, silky, moderately dense, appressed, cover-
ing all of elytra. Dull rufescent to brown, elytra with dark
midband and apex, markings usually connected along su-
ture ,19
216
Psyche
[Vol. 90
— Pubescence very sparse, or dense pubescence confined to
postbasal transverse impression of elytra 20
19. $ fifth visible abdominal sternum excavated on disc; lobes
of visible sternum 6 moderately broad. 2.6-3. 2 mm. Cuba,
Puerto Rico Acanthinus quinquemaculatus (LaFerte)
— $ fifth visible abdominal sternum simple; lobes of visible ster-
num 6 narrow. 2. 4-3.0 mm. Hispaniola
Acanthinus concinnus (LaFerte)
20(18) Elytra with a dense patch of white pubescence in postbasal
transverse impression. Dark brown, shiny, otherwise gla-
brous with erect tactile setae; head triangular, it and prono-
tum longitudinally strigose. Ca. 2.8 mm. Cuba?
Acanthinus ebeninus (LaFerte)
— Elytra without patch of dense pubescence in postbasal
transverse impression 21
21. Head and prothorax strongly sculptured 22
— Whole dorsal surface smooth, shiny, punctures fine and
indistinct, setae very short, sparse and inconspicuous, only
erect tactile setae evident 23
22. Dark brown with quadrate yellowish white mark laterally in
cuticle of postbasal transverse of elytra; head unusually
large, it and prothorax with some longitudinal striations.
2.0-2. 8 mm. Jamaica, Cuba
Acanthinus angusticollis (LaFerte)
— Head and prothorax rufescent, elytra rufescent at base, with
a complete luteous band in postbasal impression, brown
behind. Head and prothorax rugose-punctate. 2.4-2.8 mm.
Cuba Acanthinus schwarzi Werner
23(21) Rufescent, elytra paler with brownish to almost black mark-
ings, at least in narrow, interrupted bands at basal and apical
thirds, to dark with postbasal impression and a narrow post-
median band pale. Prothorax with a strong constriction that
continues weakly across dorsum. Edge of mesosternal shelf
visible from above, in front of elytral humeri. Head nar-
rower than semicircular behind eyes. 2.0-2.9 mm. Cuba . . .
Acanthinus scitulus (LeConte)
— Pale rufescent, elytra pale rufescent at base, dark brown on
humeri and behind postbasal transverse impression. Pro-
1983]
Werner — Anthicidae
217
thorax weakly constricted, almost evenly globular from
basal impressed line to collar. Only fringe setae of mesoster-
num visible from above. Ca. 2.0 mm. Hispaniola
Anthicus russoi Krekich
24(3) $ fifth visible abdominal sternum shallowly dished out on
disc, the excavation flanked with some erect setae. Ca. 2.0
mm. Jamaica, Cuba, Hispaniola, Puerto Rico, Virgin
Islands Sapintus teapensis (Champion)
— (5 visible sternum 5 simple. Ca. 2.7 mm. Jamaica
Sapintus similis Werner
25(1) Each side of prothoracic horn with 3 teeth, the apex about
equal to a tooth. Visible abdominal sternum 1 without a
pubescence-lined invagination behind each hind coxa.
Length ca. 2 mm. Mecynotarsus elegans-group 26
— Each side of prothoracic horn with 3-7 teeth, the apex con-
siderably broader than any tooth. Visible abdominal sternum
1 with a pubescence-lined invagination behind each coxa.
Prothoracic horn with a few ventrolateral pits. Notoxus monodon-
group 27
26. Elytra slightly inflated. Pubescence cinereous but with some
intermixed ferrugineous scales on disc of pronotum and
basal half of elytra; elytra piceous at base, in a postmedian
band, and in a large posterior triangular marking on each.
Jamaica Mecynotarsus jamaicanus Werner
— Elytra strongly inflated. Pubescence cinereous, with diffuse
slightly darker to pale rufescent markings on disc of pronotum,
dull brown on elytra from base along a broad zone to an
apical pale cordate mark, and in a feeble slightly postmedian
band and subapical band delimiting the cordate mark.
Markings very weak in some individuals. Hispaniola
Mecynotarsus hispaniolae, sp. n.
27(25) Dark elytral markings usually including some on sides that
curve inward toward suture at apex. Tip of $ aedeagus trun-
cate. Jamaica Notoxus jamaicus Pic
— Dark elytral markings not including any on sides behind an
irregular transverse midband. Tip of $ aedeagus deeply split.
Puerto Rico Notoxus bipunctatus Chevrolat
218
Psyche
Description of Species
[Vol. 90
Measurements are given in 0.01 mm as head: length from vertex
to clypeofrontal suture over width across eyes and behind; pro-
thorax: length including collar over width at collar, maximum, at
constriction, and across base; elytra: length over width at humeri
where 45° angle would touch them, and maximum. Total length as
given is the sum of head, prothorax and elytra.
Anthicus antilleorum, sp. n.
Fig. 9, 19.
2.01-2.24 mm, rufescent, the legs luteous, elytra with a brown
midband, the base and an oblique apical mark on each luteous.
Head quadrate, antennae moderately thick toward apex; elytra
somewhat inflated, even in fully winged individuals. Pubescence
moderately short, almost appressed, the tactile setae evident above
the setae.
Holotype 2.11 mm. Head 39/46,42; eyes 16/13, 32 apart, 16
from base, which is straight, the temporal angles narrowly rounded;
sides almost straight behind eyes. Disc slightly flattened, with mod-
erately large, deep punctures ca. 3 apart, except on midline of front;
pubescence almost appressed, moderately conspicuous. Antennae
ca. 77 long, 7 thick at segment 10, segments 7-1 1 forming a feeble
club that is thicker than segment 1. Prothorax 47/24,44,32, its sides
just perceptibly concave at usual level of constriction; anterolateral
portion narrowly curved. Punctures strong, denser than on head, ca.
2 apart. Elytra perceptibly swollen, 125/53,73; punctures strong, ca.
3 apart; setae 5 long, slightly curved, almost appressed; tactile setae
5, suberect, slightly curved. Legs unmodified. Apical margin of vis-
ible abdominal sternum 5 just perceptibly convex.
Holotype: cuba: Baragua (IV-25-28, at light, C. F. Stahl) in
MCZ. Paratypes: Cuba: Baragua (same data, 6; 11-10-26, L. C.
Scaramuzza), Soledad nr. Cienfuegos (Apr. 1936, P. J. D.; V,VI-’39,
C. T. Parsons), Cayamas (Mar.-May, E. A. Schwarz), Camagiiey
Prov.: Monte Imias nr. California (at light, June 7, 1959, M. W.
Sanderson). Hispaniola: rep. dom.: Bani (65m), Boca Chica (10m),
and Ocoa (475m), all J. & S. Klapperich, 1971-73. Paratypes in
MCZ, USNM, Basel Museum and collection of FGW. Not desig-
nated as paratypes: Jamaica: Morant Bay (Chapin and Blackwelder,
4). VIRGIN islands: Tortola (BVI, Brandywine Bay, J. F. G. Clarke,
1983]
Werner — Anthicidae
219
1). BAHAMA islands: Cat Island (Bennets Harbour, E. B. Hayden &
L. Giovanolli, 2).
The specimens from Jamaica are darker than those from Cuba
and Hispaniola, only one of them having the posterior pale elytral
mark. The pubescence may be less appressed but the specimens are
so abraded that they were identified with difficulty. This species is
probably most closely related to A. pauxillus Champion, panamen-
sis Werner, and margaritae, sp. n., from Guatemala, Panama, and
eastern Venezuela, respectively. The color pattern is similar. The $
genitalia are similar, but the simple internal sac provides few clues
to relationship. A. panamensis has the tegmen step-tapered.
Anthicus margaritae, sp. n.
Fig. 8, 20.
1.90-2.20 mm, of form of Anthicus panamensis Werner and antil-
leorum, sp. n., differing from both species in having the head
broader behind the eyes and gradually widened to the narrowly
rounded temporal angles. Tegmen of $ genitalia convexly tapered
as in antilleorum, but with the apex slightly more pointed. Antero-
lateral angles of prothorax quite narrowly rounded, as in the above
2 species, differing mainly in this feature from A. exiguus Champion.
Holotype S, 1.92 mm; head 33/44,40; eyes 16/14, separated by
29,15 from base. Punctures of head ca. 3 apart, on slightly convex
disc. Antennae 75 long, 7 thick at segment 10. Prothorax
46/20,42,31. Elytra slightly swollen but with very distinct humeri,
1 13/51,65. Punctures slightly sparser than on head, intervals smooth
and slightly convex; setae ca. 9 long, not quite so decumbent as in
antilleorum, tactile setae 9 and erect.
Holotype: Venezuela: I. Margarita: Puerto Fermin (12.48,
Marcuzzi), in CASC, San Francisco. Paratypes: Venezuela: I Mar-
garita: Puerto Fermin (same data, 10), Juan Griego (3.48, 2). Sucre:
Carupan (9.48, 2). I am indebted to K. S. Hagen for the loan of
these specimens, and for some additional specimens without labels.
Paratypes in CASC and collections of KSH and FGW.
Anthicus blackwelderi, sp. n.
Fig. 10, 13,21.
2.27-2.55 mm, of aspect of a Vacusus species, head truncate,
prothorax without a constriction, and elytra subparallel. Jamaican
220
Psyche
[Vol. 90
individuals brown, with slightly paler legs, antennae and palpi.
Cuban individuals with pale marking at base and apex of elytra.
Moderately coarsely punctured.
Holotype 2.34 mm. Head 44/51,47; eyes 19/ 15, 35 apart, 20
from base, which is truncate with a slight impression at midline, the
temporal angles narrowly rounded; disc slightly flattened, smooth,
with strong punctures ca. 4 apart except on midline of front; setae
decumbent. Antennae ca. 97 long, 7 thick at segment 10, which is
slightly longer than thick. Prothorax 49/ 16,40,31, with even punc-
tures denser than on head, ca. 2 apart, about as wide as intervals.
Elytra 141/56,73, with feeble omoplates, as deeply punctured as
head and prothorax, punctures ca. 3 apart; setae decumbent, 8,
slightly curved; tactile setae 4, suberect. Underside of thorax with
punctures slightly smaller than above; front part of prosternum,
anterior to coxae, smooth in front half, with some punctures and
suberect setae in back half. First visible abdominal sternum finely
punctured, rest punctulate. Visible sternum 5 with its apex gently
convex, as in $; 6 with no indication of even an emargination; last
visible tergum shiny, its edge beaded, almost concealed by the ter-
gum before it, which is densely short-pubescent and has an almost
evenly rounded apex, as in 9-
Holotype: Jamaica: Kingston (no date, Chapin & Black-
welder) in USNM. Paratypes: Jamaica: Kingston (C & B, 4; P J D,
1), Morant Bay, Gordon Town, Trinityville, Bath St. Thomas, Blue
Mts. (nr. 4500', P.J.D.). Paratypes in USNM, MCZ and FGW col-
lection. Not designated paratypes: Cuba: Oriente Prov.: coast below
Pico Turquino (1); Soledad nr. Cienfuegos (2). Hispaniola: Rep.
Dom.: Constanza (19)- The Constanza specimen has very reduced
dark elytral markings, with rounded posterior emargination. Even
teneral Jamaican specimens have uniformly colored elytra.
Anthicus russoi Krekich
Fig. 5, 22.
Anthicus russoi Krekich in Menozzi 1930: 93 (type-locality: Moca, Rep. Domini-
cana).
Stricticomus russoi: Bonadona 1981: 275.
(5, Jarabacoa, 2.20 mm, very smooth, shiny, appearing somewhat
glabrous except for long, erect tactile setae; body and basal 36 of
elytra pale rufescent (abdomen brown in another specimen); humeri
1983]
Werner — Anthicidae
221
and apical area of elytra brown. Head semicircular behind eyes;
prothorax evenly swollen in profile at level of widest portion.
Head 42/45,39; eyes prominent, 17/13, 27 apart, 20 from base.
Disc evenly convex, punctures ca. 4 apart, small but distinct on
front, very fine behind; setae ca. 1, decumbent, almost invisible,
tactile setae erect, 7, fine. Antennae 105 long, 7 thick at segment 10,
gradually thickened, with moderately conspicuous suberect curved
setae ca. 4 and erect, nearly straight tactile setae ca. 7. Prothorax
47/18,36,25,28; portion anterior to strong basal impressed line
almost globular, rising 1 1 above line from top of base to top of
strong collar. Elytra 131/50,67; humeri well defined, omoplates
slightly swollen; postbasal transverse impression well indicated but
with punctures and pubescence like rest of elytra; punctures very
fine, ca. 5 apart, setae decumbent, fine, ca. 1, barely visible, tactile
setae erect, nearly straight, 11. Mesosternum extremely smooth,
flat, with lateral expansion 15 wide and reaching almost to epi-
pleura of elytra, bearing a fringe of slightly curved setae ca. 1 1 long,
partly visible from above, the lateral and posterolateral setae lap-
ping onto sides of elytra and mesepisterna. Metasternum, abdomen
and legs with sparse, decumbent setae ca. 4 long, slightly denser on
tibiae. Visible sternum 5 with disc evenly convex, its apex shallowly
emarginate and bearing several long setae; 6 ca. 11 wide, divided
into almost parallel, deeply separated lobes, which are deeply
grooved mesally. Last visible tergum thin, nearly flat. Wings ap-
parently absent. Cuticle very translucent, some parts almost
transparent.
Records: Hispaniola: rep. dom.: Jarabacoa (530m, 23.1.1972),
and Boca Chica (10m, 6. X. 1971), both on single $ specimens, col-
lected by J. & S. Klapperich, and in the Basel Museum. These
specimens agree in general with the original description, which may
have suffered from being translated from German into Italian, and
finally from my translation to English. The original figure is not
helpful. Professor M. Princippi informs me that there is a specimen
of russoi in the Menozzi Collection at the Istituto di Entomologia of
the Universita di Bologna. This must be the holotype, since the
species was described from a single specimen.
I am leaving russoi in Anthicus for lack of a better place to put it.
The mandibles and gonopore armature are different from Acanthi-
nus, and the mesothorax differently designed from Formicilla. In
222
Psyche
[Vol. 90
that genus the setae on the sides of the mesothorax arise from a
ridge above the side of the expanded mesosternum. Bonadona has
placed it in Stricticomus, an Old World group characterized by the
shape of the prothorax. While this is a convenient way to split up
the numerous species of Anthicus, the division has not been defined
on a phylogenetic basis.
Anthicus subtilis-group
Five species of Anthicus in the Greater Antilles form a very dis-
tinctive group. The males have a unique tuft of long setae on the
sides of the tegmen and the species share enough external features
that two of them are indistinguishable in the female sex. Of the five,
two have been taken only on Hispaniola, one only on Cuba, one on
Hispaniola and Cuba, and one on Cuba and Jamaica, the last with
some geographical variation on the two islands. All three species on
Hispaniola are at least partly sympatric, as indicated by the labels,
as are two on Cuba.
Anthicus subtilis LaFerte
Fig. 1, 18.
Anthicus subtilis LaFerte 1848: 135-6 (type-locality: LaFerte states it as Colombia,
collected by Moritz, but the specimens probably originated in the Greater
Antilles).
2.47-2.76 mm, pale rufescent, legs, antennae and palpi dull lute-
ous, tibiae obscurely darker at base, elytra with pale brown median
marking widely interrupted at suture and more or less triangular
with a mesal point, and a narrow, usually paler, diagonal subapical
band. Pubescence short, fine, almost appressed, dulling the gener-
ally shiny surface; punctures fine and not very evident except on base
of pronotum. On the elytra the pubescence in this and the other
species of the subtilis-group is slightly diagonal over most of the
surface, to ca. 45° in the postbasal transverse impression and nearly
transverse on rear of the weak omoplates.
(5, Ennery, Haiti, 2.66 mm. Head 44/53,47, almost semicircular
behind prominent eyes, with a slight impression at midline. Eyes
22/16, 35 apart, 16 from base. Disc evenly convex, shiny, with
small, well-defined punctures ca. 5 apart, and more numerous very
fine punctures on intervals, punctures collectively ca. 1 apart. Setae
fine, silky, decumbent. Antennae unusually slender, segments 16/7,
1983]
Werner — Anthicidae
223
9/5, 11/5, 15/5, 16/5, 16/5, 15/5, 14/6, 14/7, 13/7, 17/6, base to
apex. Prothorax 54/20,44,33,35, with well-defined collar and slight
constriction. Collar without dense pubescence ventrally. Disc evenly
convex, punctures ca. 1 apart, finer and with intervals nearly flat on
anterior 1 / 3, larger and grading to finely rugulose in region of basal
impressed line. Elytra 169/64,86, widest near middle, tapering to
moderately narrow apex; omoplates distinct, transverse impression
weak. Surface almost evenly covered with fine, slightly elevated
punctures ca. 2 apart, intervals flat; setae fine, appressed, ca. 3,
tactile setae suberect, 7. Setae of 2 slightly different lengths and
thicknesses, the longer and thicker slightly less appressed and dis-
cernible with backlighting. Punctures and setae of impression no
different from those of adjacent areas except for the setae being
more perpendicular to the midline. Legs slender, not modified. Vis-
ible sternum 5 simple, its apex truncate.
The median dark elytral markings on this individual are 37 long,
separated by 30 across suture, and 9 from side margin; subapical
band ca. 1 1 wide, paler than median marks, slightly oblique, extend-
ing forward along suture for ca. 18, pale and evanescent laterally, to
7 from margin. All of the specimens have rather similar markings,
and none has the median markings connected across the suture.
Records: All individuals are fully winged and apparently capable
of flight. HISPANIOLA: HAITI: Ennery (nr. 1000' (4(J, 5?), Camp Per-
rin (nr. 1000', 2(5), N.E. foothills of La Hotte (3000', 1(5). rep. dom.:
Villa Altagracia (1(5), Pto. Plata (25 km. S. of, 2?), San Jose de las
Matas (1-2000', 1$). cuba: Loma (Pico) del Gato (Sierra Maestra,
Oriente Prov., 2(5), Soledad nr. Cienfuegos (1$). Almost all col-
lected by P. J. D.
I am applying LaFerte’s name to this species largely on the basis
that his description matches it quite well and that he particularly
noted unusually slender antennae. He had two specimens to study,
one in the Dejean collection and one in his own, the source of both
being a series in the museum at Berlin, and ultimately the collecting
of Moritz. I have seen the specimen in the LaFerte collection and
compared it with West Indian material, but did so before I realized
that there are several species in the subtilis-group. I have never seen
a specimen of this group from a continental area. According to W.
Horn’s Entomologische Sammlungen, C. Moritz collected in both
Colombia and Puerto Rico in the 1830’s. It is likely that some labels
got mixed.
224
Psyche
[Vol. 90
Anthicus darlingtoni, sp. n.
Fig. 3, 17.
Generally similar to subtilis but smaller, 2.02-2.42 mm, head
slightly truncate and with more distinct punctures, antennae not so
slender, elytra more rounded at apex, median elytral markings usu-
ally darker and barely narrowed mesally, and subapical band very
faint. Some individuals, including the holotype, lack wings and have
the elytra slightly inflated.
Holotype <5, 2.06 mm. Head 36/46,41, subtruncate with broadly
rounded temporal angles. Eyes 17/13, 31 apart, 15 from base. Disc
evenly convex, shiny, with evenly distributed punctures ca. 1 apart,
small but well defined; diameter of punctures, including down-
curved borders, about equal to intervals. The larger punctures de-
scribed in subtilis are barely larger than those on the intervals.
Antennae not unusually slender, segments 13/6, 7/5, 9/5, 10/5,
11/5, 11/5, 11/5, 11/5, 11/6, 10/7, 16/7, base to apex. Prothorax
similar to subtilis, 44/16,41,27,31. Elytra 126/49,69, similar but
apex more rounded and impression weaker. Fully winged individu-
als are more similar. Surface slightly more deeply punctured, punc-
tures ca. 2 apart; setae similar, ca. 4, tactile setae ca. 7. Legs not
modified. Apex of visible sternum 5 very feebly excavated. Median
elytral markings 29 long, separated by ca. 18 across suture, 4 from
side margin; subapical band much paler, barely a cloud, ca. 1 1 wide.
Holotype, Haiti; Etang Lachaux (under 1000', Oct. 26-21, ’34,
P. J. Darlington, WL) in MCZ. Paratypes: Haiti: Etang Lachaux
(same data, 2 WL 1 WL 9), Camp Perrin (nr. 1000', 1 F 5, 1 WL
5, 2 F 9), Damien (2 F 9), Port-au-Prince (1 WL $), Miragoane (2
WL 9), Ennery (nr. 1000', 2 WL 9), Mt. La Hotte (Tardieu, 3000', 1
WL 9), Kenskoff (nr. Port-au-Prince, 4-6000', 1 F 9). All speci-
mens were collected by P. J. Darlington between September and
November, 1934.
In at least two localities this species is sympatric with subtilis, but
it appears to have a narrower range. Six of the specimens have full
wings (F) and 1 1 are entirely wingless (WL).
Anthicus hispaniolae, sp. n.
Fig. 2, 16.
Larger than subtilis and the other species of the group, 2.68-3.1 1
mm, and with more extensive and darker markings on the elytra.
1983]
Werner — Anthicidae
225
these tending to be connected along the suture but not along the
sides. Most of the head and prothorax brown, elytra with base to
transverse impression, a midband and an oblique subapical band
brown, these connected at least narrowly along suture; subapical
band paler in part of the series. Rest of elytra, legs, palpi, antennal
segments 1 & 2, and usually labrum, mandibles except for tips, and
head adjacent to antennal insertions luteous. Dark midband and
subapical band not reaching side margins. Underside and abdomen
pale brown. Head and prothorax densely, finely punctured. Tegmen
of $ genitalia very slender and tapering almost evenly to narrow tip.
Holotype 2.68 mm. Head 42/56,49. Eyes 25/18, 36 apart, 15
from base, which is subtruncate with a shallow median impression,
the temporal angles broadly rounded. Disc evenly convex, shiny,
but punctures ca. 2 apart and broader than intervals. Antennal
segments 1-2 pale, 1 heavier than usual; segments 18/11,9/5/ 12/5,
13/5, 15/5, 17/6, 15/6, 14/6, 13/7, 13/7, 18/7, base to apex. Pro-
thorax 52/22,47,37,42; punctures very dense, 2 apart, intervals very
narrow, especially in back half. Elytra 175/69,95, with distinct
omoplates and postbasal impression; punctures small, ca. 2 apart,
intervals flat and about as wide as punctures; setae moderately
dense, decumbent, 4, part slightly less decumbent, 5; tactile setae ca.
6. Legs simple; visible sternum 5 truncate.
Holotype, <5, rep. dom.: Constanza to Jarabacoa (2-4000', Aug.,
38, P. J. Darlington) in MCZ. Paratypes: rep. dom.: same data
(2(5), foothills of Cordillera Central (S. of Santiago, 1^). Haiti:
N.E. foothills of La Hotte (2-4000', 1(5). The last locality is almost
the same as where one subtilis was collected. All collected by P. J.
Darlington in Oct., 1934, and June and Aug. 1938. All specimens
are fully winged, and apparently capable of flight.
Anthicus soledad, sp. n.
Fig. 4, 14.
Generally similar to subtilis but smaller, 2.22-2.53 mm, elytral
markings darker, median elytral markings nearly or quite a com-
plete band in Cuban individuals, interrupted at suture in those from
Jamaica. Head slightly more truncate and deeply punctured, anten-
nae not unusually slender. Unique in having the apex of the $
tegmen nearly truncate, with a median point. Cuban specimens are
so similar to macgillavryi Buck that females cannot be identified.
226
Psyche
[Vol. 90
Holotype 2.33 mm. Head 40/49,44, subtruncate with broadly
rounded temporal angles, slightly impressed at middle. Eyes 19/14,
33 apart, 16 from base; surface similar to subtilis but with fine but
distinct punctures ca. 2 apart, most slightly narrower than intervals,
with gradually downcurved borders. Antennal segments 13/6, 7/5,
9/5, 11/5, 12/5, 13/5, 14/5, 13/5, 13/6, 11/7, 16/6, base to apex.
Prothorax similar, 47/18,39,27,33. Elytra 145/55,79, shiny, punc-
tures distinct, ca. 3 apart and almost as wide as intervals; setae ca. 4
long, tactile setae 6. The midband on this and other Cuban speci-
mens is complete, slightly paler at suture; subapical band broad and
dark, connected to midband at sides and narrowly at suture, leaving
a diagonal mark on each elytron and apex pale; base onto omo-
plates somewhat darkened. Jamaican individuals lack the basal
darkening, have the midband interrupted at the suture, and the
subapical band connected to it only at the sides. Legs unmodified.
Apex of visible sternum 5 feebly excavated.
Holotype, Cuba: Soledad nr. Cienfuegos (Oct. 21, ’26, P. J.
Darlington, F) in MCZ. Paratypes: Cuba: Soledad (2 F Caya-
mas (5 R 5). Jamaica: Rio Cobre (5 mi. above Spanishtown, 1 F
1 R 5, 1 WL 5), Ocho Rios (1 WL 5), Blue Mts. (Whitefield Hall,
nr. 4500', 1 R (J), Milk River (1 F ^). Five of the males are fully
winged (F), 7 have reduced wings (R), and 2 are wingless (WL). In
addition 5 fully winged females from Jamaica are identified with
this species but not included as paratypes: Whitefield Hall (2), Milk
River (2), and Mandeville (1, dead in light globe). Paratypes in
MCZ, USNM and collection of FGW.
Anthicus macgillavryi Buck
Fig. 12, 14.
Anthicus macgillavry Buck 1960: 69-70 (type-locality: Manicaragua, cuba, but holo-
type is a 9 and not conclusively identifiable as the species redescribed here).
2. 1 1-2.24 mm, extremely similar to sympatric soledad individuals
on Cuba, $ differing in having the front tibiae excavated in apical
2/5 and in having the tegmen of the genitalia slightly constricted
beyond middle, similar to subtilis and darlingtoni. Elytra with dark
midband complete in all specimens identified.
5, Soledad, 2.20 mm. Head 36/48,41; eyes 18/15, 31 apart, 13
from base; antennal segments 13/7, 8/5, 9/4, 11/5, 13/5, 13/5, 13/5,
1983]
Werner — Anthicidae
221
12/5, 12/5, 11/7, 17/5, base to apex. Prothorax 47/ 17,40,27,31; ely-
tra 138/56,75; setae ca. 4, tactile setae 5. Front tibiae gradually
thickened from base to 6 thick at 16 from base, zone beyond thickest
portion moderately abruptly thinned to slightly more than 4 in a
gently concave, flattened zone ca. 6 wide, this lined with moderately
dense, pale, decumbent setae. Front tarsi not modified. Apex of
visible sternum 5 feebly excavated.
Records: Cuba: Soledad, nr. Cienfuegos (5 F 5, 5 WL 5), Bara-
gua (at light, 1 F <5), Cayamas (2 F 5, 6 R 5), Limones ( 1 WL 3)- Of
the 20 specimens identified, 8 have full wings, 6 reduced wings, and
6 are wingless.
This species is more abundant than soledad on Cuba, so is the
more likely one to be associated with Buck’s name. The holotype
and all 15 paratypes sent from the Amsterdam collection are
females, so no part of the type series can be included in the records.
Mecynotarsus hispaniolae, sp. n.
Fig. 6, 7.
1.56-2.04 mm (elytra plus prothorax including horn). Brown,
appendages rufescent, surface largely concealed by appressed scales,
which are cinereous but with a median rufescent cloud on the pro-
notum and dull brown markings on the elytra. The darkest of the
elytral markings are lateral, one rounded and close to middle, the
other larger, oval and subapical, both isolated from sides by a broad
cinereous zone. Paler brown markings extend from the omoplate
area to the level of the front of the subapical mark, with vague
connections to both sets of dark marks. The background color of
the elytra is slightly rufescent dorsally. Prothorax with a sparse
fringe of long, erect, flattened, slightly clavate setae, on sides and
onto base. Elytral scales of 2 different widths, the wider ca. 1 Vi times
as wide, the 2 widths tending to be in alternate rows and the wider
just perceptibly elevated.
Holotype: 2.04 mm; head 39/45,45; eyes small, 12/9, their curved
scales ca. 1.5; 29 apart, 12 from base of head. Upperside of head flat,
with sparse setae and some 12 long, suberect setae and well-
developed erect, flattened setae on horn outline, 10-14 long. Pro-
thorax 39 long, 82 with horn, 63 wide; horn 31 wide at widest, 12
thick. Marginal setae 9 long, the ones on base slightly shorter. Horn
Figures 1-13. Fig. 1. Anthicus subtilis, described specimen. Fig. 2. A. his-
paniolae, holotype. Fig. 3. A. darlingtoni. holotype. Fig. 4. A. soledad, holotype.
Fig. 5. A. russoi, described specimen. Fig. 6. Mecynotarsus hispaniolae, holotype.
Fig. 7. Same specimen, oblique lateral view of elytra. Fig. 8. Anthicus margaritae,
holotype. Fig. 9. A. antilleorum, holotype. Fig. \0. A. blackwelderi, holotype.
Fig. 11. Thicanus texanus, Barahona, Rep. Dom., forebody. Fig. 12. Anthicus
macgillavryi, front leg of described S- Fig- 13. A. blackwelderi, elytral markings of
Cuban population, from coast below Pico Turquino.
1983]
Werner — Anthicidae
229
Figures 14-22. $ genitalia of Amhicus spp., in ventral view, most with tegmen in
left lateral view, details of internal sac and gonopore armature to sides. Fig. 14. /I.
soledad, paratype, Soledad, Cuba. Fig. 15. A. macgillavryi, Soledad, Cuba.
Fig. 16. A. hispaniolae, paratype. Fig. \1. A. darlingtoni, paratype, Damien, Haiti.
Fig. 18. A. subtilis. Villa Altagracia, Rep. Dom. Fig. 19. A. antiUeorum, paratype,
Cayamas, Cuba. Fig. 20. A. margaritae, paratype, Carupano, Venez. Fig. 21.
A. blackwelderi, paratype, Kingston, Jam. Fig. 22. A. russoi, Boca Chica, Rep.
Dom.
230
Psyche
[Vol. 90
with a well-developed crest of 2 ridges, these up to 8 apart, and with 3
strong teeth on each side. Underside of horn with a sparse brush of
suberect, anteriorly directed simple setae 10 long. Elytra 122/65,87,
strongly inflated, punctures ca. 3 apart but obscured by dense scales
ca. 4 long; no tactile setae discernible. Hind tibia 47 long, tarsus 61,
front tarsus 29.
Holotype, $, rep. dom.: Las Salinas b. Bani ( 10. X. 1979, J. & S.
Klapperich) in Natural History Museum, Basel, Switzerland. Para-
types; 2 9, same data, Basel and FGW collection.
Relationships: The species of Mecynotarsus in the elegans-group
seem assignable to at least 3 subgroups. The first, already noted
(Werner 1962), has the sutural area of the elytral apex pale, this
zone restricted anteriorly by oblique dark bands. To this group
belong elegans LeConte, intermixtus jamaicanus Werner,
and probably falcatus Chandler. In this group the male genitalia are
distinctive, the phallobase bearing rounded lateral lobes. The male
antennae are not expanded and the prothoracic horn is relatively
narrow.
A second subgroup has the markings at the tip of the elytra based
on a pale sutural mark and lateral spots, with a narrow extension
from the oblique subapical bands tending to reach the very apex on
each side, where there may be a tiny development of a pit in the
male. This subgroup contains balsasensis Werner and salvadoren-
sis Werner. These 2 species have a distinctive pale strip through
discal clouding on the pronotum. Werner (1962) indicates that the
phallobase is simple but Chandler (1977) states that there are lateral
lobes in salvadorensis. Very small size of the genitalia makes inter-
pretation difficult. The antennae are simple in the male and the horn
is relatively narrow.
Finally, a third subgroup has each elytron pale at the apex, with a
convex anterior border to the pale zone. The most distinctive fea-
ture is expansion of the intermediate antennal segments in the male,
and simple phallobase of the male genitalia. The prothoracic horn is
broader than in the other 2 subgroups, and any clouding on the
pronotum lacks a median pale stripe. This last subgroup contains
nevermanni Werner, alvarado Chandler, and vafer Chandler, with
sexnotatus Champion assignable to it on male characters but having
the elytral markings so reduced that they are difficult to interpret.
M. hispaniolae is probably a member of this third subgroup, but
1983]
Werner — Anthicidae
231
no males have been collected. However, the more posterior dark
mark on the elytra shows no sign of a posterior excavation, as is
present in alvarado and vafer. The distinctive erect setae on the sides
of the prothorax are matched in alvarado and approached in vafer,
but are also approached in salvadorensis in the second subgroup.
No other species has such differences between the broad and narrow
scales, but there is some difference in vafer, alvarado and salva-
dorensis; the tendency may be more a function of denseness of scales
than relationship. The long setae on the underside of the horn are
matched in vafer and hinted at in some others in the third subgroup,
salvadorensis in the second, and intermixtus in the first. The setae
on the horn, and matching setae on top of the head, as well as the
erect setae on the top of the head that outline the horn, probably
have an adaptive value in keeping sand grains out of the space
between head and horn when the beetle is digging. Degree of devel-
opment might very well be habitat-related.
Checklist of Species and Greater Antilles Records
Acanthinus angusticollis (LaFerte) 1848: 120-1. Werner 1966b:
747-9, fig. 1, 3, 6, synonymy. Southern Brazil to northern
South America. Introduced? Cuba: Bahia Honda, Camaguey,
Cayamas, Havana, Santa Clara. Jamaica: Kingston.
Acanthinus concinnus (LaFerte) 1848: 123. Werner 1970a: 123, fig.
7, 21. Bolivia to eastern Mexico. Introduced? Cuba: on ship
from Cuba. Hispaniola: Rep. Dom.: Boca Chica, Colonia
(1000 m), Haina, San Cristobal (35 m), San Francisco Mts.,
San Jose de las Matas, Trujillo Valdes (Boni), Villa Altagirica.
Acanthinus ebeninus (LaFerte) 1848: 117. Werner 1970a: 119, fig.
17.
Pseudoleptaleus cubanensis Pic 1917: 8 (type-locality: Cuba).
Venezuela and Colombia; reported from Guatemala without
exact locality (specimen not seen). Cuba: only the Pic specimen,
without specific locality.
Acanthinus quinquemaculatus (LaFerte) 1848: 115-6. Werner
1970a: 121-2, fig. 6, 20. Bolivia to eastern Mexico. Introduced?
CUBA: Sabanilla. Hispaniola: Rep. Dom.: Boca Chica, Colonia
(1000 m). PUERTO Rico: Flamboyant, Puerca Bay.
232
Psyche
[Vol. 90
Acanthinus schwarzi Werner 1967: 1232, fig. 10, 23. Probably
endemic. Cuba: Cayamas, Pinar del Rio, Soledad nr.
Cienfuegos.
Acanthinus scitulus (LeConte) 1852: 94-5. Werner 1970b: 724-5,
fig. 20-22, 34.
Formicilla cubana Pic 1944: 9-10 (type-locality: Cuba).
Formicillia gracillipes (sic): Buck 1960: 64, in part, Cuban
specimens.
Honduras to southeastern U.S.A. Probably a recent intro-
duction. CUBA. Hormiguero, Pinar del Rio. Through the
courtesy of Ben Brugge, of the Zoological Museum of Am-
sterdam, I have examined most of the specimens reported by
Buck. His specimen from Colombia belongs to Acanthinus
leporinus (LaFerte). Hispaniola: Rep. Dom.: Boca Chica
(10 m).
Amblyderus sp. Wolcott 1936: 210. Puerto rico: Ponce (on Randia
mitis and other flowers). Identification was provided by H. S.
Barber, but specimens cannot now be located. The blossom
association makes the identification suspect, since the usual asso-
ciation of Amblyderus is sand dunes.
Anthicus antilleorum Werner. Native. Also in Virgin and Bahama
Islands. Cuba, Hispaniola.
Anthicus blackwelderi Werner. Probably endemic. Jamaica, cuba,
HISPANIOLA.
Anthicus crinitus LaFerte 1848: 204-5. Werner 1975b: 472-3, fig. 2,
5. Old World, becoming cosmopolitan. Hispaniola: Rep. Dom.:
Bani, Mao Val-Verde, San Cristobal, St. Domingo, all near sea
level. PUERTO Rico: Fortuna A. E. S., La Parguera, Ponce.
Anthicus darlingtoni Werner. Endemic. Hispaniola.
Anthicus floralis (L.) 1758: 420. Werner 1964: 233-4, fig. 18, 71.
Cosmopolitan. Jamaica: Trelawney. hispaniola: Rep. Dom.:
San Cristobal, St. Domingo. Puerto rico: Ponce.
Anthicus formicarius (Goeze) 1977: 705. Werner 1964: 234-5, fig.
19, 72. Cosmpolitan. Jamaica: St. Andrew.
Anthicus hispaniolae Werner. Endemic, hispaniola.
1983]
Werner — Anthicidae
233
Anthicus macgillavryi Buck. Endemic. Cuba.
Anthicus pallidus Say 1826: 245. Werner 1964: 230-1, fig. 1, 2, 64,
synonymy. Coastal areas, Florida to northern South America;
Lesser Antilles. Probably native, cuba: Maisi in Oriente Prov.
HISPANIOLA: Haiti: Grande Anse. Rep. Dom.: Barahona.
PUERTO Rico: Bayamon.
Anthicus russoi Krekich. Probably a myrmecophilous endemic.
HISPANIOLA.
Anthicus soledad Werner. Endemic, cuba, Jamaica.
Anthicus subtilis LaFerte. Endemic, cuba, Hispaniola.
Anthicus tobias Marseul 1879: 125. Werner 1964: 235, fig. 12. Old
World, becoming cosmopolitan; Virgin Islands. Jamaica: Gor-
don Town, Morant Bay, Spanish Town. Hispaniola: Rep.
Dom.: Boca Chica, San Cristobal, Santo Domingo, all near sea
level.
Mecynotarsus hispaniolae Werner. Endemic. Hispaniola.
Mecynotarsus jamaicanus Werner 1962: 84, fig. 3, 10. Probably
endemic. Jamaica: Kingston.
Notoxus bipunctatus Chevrolat 1877: ix. Chandler 1978: 35, fig. 26,
57. Probably endemic. Puerto rico: Alsina, Anaso District,
Coama Springs, Ponce, Rio Piedras, San Juan.
Notoxus jamaicus Pic 1913: 8-9. Chandler 1978: 36, fig. 27, 57.
Probably endemic. Jamaica: Alligator Pond Bay, Bull Run in
St. Andrew Parish, Milk River, Morant Bay, Santa Cruz, Span-
ish Town, Trelawney.
Sapintus similis Werner 1983: 420. Mexico to Panama. Introduced?
JAMAICA: Spanish Town.
Sapintus teapensis (Champion) 1890: 249. Southeastern Mexico to
southern Brazil. Introduced? cuba: Baracoa, Cayamas, Vinales.
HISPANIOLA: Haiti: Desbarriere-Mt. La Hotte, Port-au-Prince.
Rep. Dom.: Bani, Haina, La Romana, Monte Cristi, Puerto
Plata. JAMAICA: Orange Bay, Santa Cruz, Spanish Town.
PUERTO Rico: Tortuguero Lake.
234
Psyche
[Vol. 90
Thicanus texanus (LaFerte) 1848: 301. Werner 1975a: 290, synony-
my. Southeastern U.S.A. to eastern Texas, primarily coastal.
Probably native. Hispaniola: Rep. Dom.: Barahona, Lake
Enriquillo. Puerto rico: Ensenada.
Vacusus holoxanthus (Fairmaire & Germain) 1860: 3. Werner 1961:
808-9; 1966a: 219, synonymy.
Vacusus jamaicanus Werner 1961: 809.
Chile to southern Brazil. Probably introduced. Jamaica:
Gordon Town, Milk River, Morant Bay, Spanish Town.
Vacusus vicinus (LaFerte) 1848: 157-8. Werner 1961: 799-801,
synonymy. Southern U.S.A. to Venezuela. Lesser Antilles.
Introduced? Cuba: Baragua, Camaguey, Cayamas, Havana,
Hormiguero, Jatabonica, Manicaragua, Soledad nr. Cien-
fuegos. HISPANIOLA: Rep. Dom.: Bani, Boca Chica, Mao Val-
Verde, San Cristobal, Santo Domingo. Jamaica: Bath St.
Thomas, Clarkstown, Milk River, Morant Bay, Santa Cruz,
Spanish Town, Trinityville. Puerto rico: Ensenada, La Gua-
nica, Lajas, Mayaguez, Sabena Grande, Salinas, virgin
ISLANDS.
References Cited
Buck, F. D. 1960. Anthicid beetles from Venezuela, Colombia, Cuba, and the
Netherlands Antilles. Studies on the Fauna of Curasao and other Caribbean
Islands: No. 47, pp. 64-71. Martinus Nijhoff, The Hague.
Champion, G. C. 1890. Anthicidae (pp. 203-50, pis. 9-10). In: Biologia Centrali-
Americana. Insecta, Coleoptera Heteromera, vol. 4, part 2 (1889-1893). R. H.
Porter, London, x + 494 pp.
Chandler, D. S. 1977. New with a key to the New World species
(Coleoptera: Anthicidae). Coleop. Bui. 31: 363-70.
1978. A revision of the Central and South American Notoxus and de-
scription of a new genus, Plesionotoxus .... Contrib. American Entomol. Inst.
16(3): i-iv, 1-83 (as 1977).
Chevrolat, a. L. a. 1877. Description d’espfeces nouvelles d’Heteromferes
provenant Hie de Porto Rico .... Bui. Soc. Entomol. France 7: viii-ix.
Fairmaire, L., and P. Germain. 1860. Coleoptera Chilensia. Part 2. Sect. 1.
8 pp. Paris,
Goeze, a. E. 1777. Entomologische Beitr^ge zu des Ritter Linne 12. Ausgabe des
Natursystems, vol, 1. Weidmann, Leipzig. 736 pp.
LApERTfe-SfeNECTfeRE, F. T. de. 1848. Monographie des Anthicus et genres voisins
... , De Sapia, Paris, xxiv + 340 pp., 16 pis.
1983]
Werner — Anthicidae
235
LeConte, J. L. 1852. Synopsis of the anthicites of the United States. Proc. Acad.
Natur. Sci. Philadelphia 6: 92-104.
Linnaeus, C. 1758. Systema Naturae ... , ed. 10, vol. 1. Laurentii Salvii, Holm.
2 + 824 pp.
Marseul, S. a. de. 1879. Monographie des anthicides de I’ancien-monde.
L’Abeille 17: 1-268, 2 pis.
Menozzi, C. 1930. Una nuova specie, probabile mirmecophila, di Anthicus . . .
della R. Dominicana. Boll. Lab. Zool. gen. et agraria del R. Inst. sup. agrario
— Portici 25: 92-4.
Pic, M. 1913. Descriptions de 29 especes. Mel. Exotico-Entomol., fasc. 5: 8-9.
1917. Descriptions abregees diverses. Mel. Exotico-Entomol., fasc. 22:
2-20.
1944. Opuscula Martialia, Xlll. Echange, Numero special: 1-16. Les
Imprimeries Reunies, Moulins.
Say, T. 1826. Descriptions of new species of coleopterous insects inhabiting the
United States. J. Acad. Natur. Sci. Philadelphia 5: 237-84, 293-304 (continued
from 1825, op. cit. 5: 160-204.).
Werner, F. G. 1961. A revision of the genus Vacusus .... Ann. Entomol. Soc.
America 54: 798-809.
1962. The species of Mecy no tarsus related to elegans .... Proc. Entomol.
Soc. Washington 64: 79-86.
1964. A revision of the North American species of Anthicus, s. str. Misc.
Publ. Entomol. Soc. America 4(5): 195-242.
1966a. Notes on the South American species of Vacusus ... . Ann.
Entomol. Soc. America 59: 218-22.
1966b. A revision of Acanthinus ... II .... Op. cit. 59: 746-51.
1967. A revision of Acanthinus ... VI .... Op. cit. 60: 1217-34.
1970a. A revision of Acanthinus ... VII. Op. cit. 63: 1 1 1-28.
1970b. A revision of Acanthinus ... IX ... . Op. cit. 63: 718-31.
1975a. New synonymy in the nearctic Anthicidae .... Proc. Entomol.
Soc. Washington 77: 290.
1975b. Additions to the nearctic Anthicus .... Op. cit. 77: 472-7.
1983. Neotropical Sapintus, with a general key to species. Proc. Entomol.
Soc. Washington 85: 405-425.
Wolcott, G. N. 1936. Insectae Borinquenses .... Journ. Agric., Univ. Puerto
Rico 20: 1-627.
,4! *>. \il j^
I/. , ' *-<.> ' • s"-*^ '■'■
.■^i.id;ji>#E'-’;‘'£#‘
> . .-vV'.i'
,• • :,][, ■.j^ikfi' ; * 'i^' •■* *■ • i‘*'' —
^■>
■ '^‘'¥fe*'r:'' .:if-‘ ;=
•■■ - ..•'T^y •
, ^;' . v;
*V'
/-€' t ■
NEST ARCHITECTURE AND BROOD DEVELOPMENT
TIMES IN THE PAPER WASP, POLISTES EXCLAMANS
(HYMENOPTERA: VESPIDAE) *
By J. E. Strassmann and M. C. Ferreira Orgren
Biology Department,
Rice University,
Houston TX 77251
One of the distinctive features of social insects is that they rear
their brood in nests. In the Vespidae these nests are typically con-
structed of paper; they have one or several layers of cells, and may
have an outer envelope of paper (Jeanne 1975). Nest architecture
has been interpreted as a means of minimizing vulnerability to nest
predators, particularly ants (Jeanne, 1975; 1979). Another factor
that may contribute to nest design is a limitation on efficient food
distribution to larvae when there are many cells in a single layer. For
example if foragers tend to land on one part of the nest and then
begin feeding the nearest larvae, unequal food distribution would
result. The purpose of this study was to examine the influence of cell
location on brood development times of Polistes exclamans Vie-
reck. Also examined were the roles of time of year, numbers of
workers, and larvae per worker as factors influencing development
times. P. exclamans was chosen as a study organism because all cells
are in one layer, without an envelope; nests are approximately
circular, and have a single off-center pedicel usually located
towards the top of the nest. Cells near the pedicel are the
oldest. These features make nests of P. exclamans among the
more simple types of nests. In central Texas nests of P. exclamans
vary greatly in size, reaching an upper limit of about 500 cells
(Strassmann, personal observation).
Methods
Three nests representing small, average and large nests were
chosen for observation at Brackenridge Field Laboratory of the
University of Texas at Austin, Texas. These nests appeared to be
* Manuscript received by the editor May 15, 1983.
237
238
Psyche
[Vol. 90
generally representative of their size classes. In 1978 nest 83 (44
cells) and nest 22 (220 cells) and in 1979 nest 27 (345 cells) were
observed (Fig. 1). Contents of cells were scored every other day on a
cell map, and numbers of females associated with the nest were
marked and counted. Nest 22 was observed from 12 June to 14
August; nest 83 was observed from 19 June to 14 August; and nest
27 was observed from 6 June to 18 August. From mid June to mid
August mean daily termperatures changed very little. Average
monthly temperatures were 23.5° C for June, 24.7° C for July and
24.8° C for August (30 year averages for Austin, Texas, National
Weather Service). When an adult emerged from a cell an egg was
laid in it so all cells contained brood. In the rest of this paper the
nests will be referred to as the small (nest 83) medium (nest 22) and
large (nest 27) nest.
For the purposes of analysis the medium and large nests were
divided into 4 regions, and the small nest was divided into 3 regions
(Fig. 1). The regions were chosen by first mapping development
times of brood in every cell on a cell map, and then choosing regions
that were homogeneous within themselves and as different as pos-
sible from other regions. This technique maximized the probability
that differences among regions would be found. On all nests region
1 is the oldest, directly in front of the nest pedicel. Region 2 is the
center of the nest. Regions 3 and 4 are edge regions. The medium
nest may appear in the figure to have two lobes to it but they were
actually contiguous. The cells were deformed somewhat due to
contact with 1 inch chicken wire mesh which ran down the center of
the nest.
Large sample sizes, normal distribution of data and nearly equal
variances allowed us to use parametric statistics in this study.
Because of its size the small nest was omitted from some of the
analyses. All statistical analyses were performed using Statistical
Analysis System (SAS) or Statistical Package for the Social
Sciences (SPSS).
Results
Average development times of eggs varied from 9 to 14 days
depending on the nest (Table 1). Eggs took significantly longer to
develop on the small nest (small compared to medium nest t = 5.45.
1983]
Strassmann & Orgren — Polistes
239
SMALL NEST
Figure 1. Cell maps of nests in the study indicating nest regions.
240
Psyche
[Vol. 90
df = 38 1 , p <0.00 1 ; small compared to large nest, t = 8.55, df = 542,
p <0.001). There were no differences in egg development times
between the medium and large nests (t = 0.57, df=839, n.s.). Aver-
age larva development times varied from 13 to 18 days depending on
nest (Table 1). The small nest and the large nest did not differ in
average larva development times (t = 0.68, df=343 n.s.). However
they both had longer larva development times than did the medium
nest (small compared to medium, t = 4.84, df=322, p <0.001;
medium compared to large, t = 8.89, df=611, p <0.001). Pupa
development times averaged 13 days on all nests, and there were no
significant differences among the nests.
Development times of eggs, larvae and pupae did not vary signifi-
cantly from one region of the nest to another on the small and
medium nests (Tables 2-4, Fig. 2). In the large nest both eggs and
larvae developed most quickly in region 1 (the oldest part of the
nest), and most slowly in the edge regions 3 and 4 (Tables 2-4, Fig.
2). In no case did region of nest explain more than 10% of the
variance in development time.
Date did not have a consistent effect on development times. Eggs
developed more slowly towards the end of the season in the large
nest, and more quickly towards the end of the season in the small
nest (Table 5). There was no change in egg development time with
date in the medium nest. Larva development time increased with
date in both the large and the medium nest, and decreased with date
in the small nest (Table 5). Development times of pupae did not
change with date in the large nest, but decreased with date in the
medium and small nests (Table 5). Date explained 35% to 51% of
the variance in larva development times depending on nest. It
explained smaller percentages of the variance in egg and pupa
development times except on the small nest (Tables 2-4).
Interaction between date and region of nest was examined using a
2-way ANOVA (Tables 2-4). There was a significant (p <0.05)
interaction between date and region of nest for egg and larva devel-
opment times in the large nest that explained 7% and 5% of the
variance respectively (Tables 2-4). This interaction thus explains a
trivial amount of the variance in development compared to that
explained by date.
The effect of numbers of females tending the nest was found to
be quite variable. Looking only at the medium and large nests, it
1983]
Strassmann & Orgren — Polistes
241
TABLE 1. Average brood development times per nest.
X
S.D.
N
Egg development time
Small nest
13.81
5.67
43
Medium nest
9.58
4.67
340
Large nest
9.43
2.92
501
Larva development time
Small nest
18.18
3.71
28
Medium nest
13.39
5.09
296
Large nest
17.39
6.03
317
Pupa development time
Small nest
12.74
2.18
19
Medium nest
13.05
4.31
218
Large nest
13.00
3.83
268
TABLE 2. ANOVA of the effects of date and location in nest on egg devel-
opment times.
Main Effects Interaction
Location in nest
Location in nest
Date
X date
Small nest
Sum of squares
68
648
76
F
2.1
1.31***
1.5
df
2
3
3
% of variance explained
5
48
6
Medium nest
Sum of squares
56
1184
319
F
1.0
16.3***
1.5
df
3
4
12
% of variance explained
16
4
Large nest
Sum of squares
105
100
278
F
4.5**
3.2*
3.2***
df
3
4
11
% of variance explained
3
2
7
df = degrees of freedom, *p <0.05, **p <0.0 1 , ***p <0.00 1
242
Psyche
[Vol. 90
was found that egg development time increased with number of
females on the large nest and decreased on the medium nest (Table
5). Larva development times increased with numbers of females on
both large and medium nests. Pupa development time decreased
with increasing numbers of females on the large nest, and did not
change significantly on the medium nest (Table 5).
Partial correlations of development time with date were calcu-
lated controlling for numbers of females, since numbers of females
increased with date. Development times of larvae increased signifi-
cantly (p <0.01) with date on the large and medium nests when
numbers of females were controlled for (Table 5). Development
times of eggs decreased with increasing numbers of females on the
medium nest when date was controlled for. Development times of
larvae and pupae decreased with increasing numbers of females on
the large nest when date was controlled for.
The ratio of larvae to females on the large nest was 3. 1 1 ± S.D.2.05,
and on the medium nest it was 1.23 ± S.D.0.47. Development time
of larvae was slower when there were more larvae per female on the
large nest (r = -0.13, p<0.03, N = 317). The correlation was in the
same direction on the medium nest, but was not significant
(r = -0.10, p >0.1, N = 296) perhaps because there averaged more
females per larva.
Discussion
The results presented here do not offer strong support for the
hypothesis that location of brood affects development rates. Cen-
trally located brood developed no more quickly than did edge
brood. Feeding efficiency does not appear to be limiting the size of
nests. Foragers arriving on the nest with prey typically share it with
3 or 4 other females who each visit many larvae (Strassmann,
unpub.). The result seems to be even distribution of food. Though
the largest nest did show a significant region effect for egg and larva
development, it explained less than 7% of the variance. Perhaps an
even larger nest would show a more marked effect. West Eberhard
(p. 38, 1969) suggested that 7 pupae in the center of a nest she
observed and 8 pupae towards the edge of the nest had long and
short pupal periods respectively because of differences in larva
nutrition. She suggested that better-fed larvae in the center of the
nest would have longer pupa development periods. This study does
LARVA DEVELOPMENT TIME IN DAYS
1983]
Strassmaun & Orgren — Polistes
243
123 1234 1234
SMALL NEST MEDIUM NEST LARGE NEST
NEST REGION
Figure 2. Development times of larvae in different regions of each nest. Bars
indicate means and lines indicate standard deviations.
244
Psyche
[Vol. 90
TABLE 3. ANOVA of the effects of date and location in nest on larva
development times.
Main Effects
Location in nest Date
Interaction
Location in nest
X date
Small nest
Sum of squares
25
148
22
F
1.4
5.6
1.3
df
2
3
2
% of variance explained
7
40
6
Medium nest
Sum of squares
35
2631
49
F
0.7
37.0***
0.3
df
3
4
II
% of variance explained
I
35
1
Large nest
Sum of squares
821
5842
568
F
19.4***
138.1***
5.0***
df
3
3
8
% of variance explained
7
51
5
df = degrees of freedom, *p <0.05, **p <0.0 1 , ***p <0.00 1
not support her conclusions.
High mean daily temperatures characterized the entire period of
this study, essentially eliminating temperature as a variable. The
slight increase in temperature over the season would be expected to
speed up development if it had any effect at all. The increase in
development time with date may best be explained by a gradual
seasonal decrease in abundance of prey. Later in the season larvae
may take in less nutrition per day, which results in longer times
spent as larvae. Also there is usually a gradual increase in size of
adults over the season in P. exclamans (Strassmann, unpub.).
Larvae destined to become larger adults may require longer feeding
periods. In P. metricus midsummer workers are as large as queens
(Haggard and Gamboa, 1980). These large workers were larvae
when worker to larva ratios were at their maximum (Haggard and
Gamboa, 1980).
Development times of larvae were shortest on the medium nest
which had the fewest larvae per worker. There are probably advan-
1983]
Strassmann & Orgren — Polistes
245
TABLE 4. ANOVA of the effects of date and location in nest on pupa development.
Main Effects
Location in nest Date
Interaction
Location in nest
X date
Small nest
Sum of squares
7
33
3
F
1.1
11
1.1
df
2
1
1
% of variance explained
8
39
4
Medium nest
Sum of squares
21
333
180
F
0.4
4.8**
1.0
df
3
4
10
% of variance explained
8
5
Large nest
Sum of squares
57
278
105
F
1.4
5.0
0.8
df
3
4
9
% of variance explained
2
7
3
df= degrees of freedom, *p<0.05, ♦♦p<0.01, ***p <0.001
tages to flexibility in development time of larvae which allow more
time for development when food is limiting, either because there are
fewer females to harvest it, or because of a general scarcity of prey in
the environment.
Data on development times were found in the literature for 4
populations of P. fuscatus, and 1 each of P. hunteri, P. annularis, P.
gallicus and P. exclamans (Table 6). Development times varied from
10 to 25 days for eggs, from 15 to 25 days for larvae and from 13 to
22 days for pupae (Table 6). P. exclamans in this study falls towards
the faster end of this range, particularly for pupa development time.
All reports of pupa development times averaged over 18 days except
Rabb’s study of P. exclamans in N. Carolina. P. exclamans was
studied in the most southern climate, so it is possible that the differ-
ences are due to temperature. There is a trend in P. fuscatus towards
shorter pupa development times in more southern populations. P.
exclamans and P. annularis are the only two members of the subge-
nus Aphanilopterus represented here. It is possible that Aphanilop-
terus, which generally has larger nest sizes and more adults tending
246
Psyche
[Vol. 90
TABLE 5. Correlations between development time, date and number of females on
the nest.
(U
(U
0>
X)
E
c
Vi
S3
•o
E
3
3
3
"o
Si
j=
CQ
C
X
i
£
b.
<2
3
O
c
c
o
c
_o
&0
c
Vi
C
u
Vi
.rt
k-
“E
"3
.3
JD
o
E
U
E
k.
b.
o
U
date
b.
k.
o
U
E
3
C
k.
o
U
c
o
o
2
‘o
O
U
kM
o
Egg development time
Small nest
-0.31*
43
Medium nest
-0.07
-0.15**
0.13**
-0.18***
340
Large nest
0.10*
0.10*
0.01
0.04
501
Larva development time
Small nest
-0.58**
28
Medium nest
0.46***
0.40***
0.25***
-0.04
296
Large nest
0.65***
0.30***
0.96***
-0.93***
317
Pupa development time
Small nest
-0.59**
19
Medium nest
-0.15*
-0.11
-0.11*
0.05
218
Large nest
-0.04
-0.16**
0.25**
0.29***
268
*p <0.05; **p <0.01; ***p <0.001
the nest, also has faster development times (Strassmann, unpub.).
This is not contradicted by the very long development times
reported by Jeanne for P. annularis since these were early spring
data, and are therefore not strictly comparable. Another factor that
may result in selection for fast pupa development times is that nests
of P. exclamans are very vulnerable to predation, and to loss of the
nest due to death of all workers (Strassmann, 1981). Short pupa
development times may reduce the probability of nest loss. But this
1983]
Strassmann & Orgren — Polistes 247
U CQ
E ^
'Z1 c
c >
fc 60
(U (U
> 60
U C
T3 es
y. «
§ :£
CO
^ t
tU
-j a;
< o
H U,
Ov
SO
Os
■H ■§ xi
ca 3 o. 3
d. o ^ o ^ o — ^
reference
West Eb(
Klahn, u
Rabb, 19
Morgan,
Rabb, 19
Jeanne, u
Rabb, 19
Pardi, 19
Z
124
96
6
192
8
15
6
106
Os rn \0 T}-
-sa- <N — : _■ —
ea
+1 +1 +1 +1 *7
o.
<Nrsj»oosTtr^r^^
3
D.
pi Os" od Os’ r^’ CTs’ Tj- m
z
58
05
4
92
6
42
6
17
~ — —
— <N <N ^
r- m «ri r*~i
ea
+1 •+! +1 +1 7
>
«^so«n(N<NTtr^^
-2
^ vS Os' vo
(N
Z
36
22
5
95
6
96
8
01
Os -- OS Tt
fN <N — ' Tt
+1 +1 +1 +1 7
60
OfNOOinoosTtw
60
2 o — Tt f*S o
U
3 ca
3 -s .£ .E .£
c
_o
.3) 2 = 2-22
ta
u
•i|U§U"5)U>s
^ z 5 z z 2
c/3
i/>
u
‘o
(U
D.
a ^ c/3 cn 'c a
5HHa"=fl|s
2 S S .2 « 3 « .2
;SJ303C><15
^'*.v,>JS(g4^6U
^ 0." CL CL CL CL CL O,"
248
Psyche
[Vol. 90
may also be the case for the other species of Polistes where data on
nest failure rates are not available.
Polistes has slightly longer development times than other social
wasps. Paravespula vulgaris, Dolichovespula sylvestris and Vespa
crabro all have summer egg development times of about 5 days,
larva development times of 10 to 15 days and pupa development
times of 1 1 to 15 days (Spradbery, 1973). The shorter development
times in these species as compared to Polistes may be due to the
much larger colony sizes found in these species, as well as their
ability to eat a greater variety of arthropods and carrion.
Acknowledgements
We thank Christi Steinbarger and Dana Meyer for help with field
work, and Bill Mueller and Colin Hughes for their comments on the
manuscript. This research was supported by NSF Postdoctoral Fel-
lowship #SPI-7914902 and NSF #DEB80-05739 to JES.
References Cited
Haggard, C. M. and G. J. Gamboa
1980. Seasonal variation in body size and reproductive condition of a paper
wasp, Polistes (Hymenoptera; Vespidae). Canadian Entomolo-
gist 112:239-248.
Jeanne, R. L.
1975. The adaptiveness of social wasp nest architecture. Quarterly Review of
Biology 50:267-287.
1979. A latitudinal gradient in rates of ant predation. Ecology 60: 121 1-1224.
Pardi, L.
1951. Studio dell attivita e della divisione di lavoro in una societa di Polistes
gallicus (L.) dopo la comparsa delle operaie. (Ricerche sui Polistini XII).
Archivio zoologico italiano 36:361-431.
Rabb, R. L.
1960. Biological studies of Polistes in North Carolina (Hymenoptera: Vespi-
dae). Annals of the Entomological Society of America 53:111-121.
Spradbery, J. P.
1973. Wasps: an account of the biology and natural history of social and
solitary wasps. University of Washington Press, Seattle, Washington,
408 pp.
Strassmann, j. E.
1981. Parasitoids, predators, and group size in the paper wasp, Polistes excla-
mans. Ecology 62:1225-1233.
West Eberhard, M. J.
1969. Social biology of polistine wasps. University of Michigan, Museum of
Zoology, Miscellaneous Publications No. 140: 1-101.
NEW SPECIES OF THE ANT GENUS MYOPIAS
(HYMENOPTERA: FORMICIDAE: PONERINAE)
By Robert B. Willey' and William L. Brown, Jr.^-^
The work reported upon here began in the early 1950’s as a
revision of genus Myopias, including as a synonym Trapeziopelta.
For a year or more it served as the trial focus of RBW’s doctoral
thesis research, until his interests shifted into other channels, and he
laid the revisionary work aside. Meanwhile, WLB’s interest in the
revision continued, but he had no opportunity at that time to do
much more than supervise the drafting of a set of illustrations by
artist Nancy Buffler — many of which are now offered here — and to
make some of the dissections of mouthparts, etc.
As WLB’s work on the reclassification progressed for over 25
years through the tribes of subfamily Ponerinae, much new material
was added to what had been available for the original Myopias
study, and additional new synonymies and new species were discov-
ered, as well as valuable information on the larvae, males, distribu-
tion and bionomics of species new and old. Even the status of
Myopias as a genus apart from Pachycondyla came into question.
Although in some ways it would be best if the old findings to which
we both contributed could simply be incorporated in the reclassi-
fication part dealing with tribe Ponerini 5. str., there seemed in this
course no convenient way to recognize the legitimate claim of RBW
to authorship based on the considerable amount of work he had
done on Myopias in 1955.
The compromise reached sees the larger Myopias review, with
keys to species and discussions of synonymy, biology, etc. to be
included in Brown’s forthcoming Part VII of “Contributions toward
a Reclassification of the Formicidae,’’ while descriptions of the new
species included in various drafts of our joint manuscript of the
'The University of Illinois at Chicago Circle, Biological Sciences, P.O. Box 4348,
Chicago, IL 60680.
^Department of Entomology, Cornell University, Ithaca, NY 14853. (Address cor-
respondence here.)
3A report of research from the Cornell Agricultural Experiment Station. Research
supported by National Science Foundation Grant DEB-8003722.
Manuscript received by the editor March 15, 1983
249
250
Psyche
[Vol. 90
1950’s are presented here, together with a few notes on variation, on
bionomics, and on the distribution of certain species. Figures of
some old species are included with those of the new ones.
Collections and Collectors, with Abbreviations
The main collection used is that of the Museum of Comparative
Zoology at Harvard University, Cambridge, Massachusetts (MCZ),
for Myopias based mainly on collections by Eric Mjoberg, Edward
O. Wilson, James W. Chapman, William L. Brown, Jr. and Philip
S. Ward. Secondary sources were the British Museum (Natural His-
tory) in London (BMNH), collected by Barry Bolton and others,
and the Australian National Insect Collection at Canberra (ANIC),
collected by Robert W. Taylor and others. For the collectors named
above, only surnames are cited in the text. Our thanks go to all who
provided us with specimens.
The drawings provided here were mostly done during the mid-
1950’s by Nancy Buffler. Fig. 4 is by James S. Miller. We are also
grateful for a copy set of Edward Wilson’s wonderful New Guinea
field notes of 1955, which have yielded most of what we know about
Myopias bionomics, here published for the first time.
Measurements and Ratios
Where series were available, measurements were usually taken on
the largest and smallest (worker) specimens in each locality-series.
The measurements and indices are mostly those standard in ant
taxonomy for the past 30 years.
TL (total length) axial length of body, including closed mandi-
bles; summed ML + HL + WL + petiole L + length of
gaster.
HL (head length) maximum measurable length of head as seen
in dorsal full-face view, using the anterior edges of the fron-
tal lobes as the anterior reference point, and the posterior-
most point or points of the cranial outline as the posterior
reference point.
HW (head width) maximum measurable width of head, not
including the eyes, as seen in dorsal full-face view.
Cl (cephalic index) HW X 100/ HL.
ML (mandibular extension) maximum measurable distance be-
1983]
Willey & Brown — Genus Myopias
251
tween the most distal apex of the closed mandibles and the
anterior edges of the frontal lobes, as seen in same (dorsal
full-face) view from which HL is taken.
Ml (mandibulo-cephalic index) ML X 100/ HL.
MLO (mandibular outside length) maximum absolute chord
length of left mandible measured from lateral insertion to
apex.
CLL, (length, width of median clypeal lobe) as measured in dorsal
CLW full-face view.
SL (scape length) chord length of antennal scape, excluding
radicle.
SI (scape index) SL X 100/ HW
EL (eye length) maximum measurable length of facetted part of
eye.
WL (trunk length) diagonal length of trunk as measured from
side view, from anterodorsal slope of pronotum (excluding
cervix) to most posterior extremity of propodeum.
Myopias gigas, new species
(Figures 1, 12)
Diagnosis, worker: A very large species of the M. loriai group,
even larger than M. loriat, with proportionately longer mesonotum
and petiolar node, and with the head dorsally, trunk dorsum and
pleura of posterior section of trunk sharply and regularly striate;
body otherwise prevailingly smooth and shining. Funicular segment
II very long, longer than I.
Worker, holotype: TL 16.9, HL 2.50, HW 2.60 (Cl 104), ML 2.26
(MI 90), SL 2.62 (SI 101), EL 0.45, WL 4.61, petiole L 1.7, hind
femur L 3.7, hind tibia L 3.16 mm.
This, the largest known species of Myopias, has the broad, poste-
riorly narrowed head of the loriai group; long, slender, curved
mandibles and rather large eyes with many fine facets. A scape,
when held straight back as seen in full-face view, surpasses the
posterior border of the head by nearly IVi times the apical scape
width. The posterior border of the head is transverse and nearly
straight, varying from very feebly concave to subsinuate in slightly
different views. As in M. loriai, the median clypeal lobe is apically
biconvex, with a shallow median notch; the lobe is shorter and
broader than in M. loriai, and tapers slightly from base to apex.
252
Psyche
[Vol. 90
Antennal funicular segment 11 is longer than 1, 111 and all other
funicular segments except the apical, and it is 2.5 times longer than
its maximum (apical) width. No differentiated antennal club.
Labrum with a sharp erect tooth at the apex of each labral lobe;
no median labral tubercle. Palpi concealed, not seen. Mandibles as
shown in Figure 1; apical tooth followed closely basad by 2 coarse
denticles and a blunt tooth; middle tooth followed basad by a low,
rounded basal angle. Strix (mandibular groove) well-developed
from base to apex.
Trunk long and robust; mesonotum longer than in loriai, but
wider than long (L/W ~ 0.7). Metanotum present as a deeply
impressed groove, widening laterad on each end. In side view pro-
file, pronotum strongly convex, although transversely impressed
just in front of the raised, cariniform posterodorsal margin; meso-
notum feebly convex and sloping downward behind, but its anterior
margin raised slightly above the posterior pronotal margin, espe-
cially (as in the type) when the two somites are flexed against each
other. Promesonotum (without cervix) and propodeum subequal in
length; propodeum broadly convex from front to rear, with its de-
clivity steeper than its dorsum, but passing into dorsum through a
gentle curve. Mesopleural suture distinct and complete, moderately
deeply impressed (more distinct than in M. loriai). Propodeal spira-
cle elongate and oblique, its opening about IVi times longer than
wide.
Petiole (Fig. 12) loaf-shaped, longer than broad and longer than
high; exact shape of subpetiolar process, if any, not determined
because the extreme anterior end of the segment is hidden by the
coxae. Gaster long, gently downcurved, with a distinct constriction
between first and second segments; dorsally viewed, second seg-
ments longer and a little wider than first. Sting long and strong,
distinctly upcurved.
Dorsum of head completely finely and regularly striate in a longi-
tudinal direction, the striae mesal to and behind the eyes tending to
curve slightly outward. Dorsum and declivity of trunk similarly
striate, but in a transverse direction, arching on pronotum. Sides of
trunk behind pronotum with similar, oblique striation, continued
from the propodeal dorsum through a curve. Remainder of head,
body and appendages smooth and shining, including mandibles,
cervical border of vertex and sides of pronotum. Coarse, spaced.
1983]
Willey & Brown — Genus Myopias
253
Figs. 1-3, Myopias spp., heads of workers, sculpture and pilosity omitted. Fig. 1,
M. gigas holotype in full-face view. Fig. 2, M. lobosa, paratype in full-face view. Fig.
3, M. lobosa, another paratype in side view. Scale bars for Figs. 2 and 3 are 0.5 mm.
254
Psyche
[Vol. 90
piligerous punctures are conspicuous in smooth areas, particularly
the sides of the head, mandibles, femora and tibiae, petiole, and
normally exposed surfaces of gastric terga.
Pubescence appressed and decumbent, generally very sparse,
except on antennal flagella, coxae, tarsi, flexor surfaces of fore
tibiae, flexor surfaces of mid femora, extensor surfaces of mid
tibiae, and apex of hypopygium. Rather abundant erect or suberect,
fine, tapered hairs, from short to over 0.5 mm long, occur on almost
all normally exposed surfaces of body and appendages. Color deep
reddish brown, appendages mainly clear light red.
Holotype (MCZ) a unique worker from Dobodura, Papua New
Guinea (P.J. Darlington leg.).
This magnificent species is even larger than M. loriai, and has
very different sculpture, but the two forms are obviously closely
related. Because of the long mandibles and large size, we guess that
M. gigas may be a millipede predator, but we have no direct evi-
dence of feeding behavior for this species.
Myopias julivora new species
(Figs. 5, 22)
Diagnosis, worker: Similar to M. tenuis, but larger (HW 0.80-
1.01), with relatively longer mandibles and antennae, MI > 65,
scapes overreaching posterior border of head (when held straight
back, full face view) by about their own apical width to nearly twice
their apical width; all antennomeres longer than broad. Shafts of
mandibles approximately straight over middle half of their length.
Worker, holotype: TL 6.2, HL 1.04, HW 0.94 (Cl 90), ML 0.73
(MI 70), MLO 1.01, SL 0.90 (SI 96), EL 0.09, WL 1.74, hind femur
L 1.00, hind tibia L 0.94 mm.
Worker, paratypes (n = 6 of 34 representing 7 colonies from 6
localities, including largest and smallest specimens): TL 5. 8-6. 7,
HLO.91-1.14, HW 0.81-1.01 (Cl 88-90), ML 0.62-0.83 (MI 66-73),
MLO 0.86-1.14, SL 0.86-1.09 (SI 96-108), EL 0.06-0.10, WL
1.66-1.93, hind femur L 0.89-1.15, hind tibia L 0.87-1.12 mm.
Description limited to details not covered in diagnosis and mea-
surements. Median frontal sulcus extends approximately to middle
of HL, followed posteriad after a gap by a shallow pit marking
location in queen of anterior ocellus; this pit is usually absent in M.
tenuis, but is occasionally faintly indicated there. Compound eye
1983]
Willey & Brown — Genus Myopias
255
essentially reduced to a single convex lens, but at high magnifica-
tions, traces of an ommatidial grid can be made out; reduction
approaches the state in M. tenuis, but does not go quite so far.
Median clypeal lobe trapezoidal, widest near apex (CLL 0. 12, CLW
0. 16 mm), but by optical illusion may seem as long as or longer than
wide; free corners rounded; anterior margin straight, convex, or
even slightly sinuate. Basal oblique mandibular groove (strix) sub-
lateral in origin, difficult to see in dorsal view, but distinct with its
ventrolateral extension in side view. Submedian tooth situated in
seventh tenth of the shaft length, counting from base. Basal angle
obsolete.
The upturned tooth on each labral lobe and 3,3 palpal segmenta-
tion formula are as in tenuis.
Trunk formed much as in M. tenuis; promesonotum subequal in
length to propodeum; side view outline rather low and weakly con-
vex, with a distinctly, but not deeply, impressed metanotal groove;
propodeal dorsum only feebly convex, and sometimes very feebly
impressed near midlength. Petiolar node slightly longer than broad,
about as broad as long, or slightly broader than long, in different
series (as in M. tenuis also), summit convex, slightly higher behind.
Caster with first segment strongly rounded above, tergum rising
caudad; segment II distinctly constricted in front at juncture with its
acrotergite; about as high at maximum height as segment I, and
slightly wider. As seen from above, anterior margin of segment I
straight or feebly convex; shallowly concave in Vanimo worker (and
queen). Sting long (extruded up to 0.6 mm), sharp, upcurved.
Sculpture prevailingly smooth and shining; punctures minute and
widely spaced, more numerous and coarser on head, especially in
Vanimo worker and queen, and on propodeum, but even here still
obscure. Pilosity of uneven length, fine, tapered, erect to suberect
hairs, mostly 0.05 to 0.30 mm long; pubescence decumbent to sub-
erect, very dilute on anterior dorsum of head, but more abundant on
antennae and legs, especially extremities.
Color averaging lighter than in fully pigmented M. tenuis
workers, light to medium brownish red to dark brownish red, light
orange brown in some workers, possibly callow. Appendages usu-
ally lighter, more yellowish, than basic body color.
Worker variation, as mostly discussed already above, involves
mainly size-related features and shape of clypeal lobe, distinctness
256
Psyche
[Vol. 90
and density of the obscure puncturation, length and degree of apical
taper of petiolar node, size and pigmentation of compound eyes,
length of antennal scapes, and depth of body color.
Queen, dealate (from type nest series, Wilson No. 905), TL 7.1,
HL 1.10, HW 0.97 (Cl 88), ML 0.74 (MI 67), MLO 1.02, SL 0.96
(SI 99), EL 0.26, WL 2.03 mm. Combined measurements for the
largest queen specimen (above), another queen from the type local-
ity, colony No. 1048, and a smaller queen from near Vanimo, are:
TL 5.6-7. 1 , HL 0.93-1 . 10, H W 0.84-0.97 (Cl 88-90), ML 0.63-0.76
(MI 67-72), MLO 0.87-1.03, SL 0.87-0.96 (SI 99-104), EL 0.22-
0.26, WL 1.78-2.03 mm.
The queen differs from accompanying workers by the usual pone-
rine characters, and is also darker in color, prevailingly piceous, or
even blackish in the Vanimo specimen. On trunk, centers of scutum
and scutellum are infuscated, while marginal areas of these and
other sclerites are lighter and more reddish. Appendages lighter,
more yellowish.
Male unknown.
Described from material representing seven separate collections
from six localities in Papua New Guinea. Holotype (MCZ) from
Wilson’s colony No. 905, lower Busu River, Huon Peninsula, Papua
New Guinea, 3 May 1955, a nest in rain forest in a small Zoraptera-
stage rotten log, in a part of the log somewhat raised off the ground,
containing one queen, about 30-40 workers, and brood of all stages,
with pupae predominating. Abundant remains of millipedes were
found in the brood chamber and galleries leading away. One fresh
millipede corpse was among larvae; the prey all seemed to belong to
one kind.
Another colony (Wilson No. 1048) also came from the lower Busu
River tract, 15 May 1955, from cavities in an old, hard polypore
fungus growing on a large Passalus-^idigt log, containing a queen
and about 75 workers, plus abundant brood of all stages, without
notable preponderance. Half of a freshly dead millipede was found
with the brood; the midden remains were collected (but later lost
with the nest residue in alcohol).
A worker and a dealate queen were found in lowland (40 m) rain
forest next to the quarry at Km 2 on the Bewani Road, near
Vanimo, West Sepik District, Papua New Guinea, 27 February
1981, leg. Brown (No. 81-48). The nest was in a small rotten stick
1983]
Willey & Brown — Genus Myopias
251
lying on the ground, and contained larvae as well as the remains of
small millipede prey. (Paratypes in MCZ, BMNH, ANIC, etc.)
In addition, single strays come from three widespread localities:
Dobodura, March to July 1944, leg. P. J. Darlington, Jr.; lora
Creek, 17 km. S. of Kokoda at 1400 m, leg. Ward (No. 1831) rotten
log, montane rain forest; Baiyer River, Western Highlands, about
1200 m, 6 July 1974, leg. S. Peck, berlesate B-28 1 . The last specimen
is the largest one of the species seen; it is also the darkest in color,
has somewhat coarser punctures than usual on the head, and has the
longest scapes, so that it might be thought transitional to M. media,
but the form of the mandibles and clypeal lobe is typical for
julivora.
The name of this species derives from the Latin julus, a millipede,
and vorare, to devour. The new species is close to the very variable
M. tenuis, but seems constantly distinct from it, even where the two
species occur in intimate sympatry, as they do in the Busu River
tract. For relationship to M. media, see under that species below.
Myopias media new species
(Figs. 6, 23)
Diagnosis, worker: member of tenuis group, very similar to M.
julivora in habitus, color, etc., but larger, head wider, with more
robust and more strongly curved mandibles, the submedian tooth
situated closer to the midlength (at the seventh twelfth from base
along MLO). Antennae long; scapes overreaching posterior border
of head (when held straight back) by nearly twice their apical width.
Worker, holotype: TL 7.6, HL 1.25, HW 1.24 (Cl 99), ML 0.93
(MI 74), MLO 1.26, SL 1.27 (SI 102), EL 0.13, WL 2.20, hind femur
L 1.40, hind tibia L 1.35 mm.
Details additional to diagnosis: Viewed at apparent full-face, pos-
terior border of head feebly convex, almost straight, but even a
slight tilting of the cranium forward yields a concave border, and an
increase in HL to 1 .30, so that from this view. Cl would be about 95.
Anterolateral corners of head more prominent (at a lower level of
focus), so that, excluding eyes, head is widest just behind clypeus.
Median frontal sulcus continuing past mid-HL to include anterior
clypeal pit. Eyes as in M. julivora, but relatively a little larger, and
with remnants of facetting a bit more evident. Median clypeal lobe
very obviously broader than long (CLL 0.13, CLW 0.20), with
258
Psyche
[Vol. 90
weakly concave sides, nearly straight apical margin, and one free
corner rounded, the other rectangular. Mandibles thicker, particu-
larly in the stretch between the obsolescent basal angle and the
submedian tooth, which is also feebly convex mesally (concave or
straight along mesal margin in M. julivora). Labrum toothed as in
M. julivora.
Sculpture as in M. julivora, but small, widely spaced punctures
(diameter 0.01-0.02 mm) are perhaps more distinct on head and
trunk. A small patch of longitudinal costulation lies below spiracle
on side of propodeum (as in M. julivora). Posterior corners of
propodeum less broadly rounded, tending more towards angularity,
both in side and dorsal views, than in M. julivora, and both the
pilosity and pubescence seem to be less copious and a trifle longer.
Color deep brownish red; legs yellowish red; antennae and man-
dibles dark yellowish brown.
Holotype and only known specimen (MCZ) a stray collected from
rotten wood at Joangeng, a village in the Mongi River Watershed of
the Huon Peninsula, Papua New Guinea, at about 1500 m, 7-8
April 1955, in montane rain forest, leg. Wilson, No. 752.
We describe this species with some misgiving because it is based
on a unique, and because it is so similar to M. julivora, especially to
the largest (Baiyer River) specimen of the lattter. The mandibles,
however, differ enough that we feel inclusion of the big Joangeng
specimen in M. julivora would unduly strain the concept of that
species. Further collections will of course help to demonstrate
whether our decision is correct or not. The name media refers to the
size of the body, intermediate in the tenuis group between M. tenuis
and such large forms as loriai and gigas.
Myopias concava new species
(Figs. 4, 18)
Diagnosis, worker and queen: A medium-sized, stout-bodied spe-
cies with head slightly broader than long, widest just behind eyes.
Median labral tooth absent, but an erect apical tooth on each labral
lobe. Eyes of worker large and multifacetted, occupying more than a
quarter of the length of the sides of the head. Posterior margin of
head weakly concave; sides convex. Median lobe of clypeus distinct
but very short, rectangular. Mandibles short and stout, each with 2
1983]
Willey & Brown — Genus Myopias
259
small teeth at apex, 2 large blunt teeth basad of these, and an
obtusely rounded basal angle. Antennal scapes overreaching poste-
rior margin of head. Trunk compact, promesonotum and propo-
deum subequal in length, forming separate weak convexities
meeting at a distinct and depressed metanotal groove. Petiolar node
massive, subcuboidal, broader than long. Anterior face of gastric
segment I weakly concave as seen from dorsal view. Integument
prevailingly smooth and shining, but with abundant, coarse piliger-
ous foveolae, sometimes contiguous on head, and tending to
become elongate on first two gastric terga. Color brownish red.
Worker, holotype: TL 7.1, HL 1.25, HW 1.31 (Cl 105), ML 0.71
(Ml 57), MLO 1.26, SL 1.1 1 (SI 85), EL 0.33, WL 2.16, hind femur
L 1.25, hind tibia L 1.20 mm.
Worker, paratypes (n = 6 of 42 from 4 colonies, including largest
and smallest specimens): TL 6.5-S.6, HL 1.17-1.43, HW 1.21-1.47
(Cl 100-105), ML 0.67-0.81 (Ml 53-66), SL 1.00-1.24 (SI 83-88),
EL 0.30-0.40, WL 2.00-2.46 mm.
Head broader than long, with sides convex, broadest immediately
behind eyes, and narrowed slightly in front of eyes; posterior border
broadly and shallowly concave. (The head can be lengthened
slightly by tilting it forward from the full-face plane; this has the
effect of foreshortening the mandibles and deepening the concavity
of the posterior margin, and of course decreasing CL) Eyes large
and convex, with about 18-19 ommatidia in the longest diagonal
row, each eye occupying nearly 3/10 of the length of its side of the
head, situated about 2/3 its own length from mandibular insertion.
Clypeal lobe distinctly projecting but short, rectangular, more
than twice as broad as long, with parallel sides, a nearly straight
anterior margin, and subrectangular free corners (in Wau Creek
series, anterior margin weakly convex, free corners more rounded).
Labrum with the transverse ridge feebly sinuate in front view, lack-
ing a median tubercle; labral lobes each with a small upturned apical
tooth. Maxillary palpi each 3-merous; basal segment broadest, with
one subapical lateral sensillum; apical segment with a single apical
sensillum. Labial palpi each with 3 subequal segments; basal seg-
ment with 2 adjacent submedian sensilla; II with one subapical lat-
eral sensillum; III with the same, plus 2 apical sensilla.
Mandibles stout, gently bowed, each with two small teeth at apex,
a blunt tooth near apical quarter of ML, a large, blunt submedian
260
Psyche
[Vol. 90
Fig. 4, Myopias concava, head of worker paratype in full-face view. Scale bar is 0.5
mm.
tooth, and a distinct but rounded basal angle. Oblique groove at
base continued as a broad lateral-marginal groove (strix) to apex.
Median frontal sulcus of head extends to or nearly to posterior
quarter of head length. Scapes gently curved, moderately incrassate
apicad, overreaching posterior border by more than their apical
width when head is viewed full-face. Funiculus relatively slender, all
segments longer than broad; apical segments not forming a club;
1983]
Willey & Brown — Genus Myopias
261
pedicel (funiculus 1) longer than II as 4:3.
Trunk robust, with a weakly convex dorsal profile as seen from
the side; propodeum subequal in length to promesonotum; mesono-
tum convex, about half as long as propodeal dorsum, and separated
from it by a distinct but only moderately impressed metanotal
groove. Propodeal dorsum only very feebly convex, passing into
declivity through a rounded obtuse angle. Declivity almost flat, with
bluntly subangular lateral edges, densely punctate in upper 2/5,
smooth and shining below this.
Petiolar node massive, subcuboidal, slightly higher and broader
behind than it is long (disregarding sternital keel); front and rear
faces flat, vertical, dorsal face gently convex and sloping slightly
anteriad. Sternite forming a sharp, recurved (hooklike) anterior
subpetiolar process, followed by a short concavity and then by a
long, low, feebly convex keel.
Postpetiolar segment (gaster I) wider than long (roughly about as
4:3) and very slightly wider than gaster II; anterior face abruptly
vertical, its dorsal margin gently concave as seen from above. Gaster
II (ignoring acrotergite normally covered by gaster I) longer than I,
but still not quite as long as wide. In side view, these two segments
are about equally high. Apical gastric segments short, as usual; sting
very long (and is found extended up to 1.1 mm in some paratype
workers), gently upcurved. Gonostylus (in paratypes) long,
2-merous.
Body basically smooth and shining, but sown with deep, conspic-
uous, piligerous foveolae, mostly round or oval on the head (here
0.03 to 0.09 mm in diameter), trunk and petiole, becoming more
elongate axially on first two gastric terga. Foveolae on head smaller
and more crowded, forming oblique chains interspersed with costu-
lae between eyes and frontal lobes, but those caudad of eyes larger,
forming vague, oblique chains, separated on the average by their
diameters near the cephalic midline, but smaller and more crowded.,
often subcontiguous laterad and caudad. Foveolae more widely
spaced on trunk and petiole, especially near midline and on sides of
pronotum; metapleura with a few coarse longitudinal-oblique cos-
tae. Petiole and postpetiole (first gastric segment) with smaller,
crowded foveolae on sides and ventrad, but on second gastric seg-
ment, the foveolae become very sparse apicad and ventrad, the sur-
faces here virtually smooth, except for a crowded band of small
foveolae along the apical margin. Apex of gaster, antennal scapes.
262
Psyche
[Vol. 90
Figs. 5-7, Myopias spp., heads of workers in full-face view, sculpture and pilosity
omitted. Fig. 5, M. julivora paratype. Fig. 6, M. media holotype. Fig. 7, M. ruthae
holotype. All to same scale; bar is 0.5 mm.
mandibles and legs prevailingly smooth, with spaced piligerous
punctures.
Body and appendages clothed with numerous fine, tapered,
decumbent to subdecumbent hairs, mostly each issuing from a
foveola, and nearly all 0.10 to 0.25 mm long (up to 0.30 mm on
anterior clypeal lobe).
Color rich, deep brownish red; legs a little lighter reddish.
Worker variation; apart from size, mainly involves slight differ-
1983]
Willey & Brown — Genus Myopias
263
ences among nest series in the shape of the median clypeal lobe
(convex vs. straight apical margins), density and size of individual
foveolae of sculpture, and depth of coloration.
Queen, dealate, from holotype nest series, Wamuki: TL 8.2,
HL 1.37, HW 1.50 (Cl 109), ML 0.80 (Ml 58), SL 1.20 (SI 80),
EL 0.45, WL 2.61. Four additional queens range downwards in size
slightly from this (Collection Nos. 887 (n = 3) and 990 (n = 1) from
Busu River, the smallest having HW 1.33. A female from the Wau
Creek series is ergatoid, but has HW about 1.50; this specimen lacks
ocelli, but has small, blackened forewing stumps. The queens
resemble the workers except in the caste difference usual for
ponerines.
Male unknown.
Described from material from four separate nest series, all from
what is now Papua New Guinea: holotype from Wamuki, 800 m, on
the Mongi River watershed, Huon Peninsula, 19-20 April 1955
(Wilson No. 844; MCZ). No. 844, a colony containing one queen
and about 20 workers, was taken from a Zoraptera-stage rotten log
in hill rain forest. Two colonies came from the area between the
lower Busu and Bupu rivers, near Lae, at the base of the Huon
Peninsula, in lowland rain forest (Wilson Nos. 887 and 990). No.
887 was a nest in a small Passalus-siSLge log, 28 April 1955, and
included at least three queens. No. 990 was in a small (10 cm
diameter) rotten log with interior crumbling, but bark intact. It held
50-60 workers, two queens, eggs, larvae up to half-grown (no larger
larvae) and one cocoon. The brood chamber contained an unidenti-
fied insect larva, also an adult (cucujoid?) beetle that was still alive
and feebly moving; this beetle could possibly have fallen or walked
in during excavation of the nest. (Unfortunately, the residues from
Wilson’s collections were eventually lost.)
The fourth collection comes from Wau Creek, at about 1200 m
elevation in a “Stage HI” [rotten] log (leg. D.H., A.C. and A.H.
Kistner, No. 1213); it contained at least 10 workers and a more or
less ergatoid queen.
This very distinct species shows some affinities with the tenuis
group in the presence of upturned teeth on the labral lobes and lack
of median labral tooth, but it is different in its robust build, very
prominent foveolate sculpture, shorter mandibles, and the concave
anterior face of the first gastric tergum, which gives the name
concava.
264
Psyche
[Vol. 90
Myopias chapmani new species
(Figs. 10, 26)
Diagnosis, worker: A modest-sized member of the tenuis goup;
head large, nearly square, with sides almost straight and nearly
parallel; posterior margin concave. Eyes small but distinctly facet-
ted. Mandibles rather short and stout; antennal scapes distinctly
overreaching posterior border of head. Trunk robust, with broad
and deeply impressed metanotal groove; propodeal dorsum less
than twice as long as mesonotum. Node of petiole higher and wider
than long, convex above. Gaster distinctly constricted between first
and second segments. Sculpture predominantly smooth and shining,
with spaced, indistinct punctures, especially on head, but sides of
propodeum obliquely costulate, subopaque; dorsal propodeal sur-
face finely roughened in part, and bearing a few, coarse, indistinct
grooves and punctures, as well as a weak impression just caudad of
its midlength. Color light ferruginous red.
Worker, holotype: TL 5.7, HL 1.12, HW 1.07 (Cl 96), ML 0.57
(MI 51), MLO 0.87, SL 0.95 (SI 89), EL 0.1 1, WL 1.77, hind femur
L 1.06, hind tibia L 1.00 mm.
Worker paratypes (21 from type nest series) range downward
from the size of the holotype to the smallest individual, which has
TL 5.2, HL 1.03, HW 0.98 (Cl 95), ML 0.54 (MI 52), MLO 0.81,
SL 0.86 (SI 88), EL 0. 1 1 , WL 1 .63, hind femur L 0.94, hind tibia L
0.86 mm.
Head massive, sides only feebly convex, widest at posterior edge
of eyes and tapering almost imperceptibly to rather abruptly
rounded posterior corners; posterior border moderately concave
across its middle half. Eyes almost round, with about 17 or 18
facets, each situated nearly twice its own diameter away from man-
dibular insertion; an indistinct groove extends the dorsal (mesal)
orbital groove forward onto clypeal wing. Median clypeal lobe
slightly longer than wide (CLL 0.14, CLW 0.13 mm), widest near
apex, its anterior margin convex and free corners rounded. Median
frontal sulcus wide and deep, extending back to posterior third of
HL.
Antennal scapes slender, gently bowed, slightly incrassate apicad,
overreaching the posterior border of the head by about the same as
their apical width when held straight back in dorsal full-face view of
head. Funiculus slender, but with an indistinctly 4-merous club; all
1983]
Willey & Brown — Genus Myopias
265
antennomeres longer than wide; funiculus I about twice as long as II.
Mandibles robust, rather short, with a sharp apical tooth and a
minute adjacent companion tooth; one subapical and one subme-
dian tooth each isolated, blackened and rounded; basal angle pres-
ent, but low and rounded. Oblique basal groove and its lateral
continuation very distinct. Labral lobes each bearing a delicate,
upturned apical tooth, practically impossible to see without dissec-
tion. Palpi segmented 3,3; basal maxillary palpomere short and
broad, last two subequal in length, but apical broader, fusiform,
with apical sensillum; labial palpomeres all slender, the apical
slightly longer and thicker than the basal two, and with an apical
sensillum.
Trunk robust, divided by a broad and deeply impressed metano-
tal groove into a promesonotal portion and a shorter propodeal
portion. Mesonotum convex, rising above pronotum, sloping cau-
dad, nearly 2/3 as long as propodeal dorsum; propodeum weakly
convex, but with a feebly impressed area in the posterior half of its
dorsum (variably distinct in paratypes); dorsum rounded unevenly
into declivity and with a feeble median impression at the point
where they meet; declivity more or less flat, with lateral boundaries
distinct, almost submarginate.
Petiolar node distinctly higher than long, its curved dorsal face
highest behind the midlength; anterior face in side view straight or
feebly concave, sloping caudad; posterior face convex in side view,
sloping cephalad. In dorsal view, anterior cornuae of node very
prominent; node widest behind, with convex sides, slightly wider
than long. Postpetiolar (gastric I) segment slightly broader than
long; its anterodorsal border feebly concave in the middle; gastric II
a little wider than I, but equal in depth in side view after a distinct
constriction between the two that is boldly scrobiculate. Sting long
and sharp, gently upcurved (found extended up to 0.60 mm in var-
ious specimens.
Sculpture mainly smooth and shining, with mostly inconspicu-
ous, separated, piligerous punctures, distributed as follows: on dor-
sum of head, on each side of midline, numerous small ones,
averaging about 0.01 mm in diameter, or smaller, mostly in the
space between eye and median sulcus; small punctures distributed
sparsely on mandibles, back and sides of head, fore coxae, prono-
tum, mesonotum, and gastric tergum II. Moderately coarse, often
266
Psyche
[Vol. 90
elongate, punctures on propodeal dorsum, petiolar node and first
gastric tergum. Antennae and legs largely smooth and shining, but
with very fine punctulation, increasing toward extremities. Sides of
propodeum with fine, partly broken, oblique costulation, rising
caudad, surface here subopaque, giving way on dorsal surface to
some partial, roughened microsculpture that renders the surface
between coarse punctures only weakly shining. Upper propodeal
declivity feebly, finely, transversely strigulose, smooth and shining
below.
Pilosity consisting of fine, tapered, erect to suberect setae of
uneven length, mostly 0.03 to 0.25 mm long, distributed abundantly
over dorsal surfaces of body, venter of head, and gaster, fore coxae,
and most surfaces of appendages. Decumbent pubescence is dilute
on dorsum of head, directed mesad; more dense on anterior surfaces
of mid coxae, and on all tibiae and tarsi.
Color rich, light, ferruginous red; appendages slightly lighter.
Worker variation, apart from the slight mensurable spread, is
weak. As already mentioned, the feeble impression, or “saddle,” in
the posterior dorsal surface of the propodeum varies from distinct
to almost absent in different workers.
Queen, dealate: TL 5.2, HL 1.00, HW 1.00 (Cl 100), ML 0.55
(MI 55), MLO 0.77, SL 0.84 (SI 84), EL 0.23, WL 1 .65, hind tibia L
0.81 mm. Notable for size being slightly smaller than for workers of
the same colony. Otherwise, differences are those usual between
castes in Ponerini. Nota of pterothorax smooth and shining, with
dispersed, small punctures. Propodeum more completely and
strongly sculptured than in worker, subopaque, finely transversely
strigulose, with a short, longitudinal, median sulcus or impression;
lower declivity smooth. Color slightly darker than in worker, espe-
cially lightly infuscated parts of cranium, pronotum, median scu-
tum, propodeum, petiole, and first two gastric segments.
Male unknown.
Holotype worker (MCZ) and paratypes (MCZ, BMNH, ANIC)
from a small nest in a thick fragment of a rotten branch lying on the
ground in wet rain forest along Obi Obi Creek, below and just west
of Montville, Blackall Range, Queensland, Australia, 20 May, 1951,
leg. Brown. The nest contained 20-30 workers, larvae (since lost
together with prey remains), and two dealate queens. The forest at
the type locality has since been destroyed (fide P.J. Darlington,
1983]
Willey & Brown — Genus Myopias
267
Figs. 8-11, Myopias spp., heads in full-face view, sculpture and pilosity omitted.
Fig. 8, M. delta, paratype worker. Fig. 9, M. nops, holotype worker. Fig. 10, M.
chapmani, paratype worker. Fig. 1 1, A/, densesticta, paratype queen from Kuranda,
Queensland. All to same scale.
268
Psyche
[Vol. 90
personal communication). The species is named for the late Dr.
James W. Chapman, who collected many Myopias series in the
Philippines.
Myopias densesticta, new species
(Figs. 11,29)
Diagnosis, worker and queen: A member of the M. tenuis group,
similar to M. chapmani, but with much more distinct and abundant
foveolate sculpture and a shorter, wider median clypeal lobe and
shorter antennae. Also, the trunk is not deeply divided at the
metanotal groove, the petiolar node is more massive, and the gaster
is gently tapered, not sharply constricted, behind the first segment.
Worker, holotype: TL 5.6, HL 1.03, HW 0.97 (Cl 94), ML 0.55
(MI 53), MLO 0.82, SL 0.85 (SI 88), EL 0.09, WL 1 .62, hind femur
L 0.87, hind tibia L 0.83 mm.
Worker, paratypes (n = 3 of 9 from two colonies, including largest
and smallest, the holotype): TL 5.6-5.7, HL 1.03-1.07, HW
0.97-1.03 (Cl 94-97), ML 0.55-0.56 (MI 51-54), MLO 0.82-0.83,
SL 0.85-0.86 (SI 83-88), EL 0.09-0.1 1, WL 1.62-1.71, hind femur L
0.87-0.89, hind tibia L 0.83-0.84 mm.
In overall size, proportions of head, and mandibles, this species is
very similar to M. chapmani, but the sides of the head are a trifle
more convex, and the basal angle of the mandible is a little less
distinct; also the following, more definite differences from M.
chapmani:
(1) Antennae shorter; scapes overreach posterior border when held
straight back by only a slight amount, less than their apical width.
Segments II through VIII of funiculus wider than long; I (pedicel)
more than twice as long as II.
(2) Median clypeal lobe shorter, wider (CLL 0.08-0.10, CLW
0.16-0.17 mm), with sharply angular free corners terminating the
divergent carinae that form the lateral edges of the lobe.
(3) Promesonotum shorter than propodeum; propodeal dorsum
about twice as long as mesonotum, and nearly on the same level;
both only weakly convex and meeting at a distinct but not deeply
impressed metanotal groove, so that the side-view dorsal profile is a
nearly smooth, gently convex outline from top of front pronotal
incline to top of propodeal declivity.
(4) Petiolar node more massive and more nearly cuboidal, less no-
1983]
Willey & Brown — Genus Myopias
269
tably longer than high; seen from above wider than long, but sides
convex; widest near midlength.
(5) Second gastric (true abdominal IV) segment narrower and lower
than first segment (postpetiole), so that the gaster is gradually
tapered caudad of I, and not constricted and recovering after.
(6) Body, especially head, trunk and petiolar node, with deeper and
much more distinctly developed foveolate sculpture, the punctures
mostly 0.01-0.02 mm in diameter and densely crowded, contiguous
on front of head between eyes and frontal lobes, becoming larger,
mostly 0.02-0.03 mm in diameter and narrowly separated on poste-
rior half of head, at times with intervening, indistinct, longitudinal
strigulosity, and still coarser and more widely spaced on sides and
underside of head and near median frontal sulcus (which reaches
back to near the posterior quarter of HL). Trunk and petiole with
abundant foveolae, mostly 0.02-0.04 mm in diameter, separated on
the average by a little more than their own diameters, but more
crowded and more elongate on sublateral strips of propodeal dor-
sum; truncal midline strip partly open, with few foveolae. In gen-
eral, interfoveolar surfaces smooth and shining, but lower sides of
propodeum indistinctly, longitudinally costulate, and sides of petio-
lar node coarsely and densely foveolate and minutely roughened,
more or less opaque. Gaster I smooth, with scattered coarse punc-
tures, and these become fewer and smaller still on gaster II. Mandi-
bles smooth and shining, with scattered punctures. Antennal scapes
and legs smooth and shining, but with fine punctulation. Propodeal
declivity nearly smooth, but peppered with many small foveolae.
Clypeus smooth and shining.
(7) Pilosity and pubescence more abundant than in M. chapmani,
most notably on mandibles and antennal scapes; decumbent pubes-
cence on head more conspicuous, directed mesad.
(8) Color perhaps averaging slightly darker than in M. chapmani,
but legs and antennae tending to be lighter, more yellowish red. As
in chapmani, the palpi are segmented 3,3, and upturned teeth are
present, one on each labral lobe. Worker variation is very slight
overall. The Koombooloomba series averages very slightly larger,
and the compound eyes may be a trifle larger than in the Shipton’s
Flat colony.
Queen, dealate, a unique taken in rotten wood in a rain forest
patch near Kuranda, Queensland, 31 October 1950, leg. Brown: TL
270
Psyche
[Vol. 90
6.7, HL 1.07, HW 1.05 (Cl 98), ML 0.52 (Ml 49), MLO 0.81, SL
0.85 (SI 81), EL 0.24, WL 1.92 mm.
Male unknown.
Holotype [MCZ] one of six workers from Shipton’s Flat, south of
Cooktown, Queensland, during June 1958, leg. P.F. and P.J. Dar-
lington. This locality is savannah woodland grading into riparian
rain forest (gallery forest), and lies at an elevation of about 300 m.
(The Kuranda queen is from a similar elevation.) A pin of three
workers comes from Koombooloomba, near the dam of the same
name south of Ravenshoe, Queensland, at about 750 m in rain
forest, “4/7/71,” leg. Taylor and J. Feehan. We have no informa-
tion about possible prey.
The name densesticta refers to the characteristic foveolate
sculpture.
Myopias tasmaniensis
(Fig. 16)
Myopias tasmaniensis Wheeler, 1923, Psyche 30:177-179, fig. I, worker. Type loc:
Hobart, Tasmania.
Trapeziopelta tasmaniensis: Brown, 1953, Psyche 60:51, records from Dandenong
Range, Victoria, Australia.
Trapeziopelta diadela Clark, 1934: Mem, Nat. Mus,, Melbourne, Australia, 8:54-55,
pi. 4, f., 7,8, worker, queen. Type loc., Turton’s Track, Otway Ranges, Victoria.
(Syn, by Brown, 1953).
Two collections made by Father Bede Lowery extend the range
far to the north in eastern Australia; Minnamurra Falls, near
Kiama, New South Wales, nest in soil of very moist forest, behind
rock slab set in a vertical bank, 22 Dec. 1959; Cunningham’s Gap,
southeastern Queensland, at about 600 m in rain forest,. 22 Jan.
1961.
Myopias tenuis new combination
(Figs. 15, 17)
Trapeiiopelta tenuis Emery, 1900, Termeszetr, Fliz. 23:313-314, dealate queen. Type
loc.: Beliao Island, near Berlinhafen (now Aitape), Papua New Guinea. 1902, 155,
worker, Sattelberg, Huon Peninsula, Papua New Guinea.
Trapeziopelta tenuis \ar. fulvescens Emery, 1902:155, worker, dealate queen. Type
loc,: Sattelberg, Huon Peninsula, Papua New Guinea. New synonym.
This is the smallest of the known Melanesian Myopias species,
and also the most common and widespread. The typical form is
1983]
Willey & Brown — Genus Myopias
271
Figs. 12-19, Myopias spp., petiolar nodes in lateral and dorsal view, sculpture and
pilosity omitted, all to same scale. Fig. 12, M. gigas holotype worker. Fig. 13, M.
loriai worker fro Gemeheng, Huon Peninsula, Papua New Guinea. Fig. 14, M.
cribriceps worker from Bubia, near Lae, Papua New Guinea. Fig. 15, M. tenuis
queen from Bubia, near Lae, Papua New Guinea. Fig. 16, M. tasmaniensis from
Olinda, Victoria, Australia. Fig. 17, M. tenuis worker from Bubia, near Lae, Papua
New Guinea. Fig. 18, M. concava paratype worker from lower Busu R., near Lae,
Papua New Guinea. Fig. 19, M. levigata worker from Nganduo, Huon Peninsula,
Papua New Guinea.
272
Psyche
[Vol. 90
small, has slightly convex sides of the head, scapes that just fail to
reach (or just barely reach) the occipital border when held straight
back, and a median clypeal lobe that is as long as, or slightly longer
than, wide at its widest (near apex). The end of the lobe is convex or
straight, and the free angles may be rectangular or rounded. Mea-
surements for Papua New Guinea North Coast workers are: TL
2.S-3.7, HL 0.53-0.71, HW 0.45-0.60 (Cl 81-91), ML 0.30-0.43
(MI 81-92), MLO 0.40-0.57, SL 0.42-0.58 (SI 87-95), EL 0.03-0.06,
WL 0.95-1 .25 mm. Workers of a colony series from Salawati Island,
at the western end of New Guinea, fall within these dimensions and
proportions. Workers from Bisianumu, in the hills above Port
Moresby, fit the North Coast dimension range, while a sample from
Karema, in the lowlands north of Moresby, tends slightly to exceed
the North Coast samples in size.
Samples from the Cape York area of North Queensland average
larger than any of the New Guinea series; a large worker from the
Black Mt. Road, north of Kuranda, measures TL 4. 1 , HL 0.74, H W
0.67 (Cl 91), ML 0.43 (MI 91), MLO 0.58, SL 0.59 (SI 88), EL 0.04,
WL 1.34 mm. The Australian samples often have the laid-back
scapes reaching the posterior border of the head, and the posterior
border is more distinctly concave. In addition, the median clypeal
lobe tends to be wider, often as wide as or wider than long, and the
minute punctures, especially on the head, are a trifle coarser and
more distinct. Several of these series have sordid yellowish individu-
als, undoubtedly partly callow, that correspond to Mdcc.fulvescem.
New locality records: papua new guinea: Karema, Brown R.,
rotten log, lowland rain forest, leg. Wilson, No. 552. Bisianumu,
near Sogeri, about 500 m, hill rain forest, Wilson Nos. 637, 637A,
litter and rotten wood, strays. In the vicinity of Lae (Didiman
Creek, Bubia and lower Busu R.), several nests and litter strays,
Wilson Nos. 688, 689A, 690, 716, 939, 962, 978, 1037, 1045, 1058, all
in lowland rain forest. No. 689 was a small colony in a Zoraptera-
stage log, with about ten workers and two queens. No. 716 was a
worker carrying an entomobryid collembolan about its own length
lengthwise beneath its body, army ant fashion. No. 1037 was a nest
in a cavity in the under surface of a hard, barkless log in leaf litter.
No. 1045, a nest in a soft Passalus-s\?igQ log, had peripheral galleries
packed with unidentified arthropodan cuticular fragments. Nadzab,
dry evergreen forest, Wilson No. 1 100. Wau north, on Bulolo road.
1983]
Willey & Brown — Genus Myopias
273
650 m, leg. S. Peck, B-278. irian jaya: near Phillips Petroleum Base
Camp, SE Salawati 1. (just off western extremity of Vogelkop),
swamp forest near sea level, leg. Brown No. 81-189, nest in rotten
wood, with at least 15 workers, a dealate queen, a male, about a
dozen pupae in tan cocoons, and a few half-grown larvae. Austra-
lia, N. QUEENSLAND: Black Mt. road N. of Kuranda, 300-600 m, leg.
P.F. Darlington, in rain forest, small colony with at least two deal-
ate queens. Mt. Cudmore Range, 1 1 mi. N. of Ingham, about 210 m,
six workers from rotten log in small roadside patch of disturbed rain
forest, leg. Taylor, Acc. No. 1706. Mulgrave Forestry Road,
17° 18'S, 145°48'E, leg. Ward No. 4366, from rotting epiphyte fern
on rain forest floor.
From the Solomon islands we have three scanty samples of
forms sent from anic that could belong to M. tenuis, or to sibling
species:
(1) A large form, extending some of the tendencies seen in Aus-
tralian series; HW 0.80, EL 0.09 mm; scapes reaching posterior
border of head. Propodeal dorsal profile a little more convex than
usual in M. tenuis. Color castaneous. Two workers from Guadal-
canal I.: Mt. Austen, Feb. 1966, leg. P.M. Greenslade, No. 21095.
(2) A small worker, also from Mt. Austen, Guadalcanal,
14/5/ 1963, leg. P.M. Greenslade, No. 6076; HW 0.55, EL 0.03 mm;
scapes very short, failing to reach posterior border of head by at
least the apical scape width; posterior border of head weakly con-
cave. Color yellowish brown. Sides of head straighter and more
parallel-sided than in the other Mt. Austen sample.
(3) A worker from San Cristoval L, Humi R. est., N.E. Wainoni,
leg. Royal Society Expedition, 1966-1, HW 0.60, EL 0.05 mm; sides
of head almost perfectly straight and parallel, posterior border fee-
bly convex; scape fails to reach posterior border of head by about
half of apical scape width; mandibles unusually short (ML 0.37,
MLO 0.47 mm) and broad; basal angle forming a distinct convexity.
Color deep brownish red.
I suspect that the Solomons will eventually yield much more vari-
ation in the tenuis complex; the available material is simply inade-
quate as a basis for understanding the complex in this archipelago.
274
Psyche
[Vol. 90
Key to Known Myopias Species of Australia,
Based on Workers
1 . Small, slender species (H W < 0.75 mm); worker compound eye
reduced to a single smoothly lenticular dot < 0.05 mm long (in
Australia, N. Queensland) tenuis
More robust species (HW > 0.75 mm); worker compound eye
0.05 or more long, with 3-5 rows of distinct, raised ommatidia
(Figs. 10, 11) 2
2. Second gastric segment lower and narrower than the first and
tapering gently apicad, not constricted at base (N. Queensland)
densesticta
Second gastric segment wider than the first and sharply con-
stricted from its own acrotergite basad 3
3. Antennae shorter, scapes not overreaching posterior border of
head when they are held straight back from insertions as head is
viewed full-face; small funicular segments (funiculus II-VII at
least) broader than long (Tasmania to SE Queensland)
tasmaniensis
Antennae longer, scapes distinctly overreaching posterior
border in full-face view of head; all antennal segments longer
than broad, or at least as long as broad
(S. Queensland: Blackall Range) chapmani
Myopias ruthae new species
(Figs. 7, 20)
Diagnosis, worker: A modest-sized Myopias; head longer than
broad, with nearly parallel but gently convex sides and weakly con-
vex posterior border; scapes curved, barely surpassing posterior
border; funicular club indistinctly 6-merous. Median frontal groove
deep and wide, reaching to the posterior quarter of head length.
Median clypeal lobe broad, short, rectangular, widest basad.
Labrum without a median tubercle. Eyes fairly large, convex, finely
facetted. Mandibles robust, gently curved, with 4 teeth and a low
basal angle. Body robust, metanotal groove distinct but weakly
impressed. Petiolar node massive, subcuboidal; gaster short and
thick. Sculpture of numerous coarse punctures or foveolae, dense
and contiguous, or nearly so on most of head and sides of petiolar
node; foveolae sparser mesad on vertex, trunk and succeeding terga.
1983]
Willey & Brown — Genus Myopias
275
the surface here prevailingly smooth and shining. Color piceous,
nearly black, with contrasting tan appendages.
Worker, holotype: TL 5.2, HL 0.96, HW 0.86 (Cl 90), ML 0.48
(MI 50), SL 0.73 (SI 85), EL 0. 14, WL 1 .60, hind femur L 0.80, hind
tibia L 0.83 mm.
Description mainly directed at details not fully covered in the
diagnosis and figures. Antennal scape broadly curved in basal half,
incrassate distad; apical width about 0.12 mm, or slightly less than
maximum eye length. Funicular segment I about twice the length of
II. Eye with distinct but fine ommatidia, numbering about 1 1 or 12
units in the longest diagonal row, darkly pigmented. Median clypeal
lobe about 0.05 mm long (CLL) and about 0.14 wide (CLW) at
apex, about 0.15 mm wide at base where it meets frontal lobes;
anterior border straight, free corners subrectangular. In examining
the single, intact specimen, no upturned teeth could be seen at the
apex of each labral lobe, but a dissection would be needed to make
sure that they are really absent. Mandibular armament consists of
an acute apical tooth and a small adjacent tooth, then after a gap
another large, blunt tooth, another gap and a similar-sized but more
acute median tooth, then halfway from this to the base, a low,
rounded basal angle. Oblique groove (strix) near dorsal base of
mandible distinct, continuing along lateral margin to near apex.
MLO 0.77 mm.
Trunk robust, dorsal outline in side view nearly straight, the
mesonotum feebly sunken; metanotal groove slightly impressed, but
distinct; propodeal dorsum very feebly convex, rounding obtusely
into declivity, but sides of declivity forming blunt angles with pleu-
ral faces of trunk. Propodeal spiracle small and round, situated at
mid height. Petiolar node massive, subcuboidal, slightly wider
behind than long; slightly higher than long if one ignores the small,
hooklike anterior subpetiolar process; dorsal surface convex in both
directions.
First gastric (postpetiolar) segment higher and wider (by about
4:3) than long. Succeeding (gastric II) segment about as wide as the
first, and only slightly longer, but slightly thinner dorsoventrally.
Sting long and slender, gently upcurved.
Sculpture distinctive, consisting basically of a smooth, shining
integument invaded by coarse, mostly umbilicate, piligerous foveo-
lae. The foveolae are densest and smallest (0.02-0.03 mm diameter)
276
Psyche
[Vol. 90
26 CHAPMANI
29 DENSESTICTA ¥
Figs. 20-29, Myopias spp., petiolar nodes in lateral and dorsal view, sculpture and
pilosity omitted, all to same scale. Fig. 20, M. ruthae holotype worker. Fig. 21,
M. delta, paratype worker. Fig. 22, M. Julivora, paratype worker from Papua New
Guinea. Fig. 23, M. media, holotype worker. Fig. 24, M. latinoda queen from Maffin
Bay, Irian Jaya. Fig. 25, M. lobosa, paratype worker. Fig. 26, M. chapmani, para-
type worker. Fig. 27, M. latinoda worker from Maffin Bay, Irian Jaya. Fig. 28, M.
nops, holotype worker. Fig. 29, M. densestieta paratype queen from Kuranda,
Queensland.
1983]
Willey & Brown — Genus Myopias
111
on anterior and sides of head, where most are contiguous and yield a
reticulate-foveolate surface that is subopaque in most lights. This
kind of sculpture, a bit more loosely distributed, covers the upper
sides and dorsum of trunk and petiole, except for median posterior
part of vertex, midline of trunk, and dorsal midline of petiole, which
have wide spaces free of most foveolae, and are smooth, shining.
Sides of trunk below largely smooth, with sparse foveolae, and
posteriorly, low down, with a few fine, longitudinal costulae. Sides
of petiolar node foveolate-striate. Caster with spaced foveolae,
becoming smaller (0.02 mm) and sparser caudad, interspaces
smooth and shining, but a double band of foveolae along apical
margin of second gastric tergum. Mandibles and legs with sparse
punctures, generally otherwise smooth and shining; scapes and
middle tibiae, and all tarsi, more densely punctulate, but still
shining.
Hairs numerous, fine, tapered, suberect to decumbent, mostly
0.04 to 0.25 mm long; those on head and appendages mostly short,
while those on clypeal lobe, trunk, petiole, and especially gastric
apex are longer.
The species is named for Dr. Ruth Lippitt Willey.
Holotype (MCZ) a unique worker specimen from Bubia, about 13
km NW of Lae, Papua New Guinea, about 20 m above sea level, in
high-graded rainforest, 26 March 1955, by E.O. Wilson (MCZ). The
worker was foraging under the bark of a large Zoraptera-stage (rot-
ting) log.
This species is distinct from all congeners, but difficult to place to
a group. Probably it comes closest to the tenuis group than any
other so far described, but the longish head, bulging eyes, short
scapes and coarsely foveolate sculpture will distinguish it from all
tenuis-gxoup species.
Myopias lobosa new species
(Figs. 2, 3, 25)
Diagnosis, worker and queen: Head distinctly longer than broad;
median clypeal process obsolete; labrum without median tubercle.
Mandibles much broadened, each with the two major teeth before
the apex exaggerated into triangular lobes; blades of mandibles
sharply curved ventrad and rotated so that their blades lie nearly
parallel to the sagittal plane of the head at full closure. Antennal
278
Psyche
[Vol. 90
scapes very short. Sculpture composed of distinct punctures, fine
and densely arranged on cephalic dorsum, with smooth to finely
shagreened interspaces.
Worker, holotype and six paratypes from type nest series: TL
6.0-6.3, HL 1.27-1.31, HW 0.98-1.04 (Cl 76-82), ML 0.76-0.78
(MI 75-83), SL 0.74-0.77 (SI 75-80), EL 0.22-0.25, WL 2.01-2.09
mm.
Head with gently convex subparallel sides, straight to feebly con-
vex occipital border, and broadly rounded posterior angles. Grea-
test head width slightly behind midlength. Eyes oval, only feebly
convex, their greatest diameter greather than maximum width of
scape; about 12-15 fine facets in rows across the short axis; eye
separated from anterior corner of head by about 2/3 its own length
or slightly more. Frontal sulcus extends slightly beyond midlength
of head. Median clypeal process obsolete, represented only by a
weak convexity between the frontal lobes with two or three minute
piligerous tubercles. Form of mandibles shown in Figs. 2, 3. Scapes
reaching roughly to about 3/4 the distance between their insertions
and the occipital border; segment I of funiculus distinctly longer
than II; II through X increasing rather uniformly in size, so that no
club is formed.
Trunk as seen from side forming gentle, subequal promesonotal
and propodeal convexities, separated by the impressed metanotal
groove. Petiolar node slightly higher than long seen from the side
(without subpetiolar process) and very slightly longer than broad
seen from above. Postpetiolar segment seen from above about 4/5
as long as broad, slightly shorter than the succeeding segment.
Dorsum of head densely sown with fine, uneven-sized punctures,
close together, but interspaces mostly smooth and shining; region
posteromesad of compound eyes with densest punctation. Sides and
underside of head smooth and shining and with scattered coarse
punctures. Mandibles, gastric apex, scapes and legs smooth and
shining.
Trunk, petiole, and first two gastric segments with numerous
coarse, mostly elongate piligerous punctures. Between punctures the
integument is mostly smooth to shagreened on the dorsum, becom-
ing striolate and subopaque on sides and rear of trunk and petiolar
node, and to some extent on mesonotum and propodeal dorsum.
Pilosity of fine, tapered, erect hairs, moderate in length and gen-
erally distributed. Head with abundant, mesally directed decumbent
1983]
Willey & Brown — Genus Myopias
279
pubescence over most of the dorsum; scapes and legs with similar
pubescence, directed apicad. Color bright to deep brownish-red;
mandibles and appendages lighter, more yellowish.
Holotype (MCZ) a worker from a uninidal series of 7 workers
and a queen from the Cuernos Mts., near Dumaguete, Negros, Philip-
pine Islands (Chapman). Paratype workers from the same colony
(MCZ, BMNH).
Queen, dealate: slightly larger than the worker. Flight sclerites
and wing stumps well developed. Petiolar nodes slightly broader
than long as seen from above. Mesonotum and propodeum sha-
greened (finely reticulate) above and with scattered coarse punc-
tures. Ocelli small. Compound eyes large; maximum diameter ca.
1/5 the full head length (HL).
Myopias nops new species
(Figs. 9, 28)
Diagnosis, worker: A modest-sized, depigmented (dull yellowish)
species without eyes; sculpture opaque, predominantly densely
reticulate-punctulate over head, trunk, node and first gastric seg-
ment. Mandibles short and stout, with basal angle distinct, but
rounded and close to submedian tooth. Median clypeal lobe dis-
tinct, short with subacute free corners and indented apical margin.
Antennal scapes just reaching posterior border of head. Petiolar
node thick, but tapered apicad, its sternal keel ending behind in an
abrupt angle, paired bilaterally as in Ponera.
Worker, holotype: TL 4.4, HL 0.85, HW 0.77 (Cl 91), ML 0.42
(Ml 49), MLO 0.59, SL 0.65 (SI 84), WL 1.30, petiole L 0.45, hind
femur L 0.65, hind tibia L 0.64 mm.
The two paratype workers, both dismembered, hardly differ from
the holotype by more than the usual error of measurement in the
standard dimensions. Since the three specimens of the type series are
all incomplete (one lacking head, another without gaster, various
legs missing, sculpture in part obscured on holotype), the descrip-
tion is composite.
Head oblong with nearly parallel, weakly convex sides, greatest
width a little way anterior to midlength; posterior corners rounded;
posterior border straight in full-face view, or perhaps just the slight-
est bit concave. Median frontal sulcus broad but short, not reaching
back to mid-HL. Eyes obsolete, or at least unpigmented and not
280
Psyche
[Vol. 90
distinguishable amid the sculpture in strong light at SOX. Antennae
with robust scapes that just reach the posterior border of the head
when held straight back; funiculus long, with an indistinctly 4-
merous apical club; funicular segments II through V short, wider
than long; pedicel (I) is 3-4X as long as II; VI-XI longer than wide.
Mandibles short and stout, strongly downcurved, with acute api-
cal tooth and blunt companion tooth, followed after long gaps by
two blunt teeth, of which the submedian is followed closely basad by
a distinct but rounded basal angle. Labrum without a distinct
median tooth, but the two lobes each bear a delicate, upturned tooth
at apex. Palpal segmentation not determined. Median clypeal lobe
distinct but short, CLL 0.08, CLW 0.12, with indented or concave
apical margin and subacute free corners; one side deformed in
holotype.
Trunk compact, with a weakly convex dorsal outline (in side
view) between steeply sloping pronotal and propodeal declivities;
promesonotum distinctly longer than propodeum; mesonotum
weakly convex; metanotal groove strong, but only moderately
impressed; propodeal dorsum feebly convex overall, but with a very
shallow impression near midlength. Position of metapleural suture
indicated by a vague sulcus. Propodeal declivity rather abruptly
rounded off from dorsum, weakly transversely aciculate above,
smooth and shining below, meeting sides of propodeum through
bluntly subrectangular curves. Lengths of propodeal dorsum: meso-
notum about as 5:3.
Petiolar node thick but higher and broader than long; summit
anterior, dorsal face rounded, but meeting steep concave anterior
face through an abrupt curve, rounding broadly caudad into poste-
rior face, which is low, flat and smooth. Ventral keel of petiole with
a large, obliquely truncate process in front and another, lower, rec-
tangular or obtuse angle farther caudad; this last angle is paired
bilaterally with a mate, and together they appear to be homologous
with the similar teeth or angles diagnostic of the genus Ponera.
Gaster robust; constriction behind first segment deep, broad,
scrobiculate. First segment abruptly truncate in front, the front face
vertical, flat, smooth and shining; second subequal in length to first,
but slightly wider than first.
Head, trunk, and anterior disc of first gastric segment densely
reticulate-punctulate and opaque, with a minutely pitted overlay;
1983]
Willey & Brown — Genus Myopias
281
sides of trunk, especially mesopleura and metapleura, and sides of
node, obscurely striate-punctulate; coxae minutely striate, becom-
ing smooth anteroventrad. Posterior disc of gastric tergum I, and
most of 11, densely covered with small, round punctures with
smooth, shining, but very narrow interspaces, becoming wider
behind; undersides of the same two gastric segments with scattered
coarse punctures, the interspaces in part minutely roughened (1) or
shining. In addition to the other surfaces listed above as smooth and
shining may be added the gastric apex, mandibles and femora, all
with scattered punctures. Antennae, tibiae, tarsi mostly finely punc-
tulate, but more or less shining.
Pilosity reduced to a mostly pubescence-like vestiture, abundant
but not very conspicuous, of appressed to subdecumbent, fine hairs;
only the clypeal and paired humeral setae as long as 0.10 mm, but
the specimens are badly rubbed, and probably had moderately,
abundant, but still fine and short, erect and suberect pilosity, some
of which can still be seen at times on scapes, legs, and dorsum of
trunk, as well as gastric apex.
Color dull, light brownish yellow.
Queen and male unknown.
Holotype (MCZ) and two paratypes workers (MCZ, BMNH)
from Taiwan; Rarasan (probably the same as the mountain now
called La La Shan, 24°44'N, 121°26'E, to the southwest of T’ai Pei),
31 July 1933, leg. R. Takahashi. I have no information concerning
the habitat, nest site, or prey. This is obviously a cryptic-foraging
form, probably living in the soil or in rotten wood. A related un-
described species has been found in Borneo.
The type series was originally three workers mounted on points
on a single pin; these were heavily damaged in a laboratory accident,
but the species is so interesting that we decided to describe it from
the collectively adequate remains. The name nops is from a Greek
word meaning blind.
Myopias delta new species
(Figs. 8, 21)
Diagnosis, worker and queen: A modest-sized species, completely
distinct from all congeners in possessing downcurved triangular
mandibles with distinct basal and masticatory borders meeting at a
dentiform basal angle. Head oblong, with convex sides and straight
282
Psyche
[Vol. 90
posterior border, and broadly rounded posterior corners. Worker
eyes reduced to dots. Frontal lobes and median clypeal lobe large;
clypeal lobe squarely truncate. Antennae very robust; scapes over-
reaching posterior border of head; funiculus dominated by a long,
thick, 4-merous apical club. Trunk compact, weakly convex, sepa-
rated into two subequal parts by a distinct but not sunken metanotal
groove. Petiolar node short and high, summit posterior and acutely
rounded, posterior face vertical and feebly concave. Gaster con-
stricted behind first segment. Integument smooth and shining, with
separated minute punctures. Color dark yellowish brown.
Worker, holotype: TL 4.1, HL 0.79, HW 0.70 (Cl 89), ML 0.30
(MI 38), MLO 0.54, SL 0.66 (SI 94), EL 0.04, WL 1.27, hind femur
L 0.68, hind tibia L 0.61 mm.
Worker paratypes (15 from type nest series) range downward
from size of holotype to smallest individual, TL 3.9, HL 0.75, HW
0.68 (Cl 91), ML 0.29 (MI 39), MLO 0.49, SL 0.62 (SI 91), EL 0.04,
WL 1.20 mm.
Head a little longer than broad, with parallel gently convex sides,
straight posterior border, and rounded posterior corners. Frontal
lobes broad, median clypeal lobe thick and wide (CLL 0.07, CLW
0.14), squarely truncate at apex, sides slightly convergent towards
apex. Median frontal sulcus very short, not extending rearward past
constricted ends of frontal carinae.
Mandibles basically of the ordinary ponerine, rather than Myopias-
like, form, triangular and strongly downcurved, with distinct basal
and oblique masticatory margins, each furnished with five coarse,
spaced teeth, the most basal of which, corresponding to the basal
angle, is subacutely dentiform; apical tooth the largest and most
acute; masticatory margins crossing over one another at full closure.
Basal oblique groove and its lateral extension (strix) strongly
developed.
Eyes small, round and dot-like, with indistinct facets, only
0.03-0.04 mm long, and distant from mandibular insertions by
about 0.20 mm. Antenna massive, scape thick, especially toward
apex, and overreaching posterior border of head (when held straight
back in full-face view) by more than half apical scape width; funicu-
lus with 4-merous club (which takes up more than 0.6 of funicular
length) following six short, transverse ring segments (II through
VII); pedicel (funiculus I) about 5X length of II.
1983]
Willey & Brown — Genus Myopias
283
Labrum without a median tooth, but each of its two lobes bears a
delicate, upturned apical tooth. Palpi segmented 3,3.
Trunk compact; aside from the rounded declivities of pronotum
and propodeum, the dorsal profile in side view is only weakly con-
vex, with moderate interruptions at premesonotal suture (movable)
and metanotal groove; latter is moderately wide and distinctly, but
not very strongly, impressed, and it divides the trunk into approxi-
mately equal anterior (promesonotal) and posterior (propodeal)
halves. Dorsum of propodeum gently convex, about twice as long as
mesonotum; declivity of propodeum steeply sloping, its outline con-
vex in side view, but the surface feebly concave, and weakly sub-
marginate above and laterally, as seen from above.
Petiolar node short and high, highest and widest behind at nar-
rowly rounded summit, after which the posterior face drops off
sharply and almost vertically. Anterior face nearly as steeply sloping
upward, shorter, meeting sloping dorsal face at an obtusely rounded
angle. Sternal keel prominent, with a thick, obliquely truncate ante-
rior process, pointed behind, then diminishing convexly caudad (see
fig. 21). In holotype and two of the paratypes, the posterior convex
portion bears an additional low point or tubercle, but shape of keel
is variable in any case.
Caster robust, distinctly constricted after first segment; segment
II about as high as I, and only very slightly wider. Sting long,
sharp, upcurved, capable of at least 0.4 mm extension. Seen from
above, anterior border of gaster 1 transverse, straight.
Body smooth and polished, with well-spaced, small (mostly 0.01
mm diameter or less), piligerous punctures, most numerous on dor-
sum of head and gaster. Antennae, frontal lobes, tibiae and tarsi
densely and finely punctulate. Bullae of metapleural glands obs-
curely striate.
Longer pilosity abundant on body, sparse on antennae and man-
dibles, and almost lacking on legs; mostly 0.03 to about 0.20 mm
long, appressed to erect, but mainly decumbent to suberect on
propodeum, node, and first two gastric segments; longest on clypeal
lobe, propodeum, node and gastric apex. Pubescence mostly
appressed to decumbent, inconspicuous and mesally inclined on
anterior half of head, more abundant and conspicuous on antennae
and legs.
Color dark yellowish brown or orange brown; legs, mandibles,
antennal scapes slightly more yellowish.
284
Psyche
[Vol. 90
Queen, dealate, from holotype nest series: TL 4.7, HL 0.80, HW
0.71 (Cl 89), ML 0.34 (MI 43), MLO 0.57, SL 0.70 (SI 99), EL 0.18,
WL 1.47 mm.
Showing the usual ponerine differences of caste, and darker, deep
brownish red, in color; scutum yellowish brown, with a broad, V- or
Y-shaped median fascia of deep reddish brown; mandibles, anten-
nae, legs, and indefinite patches on lateral areas of pronotum and
upper mesopleura obscure yellowish brown. Punctures a little
coarser and more conspicuous than in workers. An additional deal-
ate queen apparently belonging to this^species, taken in 1901 by L.
Biro (Hungarian Natural History Museum), comes from Friedrich-
Wilhelmshafen (now Madang, Papua New Guinea); it is notably
smaller (HL 0.62, HW 0.50 mm.) than either workers or queen from
the type colony; and was found in the Hungarian collection placed
with the M. tenuis.
Holotype (MCZ) and paratypes (MCZ, BMNH, ANIC) from a
colony collected in rain forest just west of the lower Busu River,
near Lae, Papua New Guinea on 9 May 1955, by Wilson (No. 983).
Wilson’s notes on this colony are slightly modified:
“A colony of one queen and about 30 workers, with brood at all
stages, none preponderant; in a crumbling small Passalus-sidigQ log,
diameter about 5 inches, held in shape by intact bark. Ants rela-
tively fast, nervous, similar to other [Myopias'] species. Workers and
brood scattered through a number of indistinct galleries and
chambers in the crumbling wood. In one chamber near larvae was a
fresh, decapitated worker of a small Leptogenys species. Around
another, large chamber was the kitchen midden, consisting of dis-
carded [Myopias] cocoons and numerous remains of ants, mostly or
entirely myrmicines, including at least two genera (q.v.).”The “q.v.”
refers to the vial containing the whole colony. Unfortunately, the
Myopias brood and the midden remains, left in alcohol after an
adult sample had been mounted from the vial, was lost in transit by
a colleague who had borrowed Wilson’s and Brown’s Myopias col-
lection residues for study.
From the circumstances of the collection as noted by Wilson, it
seems likely that M. delta is an ant predator specializing on Myrmi-
cinae, but perhaps occasionally accepting ponerines or other sub-
families. We need further collections and field and laboratory
1983]
Willey & Brown — Genus Myopias
285
observations to confirm this interesting possibility and to learn the
details of the M. delta behavior and ecology.
The name delta refers to the triangular mandibles. This species
cannot easily be placed to species group, and the mandibular form
even puts generic assignment into doubt. The 3-merous palpi (both
sets) and the upturned tooth on each labral lobe probably are
derived characters shared with the tenuis group, so the triangular
mandibular shape may well be secondarily derived from the
Myopias plan; it seems, indeed, that the dentition of M. delta is
more easily homologized with that of various Myopias species than
it is with the run of Pachycondyla groups. It thus becomes doubly
important to find again the larvae of M. delta.
■ f/J
-V-.
h’W'T
i9~r . '■ *■ ■' '. . , ' ,
'■■'*■ jii'jPtVr'l; -^Tii. ••««%>• ^“' Cl •«»»**i''
'1.5 f-rfe - • ir-jt^lii
LARVAE OF WRACK COLEOPTERA
IN THE FAMILIES CORYLOPHIDAE,
RHIZOPHAGIDAE, AND LATHRIDIIDAE*
By Donald S. Chandler
Department of Entomology
University of New Hampshire
Durham, New Hampshire 03824
Wrack studies in New Hampshire have revealed a number of
poorly known beetles. The temporary habitat produced by moist
rotting seaweeds presents an environment which allows certain
insects to become quite abundant for a short time. After processing
with a Berlese funnel the siftings of several square meters of wrack
from Odiorne Point State Park, I noted that large numbers of three
unfamiliar taxa of beetle larvae were present. Two of these three
taxa, Orthoperus scutellaris LeConte (Corylophidae) and Mono-
tonia producta LeConte (Rhizophagidae), were subsequently reared.
The third taxon is associated with Corticaria valida Fall, the only
adult lathridiid collected in or near this habitat. Generic characters
of another described larva of Corticaria confirm this placement.
Descriptions of these larvae are presented in this paper to aid those
workers studying wrack fauna.
All larvae were obtained from beach wrack by the author on June
15, 1982, at Odiorne Point State Park, Rockingham County, New
Hamsphire. Adults were reared by July 1, 1982. The two reared
taxa were processed with a simple program. Plastic boxes with
removable tops were supplied with a thin layer of sand, enough
water to saturate the sand, and a piece of rotting wrack. A number
of the largest larvae of the taxa were separated out, placed in the
containers, and left undisturbed except for the occasional addition
of water every 3-4 days. Examination of the gut contents of field
collected larvae indicated that all three taxa feed on the spores of
two species of Fungi Imperfecti, Helminthosporium sp. and Alter-
naria sp., which grow on the rotting kelp.
♦Scientific Contribution Number 1227 from the New Hampshire Agricultural Exper-
iment Station.
Manuscript received by the editor May 8, 1983.
287
288
Psyche
[Vol. 90
Orthoperus scutellaris LeConte
(Figs. 1,2)
There are two apparent taxa of Orthoperus found in wrack at
Odiorne Point. These run to Orthoperus s. scutellaris LeConte and
Orthoperus s. piceus Casey in the last revision of the family (Casey
1900). I have not been able to separate these two taxa when examin-
ing specimens mounted on slides. Both forms are represented in the
type series of LeConte for scutellaris at the Museum of Comparative
Zoology.
Their collection in the same habitat at the same time indicates
that the differences observed may be no more than normal variation
within the species. All of the reared adults were assignable to the
nominate subspecies.
Last instar larva: length 1. 5-2.0 mm. Body elongate, slightly flat-
tened dorso-ventrally, white with grey or brown patches in dorsal
view. Head slightly declined, two stemmata to each side, setae acic-
ulate, frontal arms of epicranial suture widely V-shaped, epicranial
stem absent, gular sutures distinct and widely separated through
length to head base; antennae two-segmented, sensorium almost as
long as terminal seta; mandibles symmetrical with several teeth at
apex, mola well-developed with series of large teeth on margin;
sclerites of maxillary base fused, palps two-segmented, mala arcuate
and blunt at apex, labial palps of a single segment; hypopharyngeal
sclerome elongate, the arcuate anterior cap may be the reduced
epipharynx. Thorax and abdominal segments densely covered with
short spicules, scattered setae are apically enlarged and truncate,
fluting is visible toward the apex, aciculate setae are found only on
the lateral margins. Prothorax with large quadrate shield formed by
dense clustering of larger spicules; medial longitudinal light area
dividing shield lacking spicules; the remaining thoracic and abdom-
inal segments each with dark lateral area formed by dense large
spicules, last abdominal segment lacking urogomphi, somewhat
explanate, darkened by dense large spicules, with alternating fluted
and aciculate setae on margin; abdominal segments 1-7 with large
glandular openings on lateroposterior margins of lateral darkened
areas. Spiracles annular. Legs well-developed, with five segments,
coxae widely separated, tarsungulus with single seta.
The only illustration of an Orthoperus sp. was by Perris in 1852 (in
Klausnitzer 1978, p. 275). The illustration indicates the general form
1983]
Chandler — Wrack Coleoptera
289
Figure I. Orthoperus scutellaris LeConte, dorsal view of larva; line equals 0.1
of members of the genus, but differs in many of the fine details.
Considering the age of the description, no attempt is made here to
contrast it with the description of O. scutellaris.
Monotoma producta LeConte
(Figs. 3, 5, 6)
Adults were identified by using the key of Horn (1879), and com-
paring specimens with the LeConte type. Last instar larva: length
4.0-4.5 mm. Body elongate, flattened dorso-ventrally, whitish, all
setae aciculate. Head with patches of darkly sclerotized tubercles,
labrum distinct, single stemma on each side, frontal arms of epicran-
ial suture lyriform, epicranial stem absent; antennae three-segment-
ed, sensorium two-thirds length of third segment; maxillary base
divided into three sclerites, palps three-segmented, mala bluntly
290
Psyche
[Vol. 90
2.
Figure 2. Orthoperus scutellaris LeConte. A, dorsal view of head; B, ventral view
of maxillae and labium; C, ventral view of hypopharyngeal sclerome; D, dorsal view
of left mandible; E, dorsal and lateral views of enlarged setae. Line equals 0.1 mm
unless otherwise indicated.
produced at apex, row of thick setae in inner margin to apex;
labium transversely divided, palps one-segmented; mandibles sym-
metrical, with large prostheca, prostheca and incisor edge of mandi-
ble serrate, accessory ventral process present on mandible base,
mola with series of fine teeth over surface, two widely separated
long setae on outer margin; the area of the hypopharyngeal scle-
rome and epipharynx very complex, only outline of sclerome is
figured. Prothorax with scattered patches of dark tubercles on ante-
rior half, notum with two transverse rows of four multiply tubercu-
1983]
Chandler — Wrack Coleoptera
291
Figure 3. Monotonia producta LeConte. A, dorsal view of head; B, ventral view
of maxillae and labium; C, ventral view of hypopharyngeal sclerome; D, dorsal view
of right mandible; E, dorsal view of right antenna. Line equals 0.1 mm unless other-
wise indicated.
292
Psyche
[Vol. 90
late processes, the processes of the posterior row being reduced,
basal lateral margins posterior to other lateral tubercles with single
tuberculate process, each process bearing a single long seta; the
remaining thoracic segments and abdominal segments 1-8 bear dor-
sally two transverse rows of six multiply tuberculate processes, each
process bearing a single long seta, segment 9 possesses an anterior
row of three and a posterior row of two processes similar to those of
the other segments, lateral margins of all segments with 2-3 large
palmate tubercles bearing 1-2 long setae; urogomphi similar in form
to the lateral processes, multiply tuberculate and bearing 2-3 long
setae; spiracles biforous, bourne on short tubes. Legs well-developed,
with five segments, coxae moderately separated, tarsungulus with
two short adjacent setae.
This is the first member of the genus to be formally described.
Peacock (1977) presents a brief description without figures. Her
diagnosis agrees with the features described here for M. producta.
Corticaria valida Fall
(Fig. 4)
This species is quite distinctive and fits the characters presented in
the key and description of Fall (1899). This identification is tenta-
tive, however, since the type localities of Fall were the Midwest and
Rocky Mountain states. Last instar larva: length 2. 5-3.0 mm. Body
elongate, cylindrical, whitish with scattered long setae abruptly
expanded and flattened at apex. Head slightly declined, with scat-
tered modified setae, the few aciculate setae on or near anterior
margin, labrum free, four stemmata to each side, three in vertical
row, the fourth posterior to the lowest stemma, epicranial suture
moderately long, frontal arms broadly V-shaped; antennae three-
segmented, sensorium as long as third segment, second segment
twice as long as first; maxillae and labium fused at base, maxiliary
palps three-segmented, mala with acute hook at apex, labial palps
one-segmented; mandibles lacking apical teeth, with lateral enlarged
fleshy lobe bearing two long setae at apex, mola enlarged, with
series of fine teeth over surface; hypopharyngeal sclerome short,
distinct. Thorax and abdomen dorsally with smoothly raised circu-
lar sclerotized patches bearing 1-4 modified setae, abdominal seg-
ments 1-8 with row of six sclerotized raised areas each bearing three
1983]
Chandler — Wrack Coleoptera
293
4.
Figure 4. Corticaria valida Fall. A, dorsal view of head; B, ventral view of
maxillae and labium; C, dorsal view of mandible and hypopharyngeal sclerome; D,
dorsal view of abdomen apex; E, dorsal view of left antenna. Line equals 0.1 mm
unless otherwise indicated.
294
Psyche
[Vol. 90
Figure 5. Monotoma producta, dorsal view of larva.
modified setae, outer sclerotized patches on lateral margin visible
dorsally, segment 9 with two lateral aciculate setae on each side,
setae on segment 10 all aciculate. Spiracles annular, not raised on
tubes. Legs well-developed, with five segments, coxae widely separ-
ated, tarsungulus with one seta.
Hinton (1945) is the only author who has provided a complete set
of figures describing Corticaria. The form of the mandibles and
mala, and the four stemmata to a side seem to characterize this
genus. The most obvious difference between species are the setal
forms. The long apically expanded setae of valida are most similar
to those in C. pubescens (Gyllenhal) (Hinton 1945). Other larvae
have been poorly or briefly described, and comparison with those
species is not attempted.
Acknowledgments
I would like to thank Dr. Alan L. Baker, Mary Lou Turner, and
David Gadoury for their efforts in the identification of the two
1983]
Chandler — Wrack Coleoptera
295
Figure 6. Monotonia producta, dorsal view of abdomen apex.
fungi. Dr. Ronald J. McGinley permitted the examination of the
LeConte types in the Museum of Comparative Zoology, Harvard
University. Dr. John F. Lawrence, C. S. I. R. O., Australia, offered
comments on the manuscript and graciously sent copies of his char-
acterizations of the three families. Drs. John F. Burger and R.
Marcel Reeves, University of New Hampshire are thanked for
checking the manuscript. Mrs. Marilyn Ecker, University of New
Hampshire, kindly provided the photomicrographs.
Summary
Wrack inhabiting larvae of three species of Coleoptera are des-
cribed for the first time. Orthoperus scutellaris LeConte (Corylo-
phidae) and Monotoma producta LeConte (Rhizophagidae) were
reared, with the third larva being associated with Corticaria valida
Fall (Lathridiidae). Spores of Helminothosporium sp. and Alterna-
ria sp. (Fungi Imperfecti) were found in the guts of all three taxa.
296
Psyche
[Vol. 90
Literature Cited
Casey, T.L.
1900. Review of the American Corylophidae, Cryptophagidae, Tritomidae
and Dermestidae, with other studies. Journal New York Entomological
Society 8:51-172.
Fall, H.C.
1899. Revision of the Lathridiidae of Boreal America. Transactions American
Entomological Society 26:101-190, plates III-IV.
Hinton, H E.
1945. A monograph of the beetles associated with stored products. Vol. 1.
Jarrold and Sons, Norwich, viii + 443 pp.
Horn, G.H.
1879. Synopsis of the Monotomidae of the United States. Transactions Amer-
ican Entomological Society 7:257-267.
Klausnitzer, B.
1978. Ordnung Coleoptera (Larven). W. Junk, The Hague, vi + 378 pp.
Peacock, E.R.
1977. Coleoptera Rhizophagidae. Handbooks for the Identification of British
Insects. Vol. V, Part 5(a). Royal Entomological Society of London. 19
pp.
THE GUEST ANT, SYMMYRMICA CHAMBERLINI,
REDISCOVERED NEAR SALT LAKE CITY, UTAH
(HYMENOPTERA, FORMICIDAE)* *
By Alfred Buschinger' and Andr^ Francoeur^
Introduction
In a series of recent papers we investigated the social structures of
Formicoxenus nitidulus, F. hirticornis, and Leptothorax provan-
cheri (Buschinger und Winter 1976, Buschinger 1979, Buschinger,
Francoeur and Fischer 1980). They are all so-called guest ants, small
species living in independent colonies within the larger nests of their
host species. Formicoxenus gains its food by soliciting it from the
Formica hosts, or by stealing food when two Formica workers feed
each other (Stager 1925, Buschinger 1976). L. provancheri are often
seen licking the head and body of their Myrmica hosts; however, it
remains uncertain how they really get their food. Our observations
revealed that these guest ants had some interesting features in com-
mon, such as a functional monogyny, a queen polymorphism with
dealate and intermorphic females, and a tendency to mate within or
on the upper surface of the host nest. The Formicoxenus species
recognized up to now have wingless, workerlike males, whereas the
L. provancheri male exhibits an ordinary winged shape.
It was a challenging task, therefore, to search for Symmyrmica
chamberlini Wheeler (1904), another guest ant with wingless males
and living together with Manica mutica, in order to study its biology
and to find out its relationship to the species mentioned above. We
took the opportunity of visiting the type area of S. chamberlini in
the vicinity of Salt Lake City, Utah, after the 9th Congress of lUSSl
in Boulder, Colorado. We were able to rediscover this ant and to
collect some new material which yielded additional support for an
incorporation of Symmyrmica into Formicoxenus.
'Fachbereich Biologic, Institut fur Zoologie, der Technischen Hochschule, Schnitts-
pahnstr. 3, D 6100 Darmstadt, FRG
^Departement des sciences fondamentales, Universite du Quebec Chicoutimi, Chi-
coutimi, Quebec, Canada G7H 2B1
* Manuscript received by the editor April 3, 1983.
297
298
Psyche
[Vol. 90
Field Observations and Collecting Site
The original description of Wheeler (1904) indicates the type
locality only inaccurately as “near Salt Lake City, Utah, in the
flood-plains of Jordan River”, where the host species, Manica mut-
ica, was said to be common in some localities. S. chamberlini, how-
ever, was found only in one particular ten-acre field and, despite an
intensive search, in no other locality. Unfortunately Wheeler’s paper
(1904) contains no further details on the exact site of that field.
On August 15, 16 and 17, 1982 we located about 30 flourishing
Manica mutica populations along the Jordan River, beginning with
our search near Lehi and working down the river to North Salt
Lake. We followed the roads and highways crossing the river, and,
always beginning at the bridges, we looked for the host species in or
near the banks. M. mutica was found near Lehi, on the eastern bank
north of the bridge of road no. 73, and in several places in West
Jordan (between 5400 South Street and 7800 South Street, east
bank), in Murray and South Salt Lake (between 5300 and 3300
South Street). Often the colonies seemed loosely concentrated. A
search in Big Cottonwood Canyon was not successful. We have
heard since then that unfortunately, late in following September, the
Jordan River heavily flooded the type area, the only known nesting
site for S. chamberlini.
The species was detected only in one locality, on the eastern bank
of the river, about 200 m south of the bridge of 3300 South Street,
South Salt Lake. Manica mutica there forms large nests in the silty
soil just in the upper edge of the steep river-bank about 2 m above
the waterline. The area is a horse pasture with poor, short vegeta-
tion, which was quite dry in August. Between two nests containing
chamberlini there was a willow brush, and in the estate adjoining to
the north, some rose bushes covered partly a private garbage dump.
One very large mutica colony with a chamberlini nest was found
there underneath a piece of concrete (50 X 18 X 15 cm).
Altogether we found chamberlini in three mutica nesting sites,
with distances of about 6 m between one other. We could not decide
whether the flourishing mutica nests belonged to separate colonies,
or whether they were parts of a large supercolony. However, two
samples of living workers from two similarly adjacent nest sites of
another locality (3900 South Street, South Salt Lake City) were
successfully mixed and became host of chamberlini colony no. 3.
1983]
Buschinger & Francoeur — Symmyrmica
299
Single mutica workers or groups with and without brood were
found nearly everywhere in that area when we dug a few centimeters
into the soil.
The first site, the southernmost one (Fig. 1), yielded just 30 chain-
berlini workers and intermorphs, but no brood (colony no. 1 in the
following). In the second site, about 6 m to the north and beyond
the willow brush, we found a chamberlini nest (no. 2) about 15 cm
below the surface, in the soil and surrounded by larger tunnels with
mutica workers and brood. The chamberlini nest contained larvae
and prepupae, about 38 workers and intermorphs, two wingless
males, and one male pupa. The prepupae from this colony were used
for a karyotype study. In the third site, again about 6 m to the north,
in the garbage dump, we found a chamberlini nest (no. 3) with about
30 workers and intermorphs, pupae, prepupae, and larvae. One
dealate female was detected but escaped capture. The relative
importance of intermorphs for our samples is given in table 1 in
comparison with Wheeler’s data.
Fig. 1. The chamberlini site on the east bank of Jordan River, looking southward
(upriver).
a) Site of chamberlini colony no. 1 within a Manica colony
b) Site of another Manica colony which extended along the willow brush to the
right
c) willow brush between chamberlini colonies 1 and 2
300
Psyche
[Vol. 90
Results of Dissecting Symmyrmica chamberlini
The three samples were kept alive for several months. However,
numerous specimens died during the first few weeks. A number of
them could be dissected following the method described by Busch-
inger and Alloway (1978).
In a total of 15 ordinary workers without any vestiges of ocelli on
their heads, the number of ovarioles was always two, except in one
specimen which has three. No spermatheca could be found in any of
these workers.
On the contrary, we found five slightly intermorphic specimens,
with between one and three more or less perceptible ocelli, with
somewhat deeper thoracic sutures, and with 6 ovarioles and a
spermatheca each. Two of these specimens, both from colony no. 2
(where males had been present), contained living sperm in their
receptacles. Their ovarioles, however, were short and transparent as
is usual in young, not yet egg-laying females.
Additional observations were made referring to the abdominal
glands of S. chamberlini. Thus, the poison gland reservoir was
always of usual size and shape, as in other leptothoracine ants. The
Dufour’s gland, however, is large both in workers and intermorphic
females. Its size exceeds considerably that of independent Lepto-
thorax species, and it reaches that of, e.g., Harpagoxenus sublaevis
(Buschinger and Alloway 1978).
A karyological study of 7 prepupae from colony no. 2 was made
following the method of Imai et al. (1977). The results, however,
were not as good as to permit the presentation of a karyotype. We
could only determine the chromosome number, which is 2n = 28.
Laboratory Observations
We were not able to take large samples of the Manica host species
with us alive. So only very few observations of interactions between
chamberlini and their hosts were possible. However, following a
method which had already worked with Formicoxenus nitidulus
(Buschinger 1976), we tried to join chamberlini brood and adults
with an unnatural host species. We chose a Leptothorax species
which was nesting within dead willow stems near to our chamberlini
site. Apparently it represents an unknown, new species belonging to
the subgenus Leptothorax (= Mychothorax Ruzsky). The following
experiments and observations were made:
1983]
Buschinger & Francoeur — Symmyrmica
301
a) After an artificial wintering, four S. chamberlini specimens of
colony no. 3 were isolated with 5 pupae of the Leptothorax species.
Honey and freshly killed Drosophila were provided. However, the
chamberlini did not survive. Two agonizing chamberlini, almost
without movement, were returned to the Manica mutica artificial
nest arena. The Manica workers immediately brought them into the
nest, and licked them all over. A few hours later, the two chamber-
lini could feebly walk. Next day, they were running normally in the
nest and its arena, having completely recovered. When an appar-
ently dead chamberlini was offered to mutica workers, they put it in
the refuse heap confirming its death. Trophallactic exchange be-
tween chamberlini nestmates was never seen, but only one was
noted between chamberlini and mutica.
b) The larvae and pupae of colonies no. 2 and 3 were put into a nest
together with 20 workers of the Leptothorax species mentioned
above. One chamberlini worker hatched, but died (or was killed?)
after two weeks. Chamberlini larvae survived an artificial hiberna-
tion from 27 October to 1st December 1982. They were easily distin-
guished from the Leptothorax larvae which developed from worker-
laid eggs: the chamberlini larvae are much hairier.
After the hibernation, the colony raised numerous alate Lepto-
thorax males, but no chamberlini. The chamberlini larvae vanished
one after the other.
c) About 20 workers and intermorphs of colony no. 2 were placed
together with 25 white and brown worker pupae and a few larvae of
the Leptothorax species on 1st September, 1982. After one week,
the first Leptothorax workers had hatched, and \2 chamberlini were
still alive. Among them an intermorph which had lost the right
antenna seemed to become fertile. This specimen, later on, was
observed several times to lay an egg. Together with a second inter-
morph it was still alive on 12 April, 1983.
The first, comparatively long-shaped eggs of chamberlini ap-
peared three weeks after the beginning of the experiment. Adult
chamberlini often fought with each other, possibly in order to elimi-
nate supernumerary reproductives. Some of the victims of these
fights were dissected, when they were not too much decomposed. In
addition, not only inseminated intermorphs but also ordinary
workers died rapidly. After the hibernation (cf. section b), only two
chamberlini intermorphs were alive, among them the one with only
302
Psyche
[Vol. 90
the left antenna. Both became fertile again, and the brood still con-
tained some hairy chamberlini larvae. Between 15 December and 26
January, in a temperature rhythm of 12 hours/ 15°C and 12 hours/
25° C, several Leptothorax males, females and workers hatched, but
no chamberlini larva reached the pupal instar.
After raising the temperature to lOh/ 17°C and 14h/28°C on 2nd
February, 1983, three chamberlini larvae became prepupae, and on
10 and 12 March two prepupae molted into apterous male pupae.
Nevertheless, it is doubtful whether breeding of chamberlini with
that Leptothorax will be as successful as the experiments with For-
micoxenus nitidulus and Leptothorax acervorum as host species
(Buschinger 1976), since both pupae and the remaining prepupa
were eaten during the following three days. In the mixed colony
chamberlini! Leptothorax sp. we observed quite amicable relations
between the two species. Often the chamberlini solicited food from
Leptothorax workers, and sometimes they were seen licking the
mouthparts of larvae. We never saw a chamberlini foraging outside
the nest, where honey and pieces of Tenebrio or Periplaneta were
offered as food. The chamberlini larvae, like those of the Lepto-
thorax species, are fed with solid particles of the insect pieces. Lep-
tothorax workers place the particles on the ventral surface of the
larvae, which then chew and eat them.
Discussion
Our knowledge of the biology of this rare ant still remains frag-
mentary. We can confirm the observation of Chamberlin, as
reported by Wheeler (1904) in that we also found this ant in mixed
colonies with Manica mutica, in the flood-plains of Jordan River
near Salt Lake City. The guest ants are living within independent
nests in the midst of prosperous Manica colonies. However, we
could not observe whether they solicit food from their hosts, or
what are the other relations of the two species. The observation
mentioned in the previous section, experiment a, raises questions of
whether the licking of chamberlini by the mutica hosts is linked to
any important cuticular secretion.
The nesting habits of S. chamberlini resemble closely those of
Leptothorax provancheri, the guest ant of Myrmica incompleta
Provancher (Buschinger et al. 1980). As was already suggested by
1983]
Buschinger & Francoeur — Symmyrmica
303
Table 1. Ratios
chamberlini
of workers and intermorphs
in colonies of
Symmyrmica
Source
Workers (%)
Intermorphs (%)
Total
Wheeler (1904)
8(38)
13(62)
21
Colony no. 1
20 (66)
10(33)
30
Colony no. 2
16(42)
22 (58)
38
Colony no. 3
13(43)
17(57)
30
2
57 (48)
62 (52)
119
Wheeler (1910), S. chamberlini is closely allied to the genus Formi-
coxenus, guest ants of Formica species in Europe and North Amer-
ica. Since the wingless male of chamberlini nevertheless is not as
workerlike as the Formicoxenus male, Wheeler may be right in
suggesting that it could represent an archaic form of Formicoxenus.
The close relationship of S. chamberlini and Formicoxenus is
further corroborated by our observations of intermorphic queens in
our new material. Such queens, which often look like ordinary
workers except that they have one or up to three vestigial ocelli and
sometimes a little bit more developed thoracic sutures, occur quite
frequently in Formicoxenus nitidulus (Buschinger and Winter,
1976), in F. hirticornis (Buschinger, 1979), and in Leptothorax pro-
vancheri (Buschinger et al. 1980). We cannot yet determine whether
S. chamberlini also has a functional monogyny like the 3 guest ants
we mentioned above. This would mean that alongside one func-
tional queen in each nest, there exists one or several inseminated but
not egg-laying potential queens. However, at least our finding of
two recently inseminated intermorphic females in S. chamberlini
colony no. 2 reveals that, as in the other guest ants, copulation takes
place within or near the mother colony, and that newly mated
females may remain for a while in the mother nest.
The analysis of intermorph composition presented in table 2
based on the classification of Plateaux (1970) for caste polymorph-
ism in Leptothorax nylanderi, revealed only few superior inter-
morphs with intermediate trunk between a fully developed gyno-
morph and a typical ergatomorph. Moreover the inferior intermorph
classes seem to be dominated by the form 4 which has 3 small or
minute ocelli in any combination, a mesothorax not, or slightly
enlarged, a promesonotal suture more or less prominent. The indi-
viduals with a potential or actual queen function capacity are found
304
Psyche
[Vol. 90
Table 2. Types of S. chamberlini intermorphs according to Plateaux’ classification
of Leptothorax nylanderi.
Source
Form 2
Form 3
Form 4
Form 6-7
Total examined
Wheeler (1904)
0
1
5
0
6
Colony no. 1
3
1
4
2+)
10
Colony no. 2
7
0
13
2++)
22
Colony no. 3
2
2
13
0
17
■'■)Both with thoracic sutures and sclerites according to Plateaux’ form 7, except for
the lack of wings. One specimen with very short forewing rudiments.
■*^)Two specimens between Plateaux’ form 6 and 7, without traces of wings.
mainly in that class of intermorphs. It is worthy to stress that Holli-
day’s (1903) data for 1000 specimens of L. provancheri includes 37%
of intermorphs without the microgynes; the intermorph composi-
tion exhibits the same trends as in chamberlini.
The karyotypes cannot yet confirm a closer relationship of all
these guest ants. However, they also do not contradict such an
assumption. F. nitidulus has a haploid number of n = 15 chromo-
somes, L. provancheri has n = 1 1, and S. chamberlini with n = 14
lies in between. For F. hirticornis and diversipilosus the chromo-
some numbers are not yet known.
Summing up the known features, queen polymorphism with alate
and intermorphic females, males with their tendency to reduce
wings and to become ergatomorphic, the presence of inseminated
young (and in Formicoxenus also old) potential queens in the nests,
and the life habits as guest ants, we believe that Symmyrmica, and
also L. provancheri, should be incorporated in the genus Formi-
coxenus. A comparative morphological study has been undertaken
in order to link the biological informations accumulated on the
guest ants mentioned above in a taxonomic revision of the genus
Formicoxenus.
Acknowledgements
We thank Karl Fischer for providing the chromosome number of
Symmyrmica chamberlini, and Robert Loiselle for laboratory
assistance. Wheeler’s specimens were loaned by the American
Museum of Natural History, New York (Mrs. A. Favreau), and by
the USNM, Washington, through Dr. D.R. Smith (USDA).
1983]
Buschinger & Francoeur — Synimyrmica
305
The field work was supported by a grant of the Deutsche
Forschungsgemeinschaft (Buschinger) and a grant of the Natural
Science and Engineering Research Council of Canada (Francoeur).
Summary
Symmyrmica chamberlini was described by Wheeler (1904) from
specimens taken by C.V. Chamberlin in 1902 in a colony of Manica
mutica (Emery) near Salt Lake City. No further records of this
species are known. In order to find out the systematic relations of
Symmyrmica to other ants like Leptothorax provancheri Emery or
those of the genus Formicoxenus, we have collected some new mate-
rial in August 1982, in the Salt Lake City area. The morphology,
female polymorphism, and wingless male together with biological
features indicate that S. chamberlini is a species that should belong
to the genus Formicoxenus.
References
Buschinger, A. (1976): Eine Methode zur Zucht der Gastameise Formicoxenus
nitidulus (Nyl.) mit Leptothorax acervorum (Fabr.) als “Wirtsameise” (Hym.,
Form). Ins. soc. 23, 205-214.
Buschinger, A. (1979): Functional monogyny in the American guest ant Formi-
coxenus hirticornis (Emery) (= Leptothorax hirticornis), (Hym., Form.). Ins.
soc. 26, 61-68.
Buschinger, A. and Alloway, T. M. (1978): Caste polymorphism in Harpa-
goxenus canadensis M. R. Smith (Hym., Formicidae). Ins. soc. 25, 339-350.
Buschinger, A., Francoeur, A. and Fischer, K. (1980): Functional monogyny,
sexual behavior, and karyotype of the guest ant, Leptothorax provancheri
Emery (Hymenoptera, Formicidae). Psyche 87, 1-12.
Buschinger, A. und Winter, U. (1976): Funktionelle Monogynie bei der Gast-
ameise Formicoxenus nitidulus (Nyl.) (Hym., Fo'rm.). Ins. soc. 23, 549-558.
Holliday, M. (1903): A study of some ergatogynic ants. Zool. Jb. Syst. Gkol.
Geogr. Tiere 19: 293-328.
Imai, H. T., Crozier, R. H. and Taylor, R. W. (1977): Karyotype evolution in
Australian ants. Chromosoma 59, 341-393.
Plateaux, L. (1970): Sur le polymorphisme social de la fourmi Leptothorax
nylanderi (FoQx^iQv). 1. Morphologic et biologic comparees des castes. Ann. Sci.
Nat. Zool. I2e S., 12, 373-478.
Stager, R. (1925): Das Leben der Gastemeise {Formicoxenus nitidulus Nyl.) in
neuer Beleuchtung. Z. Morph. Okol. Tiere 3, 452-476.
Wheeler, W. M. (1904): Three new genera of inquiline ants from Utah and Colo-
rado. Bull. Amer. Mus. Nat. Hist. 20, 1-17, pi. 1.
Wheeler, W. M. (1910): Ants, their structure, developent and behavior. Colum-
bia Univ. Press, New York and London.
« V
rv*V5
■ *■ I* ■*■■' ’ . *r* ■
;.i' ''w:,.''®’’ >teM:i0i^ .'%tfe'x
./.vy-.4;i'-,> HsTIIM '*'r?:
%$k
. ,^ 0 ^ ^ y •■ ' ■
!.-• .•a-i*....^ st-’i
IV '*-. , '•■'h\
■,'=*ky!^
f ^
J£n^
It < ^
4Ly>n^'
^"■i4 • '
. ■.
9
. ■>'• ■- * • ' -7u'^^
EMIGRATION RAIDS BY SLAVE-MAKING ANTS: A
RAPID-TRANSIT SYSTEM FOR COLONY RELOCATION
(HYMENOPTERA: FORMICIDAE)
By Ellen C. Kwait' and Howard Topoff^
Introduction
Polyergus lucidus Mayr is an obligate slave-making ant, found
throughout north temperate regions of the world. Their slave raids
are dramatic events in which columns of highly aroused workers
penetrate nests of the related ant genus Formica, and carry the
target colony’s pupae back to their own nest (Marlin 1969; Talbot
1967). Although many of these pupae are consumed, varying
numbers are reared through eclosion and become permanent mem-
bers of the mixed-species nest. During the evolution of social para-
sitism, Polyergus workers lost the ability to participate in the
ordinary chores of foraging, nest maintenance, and brood rearing,
all of which are left to the Formica slaves.
Raiding behavior in Polyergus has only been reported in the con-
text of slave raids, or of intraspecific territorial raids (Topoff et al.
in preparation). Field observations of colonies in late summer,
however, have revealed an entirely new function of raiding be-
havior: the rapid transport of colony members during emigrations
to new nests at the end of the slave-raiding season. During such
colony movements, the low level of mixed-species ant traffic is
periodically interrupted by the abrupt emergence of Polyergus
workers, and their formation into a well-organized swarm. The
Polyergus workers promptly “raid” the old nest, and transport
adult Formica individuals to the new site. For such group processes,
occurring in the context of nest relocation, we propose the term
“emigration raid.”
'Department of Biology, City College of CUNY, New York, N.Y. 10031.
^Department of Psychology, Hunter College of CUNY, New York, N.Y. 10021, and
The American Museum of Natural History, New York, N.Y. 10024
Manuscript received by the editor May 28, 1983.
307
308
Psyche
[Vol. 90
Methods
Emigration raids were observed in three mixed colonies of P.
lucidus and F. schaufussi Mayr, located in a pine-barrens habitat in
North Centereach, Long Island, N.Y. The most detailed data were
collected from a colony monitored continuously from July through
September, 1976. The emigrations occurred on September 19 and
20, as the colony moved to a previously-constructed nest, 5.9 m
southwest of the old site. Movements of individual ants shuttling
between nests were monitored with hand-held tally counters. Callow
age was estimated by comparing their degree of pigmentation with
individuals of known age in laboratory nests (Kwait 1982).
Results and Discussion
During the morning and early afternoon of each day, the popula-
tion characteristics of the emigrations were similar to those described
for Polyergus nest movements that occasionally occur in the spring
(Marlin 1971). Thus, only Formica workers functioned as transpor-
ters, carrying adults and brood of both species to the winter nest
(Fig. 1). As the afternoon progressed, however, several P. lucidus
workers periodically joined the emigration. Although these Polyer-
gus individuals made 5-23 trips between nests, the important point
to note is the relatively low level of overall activity during most of
the afternoon (Fig. 2A). But starting about 1600 hrs (EDT), at
approximately the same time as the onset of slave raids earlier in the
season (compare Fig. 2A and 2B), groups of 30-69 callow and
mature- adult Polyergus abruptly surged out of the new nest and
formed into an organized swarm. This raid swarm backtracked and
penetrated the old nest, and after several minutes the Polyergus
workers emerged carrying nestmates (Table 1). On both emigration
days, the first emigration raid was promptly followed by a second
raid (Fig. 2A). Activity levels for both species then dropped
abruptly, as they typically do during slave raids on freeliving colo-
nies of Formica.
Aside from our field observations on emigration raids, the only
other reference to Polyergus carrying adult Formica individuals is
Huber’s (1810) study in Switzerland, of an emigration into an aban-
doned Formica nest. A more recent observation of comparable
behavior stems from studies of P. breviceps- Formica gnava mixed
colonies in a desert habitat in southeastern Arizona (Topoff et al. in
1983]
Kwait & Topqff — Slave-making ants
309
511J
!!;o|
if o'
EMIGRATION DAY IT : Formica Activity
1100 1200 1300 1400 1500 1600 1700 1800
TIME
Figure 1. Activity of Formica schaufussi during colony emigration. Most of the
workers are carrying adults and brood of Polyergus and Formica to the winter nest.
preparation). On August 14, 1981, colony #1 raided another mixed
nest (#2), 30 m to the east. Fighting between resident and intruding
Polyergus workers was minimal, and only about 1 1 pupae were
captured. Early in the afternoon on August 15, Polyergus workers
emerged from colony #2 and backtracked over the previous day’s
trail towards mixed colony #1. At 1715 hrs (MST), traffic again
reversed direction, as hundreds of Polyergus workers penetrated
nest #2. This time, however, the Polyergus emerged carrying
hundreds of Formica brood, callows, and mature-adult individuals.
All of the adult Formica being transported had their appendages
closely appressed to the body, in the “pupa-like” position that is
typical during social carrying behavior (Moglich and Holldobler
1974). The adult Formica were carried into mixed nest #1, and none
had re-emerged by the end of the observation period at sunset.
The description by Huber (1810) of adult Formica transport by
Polyergus clearly took place within the context of a colony emigra-
tion. Our observation of similar behavior by P. breviceps is more
difficult to interpret, but we suggest that it occurred in the context
Table 1. Quantitative Description of Emigration Raids
Date
Raid #
Time
Polyergus
on raid
Adult Formica
retrieved
Formica pupae
retrieved
9/19/76
1
1525
56
37
3
2
1625
45
35
0
9/20/76
1
1620
69
30
0
2
1703
30
15
0
310
Psyche
[Vol. 90
of colony reunification shortly after division by budding. The
important point in both cases, however, is that adult transport of
Formica by Polyergus took place in a staggered, prolonged, emi-
gration-type column, without the intervention of an abrupt, short-
lived, and full-scale raid.
That the P. lucidus emigration raids reported in the present study
are fundamentally similar behavioral processes to their slave raids is
evidenced by the congruence of several parameters, including: (1)
the immediate organization of the emerging workers into an organ-
ized swarm; (2) the time of raid onset; (3) the occurrence of multiple
raids; and (4) the participation of recently-eclosed Polyergus cal-
lows. The number of Polyergus workers in the emigration raids was
lower than that characteristic of most slave raids. This difference is
probably not significant, because it is known that even slave-raid
participants decrease to as few as 13-50 individuals towards the end
of the raiding season (Talbot 1967). Nevertheless, the social context
of an emigration does produce at least one major difference in the
behavior of the Polyergus workers. During emigration raids, it is
principally Formica adults that are carried by the Polyergus.
Because these adult slaves were reared from the pupal stage in the
chemical and tactile environment of the mixed nest, the communica-
tory basis for social carrying behavior is well established. During
slave raids, by contrast, Formica adults respond to the intruding
Polyergus by exhibiting various forms of withdrawal behavior (Wil-
son 1971). Asa result, it is principally Formica pupae and callows
that are retrieved during slave raids.
Emigration behavior with adult transport is common in many ant
species (Smallwood 1982), including free-living colonies of Formica
schaufussi. Because F. schaufussi is considered related to Polyergus,
emigrations probably pre-date the evolution of slave-raiding behav-
ior. The secondary use of raiding behavior for Polyergus colony
relocation represents an adaptive evolutionary transition, consistent
with Simpson’s (1958) principle of “transformation.” Accordingly,
when changes at any level of organization take place during a spe-
cies’evolution, previously existing adaptations are often remodeled
and eventually serve new functions. Because group raiding in
Polyergus involves a complex recruitment process specialized for
the efficient retrieval of other ants, it is clearly advantageous for the
colony to utilize the process in all appropriate behavioral contexts.
1983] Kwait & Topoff — Slave-making ants 311
TIME
Figure 2. (A) Activity of Polyergus lucidus during colony emigration. The two
consecutive peaks in afternoon activity represent emigration raids conducted on the
old nest by Polyergus workers that were transported to the new nest by Formica
individuals. (B) Activity of P. lucidus during typical slave-raid day. Note the sim-
ilarity in timing of slave raids and emigration raids.
312
Psyche
[Vol. 90
The rapid transport by Polyergus of adult Formica slaves to an
overwintering site shortens the emigration time during a season
characterized by increasingly unfavorable weather, and quickly re-
locates the Formica slaves to the new nest where they are needed for
overall colony maintenance.
Acknowledgments
This study is part of a dissertation by the senior author, submitted
to the graduate faculty in Biology of the City University of New
York. The research was supported by NIMH Training Grant 14280.
Special thanks go to Mr. Raymond Sanwald for his field assistance.
References
Huber, P.
1810. Recherches sur les moeurs des fourmis indigenes. J. J. Paschoud, Paris.
Kwait, E.
1982. Raid site location, recruitment, and polyethism in the slave-making ant
Polyergus lucidus Mayr. Ph.D. thesis, City University of New York.
Marlin, J. C.
1969. The raiding behavior of Polyergus lucidus in central Illinois (Hymenop-
tera: Formicidae). J. Kansas Entomol. Soc. 42: 108-1 15.
1971 . The mating, nesting, and ant enemies of Polyergus lucidus Mayr. Amer.
Mid. Nat. 86: 181-189.
MOglich, M. and B. HOlldobler
1974. Social carrying behavior and division of labor during nest moving in
ants. Psyche 81: 219-236.
Simpson, G. G.
1958. The study of evolution: methods and present status of theory. In A. Roe
and G. G. Simpson, eds.. Behavior and evolution. Yale University Press,
New Haven, Conn. pp. 7-26.
Smallwood, J.
1982. Nest relocation in ants. Insectes Sociaux 29: 138-147.
Talbot, M.
1967. Slave raids of the ant Polyergus lucidus. Psyche 74: 299-313.
Topoff, H., B. Lamon, and L. Goodloe
1983. Behavioral ecology of the western slave-making ant Polyergus breviceps.
(in preparation)
Wilson, E. O.
1971. The Insect Societies. Belknap Press of Harvard University Press, Cam-
bridge, Mass.
DEFENSE OF BRACKEN FERN BY ARTHROPODS
ATTRACTED TO AXILLARY NECTARIES
By Matthew M. Douglas
Adjunct Senior Research Scientist
Snow Entomological Museum
The University of Kansas*
Introduction:
The phenotypically variable bracken fern, Pteridium aquilinum
(L.) Kuhn, is an economically important plant that establishes dense
monocultural stands by spore dispersal and by spreading subterra-
nean rhizomes throughout the world, except for hot and cold desert
regions (Page, 1976). Bracken produces a number of so-called
“secondary plant compounds” that have been shown to protect it
from some nonadapted insects (Cooper-Driver et. al., 1977). These
compounds include the cyanogenic glucoside, prunasin (Cooper-
Driver and Swain, 1976; Cooper-Driver et. al., 1977), lignins and
silica (Lawton, 1976), sesquiterpene pterosins (Jones and Firn,
1979a), phytoecdysteroids (Jones and Firn, 1978), and the protein
thiaminase (Evans, 1976). Tannins, flavonoids, and phenolics have
also been implicated as possible defensive compounds in bracken
fern (Cooper-Driver et. al., 1977; Jones and Firn, 1979b).
Despite bracken’s well-developed biochemical arsenal, adapted
and nonadapted herbivorous insects in experimental plots located in
Michigan and Massachusetts often destroy up to 30 percent of a
frond’s biomass after the pinnae are completely expanded. In addi-
tion to these herbivores, stands of Michigan bracken also support a
diverse community of ectoparasites, parasitoids, and predators of
bracken herbivores, including nearly 20 species of ants and spiders
that form temporary symbiotic relationships with the bracken
croziers.
♦Research Address: 1503 Woodland St., Jenison, Michigan, 49428
Manuscript received by the editor May 2, 1983
313
314
Psyche
[Vol. 90
Methods and Study SItes
During 1980-1982, Gordon VanWoerkom (Hope College) and I
observed the relationships between bracken fern and its associated
arthropod community in Michigan. Our original intent was to
determine to what extent seasonal patterns of insect species diversity
are a reflection of quantitative or qualitative changes in the chemical
composition of their host plant. We uncovered an arthropod com-
munity associated with bracken that is much more complex than
expected. These arthropods were identified to species whenever pos-
sible, and their behavior was recorded by 16mm and 35mm cameras.
The bracken-arthropod study was conducted on four lOO-m^ plots
of bracken occupying different environments within the confines of
the Hope College Field Station. This 80 acre field station is located
in Allegan County, 2 miles south of Holland, Michigan. The mild
climate of the preserve is due to the thermal moderating effect of
Lake Michigan. Primary vegetation consists of virgin forest, mature
(second generation) deciduous forest, as well as open fields punctu-
ated with sand “blow-outs” in various stages of ecological suc-
cession.
In upland areas, where the bracken plots are located, the soil
consists of a relatively thin layer of sandy-loam topsoil overlying
what was formerly extensive sand dunes from Lake Nipissing.
Bracken fern has established extensive stands in discrete patches
throughout the preserve, some under forest canopy and others
under open field conditions.
Results
Although predaceous ants and spiders are abundant during the
spring, the diversity of herbivorous species is low, supporting Law-
ton’s (1976) observation that, “The evidence strongly suggests that
bracken in May and June may not be an easy resource for herbi-
vores to exploit.” However, protein levels are highest in the spring
(about 25 percent of the dry weight) and 40 percent lower by
August-September (about 10 percent of the dry weight). Further-
more, concentrations of lignins, tannins, and silicate are lowest in
the early spring and tend to increase throughout most of the season,
all of which might be expected to make the plants tougher and
therefore less palatable. As Lawton (1976) further points out.
1983]
Douglas — Defense of bracken fern
315
“...bracken-feeding herbivores face progressive deterioration in
“food quality” as the growing season progresses.” Yet, we have
found that herbivorous species diversity and abundance, particu-
larly of adapted herbivores, increase dramatically after the second
week of June, and peak during late July and August.
Bracken’s palatability early in the spring may also be affected by
the production of thiaminase and the cyanogenic glucoside, pru-
nasin (Jones and Firn, 1978). But bracken in England is polymor-
phic for the production of HCN: some bracken clones contain the
B-glucosidase enzyme and not prunasin, while others contain
neither enzyme nor the glucoside (Cooper-Driver and Swain, 1976;
Cooper-Driver et al., 1977). Likewise, Zavitkovsky (1979) found
uniformly negative results for cyanogenesis in Massachusetts brack-
en fern.
Thiaminase may be the only known chemical deterrent in bracken
with any potential for disrupting normal insect development, since
thiamine is essential to insect development (Dadd, 1973). However,
bracken does contain thiamine (Berti and Bottari, 1968), and thiam-
inase activity in English bracken drops from a high of nearly 30 ug
to 7 ug thiamine destroyed per min/g dry weight between the last
week of April and the second week of May (Evans, 1976). Even so,
populations of adapted bracken herbivores increase only after the
second week of June in Michigan, perhaps weeks after cyanogenic
glucosides (if functional) and thiaminase activity have fallen to low
levels. By contrast, most other diapausing and temperature sensitive
insects such as butterflies have eclosed from the pupae by early May
in Michigan, and their larvae can be found by the second and third
weeks of May.
Our study in Michigan suggests that bracken has evolved another
line of defense that complements its biochemical defense system,
and protects it from serious herbivorous damage during the rapidly-
developing crozier stage. This second line of defense consists of
predaceous arthropods, particularly ants and spiders attracted to a
sweet, viscous fluid secreted by a number of axillary nectaries. The
nectaries are dark oval enlargements that appear in the axils where
the pinnae and major pinnules branch off from the rachis. Large
numbers of “nonassociated” arthropods which utilize bracken for
purposes other than food are also attracted to the developing
bracken canopy and their oozing nectaries early in the spring. In
316
Psyche
[Vol. 90
turn, these nonassociated arthropods are preyed upon by the pre-
daceous ants and spiders. Biochemical analyses of nectary secretions
in California bracken indicate the presence of relatively large con-
centrations of glucose and fructose, minute concentrations of
sucrose and maltose (Irene Baker and Peter Atsatt, pers. comm.),
and an undetermined number of free amino acids.
By mid-May, when the primary nectaries at the axils of the pinnae
are secreting microliters of “nectar” daily, the thiaminase activity is
declining rapidly and the cyanogenic glucoside may or may not be
offering protection. In addition, the levels of tannins and silicate are
still at low levels, while proteins are at optimal level (Lawton, 1976).
Thus, the crozier stage of bracken may at once present a nutritious
stage to attack as well as a “loophole” in the biochemical arsenal.
The potential for attack at this time by nonadapted polyphagous
herbivores is great, and we have established that several nonadapted
herbivores will feed readily on bracken croziers without ill effect.
These include the gypsy moth, Porthetria dispar (L.) (Lepidoptera:
Liparidae), the large milkweed bug, Oncopeltus fasciatus (Dallas)
(Hemiptera: Lygaeidae), the common rose chafer, Macrodactylus
subspinosus (Fabricius) (Coleoptera: Scarabaeidae), and a large
tropical cockroach, Blaberus giganteus (L.) (Orthoptera: Blaberi-
dae).
Many croziers are not attacked because their actively secreting
nectaries are quickly located by at least five species of ants in Michi-
gan: Formica subsericea Say, Formica obscuriventris Mayr, For-
mica pallidefulva nitidiventris Emery, Camponotus pennsylvanicus
(De Geer), and Camponotus nearcticus Emery. Camponotus pen-
nsylvanicus, Formica obscuriventris, and Formica subsericea in
particular defend the nectaries and the developing crozier by patrol-
ling the plant in a very systematic manner. An ant patrol typically
begins after both primary nectaries at the base of the pinnae are
antennated and sampled with the mouthparts for 1-3 minutes. The
ant then proceeds up and down each pinna, investigating the pin-
nules and attenating the smaller nectaries even though these rarely
secrete visible quantities of nectar. A single patrol, covering the
entire frond, may last from 2 to 15 minutes, depending upon the size
of the crozier and the length of the time spent at each nectary.
One ant, or several ants from the same colony may patrol a given
crozier, but all other “intruding” arthropods are bitten and stung
1983]
Douglas — Defense of bracken fern
317
until they are driven off or killed by the patrolling ant(s). Corpses
are removed and taken down to the nest. Patrolling predaceous ants
thus obtain both a nutrient-rich secretion in addition to arthropod
prey attracted directly to the plant tissues or to the nectaries. The
result is that the croziers are protected (to an unknown extent) from
adapted and nonadapted herbivores during this most susceptible
stage of growth).
Species of smaller ants such as Tapinoma sessile (Say), Lepto-
t borax curvispinosus Mayr, Leptothorax muscorum (Ny lander) and
Lasius alienus (Foerster) engorge at the nectaries as well, but often
in non-aggressive interspecific groups. However, none of these spe-
cies appears to defend the bracken croziers from other arthropods,
and thus they may be “parasitic” in the broad sense of the concept
since the ants obviously benefit by gathering nectar, but the bracken
potentially suffers because it loses its attraction to more aggres-
sively-defensive ants. Although we have not documented any defen-
sive role by these small ants in Michigan, it is possible that they
remove the eggs of herbivores (Susan Koptur, pers. comm.).
The immatures of at least 10 species of spiders also imbibe at or
are attracted to the axillary nectaries, while several others are inti-
mately associated with the unfurling pinnae. For example, when the
immatures of Enoplognatha ovata (Clerck) bind the 3 pinnae
together, the nectaries become effective “baits” within the pyramidal
web, ensnaring many smaller nonassociated dipterans and parasitic
ants. Other spiders such as the thomisid Tibellus sp. extend their
body against the rachis, cepthalothorax pointing towards the nec-
tary, and ambush other arthropods as they arrive to extract nectar.
Finally, several species of salticid spiders (e.g. Metaphiddipus pro-
tervus Walckenaer), prowl the developing bracken canopy, leaping
from pinna to pinna in search of prey. Encounters between spiders
and patrolling ants are not uncommon, but it is not certain which
factors predispose one or the other to dominate a given frond. How-
ever, the small parasitic ants are common prey items of the patrol-
ling spiders in Michigan.
By the second week of June, the pinnae have completely opened,
and the nectaries darken and desiccate. Even so, patrolling ants
continue to defend the mature plants for a few more days, perhaps
attracted to the lingering odor of the nectaries. But the ant patrols
are erratic and the ants stay on the plants for far shorter periods.
318
Psyche
[Vol. 90
usually less than 30 seconds. At the end of June, defending ants and
spiders cease patrolling the fronds entirely, although mature Meta-
phiddipus protervus spiders and Formica subsericea ants can occa-
sionally be found on new croziers which emerge periodically
throughout the growing season. As with spring croziers, these
summer fiddleheads are also patrolled systematically by ants, even
though their nectaries appear to dry quickly under the hot sun.
Concurrently with the declines in the ant patrols and the spider
populations associated with bracken, there is a significant increase
in herbivore damage, largely from adapted insects. Damage in-
creases as adapted insect populations peak in mid to late summer,
despite the “toughening” of the bracken with increasing concentra-
tions of tannins and silicate, and despite the fact that available
protein has declined by over 40 percent (Lawton, 1976). Any herbi-
vore damage done at maturity, however, will affect proportionately
less of the plant biomass than if the plant had incurred the damage
during the crozier stage. Even minor chewing or sucking damage on
the newly-emergent croziers can destroy part or all of the apical
bud, or cause lodging of the plant at maturity if the rachis is wea-
kened. Lodging or individual pinna or the entire frond is especially
common in bracken plants attacked by minute gall-forming/ mining
microlepidopterans, as yet unidentified. These mining insects can
stunt 50 to 90 percent of the potential growth of a given pinna,
possibly because they feed on internal vascular tissues and cannot be
reached by patrolling ants or predaceous spiders.
Summary
Darwin (1877) was among the first to point out that the secretion
of the bracken nectaries is very attractive to ants and that the ants
may thus serve in some capacity to defend the ferns (Lawton, 1976).
The arthropod defense system found in Michigan may help to
explain why herbivorous damage from both adapted and non-
adapted insects is minimal in the crozier stage. Bracken fern has a
well-developed “arsenal” of potentially toxic secondary plant com-
pounds that may also serve to deter or inhibit insects from attack.
Yet, despite a relatively herbivorous-free period during its early
growth stage, a diverse community of adapted herbivorous insects
inflicts moderate to heavy damage later in the summer months.
1983]
Douglas — Defense of bracken fern
319
Bracken may (in part) be protected by compounds such as thiam-
inase and cyanogenic glucosides, but our research of the past three
years shows that bracken at the very least supplements its passive
chemical defenses with a mobile, predaceous arthropod defense
community. This community includes at least 5 species of ants and
10 species of spiders that are initially attracted to “axillary nectar-
ies” (AN) secreting a nutrient-rich sap of sugars and amino acids.
Our research shows that bracken “turns on” these nectaries during
the rapidly-growing crozier stage, and turns them “off” after the
pinnae are fully expanded. During the secretory stage, ants patrol
and defend the pinnae from all intruders, including potential herbi-
vores, other species of ants, and other predaceous arthropods such
as spiders. However, the immature spiders also utilize the AN secre-
tions, stalk arthropods within the developing canopy, or construct
webs over the opening pinnae, turning them into effective traps with
the AN enclosed as “baits.”
Bracken’s apparent immunity to insect attack during the crozier
stage may be due not so much to the toxicity of its secondary
compounds, but to the continuing coevolution of the AN and their
attendant, predaceous arthropods that patrol the pinnae or other-
wise rid them of herbivores during bracken’s crozier stage. The
bracken-arthropod system may be one of the most unique and com-
plex hierarchies of symbiotic relationships to be found in a primitive
plant-arthropod system.
Acknowledgments
This study was supported in part by NSF grant DEB-8005581
(Gillian Cooper-Driver, Principle Investigator), and conducted at
the Hope College Research Station, Holland, Michigan. I thank
Gordon VanWoerkom and Annmarie Baldiga for their invaluable
assistance in the field and laboratory. I also thank Drs. Irene Baker,
Susan Koptur, and Peter Atsatt for use of unpublished information.
Drs. O. Taboada, H. D. Blocker, R. E. Beer, M. DuBois, C. D.
Michener, A. Brady, G. Byers, O. R. Taylor, W. H. Wagner and
especially G. Cooper-Driver assisted me in various stages of this
project and the preparation of the manuscript.
320
Psyche
[Vol. 90
Literature Cited
Berti, G. and F. Bottari
1968. Constituents of ferns. In L. Reinhold and Y. Liwschitz (Eds.), Progress
in Phytochemistry. 1: 589-685, London: Interscience.
Cooper-Driver, G., Finch, S., Swain, T. and E. Bernays.
1977. Seasonal variation in secondary plant compounds in relation to the
palatability of Pteridium aquilinum. Biochem. Syst. Ecol., 5: 21 1-218.
Cooper-Driver, G. and T. Swain
1976. Cyanogenic polymorphism in bracken in relation to herbivore preda-
tion. Nature (London), 260: 604.
Dadd, R. H.
1973. Insect nutrition: current developments and metabolic implications. Ann.
Rev. Ent., 18: 382-419.
Darwin, F.
1877. On the glandular bodies of Acacia sphaerocephala and Cecropia peltata
serving as food for ants, with an appendix on nectar-glands of the com-
mon braken fern, Pteris aquilina. J. Linn. Soc. (Bot.), 15: 398-409.
Evans, W. C.
1976. Bracken thiaminase-mediated neurotoxic syndromes. J. Linn. Soc.
(Boty 73: 113-132.
Jones, C. G. and R. D. Firn
1978. The role of phytoecdysteroids in bracken fern Pteridium aquilinum (L.)
Kuhn as defense against phytophagous insect attack. J. Chem. Ecol., 4:
117-138.
1979a. Resistance of Pteridium aquilinum to attack by non-adapted phytopha-
gous insects. Biochem. Syst. Ecol., 7: 95-101.
1979b. Some allelochemicals of Pteridium aquilinum and their involvement in
resistance to Pieris brassicae. Biochem. Syst. Ecol., 7: 187-192.
Lawton, J.
1976. The structure of the arthropod community of bracken Pteridium aquili-
num (L.) Kuhn. J. Linn. Soc. {Bot.), 73: 187-216.
Page, C. N.
1976. The taxonomy and phytogeography of bracken — a review, J. Linn. Soc.
{Bot.\ 73: 1-34.
NATURAL HISTORY OF THE WORKERLESS
INQUILINE ANT POGONOMYRMEX COLEI
(HYMENOPTERA: FORMICIDAE)*
By Steven W. Rissing
Department of Zoology
Arizona State University
Tempe, Arizona 85287
At least 10 workerless inquiline ant species are known from North
America (Francoeur 1968, 1981; Wilson 1971, 1976; Talbot 1976;
Buschinger 1979; DuBois 1981; Snelling 1981), most only from
original collections. In this paper I present field and laboratory
observations of Pogonomyrmex colei Snelling a new, apparently
workerless, inquiline ant inhabiting a colony of Pogonomyrmex
rugosus.
P. colei appears to be a very rare species: extensive searching of
the type locality for 4 yr has resulted in discovery of only a single
colony. Nonetheless, observations on this colony provide insight
into several important aspects of inquiline ant biology. P. colei is
also of interest since it is the second apparently workerless con-
generic inquiline inhabiting colonies of P. rugosus. Cole discovered
the first inquiline species, Pogonomyrmex anergismus, near Silver
City, New Mexico apparently prior to any major flight since he
exposed “more than one hundred” inquiline reproductives upon
opening the host nest (Cole 1954, 1968). Since host species mating
flights occur soon after rain during mid to late summer (Holldobler
1976; Rissing personal observation), it seemed reasonable to suspect
P. anergismus responds to the same environmental cues for mating
as does its host. Accordingly, in an effort to rediscover P. anergis-
mus, I routinely checked most P. rugosus nests on a 25 ha study area
in Boulder City, Nevada for flight activities and possible presence of
inquilines during late summer 1978 and 1979 (study area described
in Rissing 1981). P. colei was discovered during this effort.
* Manuscript received by the editor June 7, 1983
321
322
Psyche
[Vol. 90
Observations
Mating Activities and Season. Five P. colei males were collected at a
single P. nigosus nest during the morning of 13 August 1978; a
series of thunderstorms and rain had occurred 12 hr earlier.
Frenzied host worker activity suggested a mating flight or similar
activity occurred immediately prior to my arrival. No flights of
either species occurred at any nearby P. rugosus nests observed
simultaneously.
I observed a complete inquiline and host flight at this same nest
on 15 September 1978 following an extensive rain storm the
preceding day. Flights were occurring at 2 of 23 nearby P. rugosus
nests; P. colei was not found at any other nest. Mating activities
began with accumulation of several hundred host workers in and
around the nest crater. These workers pugnaciously defended the
area throughout both flights as is typical during P. rugosus flights
(Rissing, personal observation). As ground and air temperatures
increased male P. colei climbed to the crater and were soon joined
by much larger females. While both sexes of P. colei are winged,
mating occurred at the nest entrance followed by females flying
from the area and males re-entering the nest. Such in situ mating is
common in rare ant species apparently due to very low probability
of reproductives finding individuals from other nests with which to
mate (Wilson 1963). Following copulation and departure of P. colei
females, male and female P. rugosus flew from the crater as the
temperature continued to climb. Reproductive forms of P. rugosus
fly to a site away from the nest and copulate there (Holldobler
1976). Mating activities of host and inquiline were separated by at
least 30 min and, perhaps more importantly, 3° C ground tempera-
ture (Table 1). Reproductive forms of each species were seen
occasionally in the nest entrance during the mating activity of the
other. On at least one occasion, P. colei males tried unsuccessfully
to mount a P. rugosus female. During this flight I observed no
differences in behavior of host workers to host or inquiline repro-
ductives. P. rugosus workers frequently encircled copulating pairs
of P. colei and frantically ran around them, although they never
interfered.
During 1979 routine observations were begun at the study area on
18 September. A complete P. colei flight was observed at the host
nest during the afternoon of 30 September immediately following a
1983]
Rissing — Pogonomyrmex colei
323
trace of rain. No flights of either species were observed at 35 nearby
P. rugosus nests during this time. On 8 October 1979 I poured
approximately 7.5 1 of water directly onto the host nest crater
resulting in an immediate flight of P. colei. This procedure was
repeated unsuccessfully on 17 and 18 September 1982. Viability of
the host nest (as determined by worker activity, size of crater and
refuse pile, and absence of plants growing in the crater) has
remained constant and similar to that of nearby P. rugosus colonies
from 1978 to 1982. I have never observed any forms that might be
considered P. colei workers.
Colony foundation. Ten newly mated P. colei females from the 15
September 1978 flight were placed into a 7.5 m high flight enclosure
made of plastic sheeting and permitted to fly. Subsequent to this all
females removed their wings but did not dig burrows when placed
into laboratory nest boxes containing moist sand. Five of these
dealate inquilines were transferred to 5 laboratory nests containing
only newly mated P. rugosus queens. These P. rugosus queens had
been collected one week earlier at a mating site 3.2 km from the host
nest making it unlikely that they were related to the host colony.
Four of these laboratory nests contained a single, mated dealate P.
rogusus queen; the fifth contained two P. rugosus queens. The P.
colei queen added to the nest with two P. rugosus queens was
immediately attacked and removed from the glass tube occupied by
the P. rugosus queens. Of the P. colei queens added to the single
queen P. rugosus colonies, one was found dead within several hours
(decapitated), and the other was found dead (entire) 5 d later. The
other two P. colei queens lived peacefully along side the P. rugosus
queens for at least a month. During this time I frequently observed
the P. colei queens grooming the P. rugosus queens; P. rugosus
queens did not reciprocate. These last two colonies ultimately failed
during (or possibly in response to) transportation from Boulder City
to Seattle.
Five other newly mated, dealate P. colei queens were released in
the field at the entrance of large, active P. rugosus colonies near the
host nest. Inquilines were always removed immediately from the
nest by one or more workers and dropped several meters from the
crater. The P. colei queens made no attempt to re-enter these nests
following removal.
324
Psyche
[Vol. 90
Discussion
Repeated (and continuing) attempts to find P. colei or P.
anergismus around Boulder City, NV, or Globe, AZ, where a single
P. colei male has been collected (Snelling 1981) have yet to be
successful. Nonetheless, observations of P. colei from the type nest
in Boulder City provide insight into several questions of general
inquiline biology including possible method of inquiline entry into
host colonies and fate of host queen.
Inquiline entry into host colonies. Newly mated P. colei queens are
accepted into 1 week old workerless host nests in the laboratory,
while they appear incapable of entering established host nests in the
field (see above). Similar observations have been made in laboratory
experiments with the inquiline Plagiolepis xene and its host,
Plagiolepis pygmaea (Passera 1964). This suggests that at least some
inquiline species enter a host colony at the founding stage prior to
production of any workers. That this may occur in the field is
supported by discovery of a workerless inquiline queen {Strumi-
genys xenos) in an incipient host colony containing one queen,
brood and a single worker of Strumigenys perplexa (Brown 1955).
If entry into host colony commonly occurs at host colony
foundation in some species of inquilines, overlap with host species
flight season would be advantageous. Since all nests of a given
species in a locality tend to have a longer “flight season” than any
single nest (e.g. for P. rugosus see Holldobler 1976), the inquiline
might further be expected to lengthen its flight season relative to
that of its host colony to take advantage of the entire flight season
and availability of founding nests in its locality. The extended flight
season of P. colei relative to that of P. rugosus may occur for these
reasons. Similarly, occurrence of P. anergismus reproductives dur-
ing mid September in the type nest reported by Cole (1954, 1968)
may also indicate inquiline-host reproductive overlap.
Fate of host queens. Simultaneous production of host and inquiline
reproductives during the 1978 flight (Table 1) strongly suggests
coexistence of host and inquiline queen(s) at that time. Continuing
existence of the host colony until at least September 1982 further
substantiates this. Estimates of maximum longevity of worker ants
is 1-2 yr (Rosengren 1971, Brian 1972, Nielsen 1972). Further, there
has never been a reported case of queen adoption in any Pogono-
myrmex species. For the host colony to have a normal foraging
1983]
Rissing — Pogonomyrmex colei
325
Table 1. Summary of mating activities of P. colei and P. rugosus in Boulder City,
Nevada, 15 September 1978.
Time
Ground
Air
08:55
Temp. °C
Temp. °C2
09:10
20.5
20.5
09:37
10:03
21.0
21.5
10:45
26.0
23.8
12:15
29.2
25.5
12:47
32.6
26.4
13:15
33.4
30.8
Activity
Reproductives of both species in nest
entrance
P. colei reproductives on crater
Number of P. colei increases
First P. colei copulation
First P. colei female flies
Last P. colei female flies
First P. rugosus male and female fly
Last P. rugosus flies
'Temperature as determined by holding tip of a Yellow Springs Instruments direct
read thermistor (YSI #405) on ground surface; temperature read on a Yellow Springs
Instruments telethermometer (YSI #43TA).
^Temperature determined as above with thermistor 30 cm above ground and shaded.
group size in 1982, the host queen must have been alive during the
1978 and 1979 inquiline flights. Although inquiline-host coexistence
has been regarded as a “primitive” inquiline trait (Wheeler 1933,
Haskins and Haskins 1964), it offers the obviously adaptive advan-
tage of a continuously renewed host worker force for the inquiline.
Coexistence occurred in the type nest of P. colei and appears
common in other workerless inquiline species where information
regarding fate of host queen(s) is available (Table 2).
Host queen elimination does occur in at least two well docu-
mented cases (Table 2). Wilson (1971) suggests such behavior may
develop in short-lived inquiline species; inquiline longevity, how-
ever, may be more of an effect than a cause of this behavior. Host
queen elimination may be adaptive only when inquiline entry is
gained by a queen after development of a host worker force. Host
workers appear to be the primary defense against inquiline entry in
many colonies. In order to be accepted by host workers, it may be
necessary for the prospective inquiline queen to first render the
prospective host colony queenless. In those cases where host queens
are known or highly suspected of being eliminated (Table 2), the
inquiline queen enters an established colony containing workers. In
at least one of these cases, Epimyrma vandeli, the inquiline must
fight with host workers until she is able to kill the host queen.
Recent discovery that E. vandeli is a degenerate slave-maker
326
Psyche
[Vol. 90
Table 2. Fate of host queen(s) for workerless inquilines. Only those species whose
host queen(s) fate is known are listed.
Inquiline species
Host species
Fate of host
queen(s)
Reference
MYRMECIINAE
Myrmecia Myrmecia
inquilina vindex
survives
Douglas and Brown 1959
Haskins and Haskins 1964
MYRMICINAE
Myrmica Myrmica
hirsuta sabuleti
survives
Elmes 1974a, 1978
Sifolinia
laurae
Myrmica
sabuleti
survive
Brian 1972
Pogonomyrmex
colei
Pogonomyrmex
rugosus
survive*
this study
Anergates
atratulus
Tetramorium
caespitum
apparently
killed by host
workers
Wheeler 1910, Crawley 1912,
Donisthorpe 1915, Creighton
1950
Teleutomyrmex
schneideri
Tetramorium
caespitum
survives
Stumper 1950+,
Kutter 1969
Leptothorax
kutteri
Leptothorax
acervorum
survive
Buschinger 1965
Leptothorax
minutissimus
Leptothorax
curvispinosus
survive
Smith 1942,
Buschinger 1981
Epimyrma
vandeli
Leptothorax
nigriceps
killed by
inquiline
Vandel 1927
Stumper and Kutter 1951
Doronomyrmex
pads
Leptothorax
acervorum
survive
Kutter 1945+, 1969+
Monomorium
pergandei
Monomorium
minimum
survive*
Creighton 1950
Doronomyrmex
pocahontas
Leptothorax
muscorum
survive*
Buschinger 1979
Monomorium
adulatrix
Monomorium
salomonis
killed by
host workers
Wheeler 1910
Forel 1930
1983]
Rissing — Pogonomyrmex colei
327
Table 2. Continued.
Inquiline species
Host species
Fate of host
queen(s)
Reference
Mononiorium
talbotae
Mononiorium
minimum
survives
Talbot 1979
Struniigenys
xenos
Struniigenys
perplexa
survive
Brown 1955, Taylor 1967
FORMICINAE
Plagioiepsis Plagiolepsis
xene pygmaea
survive
Le Masne 1956;
Passera 1964, 1966, 1972
Aporotnyrmex
ampeloni
Plagiolepis
vindohonensis
survives
Faber 1969+
♦Presence of host queen(s) determined by presence of host reproductives
+Cited in Wilson (1971)
(Buschinger 1981, Buschinger and Winter 1982) may explain this
behavior which is rather unusual among most other inquilines
(Table 2). Only the extreme inquiline Teleutomyrmex schneideh is
known to enter established host nests without having to eliminate
host queens; these inquilines may produce a substance highly
attractive to host workers (reviewed in Wilson 1971).
Comparison with P. anergismus and other workerless inquilines. P.
colei may represent an intermediate form between its host P.
rugosus and the closely related workerless inquiline P. anergismus
(for a complete discussion of morphological differences see Snelling
1981). Discovery of P. colei adds the genus Pogonomyrmex to a
growing list of ant genera with more than one workerless inquiline
species (Table 2). Such “concentration” of inquilines into a few
genera may occur either due to non-random search by myrmeco-
logists {P. colei was discovered during an intentional search for
Pogonomyrmex inquilines) or because certain genera are more
likely to give rise to inquilines. The basic biology of the inquiline-
rich genera, however, is quite variable suggesting several evolution-
ary routes may lead to workerless inquilinism. The genus Lepto-
thorax, for example, has small, ephemeral colonies subject to slave
raids from numerous species and has given rise to several closely
328
Psyche
[Vol. 90
related Epimyrma inquiline species, themselves degenerate slave-
makers (Buschinger 1981, Buschinger and Winter 1982). Myrmica,
on the other hand, has larger colonies and many species that are
highly polygynous (Brian 1972; Elmes 1974a,b); this genus has given
rise to at least 7 workerless inquiline species: Myrmica faniensis (van
Boven 1970), Myrmica hirsuta (Elmes 1974a, 1978), Myrmica
lampra (Francoeur 1968, 1981), Myrmica myrmecophila (Bernard
1968), Myrmica quebecensis (Francoeur 1981), Sifolinia karavajevi
(Kutter 1969) and Sifolinia laurae (Brian 1972), the Sifolinia species
likely being congeneric with the other Myrmica species (Elmes
1978). Monomorium is similar with polygynous species (Dennis
1938, Cole 1940, Gregg 1945) and a number of congeneric inquilines
(reviewed in Wilson 1971, see also Talbot 1979 and DuBois 1981).
These inquiline species may have evolved through a process of some
polygynous host queens acquiring the trait of laying only repro-
ductive eggs (Buschinger 1970, Elmes 1978). To this list must be
added the genus Pogonomyrmex whose basic biology is unlike any
of the above three host genera. Colonies are substantially larger
than Leptothorax, Myrmica or Monomorium (Lavigne 1969, Ro-
gers et al. 1972, Whitford et al. 1976, MacKay 1981), strictly
monogynous (Lavigne 1969, Holldobler and Wilson 1977, MacKay
1981), with no slave-making or similar behavior in any species.
Evolutionary processes giving rise to P. colei and P. anergismus are
likely different from those that have given rise to the Leptothorax,
Myrmica or Monomorium inquilines. Certainly, the idea of mul-
tiple evolutionary pathways leading to workerless inquilinism is not
new (see Wheeler 1919, Buschinger 1970, Wilson 1971). Continued
study and search for workerless inquilines can only serve to clarify
this challenging evolutionary process.
Acknowledgements
J. Alcock, G. B. Pollock, R. R. Snelling, G. C. Wheeler, and J.
Wheeler have constructively reviewed earlier drafts of this manu-
script. Laboratory space in Boulder City, Nevada was kindly
provided by the University of Nevada Desert Research Institute,
Desert Biology Research Center and the U.S. Bureau of Mines,
Boulder City office. Portions of this work have been supported by
an NSF Graduate Fellowship, NSF grant DEB 78-02069 (T. W.
Schoener, principal investigator), NSF grant DEB 82-07052, and an
Arizona State University Faculty Grant-in-Aid.
1983]
Rissing — Pogonomyrmex colei
329
Literature Cited
Bernard, F.
1968. Les fourmis (Hymenoptera Formicidae) d’Europe occidentale et septen-
trionale. Faune de I’Europe et du Bassin Meditteraneen, No. 3. Masson
et Cie, Paris, France.
Boven, J. K. a. van
1970. Myrmica faniensis, une nouvelle espece parasite. Bull. ann. soc. entomol.
Belg. 106: 127-132.
Brian, M. V.
1972. Population turnover in wild colonies of the ant Myrmica. Ekol. Pol. 20:
43-53.
Brown, W. L.
1955. The first social parasite in the ant tribe Dacetini. Insectes Soc. 2: 181-
186.
Buschinger, a.
1965. Leptothorax (Mychothorax) kutteri n. sp. eine sozialparasitische Ameise
(Hymenoptera, Formicidae). Insectes Soc. 4: 327-334.
1970. Neue Vorstellungen zur Evolution des Sozialparasitismus und der
Dulosis bei Ameisen (Hym., Formicidae). Biologisches Zentralblatt 88:
273-299.
1979. Doronomyrmex pocahontas n. sp., a parasitic ant from Alberta, Canada
(Hymenoptera, Formicidae). Insectes Soc. 26: 216-222.
1981. Biological and systematic relationships of social parasitic Leptothoracini
from Europe and North America. Pages 211-222 in P. E. Howse and J.-
L. Clement, eds., Systematics Association Special Volume No. 19,
Biosystematics of Social Insects. Academic Press, New York.
Buschinger, A. and U. Winter.
1982. Evolutionary trends in the parasitic ant genus Epimyrma. in: M. D.
Breed, C. D. Michener and H. E. Evans, eds.. The biology of social
insects, pp. 266-269. Westview Press, Boulder, Colorado.
Cole, A. C.
1940. A guide to the ants of the Great Smoky Mountains National Park,
Tennessee. Am. Midi. Natur. 24: 1-88.
1954. Studies of New Mexico ants. VII. The genus Pogonomyrmex with
synonymy and a description of a new species (Hymenoptera: Formici-
dae). J. Tenn. Acad. Sci. 29: 115-121.
1968. Pogonomyrmex harvester ants. University of Tennessee Press, Knox-
ville.
Crawley, W. C.
1912. Anergates atratulus Schnek., a British ant, and the acceptance of a $ by
Tetramorium caespitum, L. The Entomol. Rec. 24: 218-219.
Creighton, W. S.
1950. The ants of North America. Bull. Mus. Comp. Zool. Harvard 104: 1-585.
Dennis, C. A.
1938. The distribution of ant species in Tennessee with (special) reference to
ecological factors. Ann. Entomol. Soc. Am. 31: 267-308.
Donisthorpe, H.
1915. British ants, their life-history and classification. Wm. Breden & Son,
Plymouth, England.
330
Psyche
[Vol. 90
Douglas, A. and W. L. Brown.
1959. Myrmecia inquilina new species; the first parasite among the lower ants.
Insectes Soc. 6: 13-19.
DuBois, M. B.
1981. Two new species of inquilinous Monomorium from North America
(Hymenoptera:Formicidae). Univ. Kansas Sci. Bull. 52: 31-37.
Elmes, G. W.
1974a. The effect of colony population size in three species of Myrmica
(Hymenoptera, Formicidae). Insectes Soc. 21: 213-230.
1974b. Colony populations of Myrmica sulcinodis Nyl. (Hym., Formicidae).
Oecologia (Berl.) 15: 337-343.
1978. A morphometric comparison of three closely related species of Myrmica
(Formicidae), including a new species from England. Syst. Entomol. 3:
131-145.
Faber, W.
1969. Beitrage zur Kenntnis sozialparasitischer Ameisen. 2. Aporomyrmex
ampeloni nov. gen., nov. spec. (Hym. Formicidae), ein neuer per-
manenter Sozialparasit bei Plagiolepis vindobonensis Lomnicki aus
Osterreich. Pflanzenschutzberichte 39: 39-100.
Forel, a.
1930. The social world of the ants compared with that of man. (Tr; C. K.
Ogden). Albert & Charles Boni, New York.
Francoeur, a.
1968. Une novelle espece du genre Myrmica au Quebec (Formicidae, Hy-
menoptera) Naturaliste Can. 95: 727-730.
1981. Le groupe Nearctique Myrmica lampra (Formicidae, Hymenoptera).
Can. Ent. 113: 755-759.
Gregg, R. E.
1945. Two new forms of Monomorium (Formicidae). Psyche 52: 62-69.
Haskins, C. P. and E. F. Haskins.
1964. Notes on the biology and social behavior of Myrmecia inquilina. The
only known myrmeciine social parasite. Insectes Soc. 11: 267-282.
Holldobler, B.
1976. The behavioral ecology of mating in harvester ants (Hymenoptera:
Formicidae: Pogonomyrmex). Behav. Ecol. Sociobiol. 1: 405-423.
HOlldobler, B. and E. O. Wilson.
1977. The number of queens: an important trait in ant evolution. Natur-
wissenschaften 64: 8-15.
Kutter, H.
1945. Eine neue Ameisengattung. Mitt. Schweiz. Entomol. Ges. 23: 81-94.
1969. Die sozialparasitischen Ameisen der Schweiz. Neujahrsblatt herausge-
geben von der Naturforschenden Gesellschaft in Zurich 171: 1-62.
Lavigne, R. J.
1969. Bionomics and nest structure of Pogonomyrmex occidentalis (Hy-
menoptera: Formicidae). Ann. Entomol. Soc. Am. 62: 1166-1175.
Le Masne, G.
1956. Recherches sur les fourmis parasites Plagiolepis grassei et 1’evolution des
Plagiolepis parasites. C. R. Acad. Sci. (Paris) 243: 673-675.
1983]
Rissing — Pogonomyrmex colei
331
MacKay, W. P.
1981. A comparison of the nest phenologies of three species of Pogono-
myremex harvester ants (Hymenoptera; Formicidae). Psyche 88: 25-74.
Nielsen, M. G.
1972. Production of workers in an ant nest. Ekol. Polska 20: 65-71.
Passera, L.
1964. Donnees biologiques sur la fourmi parasite Plagiolepis xene Starcke.
Insectes Soc. 1: 59-70.
1966. Fecondite des femelles au sein de la myrmecobiose Plagiolepis pygmaea
Latr. — Plagiolepis xene (Hymenopteres, Formicidae). C. R. Acad. Sci.
(Paris) 263: 1600-1603.
1972. Longevite des fourmilieres de Plagiolepis pygmaea Latr. infestees par le
parasite social Plagiolepis xene St. (Hymenoptera, Formicidae). C. R.
Acad. Sci. (Paris) 275: 409-411.
Rissing, S. W.
1981. Prey preferences in the desert horned lizard: influence of prey foraging
methods and aggressive behavior. Ecology 62: 1031-1040.
Rogers L., R. Lavigne and J. L. Miller.
1972. Bioenergetics of the western harvester ant in the shortgrass plains
ecosystem. Environ. Entomol. 1: 763-768.
Rosengren, R.
1971. Route fidelity, visual memory and recruitment behavior in foraging
wood ants of the genus Formica (Hymenoptera, Formicidae). Acta
Zool. Fennica 133: 1-106.
Smith, M. R.
1942. A new apparently parasitic ant. Proc. Entomol. Soc. of Washington 44:
59.
Snelling, R. R.
1981. The taxonomy and distribution of some North American Pogono-
myrmex and descriptions of two new species (Hymenoptera: Formici-
dae). Bull. Southern California Acad. Sci. 80: 97-112.
Stumper, R.
1950. Les associations complexes des fourmis. Commensalisme, symbiose et
parasitisme. Bull. Biol. Fr. Belg. 84: 376-399.
Stumper, R. and H. Kutter.
1951. Sur I’ethologie du nouveau myrmecobionte Epimyrma stumperi (nov.
spec. Kutter). C. R. Acad. Sci (Paris) 233: 983-985.
Talbot, M.
1976. The natural history of the workerless ant parasite Formica talhotae.
Psyche 83: 282-288.
1979. Social parasitism among ants at the E. S. George preserve in southern
Michigan. Great Lakes Entomol. 12: 87-89.
Taylor, R. W.
1967. The Australian workerless inquiline ant Strumigenys xenos Brown
(Hymenoptera-Formicidae) recorded from New Zealand. New Zealand
Entomol. 4: 47-49.
Vandel, a.
1927. Observations sur les moeurs d’une fourmi parasite: Epimyrma vandeli
Santschi. Bull. soc. entomo. Fr. 1927: 289-295.
332
Psyche
[Vol. 90
Wheeler, W. M.
1910. Ants: their structure, development and behavior. Columbia University
Press, New York.
1919. The parasitic aculeata, a study in evolution. Proc. Am. Philos. Soc. 58:
1-40.
1933. A second parasitic Crematogaster. Psyche 40: 83-86.
Whitford, W. G., P. Johnson and J. Ramirez.
1976. Comparative ecology of the harvester ants Pogonomyrmex barbatus (F.
Smith) and Pogonomyrmex rugosus (Emery). Insectes Soc. 23: 1 17-132.
Wilson, E. O.
1963. Social modifications related to rareness in ant species. Evolution 17: 249-
253.
1971. The insect societies. Belknap Press of Harvard University Press, Cam-
bridge, Massachusetts.
1976. The first workerless parasite in the ant genus Formica (Hymenoptera:
Formicidae). Psyche 83: 277-281.
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 OE PSYCHE
Requests for information about back volumes of Psyche should
be sent directly to the editor.
E. 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.
FOR SALE
Republication of Frederick Valentine Melsheimer’s 1806 “A Catalogue of Insects of
Pennsylvania”, the first separate work devoted to American insects. The facsimile
lists more than 1300 species of Coleoptera (other orders were not completed), and
includes a short biography of Melsheimer. Price: U.S., $5.00 (overseas, airmail
$6.50). Checks payable to Entomological Society of Pennsylvania, c/o Entomology
Dept., Pennsylvania State University, University Park, PA 16802, U.S.A.
ISSN 0033 2615
PSYCHE^
A JOURNAL OF ENTOMOLOGY
founded in 1874 by the Cambridge Entomological Club
Vol. 90
CONTENTS
Dedication: Philip J. Darlington, Jr. Frank M. Carpenter 333
The biology of Myrnwxenus t^ordiaf'ini Ru/sky, a slave-making ant (Hyme-
noptera. Formicidae). Alfred Buschinf^er, Ursala Winter, and
Walt her Faber 335
Temperature-Induced changes in the calls of the Green Lacewing, Chry.soperla
plorahnnda (Ncuropicra-. Chr\sop\di\c). Charles S. Henry 343
Social organization in l.eptothorax ants: within- and between- species pat-
terns. Joan M. Her hers 361
‘Protest’ sounds of a grasshopper: predator-deterrent signal? SyrH A. Blond-
heini and Fllezer Frankenherf' 387
Age poKethism: its occurrence in the ant. Pheidole hortensis. and some general
considerations. Prassede Calahi. Janies F A. Traniello. and
Michael H. Werner 395
Dail\ rh\ thms in social actix ities (d the harxester ant. Poyonotm rniex hadius.
Deborah M. Gordon 413
Behax ior of the slaxe-making ant, Harpayoxenus aniericanns (Fmery). and its
host species under “seminatural” laboratorx conditions ( Hx menoptera.
Formicidae). Thomas M. AUowavixud
Maria Guadalupe Del Rio Pesado 425
1983
Index to x olume 90
449
CAMBRIDGE ENTOMOLOGICAL CLUB
Officers for 1983-1984
Margaret Thayer
Donald S. Chandler
Scott E. Miller
Frank M. Carpenter
Andrew Spielman
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
M. D. Bowers, Assistant Professor of Biology, Harvard University
Alfred F. Newton, Jr., Curatorial Associate in Entomology, Harvard
University
E. O. Wilson, Baird Professor of Science, Harvard University
President
Vice-President
Secretary
Treasurer
Executive Committee
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: $13.00 domestic and foreign. Single copies, $4.00
Checks and remittances should be addresssed 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 required 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 $10.00
each, and for full page half-tones, $12.00 each; smaller sizes in proportion. There is
ordinarily no additional charge for setting tables of less then six columns; for tables
of six or more columns the cost is $25 per page.
Psyche, vol. 90, no. 3, for 1983, was mailed December 9, 1983
The Lexington Press, Inc., Lexington, Massachusetts
PHILIP JACKSON DARLINGTON, JR.
FHOTOGRAFIl TAKHN IN 1971
This issue of Psyche is dedicated to the memory of Philip J.
Darlington, Jr., who died in his 80th year in Cambridge, Massa-
chusetts, on December 16, 1983.
I first met Philip Darlington at a meeting of the Cambridge
Entomological Club on January 8, 1924, held at the Bussey Institu-
tion. The event was a significant one for us. We were both under-
graduates in the same class at Harvard College and for the next
333
sixty years we remained close friends, as well as colleagues in the
Museum of Comparative Zoology.
Philip was very active in the Club. He was secretary in 1931,
vice-president in 1933 and 1940, and president in 1934, 1941, and
1946, and a member of the editorial board of Psyche for thirty years.
His first talk at a Club meeting, in March, 1927, was an account of
insect collecting at the Harvard Tropical Laboratory in Soledad,
Cuba. At many other meetings over the years we enjoyed hearing
about his research and his field trips in Australia, New Guinea, and
Colombia, as well as on various Caribbean islands.
He was born in Philadelphia in 1904. After attending Exeter
Academy in New Hampshire, he entered Harvard University, from
which he received his A.B. degree in 1926 and his Ph.D. in 1931. The
following year he was appointed Assistant Curator of Insects in the
Museum of Comparative Zoology. In 1939 he became the H.C. Fall
Curator of Coleoptera, and in 1952 he assumed the position of
Curator of Insects, which he held until his appointment as Alex-
ander Agassiz Professor of Zoology. He retired in 1971.
Although Philip was primarily an entomologist and chiefly con-
cerned with Coleoptera, he had very broad interests in all aspects of
natural history. His knowledge of plants and of all vertebrate
groups was extraordinary. With such interests he was inevitably led
into studies on evolutionary theory and especially zoogeography, on
which he published several outstanding books and numerous tech-
nical papers.
Frank M. Carpenter, editor
334
PSYCHE
Vol. 90
1983
No. 4
THE BIOLOGY OF MYRMOXENUS GORDIAG/NI RUZSKY,
A SLAVE-MAKING ANT (HYMENOPTERA, FORMICIDAE)
By
Alfred Buschinger, ' Ursula Winter,' and Walther Faber-
Introduction
Myrmoxenus gordiagini was described by Ruzsky (1902) from
material which he had collected in the Akmolinsk area in Soviet
Russia, near the town of Koktschetaw. The ant was always found
living together with a newly described host species, Leptothorax
serviculus Ruzsky. The colonies inhabited narrow galleries between
and underneath small stones in the rocky slopes of a hilly region,
with some birch and spruce trees. Finzi (1924) described a subspe-
cies, Myrmoxenus gordiagini menozzii, from the Yugoslavian pen-
insula of Istria. Only one male and one female were found within
moss and soil at the foot of an oak tree, together with numerous
females and workers of Leptothorax unifaseiatus (Latreille). Finzi
therefore believed that his new subspecies was living with that host
species. Finally, in 1925, Soudek established a new genus, Myrme-
taerus, for a new species, mierocellatus, that he had collected near
Kotor in Dalmatia, Yugoslavia. Although he explicitly discussed the
close relationship of M. mierocellatus with Myrmoxenus, he des-
cribed this ant as representing a new species and genus “as a provi-
sional arrangement” (Soudek, 1925). M. mierocellatus was found
under a stone in a deciduous forest, in a mixed colony with Lepto-
'Institut fur Zoologie der Technischen Hochschule Darmstadt, Schnittspahnstr. 3,
D-6100 Darmstadt, FRG.
-Dr. Walther Faber, Vienna, died in June, 1979. Among his papers we found a
description of the colony foundation behavior of M. f'ordia/^ini, and also some
important information on localities where he had collected this species.
Manuscript received by the editor June 23, 1983.
335
336
Psyche
[Vol. 85
thorax nylancieri (Forster). All our material also was collected in
Istria, and in the Dalmatian island Krk\ Since the descriptions of all
the three forms mentioned above are nearly identical, we assume
that Myrmetaerus microcellatus and Myrmoxenus gordiagini men-
ozzii are junior synonyms of Myrmoxenus gordiagini. We have been
unable, however, to check the type material.
Nothing has been known of the biology of Myrmoxenus! Myrme-
taerus except the fact that they were always found living together
with a host species belonging to the genus Leptothorax Mayr, sub-
genus Myrafant Smith (1950), and thus apparently represent so-
cially parasitic ants. W. Faber in 1972 recorded some observations
on their colony foundation behavior. Recently we found out that M.
gordiagini is a slave-making ant. The results of our experiments are
presented in the following sections.
Material and Methods
W. Faber collected a total of three Myrmoxenus colonies on 25
May 1972, on the slopes of a small valley NW of Baska, Krk. In the
very same locality we found nine additional colonies between
September 23 and 26, 1981. Another locality, where we gathered
two colonies on 4 August 1976, and one colony on 22 September
1981, is near the ruined town Dva grada, a few kilometers east of
Rovinj in Istria, and just 35 km south of the type locality of M.
gordiagini menozzii. Ten of these 15 colonies contained a Myr-
moxenus queen; presumably the queens of the other five colonies
were either lost during collecting or were missing prior to our col-
lecting. Myrmoxenus workers were present in varying numbers up
to about 40 (exact numbers cannot be given since all colonies were
kept alive for several breeding seasons, and thus produced addi-
tional workers). Two colonies contained only a queen and no
workers; supposedly they were newly founded.
Male and female as well as worker pupae were present in the
colonies collected on August 4, 1976, and a few adult sexuals were
found in field colonies on September 22 and 23, 1981.
^Recently we found an additional population on the island Rab, south of Krk. Four
colonies were collected on October 2, 1983. in an oak forest south of Suha Punta.
They contained one queen each, and in one colony we found two additional females
that were dCcUate but not inseminated.
1983]
Buschinger, Winter, & Faber — Myrnio.xenus
337
The host species in all 15 colonies was Leptothorax liehtensteini
Bondroit 1918. Up to about 200 host workers were found in the
Myrnioxenus colonies. Nest sites were underneath small, flat stones
in the soil, or in crevices between such stones. A common, and, in
our opinion, quite important character of the Myrmoxenus habitats
is the fact that they all were situated in rather shady places in a
deciduous forest or in the underbrush. We cannot reconstruct the
exact experimental device by which W. Faber studied the colony
founding behavior. From his records we conclude that the colonies,
which he had collected in May, produced sexuals until September.
On September 9 and 22, 1972, he noted “strong flight activities,”
and numerous Myrnioxenus females were dealate in the nests and
arenas. Several times he put five dealate Myrnioxenus females
together into the feeding arenas of L. liehtensteini colonies. Others
were placed into formicaries with different Leptothorax species.
Our newly collected colonies from 1976 and 1981 were kept in
formicaries and under artificial daily and annual temperature cycles
as described by Buschinger (1973, 1974, 1982) and Winter (1979a).
For initiating slave raids, we used arenas as depicted by Winter
(1979a) and Buschinger et al. (1980). During the raids the room
temperature was about 27° C. Contrary to our experiences with
Harpagoxenus or Epimyrma, which need bright sunshine or at least
blue sky for raiding, the Myrnioxenus seem to prefer a clouded sky.
Thus, the first raid which we observed in our labroatory took place
on a cloudy day; for the second one, on a sunny day, we closed the
window shades.
Colony Founding by Myrmoxenus gordiagini
As indicated above, we rely on the quite brief notes of W. Faber,
who observed colony founding by M. gordiagini females in 1972.
According to these notes, the young Myrnioxenus queen enters a
host species colony {L. liehtensteini), apparently soon after mating
and dealation, in late summer. Most Myrnioxenus females were
attacked and often killed by workers of the host species. In a few
experiments, however, a parasitic queen survived the attacks and at
last was accepted by the host species workers. She then assaulted the
host species queen in a very characteristic manner (Fig. 1). She
grasped the Leptothorax queen’s “throat” with her mandibles, and
338 Psyche [Vol. 85
Fig. 1. A queen of M vrmoxenus f'ordiaf'ini (right) is throttling the host species
queen during colony foundation (photograph: Faber).
throttled her repeatedly, and often for several hours, as was de-
scribed for Epimyrnia ravouxi females (Gdsswald 1930). Other
Myrmoxenus queens were seen to throttle the Leptothorax queens
by seizing their necks from the back. Like Epimyrma stumperi
females (Kutter 1951) the Myrmoxenus queens also scent them-
selves by first rubbing their legs over the surface of the victims, and
then over their own backs. Sometimes the Myrmoxenus queens
throttled some host species workers, too, or stung them to death.
Furthermore, they attacked the alate Leptothorax females which
were present in the nests, and killed some of them. In one L.
Hchtensteini colony, where five Myrmoxenus females had been put
1983]
Buschinger, Winter, & Faher — Myrmoxenus
339
on September 12, the queen was dead on September 17,
and one surviving Myrmoxenus female was observed to bite the
large queen larvae of the host species.
Experiments with L. parvulus (Schenck 1852) as host species did
not succeed; the Myrmoxenus queens were all killed.
Slave Raids of Myrmoxenus gordi acini
We observed two slave raids of one Myrmoxenus colony, on June
24 and July 7, 1982. The colony was collected in September, 1981.
After an artificial hibernation from 12 December 1981, in a constant
6°C until 22 April 1982, the colony began to bring up its larvae. The
first prepupae appeared on 4 June, six weeks after the end of hiber-
nation, the first worker and sexual pupae were recorded on 1 1 June.
Sexuals hatched towards mid-July, after the raids, and sexual
activity was observed in the beginning of September. A second
colony, which was kept under identical conditions, exhibited some
scouting activity between June 4 and 25, but did not conduct a raid.
The first colony was put into an arena on 4 June. Simultaneously
a colony of the host species, L. lichtensteini, was placed into another
part of the arena, which was subdivided by a plastic wall.
No Myrmoxenus workers were seen outside the nest until 22
June.
On June 23 and 24, between one and three Myrmoxenus workers
appeared in the arena. Scouting occurred between 1000 and 1500 on
June 23. On June 24, a hole in the separating wall of the arena was
opened, and a Myrmoxenus scout found the way through to the
host species territory at 1740. At 1754 this scout ran across the
liehtensteini nest. It returned to the Myrmoxenus nest, entered there
at 1804, and suddenly a mass of ants was whirling around inside the
nest entrance.
At 1808, a file of about 20 Myrmoxenus came out of the nest
(Fig. 2) and walked across the arena towards the hole. Sometimes
the procession stopped, milling around, apparently until the leading
scout had found its way again.
At 1905 the group had reached the entrance of the target nest,
and entered it one after the other. Almost no fighting could be
observed. After 6 minutes, the liehtensteini queen and most of the
workers had left their nest, carrying a few small larvae and eggs.
Only two Leptothorax were stung. Some Myrmoxenus workers
340
Psyche
[Vol. 85
Fig. 2. A raiding party of Myniio.xenus gordiaf'ini has just arrived at the nest of
the host species (photograph: Buschinger).
returned to their own nest, and at 2030 another file of 5 Myrmoxe-
nus arrived at the Leptothorax nest. By the next morning, the Myr-
moxenus colony had moved into the former Leptothorax nest.
The arena was then subdivided again, and a new Leptothorax
nest was placed in the position of the former Myrmoxenus nest.
The second raid, in the same arena, was observed two weeks later,
on 10 July. Scouting began at 0830 and a successful scout returned
to the Myrmoxenus nest at 0912. However, in this case, a file did not
form before 0933. At 1009 a total of 14 Myrmoxenus arrived at the
target nest, entered it at 1012, and a few minutes later they had
overwhelmed the colony and were in possession of its brood. Eight
Leptothorax were immediately stung to death. Contrary to the first
raid, this time the Myrmoxenus soon began to carry pupae and
large larvae back into their own nest. One returning Myrmoxenus,
at 1 150, led a further file of 15 conspecifics to the raided nest. At
1320 the Leptothorax nest was empty except for a few eggs and one
Leptothorax male. A total of 26 dead Leptothorax workers were
counted in the arena, indicating that during this raid more fighting
had occurred than during the first one.
Discussion
Our results, despite the low number of raids observed, reveal that
Myrmoxenus gordiagini is a slave-making ant. The organization of
1983] Buschinger, Winter, & Faber — Myrmoxenus 341
the raids is essentially the same as in Epimyrma ravouxi (Andre)
(Winter 1979b, Buschinger et al. 1980), and in the North American
Leptothorax duloticus (Wesson 1940), with group recruitment and
sting fighting.
The colony foundation behavior of Myrmoxenus also corre-
sponds to that observed in several species of the genus Epimyrma
(Kutter 1951, Gosswald, 1930, Buschinger and Winter, in press),
where the queens throttle the host species queens.
The only major difference between Myrmoxenus and Epimyrma,
therefore, pertains to antennal segmentation. Myrmoxenus females
and workers have 12-segmented antennae (males 13), like their host
species group, whereas Epimyrma has 11 -segmented antennae
(males 12). We have not yet decided whether this difference can
really justify the maintenance of the two genera; however, we are
convinced that the very particular raiding and colony foundation
behaviors have a monophyletic origin.
We are not entirely certain that Myrmoxenus gordiagini is the
correct name of our ants. That they are identical with Finzi’s M. g.
menozzii (1924) seems assured, since they were collected in the same
area. We are also sure about the identity of this M. g. menozzii with
Soudek’s (1925) Myrmetaerus microcellatus. However, if a later
revision reveals that M. gordiagini Ruzsky and M. g. menozzii Finzi
were two different species, then our material should be named M.
menozzii.
The different host species recorded for the three “forms” represent
a minor problem. Slave-making ants often have more than one host
species. Thus, Epimyrma ravouxi (Andre) enslaves Leptothorax
unifasciatus (Latr.), L. nigriceps Mayr, and L. qffinis Mayr, some-
times having two slave species together within one colony (Gdss-
wald 1930). It is also quite conceivable that both Finzi and Soudek
found their ants, as we did, with L. lichtensteini as host species; L.
lichtensteini has a superficial resemblance to L. unifasciatus, and it
is quite often confused with L. nylanderi.
Summary
Myrmoxenus gordiagini Ruzsky from Dalmatia, Yugoslavia,
conducts slave raids with group recruitment and sting fighting.
342
Psyche
[Vol. 85
Young queens enter the host species colonies {Leptothorax lichten-
steini Bondroit) and kill the Leptothorax queens by throttling them.
These biological features correspond well with those observed in the
genus Epimyrma.
Literature Cited
Bi schinger, a.
1973. The role of daily temperature rhythms in brood development of ants of
the tribe Leptothoracini (Hymenoptera; Formicidae). In: Effects of
Temperature on Ectothermic Organisms, ed. W. Wieser, pp. 229-232.
1974. Experimente und Beobachtungen zur Griindung neuer Sozietaten der
sklavenhaltenden Ameise Harpaf^oxenus suhlaevis (Nyl.). Ins. Soc.
21: 381-406.
1982. Stock manipulation in leptothoracine ants. Handout prepared for
Genetics Workshop (organized by R. H. Crozier), 9th Int. Congr.
lUSSI, Boulder, Colorado.
BrscniNGER, A., Ehrhardt, W., and U. Winter
1980. The organization of slave raids in dulotic ants— a comparative study
(Hymenoptera; Formicidae). Z. Tierpsychol. 53: 245-264.
Bl SCniNGER, A., AND U . WINTER
In press. The reproductive biology of a slavemaker ant, Epimyrma ravouxi, and
a degenerate slavemaker, E. kraussei (Hymenoptera: Formicidae).
Entom. Gen. 9.
Finzi B.
1924. Secondo contributo alia conoscenza della fauna mirmecologica della
Venezia Giulia. Boll. Soc. Ent. Ital. 56: 120-123.
GbsswAi.D, K.
1930. Die Biologic einer neuen Epimyrmaart aus dem mittleren Maingebiet. Z.
wiss. Zool. 136: 464-484.
Ki tter, H.
1951. Epimyrma stumperi Kutter(Hym. Formicid.), 2. Mitteilung. Bull. Soc.
Ent. Suisse 24: 153-174.
RrzsKV, M.
1902-03. Neue Ameisen aus Russland. Zool. Jb. Syst. 17: 469-484.
Smith, M. R.
1950. On the status of Leptothorax Mayr and some of its subgenera. Psyche 57:
29 30.
SOI DEK, S.
1925. Four new European ants. Ent. Rec. 37: 33-37.
Wesson, E. G , Jr.
1940. Observations on Leptothorax duloticus. Bull. Brooklyn Ent. Soc. 35:
73 83.
Winter, U.
1979a. Untersuchungen zum Raubzugverhalten der dulotischen Ameise Har-
paf’oxenus suhlaevis Ins. Soc. 26: 123-135.
1979b. Epimyrma goessw aUh Menozzi, eine sklavenhaltende Ameise. Naturwis-
senschaften 66: 58 1
TEMPERATURE-INDUCED CHANGES IN THE CALLS
OF THE GREEN LACEWING,
CHRYSOPERLA PLORABUNDA (NEUROPTERA:
CHRYSOPIDAE)*
By Charles S. Henry
The Biological Sciences Group
Box U-43, University of Connecticut
Storrs, Connecticut 06268
Animals communicate acoustically in many different ways and
for many different purposes; in fact, a vast literature exists on the
subject, and comprehensive efforts to summarize current knowledge
have not been attempted since the early 1960’s (e.g., Lanyon and
Tavolga 1960, Busnel 1963). Insects are especially rich in singing or
noise-making species, and it seems likely that every insect order will
eventually prove to include acoustically active taxa. Neuroptera has
long been regarded as a “silent” order of primitive, behaviorally
simple insects. However, recent work suggests that many, if not
most, species of the large neuropteran family Chrysopidae are
characterized by complex courtship displays accompanied by a
specialized form of acoustical signalling (Smith 1922, Toschi 1965,
Ickert 1968, Henry 1979 and 1983b). Such lacewing calls or songs
are not acoustical in the traditional sense, but instead consist of
species-specific substrate-borne vibrations produced by vigorous,
stereotyped jerking motions of the insects’ abdomens: a phenome-
non known as tremulation in other insects (Morris 1980; Henry
1980a, c), and found also in the ancestral neuropteroid taxon
Megaloptera (Rupprecht 1975). Calling behavior is most elaborate in
Chrysoperla Steinmann; conspecific males and females of species
within that genus cannot mate without first reciprocally exchanging
similar or identical vibrational signals in a prolonged courtship duet
(Henry 1980b, 1983b). The best studied species in this regard is
Chrysoperla plorabunda (Fitch), a common North American form
that for some years was considered synonymous with the morpho-
logically identical Eurasian species Ch. cornea (Stephens) (Tjeder
1960, Henry 1983a). Both sexes of this species, when sexually
* Manuscript received hy the editor August 2. 1983.
343
344
Psyche
[Vol. 85
receptive, produce brief, repetitive volleys of low-frequency abdom-
inal vibration at (approximately) one-second intervals for up to
several minutes at a time (Henry 1979, 1980c); in heterosexual duets,
the two partners alternately and repeatedly exchange single volleys
until copulation is achieved. Each volley in Ch. plorabunda is
characterized by a smoothly increasing and then decreasing ampli-
tude envelope and by pronounced frequency modulation (see
Fig. 1): the rate of abdominal vibration gradually declines to less
than half its initial value during the course of the volley (Henry
1980c).
Green lacewings of the genus Chrysoperla are unusual among
acoustical animals in that females are just as likely to sing as males
(Henry 1983b). In contrast, males alone (or principally) sing and call
conspecific females to them in most species of the best studied
animal taxa like birds (Catchpole 1982), lizards (Frankenberg 1982),
frogs (Littlejohn 1977), katydids (Dumortier 1963), homopterans
(Ossiannilsson 1949, Young 1980), and crickets (Walker 1962,
Dumortier 1963). Such unilateral male signalling has been inter-
preted in most acoustical animals as indicative of sexual selection
operating on the higher variance in reproductive success of males
(Halliday 1978, Otte 1974, Alexander 1975). On the other hand, the
unusual presence of calling behavior in both sexes of Chrysoperla
species may stem from the proven importance of their vibrational
signals in the reproductive isolation of closely related species. For
example, neither of the sibling, interfertile species Ch. plorabunda
and Ch. downesi (Banks) will respond to the song of the other, thus
precluding courtship duets and preventing interspecific mating
under natural conditions (Henry 1983b). If such signals are to be
effective as reproductive isolating mechanisms, however, they must
remain unambiguous to recipients over the wide range of tempera-
ture typically experienced by lacewings in the field. Abundant doc-
umentation exists of gross alteration in chirp rate, wing-stroke
frequency, pulse or chirp duration, or call notes by temperature
changes in many taxonomically disparate insect groups (Brooks
1882, Hayward 1901, Alexander 1956, Walker 1962, Dumortier
1963, Shaw 1968, Booij 1982), and similar temperature-related
changes should be characteristic of lacewing songs (see Henry 1982b
for preliminary work on Chrysopa oculata Say). One would also
predict that the calls of the two sexes of a given species of Chryso-
perla should vary in a closely parallel fashion over a wide range of
1983]
Henry — Chrysoperla plorabunda
345
temperatures, since mutual recognition of highly specific call fea-
tures is so important to the reproductive success of both members of
each courting pair. Here, 1 report on the effects of temperature
change on the principal parameters of the calls of individual male
and female Chrysoperla phrahunda from North America. This
paper also contains the first complete description of the frequency
structure of the abdominal volleys of that species, together with an
experimental analysis of the effects of abdominal mass on frequency
characteristics. Regressions of call parameters against temperature
in Ch. plorabunda are compared with those described for other
singing insects, in an attempt to identify any unifying principles.
Materials and Methods
A breeding colony of Chrysoperla plorabunda was started in the
fall of 1982 from seven males and ten females collected in a field of
senescent goldenrod (Solidago spp.) at Storrs, Connecticut. Subse-
quently, adults were maintained on a WheasT“/ sucrose diet while
larvae were fed ether-killed Drosophila spp. (see Henry 1979, 1983a
for details). Males for experimentation were drawn from second-
generation laboratory stock, while females were third-generation
insects; five unmated individuals of each sex were acoustically moni-
tored at various temperatures between 19.5°C and 29.8° C. For each
of five call characteristics of interest, 1 analyzed an average of 40
volleys of abdominal vibration per individual, delivered at three to
six different temperatures between the extremes mentioned above;
in no case was a regression line for an individual based upon fewer
than 18 volleys. Temperature was monitored within 25 cm of the
calling insects, and was controlled by heating and cooling an entire
120 cubic meter room.
Lacewings were induced to call either by playing back to them
cassette tape recordings of conspecific signals or by simulating such
signals by means of a sweeping audio frequency generator (Tek-
tronix^** FG 507) gated by a physiological stimulator (Harvard
340). Patterns of abdominal vibration were detected and analyzed
with techniques and equipment described in other papers (Henry
1980a, 1982b). Details of call parameters were obtained from Pola-
roid photographs of oscilloscope tracings, using conventional
overlay methods.
Several individuals of Ch. plorabunda from a breeding colony of
different geographical origin were selected for use in experiments
346
Psyche
[Vol. 85
testing the effect of abdominal weight on vibrational frequency.
This stock was bred from five males and five females collected in the
sagebrush country of southwestern Idaho on 24 May, 1983 by Dr.
James Johnson (University of Idaho). I tested one field-caught
male, two first-generation laboratory males, and one first-gener-
ation laboratory female for the frequency characteristics of their
calls at two different temperatures; after ascertaining that all were
essentially identical to one another and to Connecticut plorahunda
in those characteristics, I weighed each individual to the nearest
tenth of a milligram on a Mettler^“ H6T or Sartorius^“ 1212 MP
balance and added weight to the middle of their abdomens, using
water-based Liquid Paper^“. Mass-loaded specimens were then re-
weighed and tested again for the vibrational frequencies of their
calls at temperatures in the 25-30° C range. All such frequency
values were also adjusted to 27° C using linear regression C in Fig. 3.
Curve-fitting of paired variables to linear, exponential, or
logarithmic functions employed a program designed for a Hewlett-
Packard^** HP-25 pocket calculator. Any reference in the Results or
Figure I. Detailed fragment of the call of Ch. plorahunda, re-drawn from an
oscilloscope tracing, showing principal parameters A-E defined in text. A = initial
volley frequency, B = median volley frequency, C = terminal volley frequency, D =
volley repetition rate, E = volley duration.
1983]
Henry — Chrysoperla plorabunda
347
Discussion sections to “significant differences” indicates that the
means of two normally distributed samples were demonstrated to
differ from one another by a 2-tailed t-test using confidence limits of
95% or better. Values following a +/- sign are one standard
deviation of the mean.
Results
I measured five different characteristics of the calls of Chryso-
perla plorabunda males and females. These are listed and defined
below and illustrated in Fig. 1
A. Initial volley frequency: The cycling rate per second of the
first eight strokes of the abdomen, at the start of a volley of
abdominal vibration.
B. Median volley frequency: The cycling rate per second of the
eight abdominal strokes following the initial period defined
above.
C. Terminal volley frequency: The cycling rate per second of
the abdominal strokes that remain in a volley after A and B
have been deleted.
D. Volley repetition rate: The rate per minute at which a calling
lacewing produces volleys of abdominal vibration.
E. Volley duration: The length of time in seconds required for
completion of a volley of abdominal vibration.
Frequency of abdominal vibration versus temperature is tabu-
lated for all males and females in Table 1 and expressed as linear
regressions for each sex in Figure 2. Since vibration frequency
decreases during the course of a plorabunda volley, it is necessary to
subdivide each volley into the three portions A, B, and C, defined
above; frequencies measured for those portions are plotted sepa-
rately on the graph. Other call parameters like volley repetition rate
and volley duration are similarly plotted separately for each sex on
the same figure, using different units on the y-axis. All data are
combined for both sexes in Figure 3, which also displays the range
of variation in the calculated regression lines for all 10 individuals.
Several obvious features emerge from these tables and plots. First
of all, it appears (Table 1) that males and females differ from each
348
Psyche
[Vol. 85
o .E o
E >
® c/2
E X) ^
^ < 9
o 2 i!
> =5 c
Q S
L , Q
C. (U ^
<u c
“ c ^
= .2 S.
O ^ 0/
> ^
+ 1
+ 1
m m
vn Tt
rsi >/-i
r*^
+1
oc —
Tt <N
oo —
+1
+1
+1
00 rn
(N
+ 1
+ 1
+ 1
+1
oo
«/^ <N
— O
sD od
X
+ 1
+1
+ 1
+ 1
+ 1
+ 1
+ 1
+ 1
X v-2
1^
+1
+1
+1 +1 +1
— o^
vO
rsi </-i
m
+1
+1
+1
U-l
a
03
C
sD
sD
r*^
rsi
ON
«/->
r'
ON
OO
CQ
c/3
OO
OO
oo
r<-2
OO
'Ct
r-
ON
C/3
(U
o
u
O'"
</d
(N
<N
(N
rb
rd
sd
rd
id
rd
<u
f-
m
+1
+ 1
+ 1
+1
Tt
+1
+1
Tl-
+ 1
+1
Table 1. Principal characteristics of the calls of males and females of Chrv.soperla plorahiinda, as measured at several
different temperatures. Mean values and their standard deviations are tabulated. Sample si/es are entered parenthetically; in
each row. the same five males or five females produced the given number of measured volleys. Parameters A E are depicted
in Fig. I and defined in the text.
1983]
Henry — Chrysoperla plorabunda
349
other very little, at any temperature, in any major characteristics of
their calls. Particularly coincident between males and females are
the initial volley frequencies and the volley durations (Table 1 and
Fig. 2A and E) of Ch. plorabunda calls. There is a consistent ten-
dency for females to vibrate their abdomens in later portions of their
volleys at somewhat lower frequencies than do males (35 cycles/ sec.
vs. nearly 40 cps at 20°C; see Fig. 2C) and for volleys to be produced
rather more slowly by males than by females (about 66 volleys/ min
vs. nearly 70/ min at 29.5°C; same figure, D), but neither of these
differences is statistically significant.
Secondly, it can be seen from Figures 2 and 3 that the slopes for
the linear regressions of frequency versus temperature differ radi-
cally from one portion of a volley to another, gradually becoming
less steep as the volley progresses. Thus, initial portions of volleys
change frequency rapidly with temperature (slope = 4.27x; Fig. 3A),
while terminal portions remain within a much narrower range of
values over equivalent temperature extremes (slope = 0.960x). X-
intercepts also differ significantly for each of the three regressions
calculated from pooled frequency data on all ten individuals: inter-
cepts of 3.60, —1.27, and — 19.09°C respectively characterize initial,
median, and terminal portions of volleys. Regression of another call
parameter, volley repetition rate, against temperature produces a
line with a slope of 3.25x and an X-intercept of 9.14°C (Fig. 3D).
Thirdly, Figure 3E demonstrates that volley duration decreases
markedly with increasing ambient temperature, and that the mathe-
matical relationship between the two can equally well be interpreted
as linear or exponential. In theory, the total number of abdominal
strokes per volley could remain constant as temperature varies,
since the increased frequency of abdominal vibration at higher tem-
peratures would automatically shorten volley duration. However,
data in Table 1 suggest that higher temperatures induce a slight but
significant reduction in the total number of abdominal strokes pro-
duced during each volley, and that this phenomenon facilitates the
volley-shortening process.
Finally, the table and figures all support the view that deviation is
relatively slight between the temperature data for each of the var-
ious call parameters and the linear regressions calculated from those
data. The closest fit to a mathematical relationship is found in the
initial frequency of abdominal vibration during a volley: for data
350
Psyche
[Vol. 85
Figure 2. Regression lines, calculated from data in Table 1 . showing the effect of
temperature on the five principal characteristics A~ E of the calls of five males (dashed
lines) and five females (solid line) of Ch. plorahuncla. The left-hand “frequency” axis
applies to parameters A C (strokes/sec) and to D (volleys/ min), while the right-hand
axis for “duration” applies only to E.
A,
males:
y
= 4.I6x - 1 1.74,
r-’ = 0.96
females:
y
= 4.37x - 18.79,
r2 = 0.95
B,
males:
y
= 2.57x+ 7.89,
r2 = 0.88
females:
y
= 2.77x- 0.55,
r2 = 0.92
C,
males:
y
= 0.87x + 21.66,
r^= 0.69
females:
y
= l.OIx + 15.40,
r2 = 0.74
D,
males:
y
= 3.l2x - 27.85,
r-^ = 0.81
females:
y
= 3.44X - 32.32,
r^= 0.88
E,
males:
y
= -40x + 1650, 1
2 = 0.80
females:
y
= -37x + 1580, 1
= 0.89
1983]
Henry — Chrysoperla plorahunda
351
pooled from all ten individuals, over 95 percent of variance is
explained by regression line A shown in Figure 3. Even for the worst
case — terminal volley frequency — nearly two-thirds of raw data var-
iance is compatible with the calculated linear regression (r- = 0.65).
And data for individuals are neither more nor less variable, on
average, than pooled samples: for the five principal call features
graphed in Figure 3, individual r- values average 0.93, 0.84, 0.59,
0.79 and 0.76 and never fall below 0.40 for any single insect. Also
shown in that figure is the close congruence of all individual lines for
each of the same five call features but particularly for initial volley
frequency, suggesting that such temperature relationships are con-
sistent, repeatable, and predictable on an individual basis.
Results of experiments manipulating abdominal mass in individ-
ual lacewings are shown in Figure 4 and Table 2. Table 2 presents
the raw temperature-frequency data taken from four insects, while
Figure 4 shows how those data relate to the linear regressions gener-
ated from the terminal volley frequencies of the ten unmodified
individuals tested earlier. Converting the frequency measurements
to their equivalent values at 27°C (Table 2) dramatically reveals
how little those data are affected by mass-loading of the abdomen:
in none of the four experimental animals was the terminal volley
frequency altered significantly by the treatment. Weight increments
(from painting) amounted to 10-27% of total body mass; however,
abdominal weights were only 2.8 mg for the female and approxi-
mately 2 mg for the three males, so increments to the mass of the
vibrating structure itself in each insect actually ranged from 36 to
50%.
Discussion
The results described above amply document the striking similar-
ity of male and female calls of Chrysoperla plorahunda, consistent
with the proven importance of such signalling behavior to the
reproductive isolation of this species from several of its morphologi-
cally identical siblings in the genus (Henry 1980b, 1983b). Females
closely resemble males in every detail of volley structure and spac-
ing, and those few differences that do exist in pooled samples tend
to break down when the characteristics of individual insects are
compared (Figs. 3 and 4). Also as predicted, males and females
352
Psyche
[Vol. 85
Figure 3. Total range of individual variation of temperature regression lines for
the five major call characteristics A-E of Ch. plorahunJa. The area above and below
each straight line includes within it the linear regresssions for all 10 individuals tested,
while the line itself summarizes temperature data for those 10 insects in a single
regression equation. Axes and labels as in Fig. 2.
A: y = 4.27x- 15.37, r-’= 0.95
B:y = 2.69x+ 3.41, r^= 0.88
C: y = 0.96x + 18.33, r^= 0.65
D; y = 3.25x - 29.72, r^= 0.83
E:y = - 39x+ 1620, r^ = 0.83
1983]
Henry — Chrysoperla plorabunda
353
change their various call characteristics in precisely parallel ways as
temperature is altered, so that the sexes’ calls remain indistinguish-
able over the range of temperatures they would typically encounter.
The only apparent exception to this conclusion concerns the slightly
lower vibration frequencies characteristic of the terminal portions of
the volleys of females at all temperatures (Fig. 2C). Such a differ-
ence is expected if the terminal volley frequency of a lacewing’s
abdomen is determined by its inherent capacity to oscillate like a
weight on a spring, since the demonstrabty heavier abdomen of a
female would resonate at a lower frequency than that of a male.
However, artificially mass-loading the abdomens of several males
and females produced no obvious downward deflections of their
frequency characteristics, suggesting that neuro-muscular mecha-
nisms actively “drive” the vibrating system for the entire duration of
each volley (Table 2) and that resonance effects have relatively little
influence on resultant frequencies. Also, as mentioned earlier, the
male-female difference in terminal volley frequencies is considerably
less impressive when the responses of individual insects are dissected
from the pooled data (Fig. 4), thus raising the suspicion that it is an
artifact of some sort.
Studies of other acoustical insects overwhelmingly support the
existence of linear functions relating temperature to most song
parameters that repeat over time, as best exemplified by and
documented for rates of wing-stroking, chirping, and “rolling” in
crickets (Alexander 1956, Walker 1962, Dumortier 1963, Prestwich
and Walker 1981) and a few katydids and homopterans (Dumortier
1963, Shaw 1968, Whitesell and Walker 1978, Booij 1982). Similarly,
most of the temperature data reported here for Ch. plorabunda
conform well to linear statistical models, although they are insufficiently
detailed to discriminate linear from exponential interpretations. The
least individual or pooled variance from the calculated regression is
found for the abdominal vibration frequency of the first eight cycles
of a volley, suggesting that this feature of the call is particularly
crucial to unambiguous communication between the sexes; otherwise,
it seems there would exist no need for such precision.
In his 1962 paper on cricket song. Walker drew attention to the
apparent convergence of many of his linear regressions on 4°C;
that is, it seemed that for many different cricket species the chirp
354
Psyche
[Vol. 85
Table 2. Effect on terminal volley frequency (C), in abdominal strokes/sec, of
artificially mass-loading the abdomen of three males and one female of Ch. plora-
hunda. Weight gains are tabulated in column 3; the last column lists n. the number of
volleys measured. Equivalent frequencies (at 27°C) in column six were calculated
from the slope of regression line C in Fig. 3. Superscript “x” denotes lab-reared
insects.
Weight.
mg.
Temp..
°C.
Vibration
frequency. ±SD
Equivalent
at 27°C.
n
Male 1
control
7.5
26.8
39.78 ± 1.83
39.98
29
mass-loaded
9.0
30.0
47.08 ± 1.84
44.00
20
Male 1"
control
6.7
27.1
44.49 ± 2.44
44.51
20
mass-loaded
8.5
27.0
44.59 ± 3.52
44.59
25
Male 2^
control
7.2
27.3
45.98 ± 2.44
45.60
36
mass-loaded
8.5
26.8
44.92 ± 2.80
45.15
36
Female 1 '
control
9.6
27.3
43.47 ± 2.31
43.20
28
mass-loaded
10.6
27.0
43. 14 ± 2.50
43.14
19
and pulse (=wing stroke) rates went to zero at about 4°C when their
temperature regressions were extrapolated downward. The same
phenomenon can be seen in other insects, as well, including several
tettigoniids studied or reported by Dumortier (1963), Shaw (1968),
and Whitesell and Walker (1978) and two delphacid homopterans
studied by Booij (1982). Temperature data for initial volley frequency
and volley repetition rate of lacewing calls are also reasonably
consistent with the concept of regression convergence, since the
x-intercepts for relevant call parameters range from 2.82° C to
9.40° C in Ch. plorahunda and average 8.3 1°C in Chrysopa
oculata (Henry 1982b). However, the flatter slopes of the temperature-
frequency regressions for middle and terminal portions of plora-
bunda volleys cause those lines to intersect the x-axis at — 1.27°C
and -19.09°C, respectively, which does not fit well with Walker’s
generalization. Moreover, other examples from the literature fail to
confirm the phenomenon, even in certain crickets and katydids
recently studied by Walker himself (and collaborators): for instance,
Anurogryllus arhoreus and winter races of Neoconocephalus triops
both display rather flat regression lines of temperature versus wing
stroke rate that intersect the x-axis well below — 5°C (Prestwich and
Walker 1981, Whitesell and Walker 1978). Thus, Walker’s “four
1983]
Henry — Chrysoperla plorabunda
355
/
Figure 4. Effect on terminal volley frequency (C) of artificially mass-loading the
abdomen of four Ch. plorabunda individuals, superimposed on a graph showing the
calculated temperature regression lines of terminal volley frequency for 10 other
insects of the same species.
356
Psyche
[Vol. 85
degree rule” is intriguing but far from universal and applies only
partially to the lacewing calls described here; why the rule should
apply at all, to any insect call parameter, is still unknown.
It should not be assumed from the discussion above that linear
regressions characterize the temperature relationships of all insect
song parameters. In fact, it is likely that all temperature regressions
are ultimately exponential in form when based upon data taken over
a sufficiently wide temperature range, since the kinetics of the phys-
iochemical processes underlying song production are non-linear
with respect to temperature. One conspicuous example is the pulse
(or chirp) duration of the French tettigoniid Ephippiger provincia-
lis (Yers.), which varies inversely with temperature in a “hyperbolic”
manner (Dumortier 1963, fig. 229). The volley duration typical of
calling Ch. plorabunda also decreases with temperature (Fig. 3E),
but the function describing that decrease seems more linear than
hyperbolic or logarithmic over the chosen range of temperatures,
and the slope of the relationship is twice as steep as that shown for
the tettigoniid. Unless thermoregulation by the larger-bodied katy-
dids accounts for these differences, one can conclude only that dis-
tinct physiological mechanisms determine pulse, chirp, or volley
durations in different insect groups.
Recently, Michelson et al. (1982) published a comprehensive
theoretical and empirical study treating general aspects of the phys-
ics, transmissibility, and energetics of the vibrational songs of
insects. This study provides a rationale for the observed frequency
modulation of Ch. plorabunda volleys, and by implication helps to
explain why lacewing singers strictly control the frequencies of all
portions of their volleys, rather than simply allowing their abdo-
mens to oscillate at their frequencies of resonance. Viewing the plant
substrates of vibrating or tremulating insects as acoustical filters,
Michelson et al. concluded that signals consisting of multiple fre-
quencies should propagate more uniformly (and effectively) through
their substrates than narrow-bandwidth calls, since a signal of rela-
tively pure tone tends to excite a certain pattern of standing waves in
its substrate and will therefore vary tremendously in its intensity
from place to place on that substrate. Broad-bandwidth or frequency-
modulated signals, on the other hand, are ideally suited for penetrat-
ing such acoustical filters, so that at least a portion of the call’s
energy reaches the receptors of the recipient individual or partner.
Although the authors reported frequency changes on the order of
1983]
Henry — Chrysoperla plorabunda
357
30-40% in the calls of their bugs and small cicadas, lacewings of the
species Ch. plorabunda alter their rate of abdominal vibration by
50-60% during each volley, suggesting that their calls can propagate
efficiently and evenly through a variety of substrate types. However,
if all this is true, it remains to be explained why other lacewing
species like Ch. dow nesi and Ch. carnea (central Europe) produce
long calls of nearly constant frequency characteristics (Henry 1980b,
1983a). Perhaps specific properties of the typical substrates utilized
by those species have shaped and narrowed the frequency ranges of
their calls over evolutionary time: for example, Ch. downesi may be
responding to some inherent acoustical property of conifer needles,
since its ecological niche is restricted to the evergreen forests of
North America.
Summary
This study assesses the effects of temperature on the major
characteristics of the vibrational song of a North American green
lacewing, Chrysoperla plorabunda (formerly Ch. carnea). Those
parameters include the volley repetition rate, the average duration
of volleys, the total number of abdominal strokes per volley, and the
vibrational frequency structure of volleys. In general, temperature
affects the call in a direct linear manner, so that linear regression
equations (of differing slopes) can be used with confidence to des-
cribe the relationship between temperature and each measurable
trait of a lacewing’s song. Additionally, variation in these equations
among individuals or between the sexes is negligible, so all members
of the species produce calls of remarkably similar construction at all
reasonable ambient temperatures — a prerequisite to unambiguous
communication among conspecifics. That the weight of an in-
dividual’s abdomen has no effect on the frequency characteristics of
its call was demonstrated experimentally by mass-loading the
abdomens of several sexually receptive insects. The relevance of
these findings to the biological function of lacewing calling, as well as
to the physics of substrate-borne vibrations, is discussed briefly.
Acknowledgments
This study was supported in part by National Science Foundation
award #DEB-79-l 1537 to the author. I thank Dr. James Johnson,
University of Idaho (Moscow), for sending me living Chrysoperla
358
Psyche
[Vol. 85
plorabunda from his home state. Drs. Theodore Taigen and Stephen
Pacala (University of Connecticut) provided useful suggestions
throughout the study, while Ms. Burma Stelmak generously typed
and edited the manuscript. Ms. Jane O’Donnell (University of
Connecticut Museum of Natural History) read and constructively
criticized an early draft of the paper.
Lithrati ri: Citkd
Alexander, R. D.
1956. A comparative study of sound production in insects, with special
reference to the singing Orthoptera and Cicadidae of the eastern United
States. Ph.D dissertation, Ohio State University. 529 pp.
1975. Natural selection and specialized chorusing behavior in acoustical
insects, hi Pimental, D., ed.. Insects. Science and Society, pp. 35-77.
Academic Press, N.Y.
Boon, C. J. H.
1982. Biosystematics of the Muellerianella complex (Homoptera, Delphaci-
dae), interspecific and geographic variation in acoustic behavior. Z.
Tierpsychol. 58: 31 52.
Brooks, M. W.
1882. Influences of temperature on the chirp of the cricket. Popular Sci.
Month. 20: 268.
Busnei , R. G.
1963. Acoustic Behavior of Animals. Elsevier Publ. Co., N.Y., 933 pp.
Catc'HFole, C. K.
1982. The evolution of bird sounds in relation to mating and spacing behavior.
In Kroodsma, D. E., and E. H. Miller, eds.. Acoustic Communication in
Birds, Vol. 1, pp. 297-319. Academic Press, N.Y.
Di'mortier, B.
1963. The physical characteristics of sound emissions in Arthropoda. In Bus-
nei, R. G., ed.. Acoustic Behavior of Animals, pp. 346-373. Elsevier
Publ. Co., N.Y.
Frankenberg E.
1982. Vocal behavior of the Mediterranean house gecko Hemidactylus turei-
cus. Copeia 1982(4): 770-775.
Halliday, T. R.
1978. Sexual selection and mate choice. In Krebs, J. R., and N. B. Davies, eds..
Behavioural Ecolof'v, pp. 180-213. Blackwell Sci. Pubis., Oxford.
Hayward, R.
1901. The katydid’s call in relation to temperature. Psyche 9: 179.
Henry, C. S.
1979. Acoustical communcation during courtship and mating in the green
lacewing Chrysopa carnea (Neuroptera: Chrysopidae). Ann. Entomol.
Soc. Am. 72(1): 68 79.
1983]
Henry — Chrysoperla plorabunda
359
1980a. Acoustical communication in Chrysopa rufilahris (Neuroptera: Chryso-
pidae). a green lacewing with two distinct calls. Proc. Entomol. Soc.
Wash. 82(1): 1-8.
1980b. The courtship call of Chrysopa downesi Banks (Neuroptera: Chrysopi-
dae): its evolutionary significance. Psyche 86(2-3): 291-297.
1980c. The importance of low-frequency, substrate-borne sounds in lacewing
communication (Neuroptera: Chrysopidae). Ann. Entomol. Soc. Am.
73(6): 617-621.
1982a. Reply to Tauber and Tauber’s “Sympatric Speciation in Chrysopa:
further discussion.” Ann. Entomol. Soc. Am. 75: 3-4.
1982b. Reproductive and calling behavior in two closely related, sympatric
lacewing species, Chrysopa oculaia and C. chi (Neuroptera: Chrysopi-
dae). Proc. Entomol. Soc. Wash. 84: 191-203.
1983a. Acoustic recognition of sibling species within the holarctic lacewing
Chrysoperla corncfl (Neuroptera: Chrysopidae). Systematic Entomol. 8:
293-301.
1983b. (in press). Sexual behavior of green lacewings. In Semeria, Y., and T. R.
New, eds.. Biology of Chrysopidae. Junk Publishers, Netherlands.
ICKERT, G.
1968. Beitrage 7ur Biologie einheimischer Chrysopiden (Planipennia, Chry-
sopidae). Entomol. Abh. Dresden 36: 123-192.
Lanvon, W. E., and W. N. Tavoi.ga.
1960. Animal Sounds and Communication. Publ., #7, A IBS, Washington,
D.C. 443 pp.
Littlejohn, M.
1977. Long-range acoustic communication in anurans: an integrated and evo-
lutionary approach. In Taylor, D. H., and S. 1. Guttman, eds.. The
Reproductive Biology of Amphibians, pp. 263 294. Plenum Press, N.Y.
Miciiei.sen, a., F. Fink, M. Gogala, and D. Trade.
1982. Plants as transmission channels for insect vibrational songs. Behav.
Ecol. Sociohiol. 11: 269-281.
Morris, G. K.
1980. Calling display and mating behavior of Copiphora rhinoceros Pictet
(Orthoptera: Tettigoniidae). Anim. Behav. 28: 42-51.
OSSIAANILSSON, F.
1949. Insect drummers. Opusc. Entomol. Suppl. X: 1 145.
Otte, D.
1974. Effects and functions in the evolution of signaling systems. Ann. Rev.
Ecol. Syst. 5: 385-417.
Prestwicii, K. N. and T. J. Walker.
1981. Energetics of singing in crickets: effect of temperature in three trilling
species (Orthoptera: Gryllidae). J. Comp. Physiol. 143: 199-212.
Rdfpre( ht,R.
1975. Die Kommunikation von Sialis (Megaloptera) durch Vibrationssignale.
J. Ins. Physiol. 21: 305-320.
360
Psyche
[Vol. 85
Shaw, K. S.
1968. An analysis of the phonoresponse of males of the true katydid, Ptero-
phylla camellifolia (Fabr.) (Orthoptera: Tettigoniidae). Behaviour 31:
203-259.
Smith, R. C.
1922. The biology of the Chrysopidae. Cornell University Agric. Expt. Stn.
Mem. 58: 1291-1372.
Tjeder, B.
1960. Neuroptera from Newfoundland, Miquelon, and Labrador. Opusc.
Entomol. 25: 146-149.
Toschi, C. a.
1965. The taxonomy, life histories, and mating behavior of the green lacewings
of Strawberry Canyon. Hilgardia 36: 391-431.
Walker, T. J.
1962. Factors responsible for intraspecific variation in the calling songs of
crickets. Evolution 16: 407-428.
Whiteseli., J. J. and T. J. Walker.
1978. Photoperiodically determined dimorphic calling songs in a katydid.
Nature 274(5674): 887-888.
Young, D.
1980. The calling song of the bladder cicada, Cysiosoma saundersii: a compu-
ter analysis. J. Exp. Biol. 88: 407-41 1 .
SOCIAL ORGANIZATION IN LEPTOTHORAX ANTS:
WITHIN- AND BETWEEN-SPECIES PATTERNS*
By Joan M. Herbers
Department of Zoology, University of Vermont
Burlington, Vermont 05405
Recent application of quantitative techniques to behavior (cf.
Colgan 1978) has resulted in new approaches to undertanding social
interactions among animals. A technique particularly widely-used
for study of ant colonies is development of the colony ethogram, or
behavioral profile. We now have ethogram information for a wide
variety of species. Most reports in the literature focus on a single
colony (Table 1); variation within a species is rarely discussed. In
addition, the colony time budget, an important second class of
information, is generally not reported (Table 1). The appropriate-
ness of behavioral comparisons across species is thereby severely
limited by availability of only one type of behavior frequency
catalog, for only one colony per species.
Caste complexity and division of labor related to morphological
or age variation comprise another type of information contributing
to an understanding of social organization. As a rule, queens have
smaller repertoires than do workers; majors have different etho-
grams than minors; and older workers display different behavior
frequencies than do younger workers. Studies of morphology affect-
ing behavior have concentrated on polymorphic species for which
descrete worker castes can be distinquished; recent work has shown
that, even for monomorphic species, worker size can bias behavior
(Wilson 1978, Herbers and Cunningham 1983).
A reasonably complete description of social organization for an
ant species should treat ethograms, time budgets, and behavioral
caste specialization, both within and between different colonies.
Here I report such details for three colonies of Leptothorax atnbi-
guus. This information is then compared to data from the closely-
related L. longispinosus to arrive at an understanding of between-
and within- species variation in social behavior.
Manuscript received hy the editor July 22. 1983
361
Table I. Ethograms for many ant species have been published, but variation between colonies is rarely reported.
362
Psyche
[Vol. 85
o
oc
O' o
— oc
c 2:
c
■y. O
C C
> U U
c
Cl
oc atj
r- ss
O' U.
S E
os
n X:
sO oc oc vO
r- r- r-
_ O' O' O' O'
atj ^
S. c c. c. o c
c -E o o o o
j- •— ^
OJ <U 3 ’-C ‘-C
■o
o
u.
C
c.
Cl
a:
c c c ^ c c
r
^ i
C 2
X
^ >
c o
c O
— — — rsj
C/5
<3
-C :t
it
I
S 2
53 ^
II
2
s: V-
Ic-S
.sn
l>5 >5 O
Ci. ^
^ £3.
C C r.
^ a, s
^ O O 53
O O, N
Zacryptocerus varians 1 1 no Cole 1980
1983]
Herbers — Social Organization in Leptothorax
363
Methods
Colonies of L. amhiguus were collected in May 1982 from the E.
N. Huyck Preserve (Albany County, NY). These colonies were
settled in artificial nest boxes and maintained according to standard
methods (Herbers and Cunningham 1983); in addition, frozen fruit-
flies were provided as a food source.
For detailed observations, three colonies were chosen on the basis
of queen and worker number to match earlier studies of L. longispi-
nosus (Herbers 1982). All Leptothorax colonies studied were of
approximately equal worker number, all had eggs and larvae, and
all produced alates by summers’ end; only queen number varied
significantly (Table 2).
Behavioral observations were conducted June 9 — August 24,
1982 through a Wild M5-A stereomicroscope as follows: a worker
was chosen at random and all her actions were recorded over a
30-minute period. In addition, activities of individuals around her in
the field of view were recorded. Head widths of the randomly-chosen
ants were measured at a standard depth of field, by use of an ocular
micrometer.
Data analysis followed methods outlined by Fagen and Goldman
(1977) for behavior catalogs; Herbers and Cunningham (1983) for
statistical evidence of polyethism and morphological bias; and Cole
(1980) for producing dendrograms.
Table 2. Colony sizes of Leptothorax used in this comparative study. Data on L
tongispinosus were reported by Herbers (1982).
Original
# of
Queens
Leptothorax amhiguus
La-A 3
La-B 1
La-C 0
Leptothorax tong isp i n o sus
Ll-A 1
Ll-B 1
Ll-C 5
Ll-D 4
Original
n of
Eggs
Larvae
Alates
Workers
Laid?
Present?
Reared?
27
yes
yes
yes
28
yes
yes
yes
20
yes
yes
yes
30
yes
yes
yes
31
yes
yes
yes
28
yes
yes
yes
36
yes
yes
yes
364
Psvche
[Vol. 85
Results and Discussion
Social Organization ofL. ambiguus colonies
A total of 60 hours were recorded over the three colonies for a
grand total of 3145 observations. Ethograms for the three colonies
of L. ambiguus are reported in Table 3. A total of 46 behaviors were
recorded for workers and 13 for queens. Despite the large catalog
size, no behavior was unique to L. ambiguus; all in Table 3 are
relatively common to many species included in Table 1.
As expected, queens were much less active than workers (Table
3). Their behavior was almost exclusively directed towards the
brood; the exceptional occasion for colony La-B occurred when a
queen was observed walking outside the nest and taking a drink; she
later returned inside. Because of the paucity of data on queen behav-
ior, analyses below concern only worker behavior.
Frequencies of observations for behaviors over three colonies are
given in Figure 1. Sample coverages were uniformly greater than
99% (Figure 1). Consequently inferences about the true colony
repertoire can safely be made. There was considerable variation
among colonies (Figure 1), yet the distributions were not signifi-
cantly different from each other (x^ = 13,80, 14 df, P > .05). Thus
distributions of observations over all behavior categories were
roughly equivalent.
Comparisons of absolute frequencies among colonies showed that
many worker behaviors were observed in all colonies (Table 3). For
some, ethogram frequencies were nearly equal (IL, CP, ATW, RW)
whereas for others the correspondences were not good (RE, IE,
ALW). A third class of behaviors included those observed for only
one or two nests (CE, FED, AAE, CR). Thus considerable inter-
nest variation existed. For a given colony, some behaviors known to
occur in the species were missing from the ethogram, some were
common or rare relative to other colonies, and some were equally
frequent to others. Despite apparent discrepancies in absolute fre-
quencies, the rankings of behaviors by frequency were similar over
the three colonies (Kendall’s coefficient of concordance; W = .864,
27 df, P < .001). That is, behaviors commonly observed in one col-
ony were also common in others whereas those rare in one tended to
be rare in all. In particular, behaviors missing from one colony’s
ethogram were generally rare in others. Therefore, although abso-
lute frequency varied from nest to nest, relative frequencies were
similar.
1983]
Herbers — Social Organization in Leptothorax 365
o
La-A
e« .9942
Lq-B
e».9974
Lo-C
e«.9980
TOTAL
e>9986
NO. OF ACTS
(OCTAVE)
Figure 1 . Abundance histograms for three colonies of L anihiguus and for pooled
data. The abscissa gives the number of observations per behavior, to the base 2; thus
octave 0 indicates behaviors observed once, octave 1 refers to exactly 2 observations,
octave 2: 3 or 4 observations, octave 3: 5 through 8, and so on. The largest octave, 10
refers to behaviors observed 513-1024 times. The value of 6 given for each data set in-
dicates the sample coverage, as described by Fagen and Goldman (1977).
Table 3. Ethograms for three colonies of L. amhiguus. Since Colony La-C was queenless, no data on queen behavior
are reported.
366
Psyche
[Vol. 85
I r- Tf r-
I o — o
< 22
CQ II
^ I
CM vC — — : vC
(NO — — 1-0 I
I '' *
I O
OC OC Tj-
r- r- oc
O
<N — r*',
oc — —
00 vO vC
O O O
O O O
fN OC O — Tt
I C^<NTtvC —
(N — O O
o o o o o
(N —
I O 'C —
— o o
o o o
O'
flD S
i7
z
— 00
O — O'
— rsi (N
— fN r<-,
m oo (N
(N
— o o
o o o
sC 'C — <N
r*-, r*-,
r*~, O O
o o o o
sC <N O' sC O'
O r*'. O —
o o o o o
o o o o o
< 2
il
O' <N
— O'
m — oc
— rsi (N
§
— oc tT
— sC 'O oc
Tt Tt o o
o o o o
'O O' O'
(N — f*-, o I
o o o o
o o o o
CO 15
aij
CJO Wj
UJ iUj
> >
> CA
J .-5 -5
- O w
^ C/2 —
c. c.
3 3
ol. a.
u -
13
o
O UJ
wJ Jj j « 5
OU-i<£OU:0U,<£o
E
D. g
w O
UJ UJ UJ ,
O U -J < -
Q UJ
"J -J — J J "J A ^
o U CJi U. < ~ o
3 5
0, <
I— 'tA
U <
UJ
Ol <
U <
Social Interactions
BC Be Carried — — .0007
ATWAntennate Worker .0454 .0472 .0570
ATB Antennate Body .0058 .0084 .0061
1983]
Herbers — Social Organization in Leptothorax
367
I I I
I 1
I I
I i
I 1
O >c oc
OC TT
r-, — O
o o o
O' r-
(N (N I
— 5 I
§ §
^ r' ^
_ — o —
o o o o
o o o o
—
o o
o o
— O <N sC sO m
1^ — Tj- O —
(N r<-. — O O O
o o o o o o
iTi — rn iTi OC O
vO — Tt —
0<N 0 0 0 —
o o o o o o
OO O' vO 'C
m 1 r>', fN 1 sO O
o o o o o
o o o o o
— tn O' vO r*~,
Tj- O' — O , —
Tt — o o o
o o o o ' o
vO sO vO
O (N — O
o o o o
o o o o
O' — sO sO vO O' sO
f*-i O' o o o — o
o o o o o o o
o o o o o o o
OC CM
— (N O'
CnI O' 'O — r' — —
rs — OOOOOOfN
ooooooooo
I I i
aij 2 o
ot) o w
<U = (D
CC < Oi
> -1 o
cjs: < oQ
aO
a E
g §
S ^
c =
< <
O' O
H -J
< <
<u
3
0
V
s
2 ^
a o
l-i,
01 lI
oa
oc u.
JJ
«
E “
§ S
^ t
— w
< U
< u
o —
— <u
■i!
^ d>
_(U ^
>. Z
S £
E
f2 X
3 W
O n
3 -3
S > z 7:
w 3 a:
tjj
t ^ c t
oj cn 3
(J UJ I U
X X
uu uu
-J u
3 a. Si
, O ^ D- ^ c
■£ o ^ g 3 £ ^
o Ll S — U UJ U.
— o
ig I
5 Q P ^ ^
2 — U uu u.
£ Q
T3 ^
t Q
Colony l.a-C was queenless.
Code Behavior
Personal Behavior
RE Rest
SC Self-Groom
MO Move inside Nesi
{N=18) (N=t4)
1519 1104 2078
.2155 .2211 .1378
.2892 .3298 .3584
Brood Care
IE Inspeci Egg
CE Groom Egg
CE Carry Egg
LE Lay Egg
AL Assist Egg-Lay
ying
GL Groom Larva
CL Carry Larva
RL Regurgitate w Larva
FLD Feed Larva Solid
ALE Assist Larval Eedysis
IP Inspect Pupa
CP Groom Pupa
CP Carry Pupa
AAE Assist Adult Eclosion
.0058
.0032
.0088
.0061
.0061
.0415 .03.36 0292
,0461 03.36 0428
.0058 .0071 0149
.0084 0052 0061
.22
.06
.07
07
.07
0026 0006
.0019 .0052
0039 0039
0006 0006
0019
.05
ALWAIIogroom Worker
BC Be Groomed
ATQ Aniennaie Queen
ALQ Allogroom Queen
RQ Regurgitate w/Queen
FQ Fight Queen
0142
0026
0006
.0013
0048
ATM Aniennate male
ALMAIIogroom Male
CM Carry Male
RM Regurgitate w/ Male
ALF Allogroom Alale Female
Physical Nest Maintenance
LN Lick Nest Walt
LOT Look Outside Nest
LEx Inspect Exuvium
CEx Carry Exuvium
EEx Eat Exuvium
HM Handle Nest Material
CR Carry Refuse
FF Forage
MOTMovc Outside
IDr Inspect Prey
CDr Carry Prev
EDr Eat Prey
FdN Feed Inside Nest
FDO Feed Outsde Nest
IFd Inspect Food
DR Drink
.07
0045
Social Organization in Lepioi
368
Psyche
Table 4. Time budgets for L. amhiguus workers.
Behavior
RE
La-A
.6362
SG
.0565
MO
.1453
IE
.0016
GE
.0003
CE
IL
.0144
GL
.0282
CL
.0022
RL
.0077
ELD
—
ALE
-
IP
.0007
GP
.0017
CP
—
AAE
.0008
ATW
.0048
ATB
.0011
RW
.0235
ALW
.0039
BG
.0055
ATQ
-
ALQ
—
RO
—
ATM
—
ALM
—
CM
RM
ALE
—
BC
—
LN
.0028
LOT
.0013
lEx
.0001
CEx
—
EEx
—
HM
CR
FF
0010
MOT
.0293
IDr
.0001
Proportion of Time (p^)
La-B
La-C
.5985
.8157
.0922
.0153
.1384
.1281
.0056
—
.0018
.0013
.0117
,0051
.0115
.0034
.0018
—
.0012
.0018
.0024
.0030
.0012
.0009
.0070
.0066
.0013
.0015
.0167
.0103
.0153
.0005
.0449
.0058
.0001
-
.0024
.0002
.0100
.0001
.0008
.0024
.0012
.0020
.0014
.0020
.0007
.0120
—
.0030
.0001
[Vol. 85
Total
.68064
.05520
.13753
.00242
.00072
.00044
.01052
.01478
.00078
.00331
.00040
.00158
.00196
.00029
.00611
.00129
.01704
.00658
.01864
.00002
.00184
.00378
.00032
.00119
.0011!
.00089
.00036
.01424
.00107
1983]
Herbers — Social Organization in Leptothorax
369
Table 4. Time budgets for L. amhiguus workers. (Continued)
Beha\ ior
Proportion
of Time (pj)
l.a-A
l.a-B
1 a-C
Total
CDr
.0042
.00139
EDr
.0291
—
—
.01021
FdN
—
.0065
.0004
.00229
FdO
.0005
—
—
.00016
IFd
.0005
—
—
.00017
DR
.0008
—
—
.00027
Time budgets for workers of the three colonies are given in Table
4. The largest elements in Table 4 correspond to resting; overall,
workers spent 68% of the time motionless. It is interesting to note
that the most sedentary colony was La-C, which had no queen;
perhaps the high rate of inactivity was related to a lower rate of egg
production or overall lack of queen stimulation. Even so, resting
was predominant for all nests. The second dominant behavior was
moving inside the nest, on average accounting for 13.8% of the
worker time budget. In addition, self grooming was a large contribu-
tor in all colonies observed, consuming on average 5% of worker’s
time. Thus personal behavior accounted for the vast majority of the
time budget; activities which can be called “social” consumed less
than 15% of the workers’ time.
Among social behaviors, time budget variation among colonies
was minimal for some types (GP, ATW, ATB). Proportions of time
spent in other behaviors were quite different among colonies; the
most extreme case was LOT, which varied by two orders of magni-
tude (pj = .0100 for La-B and .0001 for La-C). Kendall’s test for
concordance showed that, despite differences in absolute proportion
of time, rankings of behaviors by relative proportions were similar
over all colonies (W = 0.763, 13 df, P < .005). That is, behaviors
consuming a large portion of the time budget in one colony tended
to be important for other colonies, and behaviors rare in one colony
were usually rare in all. Despite quantitative differences in specific
types of activity, overall qualitative agreement in time budgets was
strong.
370
Psyche
[Vol. 85
To sum, no significant differences in relative importance of
behaviors were observed over the three colonies. Histograms of
behavior frequency (Fig. 1 ), rankings of ethogram frequency (Table
3) and relative time budget frequencies (Table 4) were not signifi-
cantly different. Therefore, in subsequent analyses of within-species
social organization below, data were pooled over three colonies. 1
will return to consideration of between-colony variation below.
Division of labor among workers was investigated by considering
the matrix of transition frequencies among behaviors listed in the
ethogram. Within this 46 x 46 matrix, elements indicate how often
each behavior followed and preceded every other behavior. To sim-
plify presentation of the results, transitions among behaviors listed
in the ethogram are synopsized in Table 5. Diagonal elements
represent the frequencies by which behaviors in the same categories
followed each other whereas off-diagonal elements represent transi-
tion frequencies between behaviors in different groups (Herbers and
Cunningham 1983). Division of labor is implicated if nonzero tran-
sition frequencies are clustered in diagonal blocks of the matrix and
zeroes occur in off-diagonal blocks.
Examination of Table 5 shows that transitions from (column 1)
and into (row 1) Personal Behavior commonly occurred. This is no
surprise, since virtually all workers displayed a form of personal
behavior. However, among social behaviors, the overall distribution
of nonzero transitions deviated strongly from random expectation
(G = 55.24, 1 6 df, P<.00 1 ), in a pattern consistent with organization
of behaviors into roles: behaviors within the group Brood Care were
positively correlated in time, as were those within the groups Social
Interactions, Physical Maintenance, and Colony Provisioning.
Between these groups, there were significantly fewer transitions than
random expectation (Table 5). The pattern of overabundance of
nonzero transitions in diagonal blocks, and under-representation in
off-diagonal blocks was absolutely consistent with expectation.
Overall, nonzero transitions clustered in diagonal blocks, thereby
providing statistical evidence of polyethism.
Worker behavior can be provisionally categorized into four roles:
brood care, social interactions, physical nest maintenance, and pro-
visioning, since transitions among behaviors within a role occurred
more often than random expectation whereas links between roles
occurred less often than by chance. Information from the single-step
Behavior i
1983]
Herbers — Social Organization in Leptothorax
371
Table 5. Synopsis of one-step transition probabilities among worker behaviors.
Entries indicate the number of nonzero transitions from Behavior i to Behavior j
observed over three colonies.
Behavior j
Personal
Brood
Social
Physical
Provision-
Category
Behavior
Care
Inter-
Main-
ing
actions
tenance
Personal Behavior
9
20
15
8
9
Brood Care
16
33^
6“
2~
0“
Social Interactions
14
8
17+
0“
2'
Physical Maintenance
7
2“
r
7+
r
Provisioning
8
0"
2~
3~
21 +
^ more than expected by chance
“ fewer than expected by chance
G = 55.24, 16 df, P< .001
transition matrix therefore gave important insights to the nature of
polyethism in L. ambiguus. However, Table 5 must be interpreted
with caution, since it does not report linkage over several acts.
Analysis of per-second transition probabilities cannot detect transi-
tions between behaviors intervened by other acts. That is, over a
relatively long period, workers may switch roles, which would not
be disclosed by single-step transition analysis (Herbers and Cun-
ningham 1983). While single-step transitions suggest patterns of
polyethism, inferences must be corroborated by long-term observa-
tions of behavior.
Information on worker behavior over 30-minute periods is given
in Table 6. There is indicated the number of ants (out of a total of
57) that executed two behaviors within a 30-minute period. It is clear
from Table 6 that behaviors provisionally assigned to different roles
were often displayed by one worker over 30-minutes’ time. Results
of statistical testing of Table 6 are summarized by Venn diagrams in
Figure 2. Behaviors intersecting in this figure co-occurred more
often than expected by chance (x^ tests) whereas sets not intersect-
ing were observed for the same individuals at a rate no different
from chance expectation. Thus behaviors involved in egg care were
related, as were those concerning care of larvae and those related to
372
Psyche
[Vol. 85
care of pupae. However, no greater proportion of egg-workers also
tended for larvae and pupae than chance expectation. That is, indi-
viduals specializing on eggs were not necessarily those specializing
on pupae or larvae. The final set of behaviors is a large but
loosely-connected cluster. Regurgitation, grooming, and antenna-
tion were closely interconnected (Figure 2); these were also
peripherally connected with inspecting the nest exterior and moving
outside, since individuals returning from an outside foray elicited
interest and grooming from nestmates. Similarly, LN and HM do
not directly intersect, since they co-occurred no more often than
chance expectation; these two behaviors were indirectly linked
through ATW and RW (Figure 2). Behaviors in the roles of social
interactions, physical maintenance, and provisioning were therefore
not strongly separated.
Thus analysis of sets of behaviors displayed over 30-minute peri-
ods illustrated the expectation that individual workers tended to
specialize, particularly within the brood care role. An ant grooming
larvae was more likely to also carry or regurgitate to larvae than
chance expectation. By a similar token, a worker guarding the nest
Figure 2. Sets of behaviors that co-occurred within 30-minutes. Two intersecting
behaviors were observed more often than expected by chance (x“ tests; for all
intersections P <.05). Behaviors not intersecting occurred as often or less than
random expectation.
1983]
Herbers — Social Organization in Leptothorax
373
entrance (LOT) was likely to move outside the nest, then to be
groomed by nestmates upon reentry. Thus some components of
polyethism were statistically verified by examination of long-term
individual worker behavior. However, the division of labor inferred
from Table 6 was considerably weaker than the instantaneous transi-
tion matrix (Table 5) suggested. Individuals involved in brood care
acts also displayed behaviors in other roles over a 30-min. period.
Likewise, co-occurrences of behaviors in other roles were very
common. The pattern emerging from consideration of all data is
that workers strongly specialized in the short-term but over
30-minutes the specialization was weakened.
A final component of social organization is morphological bias in
polyethism. Worker size is known to be correlated with behavior in
many species, including the monomorphic L. longispinosus (Her-
bers and Cunningham 1983). The range of worker sizes in L. ambi-
guus is indicated in Figure 3. Pooled data are drawn there, since
Figure 3. Morphological variation in L. amhiguus workers. Head width distribu-
tion was normal, and the largest worker was less than 1.5 times that of the smallest.
374
Psyche
[Vol. 85
distributions were not significantly different among colonies
(ANOVA, F = .09, 2 df, P < .05). Head widths were relatively
invariant in this species (x = .552 mm, s = .043 mm) such that the
largest individual was less than 1.5X the smallest (Figure 3). Thus
the potential for size-biased polyethism was quite restricted.
Head widths of workers displaying different ethogram behaviors
were considered. Data were again pooled over three colonies since
mean head widths did not differ significantly for any behavior
(ANOVA with variable df; in all cases P < .05). Figure 4 illustrates
statistics of head width for workers displaying each behavior. No
obvious differentiation of head width according to roles can be
discerned. Analysis of data in Figure 4 is summarized in Table 7,
which is based on ANOVA’s for differences between mean head
widths (LSD tests). Most of the significant differences separate
workers regurgitating with larvae from other behaviors (row and
column headed RL in Table 7). That workers displaying RL were on
average larger than others is evident from Figure 4. Starred entries
of Table 7 are sporadic; certainly patterns of differences in mean
head width showed no clear segregation by roles. Workers exhibit-
ing a brood care behavior were not more similar in size to those
displaying other brood care behaviors than they were to workers
involved in social interactions or provisioning. Thus there was no
apparent morphological bias underlying polyethism for L. am-
biguus.
The overall picture that emerges of L. ambiguus social organ-
ization is short-term specialization of individual on task according
to four roles: brood care, social interactions, physical nest mainte-
nance, and provisioning (Table 3). However, the division of labor
was rather loose, since switching between roles was often observed
over 30-minute periods (Table 4). The nonrandom co-occurrence of
sets of behaviors (Figure 2) statistically reinforced inferences about
polyethism from the transition matrix. Finally, no strong morpho-
logical bias was demonstrated for ants specializing on specific tasks.
Comparison with L. longispinosus
Results of this study were compared with data from its closely-
related congener L. longispinosus; such comparisons were valid
since all observations were conducted in the same laboratory using
standard husbandry techniques. The major difference in culture
conditions between species was the addition of fruitflies to L. ambi-
guus diets. The earlier study had not incorporated feeding insect prey
1983]
Herbers — Social Organization in Leptothorax
375
Table 6. Number of workers observed to exhibit two behaviors over a
30-minute period. Numbers in the left column refer to total number of
ants observed to display each behavior.
RG
SG
MO
IE
GE
CE
IL
GL
CL
RL
ALE
56
RG
—
45
SG
44
—
51
MO
50
43
—
6
IE
6
5
6
-
4
GE
4
3
4
4
—
1
CE
1
1
1
I
I
—
21
IL
21
18
20
3
2
0
16
GL
5
13
16
3
2
0
14
2
CL
2
2
2
1
0
0
2
2
—
4
RL
4
4
4
0
0
0
4
4
1
—
1
ALE
1
1
1
0
0
0
0
1
0
0
—
10
IP
10
8
9
1
I
0
7
5
0
1
1
5
GP
5
4
4
0
0
0
3
3
0
0
1
1
AAE
1
1
1
1
I
0
1
0
0
0
0
36
ATW
36
30
36
3
2
0
14
9
2
3
1
14
ATB
14
13
14
1
0
0
7
5
2
2
1
25
RW
24
25
25
2
1
0
12
8
2
3
1
7
ALW
7
6
7
1
0
0
3
3
1
0
1
15
BG
15
14
15
1
1
0
9
6
0
1
1
1
ATQ
1
1
1
0
0
0
0
0
0
0
0
9
LN
9
9
9
I
1
I
5
4
1
2
1
8
LOT
8
8
8
1
1
0
0
1
0
0
0
2
lEx
2
2
2
1
1
0
1
0
0
0
0
3
CEx
3
3
3
1
1
0
2
1
0
0
0
4
EEx
4
4
4
1
1
0
2
1
0
0
0
6
HM
6
2
6
0
0
0
1
1
0
0
0
1
FF
1
1
1
0
0
0
0
0
0
0
0
6
MOT
6
6
6
0
0
0
1
1
0
0
1
3
IDr
3
3
3
0
0
0
1
0
0
0
0
1
CDr
1
1
1
0
0
0
0
0
0
0
0
1
EDr
1
1
1
0
0
0
0
0
0
0
0
2
FdN
2
2
2
0
0
0
1
0
0
0
0
1
F'dO
1
1
1
0
0
0
0
0
0
0
0
1
IFd
1
1
1
0
0
0
0
0
0
0
0
1
DR
1
1
1
0
0
0
0
0
0
0
0
376
56
45
51
6
4
1
21
16
2
4
1
10
5
1
36
14
25
7
15
9
8
2
3
4
6
1
6
3
1
1
2
1
1
1
Psyche
[Vol. 85
Table 6 (continued).
IP GP AAE ATW ATB RW ALW BG ATQ LN LOT
RG
SG
MO
IE
GE
CE
IL
GL
CL
RL
ALE
IP —
GP 5
AAE I
ATW 7
ATB 2
RW 4
ALW I
BG 2
ATQ 0
LN 0
LOT 0
lEx 1
CEx 2
EEx 2
HM 2
FF 0
MOT 0
IDr 0
CDr 0
EDr 0
FdN 0
FdO 0
IFd 0
DR 0
0 -
2 1 -
I 0 13
I 1 21
1 0 7
1 0 13
1 0 I
0 0 8
0 0 7
0 I 2
0 1 3
0 I 4
1 0 6
0 0 0
0 0 5
0 0 2
0 0 1
0 0 0
0 0 2
0 0 0
0 0 0
0 0 1
II -
5 6
5 10
0 0
3 7
2 5
0 2
1 2
1 2
1 3
0 0
3 4
I 0
1 0
0 0
1 0
0 0
0 0
0 0
5 —
0 0-
2 6 0 —
2 4 0 2 —
0 0 0 0 0
0 10 0 0
0 2 0 0 0
0 3 110
0 0 0 0 1
3 5 0 2 5
0 10 0 1
0 0 0 0 0
0 0 0 0 1
0 10 0 0
0 0 0 0 1
0 0 0 0 1
10 0 0 0
1983]
Herbers — Social Organization in Leptothorax 377
Table 6 (continued).
lEx CEx EEx HM EE MOT IDr CDr EDr FdN FdO IFd
56 RG
45 SG
51 MO
6 IE
4 GE
1 CE
21 IL
16 GL
2 CL
4 RL
I ALE
10 IP
5 GP
I AAE
36 ATW
14 ATB
25 RW
7 ALW
15 BG
1 ATQ
9 LN
8 LOT
2 lEx
3 CEx
4 EEx
6 HM
1 FF
6 MOT
3 IDr
I CDr
1 EDr
2 FdN
I FdO
I IFd
I DR
1 —
1 2
0 I
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
1 0
2 -
0 0
0 0
0 I
0 I
0 0
0 I
0 0
0 0
0 0
I -
I I
0 0
I I
0 0
I I
I I
0 0
1 —
0 0
2 I
I 0
I 0
0 0
I —
I I
I I
0 0
I —
0 0
378
Psvche
[Vol. 85
HEAD WIDTH (mm.)
.46 48 50 .52 54 56 .58 60 ,62 .64 .66 .68 .70
! I 1 I 1 1 1 1 1 1 1 I I
REST
SELF-GROOM
MOVE INSIDE NEST
INSPECT EGG
GROOM EGG
CARRY EGG
INSPECT LARVA
GROOM LARVA
CARRY LARVA
REGURGITATE W/LARVA
ASSIST LARVAL ECDYSIS
INSPECT PUPA
GROOM PUPA
ASSIST ADULT ECLOSION
ANTENNATE W/WORKER
ANTENNATE BODY
REGURGITATE W/WORKER
ALLOGROOM WORKER
BE GROOMED
ANTENNATE QUEEN
LICK NEST WALL
GUARD NEST ENTRANCE
INSPECT EXUVIUM
CARRY EXUVIUM
EAT EXUVIUM
HANDLE NEST MATERIAL
FORAGE
MOVE OUTSIDE NEST
CARRY PREY
EAT PREY
INSPECT PREY
FEED IN NEST
FEED OUTSIDE
INSPECT FOOD
DRINK
•
♦
Figure 4. Head width distributions for workers exhibiting behaviors in the etho-
gram. Means and standard deviations are plotted.
1983]
Herbers — Social Organization in Leptothorax
379
Table 7.
RE
SG
MO
IE
GE
CE
IE
Gl,
CL
RE
ALE
IP
GP
AAE
ATW
ATB
RW
ALW
BG
ATQ
LN
lEx
LOT
CEx
EEx
HM
FF
MOL
Results of ANOVA tests of mean head widths for workers exhibiting
different behaviors. Starred entries indicate average head widths were signifi-
cantly different (LSD tests; *P < .05).
RE
SG
MO
IE
GE
CE
IL
GL
CL
IDr
CDr
EDr
FdN
FdO
IFd
DR
RL
K<
*
♦
*
*
*
*
*
*
*
*
*
to L. longispinosus (Herbers and Cunningham 1983); to correct for
this different rearing condition comparisons below deleted the prey-
handling behaviors reported for L. amhiguus (Table 3: IDR, EDR,
CDR). In addition, colonies had been matched with respect to
worker number in order to eliminate variation correlated with col-
ony size (Table 2). Therefore, comparisons between these studies
380
Psyche
[Vol. 85
were not confounded by variation between observers, husbandry
techniques, or colony size.
The two species are ecologically similar. Both inhabit temperate
forests throughout eastern North America, nest in small plant cavi-
ties such as acorns, twigs, and hollow roots, and scavenge for
arthropod parts. On the Huyck Preserve, L. longispinosus is more
common, preferring deep woods, while L. amhiguus appears re-
stricted to relatively open habitats; despite some microhabitat segre-
gation, the two do co-occur in many places.
A subjective analysis of their overall demeanor suggests that L.
amhiguus is the higher-tempo species (sensu Oster and Wilson
1978). That is, they are more excitable and appear to move faster
than L. longispinosus. While this study was not designed to detect
tempo differences, one set of behaviors clearly illustrated it: for the
L. amhiguus colonies, certain workers were often stationed at the
nest entrance. While there, they periodically roused to inspect the
entrance, moving a few cm outside the opening before returning to
their position. This combination (LOT, MOT) was observed for all
3 colonies (Table 3). By constrast, L. longispinosus workers only
occasionally positioned themselves near the nest entrance, and the
apparent guarding behavior was observed for only one of four colo-
nies watched (Herbers 1982). Thus the more excitable nature of the
presumed higher tempo species was evident in the colony etho-
grams.
Both species displayed a division of labor, with similar patterns of
polyethism. Roles of brood care, social organization, and nest main-
tenance were identified in each. For L. longispinosus, though,
foragers comprised a unique caste whereas in L. amhiguus foragers
displayed other provisioning behaviors as well. Moreover, in L.
longispinosus the division of labor was much tighter: very few
instantaneous transitions between roles were observed, most of
them between brood care and nest maintenance (Herbers and Cun-
ningham 1983). For L. amhiguus, transitions among roles were
more frequent (Table 5), although less common than chance expec-
tation. Over 30-min. periods, workers of both species switched roles,
but again, role-switching was far more common for L. amhiguus
than for L. longispinosus. Therefore, although specialization
occurred in both species, division of labor was considerably tighter
for one.
1983]
Herbers — Social Organization in Leptothorax
381
A startling difference between species was the strong morphologi-
cal bias underlying polyethism in L. longispinosus but lacking in L.
anihiguus. Size differentiation according to task was clear in L. lon-
gispinosus-, the pattern strongly corroborated influences of role and
caste delineation from the transition matrix (Herbers and Cun-
ningham 1983). For L. anihiguus, however, there were relatively few
differences in average worker size among behaviors, and those few
significant differences were not correlated with roles inferred from
behavior transitions. Perhaps the absence of morphological correla-
tion was due to the fact that the range of worker size was narrower
for L. anihiguus (Figure 3) than L. longispinosus (Herbers and
Cunningham 1983); a small size range of workers may have pre-
cluded task specialization by size for L. anihiguus.
Both species displayed considerable among-colony variation with
respect to behavior frequency and time budgets. To ascertain the
relative importance of within- and between-species variation, cluster
analyses were performed. These techniques involve calculating sim-
ilarity indices for all possible pairwise comparisons. Then each unit
(i.e. colony) is placed in a dendrogram based on its similarity to
every other unit. If behavior data reflect phylogeny, then the three
L. anihiguus colonies should form one cluster while the four L.
longispinosus form a second. Moreover, one might expect colonies
with similar numbers of queens to cluster more closely to each other
than colonies with different queen numbers.
The simplest comparisons used the matching coefficient, or
number of behaviors shared by two colonies relative to the total
number observed over all (Cole 1980). This similarity index utilizes
information only on presence or absence of behavior types in the
ethogram, thereby ignoring relative frequency. Analysis of matching
coefficients yielded the dendrogram of Figure 5. This simplest clus-
tering technique produced the satisfying results that L. anihiguus
colonies were more closely related to each other than to L. longispi-
nosus nests: the three formed a distinct cluster. Moveover, queen-
right L. anihiguus colonies were more similar to each other than to
the queenless nest; this result, however, was simply an artifact of the
presence of behaviors directed towards queens ( ATQ, RQ, ALQ) in
queenright but not queenless ethograms. Even so, L. anihiguus col-
onies did cluster as expected. However, the L. longispinosus nests
did not. Two (LI -A and Ll-B) clustered closer to L. anihiguus nests
382
Psyche
[Vol. 85
lOOn
80-
0-J
Ll-B La-C La-B La-A Ll-A Ll-D Ll-C
Figure 5. Dendrogram of similarity among Lepiothorax colonies based on the
simple matching coefficient.
than to conspecific nests (Figure 5). The simple matching coefficient
which weighs all behaviors equally, therefore produced a dendro-
gram that gave satisfactory results for one species but far from
pleasing results for the second. That is, differences among L. longi-
spinosus nests were stronger than differences between species, based
on simple matching coefficients.
A second type of cluster analysis used geometric distance between
ethogram frequencies of the colonies. This technique incorporated
information on frequencies of different behavior types, yielding
results more biologically meaningful than the matching coefficient
(Cole 1980). For this analysis, rest was excluded. A dendrogram of
the seven colonies produced from ethogram frequencies is shown in
Figure 6. All colonies were quite similar to each other (minimum
similarity was 97.72 on a scale of 100) because proportion data were
used. Use of frequencies changes the scale but not relative positions
of colonies within the dendrogram. Just as with the simple matching
1983]
Herhers — Social Organization in Leptothorax
383
coefficient, geometric distance between colonies failed to produce
separate species clusters (Figure 6). Interestingly, queenright colo-
nies of L. amhiguus clustered, as did monogynous colonies of L.
longispinosus. However, an anomalous cluster was comprised of
La-C and Ll-D, thereby reducing the significance of other clusters
in the dendrogram. In sum, consideration of ethogram frequencies
produced a dendrogram for which within-species variation swamped
between-species variation.
A final dendrogram was produced from comparison of time
budgets over all colonies (Figure 7), based again on geometric dis-
tance. This analysis was more discriminating than that based on
ethograms (minimum similarity = 83.17). The dendrogram shows
three longispinosus colonies clustered closely and three amhiguus
clustered closely. The single aberrant entry was Ll-D, a polygynous
lOOn
q:
<
if)
Ld
>
!<
cd
99-
98-
97
La-A La-B La-C Ll-D Ll-C Ll-A Ll-B
Figure 6. Dendrogram of similarity derived from geometric distance based on
ethogram frequencies.
384
[Vol. 85
Psyche
lOOn
96-
t:
oc:
<
in
Ld
>
92-
88-
84-
80-*
La-A La-B La-C Ll-D Ll-A Ll-C Ll-B
Figure 7. Dendrogram of similarity derived from geometric distance based on time
budget frequencies.
longispinosus clustering with the three L. amhiguus colonies. Con-
sideration of time budget data gave a reasonable but not perfect fit
to expectation.
Of the three dendrograms produced, the best fit to expectation
derived from time budget data. Even this best-fit tree, though, con-
tained an anomaly. By no statistical means could I produce a cluster
diagram that accorded perfectly to species identity. In no case did
the two species separate into discrete clusters. Variation among col-
onies within a species therefore makes separation between species
tenuous. Because most studies report data from a single colony, the
utility of cross-species comparisons of behavior is severely limited.
Moreover, ethograms themselves appear less discriminating than
time budgets for separating out variation between species. It
appears, then, that standard methods of reporting social organiza-
tion (i.e. ethogram frequencies from a single colony) neglect critical
information on between-colony variability and on time budgets.
1983]
Herbers — Social Organization in Leptothorax
385
Only with more extensive studies of within-species variability with
respect to ethogram and time budget frequencies can valid between-
species comparisons be drawn.
Acknowledgments
This work was supported by NSF grant DEB82-2361 and a grant
from the National Academy of Sciences.
Rm Ri N( I s
Brandao. C. R. F.
1978. Division of labor within the worker caste of Formica perpHosa Wheeler
( Hymenoptera:Formicidae). Psyche, 84; 229-237.
Cari IN. N. F.
1981. Pol\ morphism and division of labor in the dacetine ant Orectognathus
\r/-.v/ro/o/- (Hymenoptera:Formicidae). Psyche, 88: 231-244.
Coi r. B.
1980. Repertoire convergence in two mangrove ants, Zacryptocerus various
and Camponoius (Colohopsis) sp. Insectes Sociaus, 27: 265-275.
Coi GAN, P., ed.
1978. Quantiiaiive Ethology. Wiley-lnterscience. New York.
Corn. M. L.
1980. Polymorphism and polyethism in the neotropical ant Cephalotes airatus
(1.). Insectes Sociaux. 27: 29-33.
FaGHN, R. and R. N. GoI OMAN.
1977. Behavioral catalogue analysis methods. Anim. Behav., 25: 261-274.
Hi RHI-RS, J. M.
1982. Queen number and colony ergonomics in Leptothorax longispinosus.
pp. 238-242 in; M.D. Breed, C.D. Michener, and H.E. Evans, eds. The
Biology of Soda! Insects. Westview Press, Boulder, Colorado.
HrRBFRS, J. M. AND M. CUNNINGIIAM.
1983. Social organization in Leptothorax longispinosus Mayr. Anim. Behav.,
31: 775-791.
OsTi R, G. AND E. O. Wilson.
1978. Caste and Ecology in the Social Insects. Princeton Univ. Press, Prince-
ton, New Jersey.
Tranifi.i.o, j. F. a.
1972. Population structure and social organization in the primitive ant Am-
hlvopone pallipes (Hymenoptera:Formicidae). Psyche, 89: 65-80.
Wii SON, E. O.
1976a. Behavioral discretization and the number of castes in an ant species.
Behav. Ecol. Sociobiol., 1: 141-154.
1976b. A social ethogram of the neotropical arboreal ant Zacryptocerus varians
(Fr. Smith) Anim. Behav., 24: 354-363.
386
Psyche
[Vol. 85
1978. Division of labor in lire ants based on physical castes ( Hymcnoptera:
Formicidae^So/c/to/^.v/.s). .1. Kansas Entomol. Soc.. 51; 615-636.
1980. Caste and division of labor in leaf-cutter ants ( Hymenoptera:Formici-
ddc:Aiia). 1. The o\erall pattern in A. Behav. Ecol. Sociobiol..
7: 143-156.
Wll SON. E. O. AND R. Facu n.
1974. On the estimation of total behavioral repertoires in ants. J. N.Y.
Entomol. Soc.. 82: 106-112.
‘PROTEST’ SOUNDS OF A GRASSHOPPER;
PREDATOR-DETERRENT SIGNAL?*
S\R\L A. Blondheim and Eeiezer Frankenherg
Zoology Dept., Hebrew University of Jerusalem
Jerusalem, Israel
Introduction
Some animals emit sounds when grasped or handled. Referred to
as alarm, protest, distress or disturbance signals — the sounds
themselves, the behavior accompanying their emission and the
mechanisms responsible for their production have been described,
analyzed and discussed (Haskell 1974). But only recently have
experimental data become available in support of the oft-stated
hypothesis that these sounds may startle a predator into releasing a
noisy morsel (Bauer 1976; Smith and Langley 1978; Masters 1979;
Buckler et al 1981 ).
The grasshopper Pareuprepocneniis syriaca Giglio Tos (Acridi-
dae) when grasped, immediately begins to chirp (the biology and
acoustic behavior of this grasshopper will be described separately).
Though there are individual differences in intensity and quality of
the sounds, males have a greater tendency to squeak while females
tend to click. Emission of the sounds is easily observed to
correspond to movements of the mouthparts; if the labrum is lifted,
the mandibles can be seen rubbing against one another to the
rhythm of the chirps. Immobilization of the mouthparts prevents
sound emission.
It had been observed in our laboratory that on casual feeding of
this grasshopper to representatives of several lizard families (La-
certidae, Scincidae, Gekkonidae) the grasshopper was captured,
then promptly released. A male grasshopper introduced into the
cage of the lizard Lacerta cianfordi was caught head-first and held in
the mouth of the lizard for several seconds, after which the lizard
slowly opened its mouth and the grasshopper fell free. Several addi-
tional grasshoppers of this species were offered to two Euhlepharis
macularius, a gecko from Pakistan present in the vivarium at the
* Manuscript received by the editor July 15. 19H3.
387
388
Psyche
[Vol. 85
time. They grabbed, then released the insects. Additional observa-
tions were then made with a microphone in the cage transmitting
sounds to earphones worn by the investigator. A skink, Mahuya
vittata caught and then released a sound-producing P. syriaca male;
three geckos, P. h. guttatus each caught and promptly released
sound-producing males of P. syriaca. These rejected grasshoppers
were removed and immediately replaced by mute grasshopper
nymphs of L. m. niigratorioides, equivalent in size to the rejected P.
syriaca males; the nymphs were caught and immediately consumed,
one by each gecko. These preliminary observations raised the ques-
tion: were the sounds emitted by the grasshoppers a factor in their
release? The following experiments were designed to answer this
question.
Materials and Methods
Grasshoppers: Adult males of P. syriaca were field caught in the
hills of Jerusalem a few days prior to experiments and were main-
tained in 60 1 laboratory cages providing fresh plants, light and
heat. As males were lifted from the cage for assignment to an
experiment, the thorax was squeezed gently. Of 48 males squeezed,
only three failed to produce sound. Half the sound-producing males
were then silenced by releasing a drop of melted paraffin onto the
closed mandibles; when it hardened, these males could no longer emit
sound, though they hardly differed in appearance from untreated
males.
Fourth instar nymphs of Locusta migratoria migratorioides R &
F maintained in the gregarious state in stock cages in the laboratory,
served as additional controls. Their size, dark color and small wing
buds provided a phenotypically reasonable facsimile of the brachyp-
terous adult male of P. syriaca. These nymphs did not produce
sound when handled.
Predators: Ptyodactylus hasselquistii guttatus von Heyden was
selected as the predator for the series of experiments. The candidacy
of this gecko was supported by the following credentials: P. h. gutta-
tus, a poikilotherm like P. syriaca, is at least partially sympatric
with it and shares its biotope; it is an opportunistic insectivore; like
P. syriaca, it emerges from its retreat in rock ledges and crevices in
warm weather and has been known to feed during daylight (Werner
1965; Perry & Werner 198 1 ); juveniles could handle a grasshopper
1983] Blondehim & Frankenberg — Sounds of Grasshopper 3S9
the size of the P. syriaca male, while adults could handle even the
large female; the frequency spectrum of the sounds of P. syriaca falls
within the hearing range of P. h. guttatus (Werner 1976); and
finally, a laboratory stock of this gecko was available. Though wild-
caught, the geckos had been kept in captivity in the vivarium for
months to years. Though the memory span of this gecko species is
not known, it may well be that the long laboratory incarceration
had dimmed recollections of possible previous encounters with this
grasshopper and its ruse.
Experimental procedure: A series of three grasshoppers was
introduced simultaneously into the cage of a gecko whose habitual
diet of fly maggots had been removed at least a day previous: an
untreated P. syriaca male, a silenced P. syriaca male and a fourth
instar nymph of L. m. migratorioides. The insects were introduced
at noon, prior to the peak activity hours of the gecko (Frankenberg
1979), and observations were made every half hour from noon to
5:00 PM and from 8:00 AM to noon. The first item eaten was
assigned the number 1; the second item, 2; and the third, 3. If two
grasshoppers disappeared between any two readings, both were
assigned the same number. In the few instances in which all three
grasshoppers were alive and apparently unharmed at the end of 24
hours, it was assumed that the gecko was not hungry; the experi-
ment was not included in tallying the results. After an interval of
several days, the gecko was used again. Silenced P. syriaca were
checked at the end of the experiment to ascertain that they were
indeed still unable to produce sound.
Results and Conclusions
In the cages of the 26 geckos tested, no untreated P. syriaca was
ever the first to disappear and 69% were never eaten at all. The
silenced P. syriaca was eaten first in 46% and the L. m. migrato-
rioides nymphs in 78% of the experiments (Table I and Fig. 1). A
G-test (Sokal and Rholf 1969) was carried out to test for indepen-
dence between the three choices of prey and the order of predation.
It was found significant (G = 49.9; df = 6, p< 0.001). A sign test
(Siegel 1956) between each of the three combinations of paired
insects for all the 26 instances in which a grasshopper was eaten
showed that silenced P. syriaca were eaten before untreated ones in
18 experiments (p^O.002), L. m. migratorioides were eaten before
390
Psyche
[Vol. 85
Table I; Order of predation* by the gecko P. h. f^uiiulaius. on a choice of
grasshoppers.
serial
number
gecko
normal
P. syriaca
male
silenced
P. syriaca
male
4th instar
nymph,
L.m.ni.
1
0
1
1
2
0
1
1
3
0
1
1
4
3
1
2
5
3
2
1
6
2
0
1
7
0
1
0
8
0
1
1
9
2
3
I
10
3
1
2
11
0
2
1
12
2
3
1
13
0
0
1
14
0
1
0
15
3
1
I
16
0
2
1
17
0
0
1
18
0
1
0
19
3
1
2
20
0
0
1
21
0
2
1
22
0
1
1
23
0
2
1
24
0
0
1
25
0
2
1
26
0
0
1
*The numbers 1, 2, and 3 represent order of predation; 0 indicates that the
grasshopper was alive at the end of the 24 hr test period. The same number for more
than 1 grasshopper indicates that they were consumed between the same two
observation periods.
the untreated P. syriaca in 23 experiments (p^ 0.001) and before
silenced ones in 14 experiments (p = n.s.). It is therefore concluded
that the protest sounds produced by P. syriaca apparently reduce
predation on it by P. h. guttatus.
Discussion
To a hungry caged gecko offered a choice between fly maggots
and grasshoppers, the latter are invariably preferred. However, it is
1983] Blondehim & Frankenberg — Sounds of Grasshopper 391
apparent from the present results that the appetite for grasshoppers
may be tempered by their behavior. In the present case, mandibu-
lar sounds emitted by P. syriaca appeared to interfere with preda-
tion by this gecko.
Because of its confinement in the cage of the gecko during exper-
iments, a grasshopper which had chirped its way to freedom was
prevented from escaping its predator as it might in the wild. P.
syriaca, though it cannot fly, is an excellent jumper and under nat-
ural field conditions would probably have jumped far and hidden
itself well before the predator had recovered from its encounter.
The sound itself has a wide frequency spectrum such as that
characterizing alarm calls of birds (Marler 1957; Morton 1977). The
utility of sounds such as these might include conspecific warning,
since these grasshoppers occur in loose aggregates. However, hold-
ing a chirping male in close proximity to conspecifics, or playing the
recorded sound back into a cage of P. syriaca failed to produce any
discernible reaction.
It was observed that these grasshoppers are often seized headfirst.
It is suggested that the hollow bones of birds, or the large buccal
cavity of lizards may act as a resonating chamber, enhancing the
intensity of the insect’s sounds or vibrations.
For the few grasshopper species known to produce^mandibular
sounds spontaneously or in encounters with conspecifics, an intra-
specific communicative function has been suggested: {Paratylotropi-
dia hrunneri, Alexander 1960; Oedaleonotus fuscipes, Varley 1939;
Calliptanius italicus, Faber 1949) but no experiments have been
reported in support of this hypothesis. Henry (1942) reports that
Mesamhria duhia emits a shrill creak when seized and investigation
may reveal that this sound, like the protest sound of P. syriaca
studied here, may play a predator-deterrent role.
Whether a remnant of an intraspecific communicative cue or a
language of predator deterrence, a signal such as that presented here
has quite probably been playing a part in the evolutionary history of
the struggle for survival in this species.
Acknowledgment:
This paper is intended to answer the question of Dr. David
Blondheim, who at age 10 asked his mother (SAB) why P. syriaca
made those strange noises with its mouth when you caught it.
Thanks are extended to Dr. N. Ben-Eliahu and to Profs. R. Galun,
392
Psyche
[Vol. 85
L.m.migratorioides
P. sy r iaca silenced
P.syriaca normal
order of predation
Fig. 1 . Order ol predation on sound produeing and sileneed P. syriaca males and
mute nymphs of /.. ni. miyraiorioidc.s.
l egend: The numbers I, 2, and 3 in the abseissa represent order of predation; 0
indieates that the grasshoppers were alive at the end of the 24-hr. test period.
1983] Blondehim & Frankenherg — Sounds of Grasshopper 393
Y.L. Werner, J. Camhi, S. Friedman, E. Nevo and S. H. Blondheim
for helpful comments on the manuscript. We also thank Prof. Y.L.
Werner for making the inmates of the vivarium available for these
experiments; and P. Amitai who drew the figure. Partial support to
EF by the Center of Absorption in Science of the Ministry of
Absorption is acknowledged with thanks.
Summary
Mandibular sounds produced by the grasshopper Pareuprepoc-
neniis syriaca Giglio Tos, when seized, appear to reduce predation
on it by a probable natural predator, Ptyodactylus hasselquistii
guttatus von Heyden, an insectivorous gecko. Sound-producing
grasshoppers which had been silenced by treatment in the labora-
tory, untreated sound-producing grasshoppers, and silent Locusta
migratoria migratorioides nymphs were introduced simultaneously
to the geckos. Survival of normal, sound-producing P. syriaca far
surpassed that of both controls.
BiHI lOGRAFIlV
Ai.kxandfr. R. D.
I960. Communicative mandible-snapping in Acrididae. Science 132: 152-3.
Bauhr, I.
1976. Experimente zur Frage der biologischen Bedeutung des Stridulations-
verhaltens von Kafern. Z. Tierpsychol. 42: 57 65.
Biirni i r. E. R., T. B. Wright and E. D. Brown
1981. On the functions of stridulation by the Passalid beetle Odontotaenius
(iisjuncms (Coleoptera:Passalidae). Animal Behavior 29: 483-486.
Fabhr, a.
1949. Fine Bischer unbekannte Art der l.auterzeugung europaischer Orthop-
teren: Mandibellaut von Callipianws iialicus. Z. Naurforsch 46: 367-9.
Frankknberg, E.
1979. Influence of light and temperature on daily activity patterns of 3 Israeli
forms of Ptyodactylus (Reptilia: Gekkoninae). .1. Zool. Lond. 185:
21-30.
EIaskfi I , P. T.
1974. Sound Production in The Physiology of Insecta, Second ed. 354-405.
M. Rockstein, ed.. Academic Press.
Henry, G. M.
1942. Three remarkable stridulatory mechanisms in Acrididae (Orthop). Proc.
Roy. Ent. Soc. Series A 17: 59-62.
Marler, P.
1957. Specific distinctiveness in the communication signals of birds. Behaviour
11: 13-39.
394
Psyche
[Vol. 85
Mastfrs. W. M.
1979. Insect disturbance stridulation: its defensive role. Behav. Ecol. Sociobiol.
5; 187 200.
Morton, E. S.
1977. On the occurrence and significance of motivation- structural rules in
some bird and mammal sounds. Am. Nat. Ill: 855 869.
Pkrrv. G. and Y. L. Wfrm r
1981. Food of Ptvodaciylus hasselqiiistii f^utiatus (Reptilia: Gekkonidae) at
two locations in Israel in summer. Proc. Zool. Soc. Israel in Isr. J. Zool.
30; 98-9.
Sifc.fi,, S.
1956. Nonparametric Statistics. McGraw Hill, N.Y.
Smith, R. E. and W. M. I.angi i v
1978. Cicada stress sounds: an assay of its effectiveness as a predator defense
mechanism. Southwest Nt. 23; 187 96.
SOKAF, R. R. AND F. J. Rnoi I
1969. Biometry. Freeman, San Francisco.
Wfrnfr, Y.L.
1965. Uber die Israelischen Geckos der Gattung Piyodaci ylu.s und ihre
Biologic. Salamandra 1; 15 25.
1976. Optimal temperatures for inner-ear performance in gekkonid lizards. J.
Exp. Zool. 195: 319 352.
Vari.fv, G. C.
1939. Unusual methods of stridulation in a cicada (Clidophleps distanti
(Van D.)) and a grasshopper (Ocr/tf/co«o/M.s /w.vc//?c.v Scud.) in California.
Proc. R. ent. Soc. Fond. A 14; 97-100.
AGE POLYETHISM: ITS OCCURRENCE IN THE ANT
PH El DOLE HORTEN SIS.
AND SOME GENERAL CONSIDERATIONS.
Bv Prassede Calabi', James F. A. Traniello' % and Michael H.
Werner' -
InTRODUC TION
A main theme of eusociality is division of labor (Wilson 1971,
1975), which can be based on physiological differences (as in the
case of the reproductive queen and sterile workers), morphological
(size) differences among workers, or age differences within a physi-
cal class. In social insects both age and physical classes can comprise
castes, that is, groups of individuals which perform specialized labor
for sustained periods of time (physical castes: Oster and Wilson,
1978; Wilson, 1980a, b; Herbers, 1980; age castes: Oster and Wilson,
1978; Porter and Jorgenson, 1981; Mirenda and Vinson, 1981; See-
ley, 1982). We constructed an ethogram for the Indo-Australian ant
Pheiclole hortensis. and tested the general hypothesis of division of
labor in the worker caste by seeking to answer these questions:
1. Is there division of labor between physical castes?
2. Is there division of labor among age classes within a physical
caste?
3. And if there is age polyethism, is it continuous or discrete? (See
Wilson 1976a.)
We will consider and discuss each question separately, and then
compare our results with those from other studies on social insects.
In particular we will contrast age polyethism in Pheiclole horteusis
with that of a New World Pheiclole species, P. clentata.
Materials and Methods
Data Collec tion
Three colonies of Pheiclole hortensis were collected in July 1979
from virgin rainforest at Gilmale, Sri Lanka by Anula Jayasuriya.
'Department of Biology, Boston University. Boston. MA 02215
-Museum of Comparative Zoology Laboratories, Harvard University, Cambridge.
MA 02138
Manuscript received hy the editor August H. I9S3
395
396
Psyche
[Vol. 85
The ants were identified by E. O. Wilson, and voucher specimens
deposited in the collection at the Harvard Museum of Comparative
Zoology. Colonies thrived and produced brood in artificial nests
made of glass tubing of 5 mm diameter (approximately that of twigs
in which wild colonies have been found (Jayasuriya 1979), and fitted
with moist cotton plugs. Colonies were maintained at 26°C while
observations and experiments were carried out.
As is typical of this genus, P. hortensis has a completely dimor-
phic worker caste. And, as is true for many ant species in general,
newly eclosed P. hortensis are quite light in color, and darken as
they age. Using the method first described by Wilson (1976a), we
found that based on these color differences and the degree of pig-
mentation of body parts each physical caste could reliably be sepa-
rated into five color or age classes (see Appendix I). Using the obvious
size and color differences, ethogram data on workers of different
ages was compiled from 24 hours of observation on one colony over
a ten week period.
The nest tube and surrounding area were watched, and every
observed act was noted along with the age class and physical caste of
the ant performing it. During the 24 hours of observation 3,689 acts
of 25 behaviors were recorded for minor workers, and 256 acts of six
behaviors for majors. At the end of the study the colony consisted of
192 minors, 32 majors, brood, and the queen.
Data Analysis
Completeness of the behavioral repertory was assessed by statisti-
cal comparison with a lognormal Poisson distribution (Bulmer
1974, Fagen and Goldman 1977).
The hypothesis of age-based division of labor was tested with a
standard comparison between observed performance frequencies
by each age class for behaviors, and expected frequencies generated
with the following formula (Altmann and Altmann 1977):
(Bi)(nj)
Ejj = Expected frequency of Behaviorj by age classj
Bj = Observed frequency of Behaviorj by all age classes
n; = Number of ants in age classj
N = Total number of ants in all age classes combined
1983]
Calahi, TranieUo, and Werner — Pheidole
397
We excluded from analysis behaviors with frequencies < 1% of all
behaviors performed by that physical caste; for P. hortensis that
gives a possible frequency per behavior of about 25, or five occur-
rences per age class, under the null hypothesis.
Associations between age class and behavior were assessed by a
relative performance measure (RPM). We calculated the probability
that ants of a particular age class will perform a given behavior, and
divided those ratios by the highest such probability for that behav-
ior. Thus
X = frequency of behaviorj performance by ants of age classj,
y = frequency of all behaviors by age classj,
z = highest such frequency for that behaviorj, and RPM = (x ^ y)
7 .
Finally, any attempt at an ergonomic assessment requires that
one distinguish between task and non-task behaviors. We use the
terms as follows. “Behavior” means a logical unit like grooming,
made up of one or more physical acts, such as drawing the tibial
comb over the antennae. “Task” is used in the sense of Oster and
Wilson (1978) to denote a set of acts which achieve some purpose of
the colony. Thus there are task and non-task behaviors, and though
all tasks are behaviors, not all behaviors are tasks.
Results
1. Completeness of Repertory.
The repertory of each physical caste separately and of the species as
a whole was judged complete, based on statistical comparison with a
lognormal Poisson distribution (Bulmer 1974, Fagen and Goldman
1977). For minors, the observed repertory size is 25 behaviors, and
the estimated size is 26, with a 95% confidence interval of [23,
29]. For majors, six behaviors were observed, and six estimated,
with a 95% confidence interval of [5, 7]. For P. hortensis the
observed repertory includes 31 behaviors, with 33 estimated, and a
95% confidence interval of [30, 36].
2. Division of labor by physical castes.
Comparison of the behavioral repertories of the two physical
castes shows that there is, with the exception of trophallaxis, no
overlap in task performance (Table 1). Of the tasks carried out.
398
Psyche
[Vol. 85
brood care, food acquisition, and allogrooming are performed by
minors and defensive tasks by majors. Defense by minors was seen
only when the colony was experimentally submitted to attack by
other ant species, and even then the two physical castes performed
different tasks: minors pinioned foreign ants, making it easier for
majors to snip them up.
Minor
Major
Selfgroom
789 (.21)
1 10 (.43)
Allogroom Minor
136 (.04)
0
Allogroom Major
70 (.02)
0
Allogroom Queen
9 (.002)
0
Carry Egg
307 (.08)
0
Carry Larva
1235 (.33)
0
Carry Pupa
149 (.04)
0
Groom Egg
24 (.01)
0
Groom Larva
223 (.06)
0
Groom Pupa
225 (.06)
0
Assist Larval Eclosion
1 14 (.03)
0
Assist Pupal Eclosion
6 (.001)
0
Trophallaxis w/ Larva
25 (.01)
0
Trophallaxis w/Minor
48 (.01)
8 (.03)
Trophallaxis w/ Major
5 (.001)
0
Trophallaxis w/Queen
2 (.001)
0
Retrieve Food
186 (.05)
0
Forage
22 (.01)
0
Eat Brood/ Exuvia
37 (.01)
0
Eat Dead Adult
26 (.01)
0
Carry Brood/ Exuvia
46(.01)
0
Carry Dead Adult
7 (.001)
0
Carry Meconium
1
0
Carry Nest Material
1 -
0
Eat Solid Food in Nest
4(.001)
0
Patrol at Food
0
52 (.20)
Patrol Arena
0
22 (.08)
Guard
0
63 (.25)
Totals
3685 (1.00)
255 (1.00)
Table I. Ethogram of PheiJole horlensis. Observed frequencies are followed by
values in parentheses indicating the frequency of each act relative to the total number
of behaviors performed by a physical caste.
1983]
Calabi, Traniello, and Werner — Pheido/e
399
Both castes exchange food with minors, but trophallaxis is rela-
tively more important in the majors’ repertory. Though its actual
frequency is low, it constitutes their only non-defense task, and it
comprises 5% of the tasks they perform versus only 1% for minors.
In fact, when all categories of trophallaxis by minors are combined
(with majors, larvae, and the queen as well as with minors), trophal-
laxis still comprises only 2% of the minors’ task repertory. This
relative frequency of the behavior by majors led us to ask whether
majors serve as a replete or “cask” caste, as in Caniponotus fraxi-
nicola (Wilson 1974). Pheidole hortensis majors with full gasters
show a three-fold weight increase, but we were unable to perform
the critical experiments and test for proportionate weight gain.
However, in random surveys of the colony, replete majors (those
with distended gasters) performed virtually none of the behaviors
typcial of majors (Table 2). During experimentally induced attack
(assault with sympatric Tetramoriuni spp.), replete majors engaged
in defense only if the attack was severe (many ants) or extended in
time. How much of this inactivity by “storage” majors is due to
protecting the food supply and how much to relative inability to
move is not clear.
Another potential set of caste differences relates to the granivo-
rous habits of many Pheidole species in which minors harvest and
majors mill seeds. In an attempt to observe such caste differences in
P. hortensis, we offered grass, vegetable, and bird seeds of various
sizes and fat contents. All were ignored by both physical castes.
3. Division of labor by age classes within a physical caste.
To answer this question, we tested the null hypothesis that each
age class should perform a given behavior in proportion to the
number of ants in that age class. Thus, in a colony with three age
Patrol at Food
Patrol Arena
Attack Intruder
Guard Nest Entrance
“Replete” Majors
12
0
I
0
Non-“Repletes”
66
8
18
20
Table 2. Behavioral differences within the major worker caste of Pheidole
hortensis. The numbers of individuals observed performing various acts during
random surveys of the colony arc presented.
400
Psyche
[Vol. 85
classes comprising 30, 20, and 50% of the total colony, there is no
age polyethism if those age classes perform 30, 20, and 50%
respectively of any task. The data show that the age classes of both
physical castes do not perform tasks in proportion to their numbers
(Table 3). On the basis of X" comparisons, for most tasks with
adequate sample sizes the null hypothesis can be rejected because
there are significant differences between the observed and expected
frequencies. This indicates that there is age-based division of labor
in both physical castes (Table 4). Four tasks by minors (assist ecol-
sion, allogroom majors, trophallaxis with minors, and carry exu-
viae) are performed without apparent age bias, and eight behaviors
by minors and one by majors were observed too rarely to permit
assessment.
Thus of behaviors with an adequate sample size, for P. hortensis
minors 13 of 17, and for majors four or five, behaviors show age-
based division of labor.
4. Continuous versus discrete age castes.
Wilson (1976a) states that division of labor is discretized if there is
an exclusive association between (sets oO tasks and age class(es) and
that it is continuous under all other conditions of age class/task
association. The general question of age polyethism has two parts.
Given that some tasks are performed more or less often by certain
age classes, can adjacent age classes be combined because they show
similar performance patterns? And second, are such associations
between age class(es) and tasks exclusive? To test for associations,
we calculated relative performance measures (RPM) for each age
class by behavior. This descriptive way of treating the data controls
for variation in age class size and in number of performances
observed per age class, and it permits comparison between fre-
quently and rarely occurring behaviors, as well as comparisons of
age class performances within and between behaviors.
Figure 1 shows that there are no consistent similarities between
the relative performance probabilities for any pairs of adjacent age
classes. This implies that no pair of age classes can be combined,
and that these age classes do differ behaviorally, representing real
castes. It is also clear that the associations between age castes and
tasks or groups of tasks are not exclusive: the age-based division of
labor is continuous rather than discretized in both the minor and
1983]
Calabi, Traniello, and Werner — Pheidole
401
MINOR WORKERS
AGE CLASS
1(15)
11(27)
111(45)
1V(24)
V(81)
ROW
TOTAL
192
Selfgroom
78
159
221
95
228
781
Allogroom Minor
10
5
38
22
61
136
Allogroom Major
5
7
25
1 1
22
70
Carry Egg
1 1
121
77
30
68
307
Carry Larva
20
93
470
159
493
1235
Carry Pupa
1
18
42
24
64
149
Groom Egg
5
7
6
3
3
24
Groom Larva
8
28
94
38
55
223
Groom Pupa
8
57
65
23
72
225
Assist Eclosion
10
26
23
14
41
1 14
Trophallaxis w/ Larva
1
1
15
2
6
25
Trophallaxis w/ Minor
1
7
19
6
15
48
Forage
0
0
0
2
20
22
Retrieve Food
0
2
24
39
121
186
Eat Brood /Exuvia
0
2
22
3
10
37
Eat Dead Adult
0
1
14
2
9
26
Carry Exuvia
0
6
13
4
23
46
COLUMN TOTAL
158
540
1 168
472
1311
3654
MAJOR WORKERS
AGE Class
1(5)
11(15)
111(5)
1V(3)
V(4)
32
Selfgroom
13
58
30
8
1
1 10
Guard
3
4
14
20
22
63
Patrol at Food
0
0
2
3
47
52
Patrol Arena
0
0
0
1
21
22
COLUMN TOTAL
16
62
46
32
91
247
Table 3. Observed frequencies with which each age class (I through V) performed
various acts. Values in parentheses indicate the number of individuals in each age
class.
402
Psyche
[Vol. 85
MINORS
N
X^
P
Selfgroom
781
65.9
**
Allogroom Minor
136
13.4
.01
Allogroom Major
70
7.8
NS
Carry Egg
307
178.7
**
Carry l arva
1235
212.3
**
Carry Pupa
149
13.1
.02
Groom Egg
24
14.1
.01
Groom l.arva
223
58.6
**
Groom Pupa
225
34.8
**
Assist Eclosion
1 14
7.9
NS
Trophallaxis w Larva
25
18.8
.001
Trophallaxis w Minor
48
8.7
NS
Forage
186
79.3
**
Retrieve Food
22
22.5
**
Eat Brood Exuvia
37
27.9
**
Eat Dead Adult
26
15.0
.01
Carry Exuvia
47
5.2
NS
MAJORS
Selfgroom
1 10
23.7
**
Guard
63
13.3
.01
Patrol at Food
52
290.1
**
Patrol Arena
22
138.8
**
Table 4. X- values and significance levels for differences between observed
and expected behavior frequencies for the five age classes within each physical caste.
** indicates that the P values were so small they do not appear in the X- table and are
highly significant. NS, no significant difference. Only behaviors with frequencies ^
I9f are included. N = total number of acts observed.
1983]
Calahi, Traniello, and Werner — Pheidole
403
A E
B
F
C
G
D
H
N > > ^ A
Figure I. Relative performance measures (RPM) of various beha\iors in the
r.epertories of major (MJI-MJ4) and minor (A-Q) workers of Pheidole horiensis.
Roman numerals I-V correspond to w'orker age classes.
Additional details in Materials and Methods.
RPM RPM RPM RPM
404
Psyche
[Vol. 85
J N
K
0
L
P
Figure I . (Continued)
1983]
Calabi, Traniello, and Werner — Pheidole
405
Q
r I f f I
N > A ^
MJ-1
MINORS
A Groom Egg
B Selfgroom
C Allogr. Minor
D Allogr. Major
E Assist Eclosion
F Carry Egg
G Groom Pupa
H Eat Brood/Skin
I Troph w/ Larva
J Troph w/ Minor
K Groom Larva
L Carry Larva
M Carry Brd/Skin
N Eat Dead Adult
0 Carry Pupa
P Retrieve Food
Q Forage
MAJORS
MJ-l
Selfgroom
MJ-2
Guard
MJ-3
Patrol 0 Food
MJ-4
Patrol Arena
MJ-2
Pu
A A
V N
MJ-3
N 4s A A
V N
MJ-4
O-* 1 1 r
N > A A
Figure I. (Continued)
406
Psyche
[Vol. 85
major physical castes. For instance, in Figure 1 examine the minors’
behaviors F, G, and E, and the series H through N. The relative
performance measures for tasks F and G are highest for age class II,
yet age classes II and I have similarly high RPM for task E. Or, in
the second case, for all the tasks H through N, age class III shows
very high RPM, but the task/age class association is not exclusive.
For K, M, and N at least, other, but not necessarily adjacent age
classes, show similarly high RPM.
Discussion
P. hortensis exhibits both physical and age castes, and the latter
show continuous rather than discretized polyethism. We will com-
pare these results with results from other species, and consider some
of their general implications for the study of age polyethism.
1. Repertory size and numerical considerations.
Both repertory size and the proportion of rarely occurring behav-
iors in P. hortensis are in the same range as those of other species.
Numbers of behaviors in repertories judged complete by Fagen and
Goldman analysis are: 27 for workers of monomorphic Leptothorax
species (Wilson and Fagen 1974) and for minor workers of
Orectognathus versicolor (Carlin 1982), and 28 each for minor
Pheidole dentaia (Wilson 1976a), Formica perpilosa (Brandao
1979), and Camponotus sericeiventris (Busher 1982). Extremes may
be represented by minor repertories of Solenopsis geminata, S.
invicta, and Zacryptocerus varians: 17, 20, and 38, respectively
(Wilson, 1976b, 1978). Repertories for majors range from two
{Solenopsis geminata, Wilson 1978) to 24 {Orectognathus versi-
color, Carlin 1982); for the dimorphic Z. varians and P. dentata, major
repertories are 1 1 and nine (Wilson 1976a and b). Total repertories
for all these species range from 20 to 40 behaviors.
Given the similarity in repertory size for minor workers of both P.
hortensis and P. dentata, it may seem odd that in P. hortensis
considerably fewer behaviors ( 1 3 vs 23) are performed with age bias.
The difference results from the respective criteria used to reject
rarely occurring behaviors from analysis. Because he does not make
statistical comparisons, Wilson rejects only 2 of 28 behaviors on
grounds of insufficient data. However, when the cut-off criterion
1983]
Calabi, Traniello, and Werner — Pheidole
407
used for P. hortensis is applied (rejected behaviors with frequencies
< 1% of those performed by that physical caste) the number of
behaviors rejected for P. dentata increases from two to seven of 28
(N=l,222; Wilson 1976a). Thus the appropriate comparison of age-
biased behaviors among behaviors with frequencies ^ 1% shows a
similar situation for the two; 1 3 of 1 7 behaviors for P. hortensis. and
15 of 19 for P. dentata. Of behaviors performed without apparent
age bias, allogroom majors and trophallaxis with majors, are in
common to the two species. (In general ant repertories include high
proportions of infrequent behaviors: for minors 10 of 28 {P. dentata
— Wilson 1976a), 27 of 38 {Zacryptocerus varians — Wilson 1976b) 7
of 28 [Formica perpilosa — Brandao 1979), 14 of 27 [Orectognathus
versicolor — Carlin 1982) and 9 of 28 [Camponotus sericeiventris —
Busher 1982). Although some castes have small repertories (majors
of P. dentata and So/enopsis geminata, — Wilson 1976a, 1978),
many other castes show similar proportions of infrequent to fre-
quent behaviors (Wilson and Fagen 1974, Traniello 1978, Brandao
1979, Carlin 1982).
2. Age-based division of labor.
It is a virtual truism that among social insects older workers
forage and have little or nothing to do with brood care. Yet in P.
hortensis older workers, in addition to performing all foraging and
food retrieval (P and Q, Figure 1), show high RPM for several
brood-care tasks (K, L, M, and O). We suggest that these represent
labor cohorts, based on ant experience or colony need. They could
arise via the mechanism of task fixation, a feedback-based task
stabilizing mechanism documented in wasps (Forsyth 1978) and
suggested for the ant Amhiyopone pallipes (Traniello 1978). An
individual performs some task (e.g., trophallaxis with larvae),
receives positive feedback (continually finds hungry larvae), and
over time does not switch to other tasks because the positive
feedback does not cease. The susceptibility of individuals to task
fixation could vary so that even in a system with age-based task-
switching, task fixation might override age-based behavioral
change.
Although it remains to be demonstrated whether such fixation
occurs in P. hortensis. we wish to point out one possible con-
sequence of task fixation and the resultant caste “atypical” be-
408
Psyche
[Vol. 85
havior. A colony labor profile by age class could be an artifact of
previous colony needs and activities, specifically for the period
previous to the study by just less than an average worker life span.
For instance, suppose that a colony engaged in high brood
production several weeks previous to observations. There would
have been both need and opportunity for much brood care, and also
opportunity for fixation on brood tasks. Further imagine that the
food supply then dwindled and brood production decreased. Some
older workers are fixated on and keep performing brood care tasks,
leaving little need for younger workers to perform this task, and
therefore little opportunity for task fixation. At the time of
observation, RPM for brood care would show older castes
performing proportionately more brood care. Yet it may be
misleading and actually incorrect to draw the conclusion that older
castes “typically” perform brood care. RPM are epiphenomena of
past (and current) colony labor needs, and may say less about age
castes as such than about behavioral flexibility and colony require-
ments. Therefore, appropriate conclusions must consider this, and
include at least a time frame, plus consideration of colony age, size,
and circumstance.
3. Continuous versus discrete castes; roles
By definition, discretization of age castes is a direct consequence
of roles (a group of tasks) linked by high transition probabilities,
and exclusively or principally performed by a particular age caste
(Oster and Wilson 1978, Wilson 1976a). Our results for P. hortensis
show a continuous mode and, therefore, no roles. This differs from
results of other age polyethism studies. Wilson (1976a) and Seeley
(1982) find behavioral discretization by age, and roles. Both also
argue that spatial efficiency is its basis, with each role (suite of tasks)
involving a set of physically proximate contingencies. If that is the
case, differences among the species and especially between P.
hortensis and P. dentata, could account for the results. Colony size
in P. hortensis is a few hundred, in comparison with up to a
thousand in P. dentata. The former nests in twigs or small nuts, the
latter in logs often with underground galleries. Thus for P.
hortensis, spatial efficiency may be an irrelevant consideration.
However, other more basic considerations may also be involved.
Mirenda and Vinson (1981) elaborate on Wilson’s (1976a) use of
1983]
Calahi, Traniello, and Werner — Pheidole
409
“caste” and “subcaste.” Because the two treat their data differently,
comparisons are difficult, but we suggest that the castes and
subcastes of Mirenda and Vinson correspond to the discrete and
continuous modes of Wilson. Mirenda and Vinson consider as
subcastes “(a) groups of individuals within each caste whose
behaviour is statistically but not completely differentiated from
other such groups and (b) groups intermediate in behaviour between
two or more castes, but not completely distinct from any caste”
(1981, p. 417). Both descriptions, and especially the latter, seem to
fit the criteria for a continuous caste system — overlap in frequency
distribution of age classes performing various tasks — rather than
the discrete system, characterized by an exclusive association
between an age class and a group of tasks. If this correspondence is
indeed correct, we may be that much closer to a functional under-
standing of labor roles, spatial efficiency, caste, and how task per-
formance of individual ants sum to performance of whole castes. It
is also noteworthy that although Mirenda and Vinson do not
address the question of spatial efficiency as such, their results show
a strong correspondence between ant age, location, and “career,”
their “role” analogue.
Clearly there is age-based division of labor in Pheidole hortensis.
It does not seem to follow traditional role patterns, nor is it obvious
which pattern it does follow. Therefore, we suggest two factors
which must be considered for P. hortensis in particular, and in
studies of age polyethism in general: “atypical” behavior due to
labor cohorts, and role performance. Both have been documented
for physical castes (Oster and Wilson 1978); one for age castes
(Oster and Wilson 1978, Mirenda and Vinson 1981, herein). Because
of these specializations, we suggest that mean behavioral perfor-
mances by age classes may not be sufficiently fine-grained for
detailed ergonomic analysis, and that the study of behavioral spe-
cialization and its ergonomic consequences requires bouts of con-
tinuous observation of individually marked animals throughout
their lives.
Summary
We present evidence for and describe age-based division of labor
in the Indo-Australian ant Pheidole hortensis. Both the minor and
major physical castes exhibit age polyethism, and in both castes age
410
Psyche
[Vol. 85
polyethism is continuous rather than discretized. There is virtually
no overlap between the sets of tasks performed by the two physical
castes. These findings differ in several respects from those reported
in two other studies of age polyethism (in the New World P. dentata
and in Apis mellifera), and raise some interesting questions about
labor roles in social insects.
Acknowledgements
Thanks to L. Calabi for invaluable discussion of, and help with,
data analyses; to T. D. Seeley for a preprint of his paper and
enlightening discussion of his methods; to S. D. Porter for a critical
reading; to D. S. Gladstein for technical assistance with estimating
repertory completeness; and to W. R. Tschinkel for pointing out an
important reference.
Refhrhnchs
Al TMANN, S. A. AND J. AI.TMANN
1977. On the analysis of rates of behavior. Anim. Behav. 25: 364-372.
BrandAo, C. D.
1979. Division of labor within the worker caste of Formica perpilosa Wheeler.
(Hymenoptera:Formicidae). Psyche 85: 229-237.
Bt I.MFR. M. G.
1974. On fitting the Poisson lognormal distribution to species-abundance data.
Biometrics 30: 101-1 10.
Bt SUFR, C.
1982. Polymorphism and division of labor in Camponotus sericeiventris.
Honors Thesis, B.A., Boston University.
Cari.in, N. F.
1982. Polymorphism and division of labor in the Dacetine ant Orectognathus
v’en/Vo/or (Hymenoptera:Formicidae). Psyche 88: 231-244.
Fagfn, R. M. and R. N. Goldman
1977. Behavioural catalogue analysis methods. Anim. Behav. 25: 261-274.
Forsyth, A. B.
1978. Behavioral Ecology of Polygynous Social Wasps. Ph.D. Thesis, Harvard
University.
Hfrhfrs, J. M.
1980. On caste ratios in ant colonies: population responses to changing envi-
ronments. Evol. 34(3): 575-583.
Javaslria, a. K.
1980. The behavior and ecology of Aneureius simoni Emery. Honors Thesis,
B.A. Harvard University
1983]
Calahi, Traniello, and Werner — Pheidole
411
Mirknda, J. T. and S. B. Vinson.
1981. Division of labor and specification of castes in the red imported fire ant
So/cnop.sis invicia Buren. Anim. Behav. 29: 410-420.
OSTl R. G. F. AND E. O. Wll SON.
1978. CaMc and Ecology in ihe Social Insects. Princeton University Press,
Princeton, N.J.
PoRTI R, S. D. AND C. D. JoRGHNSf N.
1981. Foragers of the harvester ant, Pogonomynne.x owylieei: a disposable
caste? Behav. Ecol. Sociobiol. 9; 247-256.
Sin I V, T. D.
1982. Adaptive significance of the age polyethism schedule in honeybee colo-
nies. Behav. Ecol. Sociobiol. II: 287-293.
Tranifi I (), J. F. A.
1978. Caste in a primitive ant: absence of age polyethism in Anihlyopone.
Science 202: 770-772.
Wll SON, E. O.
1971. The Insect Societies. Belknap Press of Harvard University Cambridge,
MA.
1974. The soldier ant Camponotns (Colohopsis) fra.xinicola as a trophic caste.
Psyche M 182-188.
1975. Sociohiology: The New Synthesis. Belknap Press of Harvard University.
1976a. Behavioral discretization and the number of castes in an ant species.
Behav. Ecol. Sociobiol. 1: 141-154
1976b. A social ethogram of the neotropical arboreal ant Zacryptocerus varians
(Fr. Smith). Anim. Behav. 24: 354-363
1978. Division of labor in fire ants based on physical castes ( Hymenoptera:
Formicidae:5'o/c/?D/M/.v). ,1. Kansas Entom. Soc. 51: 615-636
1980a. Caste and division of labor in leaf-cutter ants ( Hymenoptera:Formici-
d'de.Atta). 1. The overall pattern in A. se.xdens. Behav. Ecol. Sociobiol.
7: 143-156.
1980b. Caste and division of labor in leaf-cutter ants ( Hymenoptera:Formici-
dae:/l//r7). 11. The ergonomic optimization of leaf cutting. Behav. Ecol.
Sociobiol. 7: 157-165.
Wll . SON, E. O. AND R. M. Faghn.
1974. On the estimation of total behavioral repertories in ants. N. Y. Entomol.
Soc. 82: 106-112.
Appendix I
Hue and degree of pigmentation for the color (age) classes of the two physical
castes of P. hortensis.
Minors.
Class /-White-yellow: head and thorax, pale white/ light grey; gaster, light
grey; petiole and femur, pale white/ light yellow; tibia, grey.
412
Psyche
[Vol. 85
Class //—light yellow: head and thorax, light yellow with amber outlines, espe-
cially dark amber edging the mandibles; gaster, medium grey; petiole, dark yel-
low with amber outlines; femur, light yellow/ brown; tibia, grey.
Class III — yellow-grey: head and thorax, dark yellow, head with grey in occipi-
tal region; gaster, dark grey; petiole, yellow; femur, dark yellow/ brown; tibia,
grey.
Class IV — amber: head and even mandibles, solid amber; thorax, amber with
some brown; gaster, dark grey/ black; petiole, amber with brown outlines;
femur, brown; tibia, grey.
Class V — amber-grey: head, dark amber with brown streaks through occipital
region; thorax, solid amber; gaster, black/dark grey; petiole, amber/brown;
femur, brown; tibia, grey.
Majors
Class / — white-yellow: head, pale white; thorax and petiole, pale white/ light
yellow; gaster, light grey.
Class // — yellow: head and thorax, yellow; petiole, light brown/ amber.
Class III — amber: head, dark amber; thorax, dark yellow; gaster, dark grey;
petiole, light brown/amber.
Class IV — medium brown: head, dark brown with lighter tinges; thorax, amber;
gaster, dark grey; petiole, dark brown.
Class V — dark brown: head and gaster, dark, dark brown/ black; thorax and
petiole, dark brown.
DAILY RHYTHMS IN SOCIAL ACTIVITIES
OF THE HARVESTER ANT, POGONOMYRMEX BA DIGS*
By Deborah M. Gordon
Department of Zoology,
Duke University,
Durham, N.C. 27706
Daily cycles in behavior are well known throughout the animal
kingdom. There is some evidence that the activities of ant colonies
are temporally organized so that, at a given time of day, a certain set
of tasks is done. This study explores that possibility by examining
temporal patterns in the social behavior of the harvester ant,
Pogonomyrmex hadius. Such patterns should be distinguished from
circadian rhythms to which conform endogenous, physiological
events exhibited by individual animals (e.g. McCluskey 1958 and
1965). The present study is concerned with daily rhythms in social
activities performed by groups of ants. Two questions are ad-
dressed: 1) Are certain tasks performed at characteristic times of
day? 2) How do activity rhythms vary among different colonies?
There have been many studies of daily temporal patterns in the
overall activity levels of ant colonies, measured as the numbers of
ants entering or leaving the nest (Levieux and Diomande 1978a,
Hunt 1974, Hansen 1978, Van Pelt 1966), or the numbers of ants in
certain areas for specified durations (Janzen 1967, Levieux and
Diomande 1978b, Levieux 1979a and 1979b, Golley and Gentry
1964). Temporal patterns of overall activity level are well docu-
mented for several Pogonomyrmex species (Holldobler 1970 and
1976a, Whitford and Ettershank 1975, Whitford et aL 1976). Some
authors have described temporal patterns of selected social activities
of various ant species (Moglich and Alpert 1979, Janzen 1967,
Levieux and Diomande 1978a and 1978b), including Pogonomyr-
mex (Willard and Crowell 1965, Holldobler 1976b). But, except for
Holldobler’s (1976b) study of mating activity, the cited work pre-
sents no systematic data on temporal patterns in behavior other
than entering and leaving the nest. In some recent field studies of
* Manuscript received by the editor November 21, 1983
413
414
Psyche
[Vol. 85
Pogonomvrniex behavior, I found activity rhythms of various col-
ony tasks (Gordon 1983b and 1983c).
P. hadius has received much less attention than other Pogono-
myrmex species, perhaps because it is geographically isolated from
them. Nevertheless, the scanty literature on P. badius behavior con-
tains some descriptive reports that suggest the existence of activity
rhythms in this species as well (Van Pelt 1966, Hangartner et al.
1970). The present study was made in the laboratory. In this way,
activity rhythms could be investigated more systematically than
would be possible in a field study.
Methods
Four queenright colonies (colonies 1, 2, 3, and 4), each containing
300-600 workers, were observed for 30 days from March 1, 1983
through April 2, 1983. Colonies were kept in open, soil-filled terra-
ria, and fed Bhaktar-Whitcomb (1970) diet or chopped mealworms.
The study colonies had all been kept in the laboratory for about one
year. They were chosen for the study because they had been consist-
ently active and healthy since being brought into the laboratory.
The laboratory temperature was maintained at 27° C (±1°).
Observations of each colony were made 5 times daily, once in
each of five 100-minute time periods, as follows: Time period 1
(TPl), 9:50-11:30; TP2, 11:30-13:20; TP3, 13:20-14:50; TP4,
14:50-16:30; TP5, 16:30-18: 10, and usually in the middle of the time
period at the following five times; 10:40, 12:20, 14:00, 15:40 and
17:20. Overhead fluorescent lights in the laboratory were on from
7:30 to 23:00. A lamp with a 60 watt bulb was placed from 30 to 50
cm above each colony as a heat source. These lamps were illumi-
nated daily from 1 1:30 until 16:30. Thus, during the first and last
observations, room lights but not individual lamps were on; during
the 2nd-4th tjrfie periods, individual lamps were on as well. Temper-
atures on the terraria surfaces, both under the individual lamps and
at other points on the opposite side of the tanks, were measured
with a thermistor (Yellow Springs Instrument Co. #408). Tempera-
ture measurements were made in the terraria of the four study
colonies and also in those of four other colonies maintained in an
identical manner. The colonies were fed every other day imme-
diately after the 12:20 observation.
1983]
Gordon — Pogonomyrmex badius
415
All behavior observed taking place outside the nest was classified
as one of five activities: Foraging, Nest Maintenance, Patrolling,
Midden Work, and Convening (Table 1). For each nest, observa-
tions noted the numbers of ants in each of the five activities. The
sum of the five numbers is the total number of ants observed outside
the nest. A total of 600 observations were made on the four colonies.
The data were analysed by profile analysis (Timm 1975) to deter-
mine whether the numbers of ants engaged in particular activities
depend significantly on both activity and time of day. Since the
times of foraging corresponded so obviously to the time the ants
were fed, foraging was not considered in the analysis.
Profile analysis is a series of 3 multivariate analyses of variance
(manova), described in detail below. Factors considered were col-
ony and date as main effects. The hypothesis that the intercept was
significantly greater than zero was also considered as a main effect.
Date was considered to be a random effect. Each analysis was made
using the data from all 4 colonies, then repeated for each colony
separately. Data were log-transformed to ensure that ratios, not
numbers of ants, were used in the analysis, making it possible to
compare colonies of different sizes.
Results
The first multivariate analysis of variance tested for significant
differences in overall activity among time periods. For each day of
observation of a given colony, a new variable was created for each
time period by adding the (log-transformed) numbers of ants doing
nest maintenance, midden work, convening and patrolling. Four
differences between time periods were used as observation variables
(Table 2, top).
The results (Table 2) show that colonies are significantly more
active in TP2 than in TPl, in TP3 than in TP4, and in TP4 than in
TP5. The overall activity level of the colony has a peak in the middle
of the day.
The next manova tested for significant differences in the numbers
of ants engaged in each activity, summed over all time periods.
A new variable was created by adding, over all five time periods
for each day of observation of a particular colony, the (log-
transformed) number of ants doing each activity. Three differences
Table 1. Classification of activities of exterior workers of P. hadius in laboratory colonies
416
Psyche
§ I
O C
^ es
• E DC
■a c Dfc
C c
c/3 H CL
c
o aJ
-o o
- .5
r: E
3 3
CL
C/5 t
c
3 -O
C oXj
C
C41
Oi)
<u
u ttc
^ X
S 5
T3 O
Dft
C (U
w -a
a c
^ 3
"O
C (U
•- ^
E ^
1i
— -O
O
M "5
lU >
-o -o
c c
3 3
O O
1_ L.
rt efl
05 05
T3
C
05
c/3
C «
O DO
II
•O CL
C lU
L» -o
. ~o ^
c -O ^
•3 E ^
."E c IF
goo
DO ^ ^
3 .E .E
— TD >.
•U 2 3
OC C/3 U
o DO
05 C
OJ • —
00 3
1 £
o
o i=
DO ^
d. d
3 3
O O
T3 -3
C C
05 05
< od U
< OQ < OQ
< CQ U
< oa U
[Vol. 85
With a group milling around slowly under the lamp, sometimes inspecting others of the
group with antennae.
1983]
Gordon — Pogonomyrmex badius
417
between activities were used as observation variables (Table 2,
middle). The results (Figure 1 for foraging, and Table 2 for the other
activities) show that the activities may be ranked as follows, accord-
ing to the numbers of ants engaged in each one: Midden work >
patrolling > convening > nest maintenance > foraging.
Figure 1 shows the activity rhythms of each of the four study
colonies. The third manova tested whether some pairs of activities
are performed by significantly different numbers of ants when the
activities are compared at particular times (Table 2, bottom).
Significant differences mean that the rate at which the colony
invests workers in a particular task depends both on the task and on
the time of day. The results (Table 2) may be best understood by
inspecting Figure 1 . For example, keeping in mind that the data are
log-transformed, activity-time period difference number 4 (Table 2)
can be stated as follows: The ratio of number of ants doing midden
work to number convening in time period 1 is significantly greater
than the same ratio in time period 3. In other words, from TPl to
TP3 convening increases faster, or has a steeper slope, than does
midden work. This difference is especially clear in the graph for
colony 4.
The overall results in Table 2 lead to the following conclusions
about slope differences in Figure 1: Convening rises to a peak in
TP2, increasing more rapidly than midden work, then declines more
rapidly than either midden work or patrolling. In general, patrolling
declines throughout the day while nest maintenance increases. The
fact that activity-time period differences 7, 8, and 9 are not signifi-
cant indicates that all 4 activities change at about the same rate from
TP3 to TP4.
The colony main effect was significant (p > 0.05) for time period
differences 1, 2, and 4, for activity differences 1, 2, and 3, and for
activity-time period differences 1, 4, 5, 6, and 10. The date main
effect was significant for time period difference 3, activity difference
3, and activity-time period differences 3, 7, 8, and 1 1.
Mean temperatures of the terraria surfaces are shown in Figure 2,
as a function of the time of day.
Discussion
The behavior of a colony clearly is temporally patterned. It has
frequently been suggested that, in harvester ants, overall activity
418
Psyche
[Vol. 85
Table 2. Results of profile analysis.
Results of test for differences in activity level by time period are shown at the top
of the table; for test for differences in number of ants in each activity, middle of the
table; for test for activity-time period differences (parallelism test), bottom of the
table. Data are log-transformed. Symbols used are p > 0.01; *, p > 0.05; #,
marginal significance. TP = time period; MW = midden work; CN = Convening; PT
= patrolling; NM = nest maintenance.
Source: Overall
Intercept as main effect
(DF= 29)
(All 4 colonies)
P
Mean
I
Difference
TPl - TP2
346.4**
-1.24
TP2 - TP3
3.9
-0.07
TP3 — TP4
17.8**
0.19
TP4 - TP5
41 1.5**
1.19
Midden Work — Convening
206.6**
1.43
Convening — Patrolling
175.8**
-1.37
Patrolling — Nest Maintenance
256.9**
2.55
1.
TPl, MW - TP1,CN -
TP5,MW+ TP5,CN
8.6**
-0.19
2.
TPl.CN -TPLPT —
TP5,CN +TP5,PT
9.5**
0.17
3.
TP1,PT -TPLNM-
TP5,PT +TP5,NM
48.1**
0.25
4.
TP1,MW - TP1,CN -
TP3,MW+ TP3,CN
115.4**
0.59
5.
TP1,CN -TPLPT -
TP3,CN +TP3,PT
132.9**
-0.60
6.
TP1,PT -TPLNM-
TP3,PT +TP3,NM
13.5
0.11
7.
TP3,MW - TP3,CN -
TP4,MW+ TP4,PT
2.9
-0.06
8.
TP3,CN - TP3,PT —
TP4,CN +TP4,PT
0.5
0.03
9.
TP3,PT TP3,NM —
TP4,PT +TP4,NM
5.7*
0.06
10.
TP4,MW - TP4,CN -
TP2,MW+TP2,CN
51.9**
0.21
11.
TP4,CN — TP4,PT —
TP2,CN +TP2,PT
28.6**
-0.21
12.
TP4,PT TP4,NM -
TP2,PT +TP2,NM
10.5**
-0.12
13.
TP4,MW - TP4,NM —
TP2,MW+ TP2,NM
11.2**
-0.12
14.
TP3,PT — TP3,NM —
TP5.pt +TP5.NM
26.9**
0.14
15.
TP2,PT - TP2,NM -
TP5.pt +TP5.NM
32.9**
0.19
level (sometimes called foraging activity) is related to temperature
(Rogers 1974, Whitford and Ettershank 1975, Bernstein 1979). My
results support this suggestion. During TP2 through TP4, when the
individual lamps were on and the soil temperatures were highest
(Figure 2), colonies were significantly more active than they were
during time periods 1 and 5. Temperatures in the field often become
so high that ants are inactive from midday until early evening. In the
1983]
Gordon — Pogonomyrmex badius
419
Table 2. (continued).
By Colony
18 14 16
13
F
Mean
Difference
F
Mean
Differen
203.8**
-1.65
106.9**
-1.1
0.3
0.04
7.9**
0.15
6.1**
0.18
13.5**
0.27
422.9**
2.05
68.8**
0.89
152.9**
2.07
86.6**
1.55
1 17.9**
-1.74
17.9**
-0.69
75.1**
1.86
152.7**
2.02
18.1**
-0.49
1.1
-0.14
12.9**
0.39
1.6
0.16
22.7**
0.31
29.1**
0.36
40.9**
0.63
1 1.6**
0.35
40.3**
-0.63
15.9**
-0.36
1.3
0.05
19.8**
0.26
1.7
0.20
1.8
-0.07
0.5
0.05
0.04
0.01
3.5
0.08
2.0
0.06
13.9**
0.23
51.9**
0.25
13.4**
-0.24
20.8**
-0.18
1.6
-0.08
16.9**
-0.22
2.4
-0.10
12.1**
-0.15
23.9**
0.26
3.3
0.10
14.6**
0.26
17.9**
0.26
F
Mean
Difference
F
Mean
Difference
65.2**
-0.96
1 13.7**
-1.22
49.3**
-0.44
0.2
-0.04
8.1**
0.19
4.1#
0.16
42.8**
0.73
89.3**
1.13
25.5**
0.89
67.1**
1.22
60.5**
-1.52
101.8**
-1.51
717.5**
3.05
152.4**
3.28
1.2
-0.09
0.04
-0.03
0.1
0.02
0.9
0.10
13.3**
0.18
2.4
0.14
10.6**
0.27
1 14.1**
1.10
23.1**
-0.39
121.5**
- 1 .03
3.8
0.10
0.3
0.04
3.2
-0.07
0.3
-0.04
0.3
0.02
0.4
0.04
0.3
0.02
2.1
0.09
46.1**
0.29
2.2
0.09
22.3**
-0.21
10.5**
-0.21
4.7*
-0.15
0.4
-0.05
1.2
-0.07
3.5
-0.16
4.1#
0.08
2.2
0.10
12.1**
0.21
0.7
0.05
laboratory conditions of the present study, temperatures never
became that high. Time of food availability has also been suggested
as a factor regulating activity rhythms of foraging ants (Hansen
1978, Hunt 1974, Levieux 1979a and 1979b, Levieux and Diomande
1978a and 1978b), and, in fact, all four study colonies foraged pri-
marily at the time of peak food availability.
However, fluctuations of temperature and of food availability
may not account completely for the activity rhythms observed here.
For example, the results show that rates of change in the numbers of
ants in each activity vary with the time of day. This means that ants
doing different tasks respond differently to environmental cues such
420
Psyche
[Vol. 85
Figure I. Daily activity rhythms.
The mean numbers of ants engaged in each activity are plotted as a function of
time of day. Times shown are the usual observation times. Error bars show standard
error of the mean. No error bar is present when the size of the error bar was smaller
than that of the symbol for a point in the graph.
as temperature and food availability. In P. badius, the five activities
described here are performed by four distinct groups of ants (mid-
den work and patrolling are done by the same individuals) (Gordon
1983d). Whether there are intrinsic physiological rhythms causing
these different groups to be active outside the nest in different
numbers, depending on time of day, is a question still to be
explored.
Temporal patterns in overall activity level are known to exist in
many species of ants. Further research on such species may reveal
more detailed patterns of particular activities. Cognizance of such
patterns is relevant to the design of further behavioral experiments.
1983]
Gordon — Pogonomyrmex badius
421
TIME PERIOD
Figure 2.
Mean temperature on the soil surface, both directly under the lamp and on the
opposite side of the tank, plotted as a function of the times of measurement.
because results may be affected by the times of day at which data are
collected (e.g., Gordon 1983d and 1983b).
The results show that, though colonies are similar to one another,
distinct colonies have distinct activity rhythms (Figure 1). Thus,
intercolony variation should also be taken into account when
designing experiments. The main point of this study, then, is not
that the activity rhythms of P. badius are those shown in Figure 1,
but that each colony exhibits some temporal pattern of activities. In
every colony, certain tasks are undertaken at characteristic times.
Clearly, we need to consider temporal patterns when we endeavor to
understand the social organization of the ant colony.
422
Psyche
[Vol. 85
Acknowledgements
1 am grateful to R. Shaw and D. S. Burdick for statistical advice,
to T. Williams for valuable discussions, to B. Holldobler for com-
ments on the manuscript, and to J. Gregg and R. Palmer for help
with all stages of the project.
Rkphrenchs
Bernstein, R. A.
1979. Schedules of foraging activity in species of ants. J. Anim. Ecol. 48:
921-930.
Buaktar, a. W. and W. Whitcomb.
1970. Artificial diet for rearing various species of ants. Fla. Entomol. 53(4):
229 232.
Goli Ev, F. B. AND J. B. Gentry.
1964. Bioenergetics of the southern harvester ant, Pogonomyrmex hadius.
Ecol. 45: 217-225.
Gordon, D. M.
1983a. The dependence of necrophoric response to oleic acid on social context
in the harvester ant, Pogonomyrmex hadius. J. Chem. Ecol. 9(1):
105 111.
1983b. The relation of recruitment rate to activity rhythms in the harvester ant,
Pogonomyrmex harhafus (F. Smith). J. Kans. Ent. Soc. 56(3): 277-285.
1983c. Species-specific patterns in the social activities of harvester ant colonies
(Pogonomyrmex). Insectes Sociaux. In press.
1983d. The organization of role behavior in colonies of the harvester ant, Pogo-
nomyrmex hadius. In preparation.
Hangartner, W., j. M. Reichson, and E. O. Wilson.
1970. Orientation to nest material by the ant Pogonomyrmex hadius. Anim.
Behav. 18(2): 331-334.
Hansen, S. R.
1978. Resource utilisation and coexistence of three species of Pogonomyrmex
ants in an upper Sonoran grassland community. Oecologia 35: 109- 1 17.
HOlldohler, B.
1970. Steatoda fulva (Theridiidae), a spider that feeds on harvester ants.
Psyche, 77: 202-208.
1976a. Recruitment behavior, home range orientation, and territoriality in har-
vester ants. Pogonomyrmex. Behav. Ecol. Sociobiol. 1: 3-44.
1976b. The behavioral ecology of mating in harvester ants (Pogonomyrme.x).
Behav. Ecol. Sociobiol. 1(4): 405-423.
Hi nt, J. H.
1974. Temporal activity patterns in two competing ant species. Psyche, 81:
237 242.
1983]
Gordon — Pogonomyrmex hadius
423
Jan/fn. D. H.
1967. Interaction of the bull’s horn acacia {Acacia cornigera 1..) with an ant
inhabitant {Pseudomyrmex fenuginea F. Smith) in eastern Mexico.
Univ. Kansas Sci. Bull. 47(6): 315 558.
Lfvifi X. J.
1979a. La nutrition des fourmis granivores. III. Cycle d’activite et regime
alimentaire en saison des pluies de Bracfu poncra senaarensis fluctuations
saisonieres. Insectes Sociaux 26(3): 232-239.
1979b. La nutrition des fourmis granivores. IV. Cycle d’activite et regime
alimentaire de Messor galla et de Messor (=Craiomyrmex) regalis.
en saison des pluies fluctuations annuelles. Insectes Sociaux 26(4):
279 294.
Lfvifi x.J. andT. Dkimandf.
1978a. l,a nutrition des fourmis granivores. 1. Cycle d’activite et regime
alimentaire de Messor galla et de Messor (=Craloniyrmex) regalis.
Insectes Sociaux 25(2): 127 139.
1978b. La nutrition des fourmis granivores. II. Cycle d’activite et regime
alimentaire de Brachyponera senaarensis (Mayr). Insectes Sociaux 25(3):
187-196.
M( Cl l SKFV, E.
1958. Daily rhythms in male harvester and argentine ants. Science 128:
536 537.
1965. Circadian rhythms in male ants of five diverse species. Science 150:
1037 1039.
MOGI I( H, M. AND G. D. Al PFRT.
1979. Stone-dropping by Conomynna hicolor a new technique of interfer-
ence competition. Behav. Ecol. Sociobiol. 6: 105-1 13.
Rogfrs, L. E.
1974. Foraging activity of the western harvester ant in the shortgrass plains
ecosystem. Env. Entomol. 3(3): 420 424.
Timm, N. H.
1975. Multivariate analysis. Monterey, CA: Brooks/Cole Publishing Co.
Van Pfi t, A. F.
1966. Activity and density of old-field ants of the Savannah River Plant, South
Carolina. J. Elisha Mitchell Soc., May: 35 43.
WllITFORD, W. G. AND G. EtTFRSMANK.
1975. Factors affecting foraging activity in Chihuahuan desert harvester ants.
Env. Entomol. 4(5): 689-696.
Whitford, W. G., P. Johnson, and J. Ramirfz.
1976. Comparative ecology of the harvester ants Pogonomyrniex harhatus ( F.
Smith) and Pogonomyrniex rugosus (Emery). Insectes Sociaux 23(2):
112-132.
Wil l ARD, J. AND H. CrOWI I I .
1965. Biological activities of the harvester ant, Pogonomyrniex owyheei, in
central Oregon. J. Econ. Entomol. 58: 484 489.
BEHAVIOR OF THE SLAVE-MAKING ANT,
HA RPA GOXENVS A MERIC AN VS (EMERY),
AND ITS HOST SPECIES UNDER “SEMINATURAL”
LABORATORY CONDITIONS
(HYMENOPTERA: FORMICIDAE)'
By Thomas M. Alloway
AND
Maria Guadalupe Del Rio Pesado
Erindale College
University of Toronto
Mississauga, Ontario L5L 1C6
Canada
Introduction
Slave-making ants are social parasites that raid the nests of host-
species colonies, capture brood, and transport it back to the parasite
colony. There, host-species workers eclosing from captured brood
become “slaves” which perform all the usual worker-ant functions in
the slave-maker colony (see review by Buschinger et ai 1980).
Harpagoxenus aniericanus (Emery) is an obligatory slave-making
parasite which forms mixed colonies with workers of three Lepto-
thorax host species: L. amhiguus Emery, L. curvispinosus Mayr,
and L. longispinosus Roger. Young H. aniericanus queens found
colonies by entering host-species nests, killing or driving off the
adults, and inducing the host-species workers which subsequently
mature from worker pupae in the nest to rear a brood of slave-
maker workers (Wesson 1939). These parasite workers then aug-
ment the slave worker force by raiding other host-species nests.
Wesson (1939) and Alloway (1979) observed H. aniericanus slave
raids in the laboratory by placing populous H. aniericanus nests in
'This research was supported by a grant from the Natural Sciences and Engineering
Research Council (Canada) to the first author and by a scholarship from CON ACYT
(Consejo Nacional de Ciencia y Tecnologia, Mexico) to the second author. The
authors would like to thank Victor Chudin for his assistance in collecting the data
and Robin Stuart, David Gibo, and James Beckwith for their constructive comments
on the manuscript.
Manuscript received hy the editor September 30. 1983.
425
426
Psyche
[Vol. 85
experimental arenas containing an arbitrarily selected host-species
target nest. Under these circumstances, H. americanus raids begin
when one or more slave-maker workers leave the parasite nest to
explore the arena. Whenever such a “scout” discovers the entrance
to the target nest, it returns to its own nest and recruits a raiding
party. After dispersing the adult residents of the target nest, the
raiders carry the captured brood back to the slave-maker nest
(Alloway 1979).
Recently, it was discovered that H. americanus, L. amhiguus, L.
longispinosus and probably L. curvispinosus form facultatively
polydomous colonies (Alloway ef al. 1982; Del Rio Pesado & Allo-
way 1983). Colonies of the three host species are also facultatively
polygynous (Alloway c/ 1982). However, H. americanus co\on\ts
apparently never contain more than one inseminated egg-laying
queen (Buschinger & Alloway 1977).
In the present paper, we augment previous findings by presenting
behavioral observations of H. americanus and its slaves interacting
with ants from other H. americanus colonies and from unenslaved
host-species colonies. These observations supplement previous find-
ings for three reasons:
1. The interactions observed were among ants from colonies col-
lected adjacent to one another in nature.
2. The ants were observed for several weeks.
3. Ants from small and “weak”, as well as populous and “strong”,
H. americanus and host-species colonies were observed.
Materials and Methods
Nests of H. americanus, L. amhiguus, and L. longispinosus were
collected on the Erindale Campus of the University of Toronto, in
Mississauga, Ontario. Since we wanted to observe the behavior of
ants from parasite nests occurring close together in nature, we
looked for places where there were at least two H. americanus nests
within less than 2 m of each other. Whenever such a spot was found,
we laid out a 2 m by 2 m quadrant centering on the parasite nests
and then collected, numbered, and mapped the location of all H.
americanus and host-species nests in the quadrant. Altogether, 19
quandrants were collected; but two pairs of adjacent quadrants were
combined to permit observation of large groups of H. americanus
1983]
Alloway & Del Rio Pesado — Harpagoxenus
427
nests. See Del Rio Pesado (1983) for a complete demographic des-
cription of the colonies studied.
In the laboratory, the ants were removed from their natural nests,
established in artificial nests (Alloway 1979), and censused. Then
they were transported to a naturally lighted, unairconditioned
room, where the field maps were used to reconstruct among the
artificial nests the same spatial relations as had existed among the
natural nests. In addition to these “natural” quadrants, we also
observed one control quadrant containing two H. americanus nests
from different collection sites. In some cases, individual ants were
marked. See Del Rio Pesado and Alloway (1983) for a detailed
description of these procedures.
Ad libitum behavioral observations were made 8 h a day, 5 days a
week during June, July, and August. Five quadrants were observed
in 1980; and 14 quadrants were observed in 1981. An assistant was
employed during 1981 to permit more detailed behavioral ob-
servations.
Results
Raiding
The slave-makers raided or attempted to raid the nests of ad jacent
colonies. Most raided nests belonged to unparasitized L. amhiguus
and L. longispinosus colonies. However, in the control quadrant
and in the two “natural” quadrants containing more than one H.
americanus colony, the slave-makers from one colony raided nests
belonging to another parasite colony.
Alloway (1979) observed that the raiding behavior of H. america-
nus is not highly stereotyped even when ants from a single parasite
nest are interacting with ants from a single target nest. In the present
study in which the slave-makers were often interacting with ants
from several naturally adjacent colonies, the results were so com-
plex and variable that their complete presentation requires a separ-
ate description of the events in each quadrant. See Del Rio Pesado
(1983) for such an account. Here we summarize those observations.
Much of the behavioral variability could be attributed to demo-
graphic variability. One demographic factor was the number of
nests in each quadrant. The initial number of slave-maker nests in
different quadrants ranged from 2 to 6, while the initial number of
428
Psyche
[Vol. 85
host-species nests ranged from 0 to 17. This variablity in nest density
is probably correlated with small-scale differences in the availability
and suitability of natural nest sites. In addition, the history of slav-
ery in a particular spot might affect nest density, since H. american-
us colonies may destroy or drive away adjacent host-species
colonies.
Another kind of demographic variability involved the number of
nests occupied by single colonies. Some H. americanus and some
host-species colonies were initially polydomous. In the laboratory,
some initially polydomous colonies moved into a single nest
(became monodomous) before any significant interactions with
members of other colonies occurred; but others remained polydom-
ous during behavioral interactions with ants from other colonies
(Del Rio Pesado & Alloway 1983). To raid host-species nests suc-
cessfully, H. americanus colonies must deploy raiding parties con-
taining several H. americanus workers. In successful polydomous
H. americanus colonies, the slaves made this possible by carrying all
or almost all the parasite workers to a single nest before raiding
began. While so doing, the slaves sometimes (but not always) moved
the entire H. americanus colony into one nest. In other polydomous
slave-maker colonies where the slaves failed to assemble the parasite
workers in this way, many slave-makers were killed during uncoor-
dinated attacks on target nests.
A third kind of demographic variability involved differing degrees
of maturity among slave-maker colonies. When collected, some of
our H. americanus colonies were incipient {i.e. initially contained
only an H. americanus queen, some slaves, and a brood), while
others already possessed slave-maker workers. Wesson (1939), study-
ing ants from the east-central United States, found that H. ameri-
canus began to raid only after the overwintered H. americanus
brood had matured. In contrast, overwintered parasite workers in
our colonies from southern Ontario began to raid before all their
overwintered brood had matured. As young H. americanus workers
eclosed, they augmented the raiding forces of mature colonies and
initiated raiding in incipient colonies. Thus, mature colonies could
start raiding earlier and had the potential to raid longer than incip-
ient colonies. In both incipient and mature colonies, first-year H.
americanus workers were involved in all phases of raiding {i.e.
scouting, attacking target nests, and transporting captured brood).
1983]
Allow ay & Del Rio Pesado — Harpagoxenus
429
In this last respect, H. americanus apparently differs from the Euro-
pean H. suhlaevis (Nylander), in which slave-makers in mature col-
onies do not begin to scout until their second year (Buschinger et a/.
1980). We also observed an apparent effect of experience on scout-
ing. On their first forays in the spring or after eclosion, scouts
ventured only a short distance from their nest. The distance trav-
elled became greater as the number of forays increased.
Alloway ( 1 979) observed that H. americanus workers could scout
either singly or in small groups. In the present study, only individual
scouting was observed. Alloway (1979) also observed that, whenever
a lone scout discovered the entrance to a target nest, it would return
to its own nest and recruit a raiding party. However, in the present
study, lone scouts sometimes attacked target nests by themselves.
Nevertheless, lone H. americanus workers rarely, if ever, captured
any brood. Invasion of a target nest by a single slave-maker excited
the target-colony workers and often caused them to attack the
intruder. Some lone intruders were killed.
The success of raider recruitment was highly variable. Upon
entering its nest, a scout that had discovered the entrance to a target
nest was immediately surrounded by a cluster of slave-makers and
slaves. Shortly thereafter, the scout would make its way back to the
nest entrance and leave. That the slave-maker was now almost cer-
tainly laying down a pheromone trail was indicated by the fact that
it conspicuously dragged its gaster along the substrate while being
closely followed by a column of other slave-makers and/or slaves.
All scouts that had located target nests excited their nestmates; and
most initiated processions. The variable success of raider recruit-
ment seemed to depend on the “steadiness” of the recruiter’s move-
ment and orientation while leading the procession. Successful
recruiters moved steadily forward without making any abrupt turns.
Less successful recruiters stopped for prolonged periods and
changed direction abruptly. Such hesitation caused nestmates to
leave the procession; and badly disoriented scouts lost all their fol-
lowers. Some initially unsuccessful individuals later relocated the
target nest and went back to their nest to try again.
The arrival of a raiding party containing several H. americanus
workers and (often) a number of slaves always caused “alarm” in the
target nest. Workers and queens would snatch up larvae and pupae
and make frenzied efforts to leave the nest. Whenever they found a
430
Psyche
[Vol. 85
place where they could safely deposit any brood with which they
had escaped, the workers returned to the invaded nest and carried
off more brood. The slave-makers countered all these efforts by
guarding the nest entrance (Alloway 1979) and by charging and
snapping at target-nest workers.
Alloway (1979) observed that target-nest workers always fled with
whatever brood they might manage to carry almost immediately
after the arrival of a raiding party. In these circumstances, the slave-
makers did not use their large, specialized mandibles against the
residents of target nests. In the present study, a broader range of
target-nest resistance and slave-maker aggression were observed.
The workers in some target nests fled very shortly after the raiders
arrived; and, in these cases, the slave-makers injured very few, if
any, target-nest residents. However, in other target nests, the
workers bit and stung the invaders. The slave-makers crushed such
resistance by employing their large mandibles to dismember their
adversaries. Slaves in raiding parties also attacked target-colony
workers, but it was apparent that the success of the always outnum-
bered raiders depended mainly upon the activities of the slave-
makers.
After all the adults had been killed or driven from the target nest,
the raiders transported the captured brood to the slave-maker nest.
Most brood was carried by slave-makers, although slaves sometimes
carried one or two larvae or pupae. Brood transport generally lasted
only a few hours, after which the raiding party vacated the target
nest. Only one H. americanus colony manifested the phenomenon
reported by Wesson (1939) of raiders requiring 2 or 3 days to com-
plete brood transport. After the raiding party had abandoned the
target nest, its previous inhabitants often returned.
Other Behavior
Our observations confirm that Leptothorax slaves do most of the
work in H. americanus colonies. The slaves forage for food, feed
and groom the parasite adults and brood, and defend the area
around H. americanus colonies by attacking foraging workers from
neighboring Leptothorax colonies whenever they are encountered
near an H. americanus nest. The slave-makers do none of these
things on a regular basis. Indeed, the parasites appear never to leave
their nests except to scout {i.e. to “look for” target nests). Since
scouting slave-makers invariably return to the same nest from which
they departed, the parasites are even dependent on their slaves to
I9K3] Alloway & Del Rio Pesado — Harpagoxenus 431
move them from nest to nest in polydomous colonies. Nevertheless,
H. aniericanus workers possess certain vestiges of non-parasitic
behavior. Inside their nest, H. aniericanus workers routinely groom
one another, periodically share regurgitated food with other slave-
makers and slaves, and occasionally engage in what appears to be
brood care. Parasite workers may even eat if they encounter food
while scouting. On such occasions, one can infer that the slave-
maker is scouting (and not foraging) from the fact that, after eating,
it continues to “look for” a target nest, instead of returning directly
to its own nest, regurgitating to nestmates, and recruiting them to
the food source. H. aniericanus workers never recruit or follow
nestmates except in the context of slave raids.
Although Leptothorax slaves generally look after the slave-
makers, we observed many instances of slave aggression against
slave-makers. In 9 slave-maker colonies, we saw slaves biting and
dragging H. aniericanus workers out of slave-maker nests. A few H.
aniericanus workers lost parts of appendages as a result of these
attacks. However, we never saw a slave-maker attack a slave; and
we never witnessed anything resembling a generalized “slave revolt”.
Individual H. aniericanus workers were attacked by individual
slaves. The same slave which attacked one slave-maker would feed
and groom another; and any slave-maker that was attacked by one
slave was cared for by others.
A somewhat different kind of slave aggresssion was seen in one of
our incipient H. aniericanus colonies. When collected, this colony
possessed a single nest containing an H. aniericanus queen, 17 L.
longispinosus workers, and a brood. Throughout the course of our
observations, the slaves fed and groomed the parasite queen and
tended her brood through the pupal instar. However, the slaves
killed all eclosing H. aniericanus workers. Similar events have been
observed in other incipient H. aniericanus colonies (R. J. Stuart,
personal communication).
As we have noted, slaves ordinarily defend the area surrounding
H. aniericanus nests against incursions by unenslaved Leptothorax
workers. Similarly, unenslaved leptothorax workers defend areas
around their nests against incursions by Leptothorax slaves. These
phenomena, together with the fact that both enslaved and unen-
slaved Leptothorax workers fight for their respective colonies during
slave raids, indicate that enslaved and unenslaved Leptothorax
workers generally recognize one another as belonging to different
432
Psyche
[Vol. 85
colonies. However, these behavioral barriers between colonies are
sometimes imperfect in the case of incipient slave-maker colonies.
For example, let us consider the situation in Quadrant 3.
When collected, this quadrant contained two incipient H. ameri-
canus colonies, each of which was located near an apparently
unparasitized L. hngispinosus nest. In both cases, some of the
slaves entered the nearest L. hngispinosus nest without being
attacked; and, reciprocally, some of the seemingly unenslaved L.
hngispinosus workers entered the H. americanus nest with impu-
nity. On one occasion, a slave picked up the H. americanus queen in
one of the parasite nests and carried her to the nearest L. hngispino-
sus nest. The arrival of the parasite female caused all the adults in
that nest to flee. Later the same day, a slave carried the H. america-
nus queen back to the nest from which she had come. Then, over a
12-day period, many of the workers which had originally fled moved
in and began to live peacefully with the H. americanus queen in her
nest.
Equally interesting events involved the other incipient parasite
colony in the same quadrant. A slave which could peacefully enter
the nearest L. hngispinosus nest began to carry brood and workers
from that nest into the //. americanus nest. Some of the in-coming
L. hngispinosus workers were accepted immediately by the other
slaves, while others were initially attacked. However, after 15 days,
all the workers from the unparasitized nest were living peacefully
with the H. americanus queen. A few days later, several L. hngispi-
nosus workers killed the L. hngispinosus queen which had been
living in the unparasitized nest.
Discussion
Both Wesson (1939) and Alloway (1979) produced slave raids by
selecting target nests and placing them in arenas with relatively
populous, single-nest H. americanus colonies. The present study
was the first in which a broader sample of H. americanus colonies
has been observed and the first in which H. americanus colonies
have been observed interacting with other colonies near which the
slave-makers had been living in nature. These procedural differences
probably account for the discrepencies between the behavioral
events observed here and those described by Alloway (1979). Sim-
ilar procedural differences, combined with possible regional differ-
1983]
Allow ay & Del Rio Pesado — Harpagoxenus
433
ences between populations, may account for differences between the
present results and those of Wesson (1939).
A number of our observations pertain to limits on the success of
slave-raiding in polydomous parasite colonies. Individual H. ameri-
canus workers can rarely capture brood and are sometimes killed by
target-colony workers. Yet, groups of 4 or 5 H. americanus workers
can successfully raid almost any target nest. Thus, H. americanus
colonies need to deploy their raiders in raiding parties containing
several parasite workers. However, polydomy sometimes prevents
such deployment. The slave-makers rely on their slaves to carry
them from nest to nest in polydomous colonies; and the slaves often
fail to assemble the slave-makers in a single nest from which success-
ful raids could be mounted. As a consequence, some polydomous H.
americanus colonies fail to organize raiding parties containing
enough slave-makers to capture brood from neighboring host-
species colonies.
This difficulty encountered by H. americanus colonies living in
more than one nest has led us to question the adaptive value of
polydomy in the slave-maker population studied. Both the Lepto-
thorax host species enslaved by H. americanus in the Toronto
region form facultatively polydomous colonies (Alloway et al.
1982). Thus, if enslaved host-species workers behave like unenslaved
conspecifics, slaves should tend to provide a polydomous colony
structure for the parasites. Perhaps, some H. americanus colonies
are polydomous because of this behavioral propensity of their slaves
and despite the fact that polydomy is detrimental to efficient
raiding.
In addition, polydomy may account for some of the overt aggres-
sion observed in the present study. By extension, polydomy might
partly explain the similar forms of slave aggression manifested by
Leptothorax slaves living in L. duloticus colonies (Wilson 1975).
Let us imagine that a slave-maker colony divides, with some of
the parasites and slaves remaining in the original nest, while others
move to another nest. Let us further suppose that the slave-makers
in the two nest raid independently. In such a situation, young slaves
maturing from captured brood in each nest might learn to recognize
as nestmates only those particular slave-makers with which they
were living. If the ants from the two nests later reunited, then the old
slaves might accept all the slave-makers, while the young slaves
434
Psyche
[Vol. 85
accepted only familiar individuals. This scenario could explain our
observations of slaves biting and dragging slave-makers out of nests.
The aggression observed was always an individual matter. Some
slaves accepted all the slave-makers, while other slaves accepted
certain slave-makers and attacked others.
A somewhat similar hypothesis might account for the imperfect
behavioral boundaries between some incipient H. americanus colo-
nies and nearby unparasitized nests. An H. americanus queen founds
a new colony by entering a host-species nest, killing or driving off
the adults, and capturing worker pupae that subsequently mature to
become her first slaves (Wesson 1939; Sturtevant 1927). If a parasite
queen founded a colony in one nest of a polydomous Leptothorax
colony, it would not be surprising if some of the parasite’s first
slaves were acceptable in other nests of the same colony. Similarly,
“free” workers from that colony might be acceptable in the slave-
maker nest. However, this hypothesis cannot explain how, under
these circumstances, a parasitized nest could unidirectionally siphon
brood and workers from an unparasitized nest or how an H.
americanus queen could become more attractive than a Leptothorax
queen. Yet, H. americanus queens and the queens of many other
socially parasitic species somehow usurp the place of host-species
queens (Wilson 1971). How parasite queens accomplish this feat
remains an important subject for future research.
Polydomy also cannot account for the case where slaves cared for
an H. americanus queen and her brood but killed all eclosing H.
americanus workers. Explaining this phenomenon would require
understanding the mechanisms of nestmate recognition in these spe-
cies; and these mechanisms are incompletely understood. However,
studies in progress (R.J. Stuart, personal communication) indicate
that apparent “mistakes” in nestmate recognition are possible in
these host species and that//, americanus may exploit these possibil-
ities. When slaves work for a parasite queen, they may be mistak-
enly identifying her as a nestmate. When the same slaves destroy the
parasite’s offspring, they may be correctly identifying them as aliens.
Gladstone (1981) discussed various theoretical reasons why slave
workers should not “revolt” against slave-makers. However, our
observations of H. americanus colonies and Wilson’s (1975) obser-
vations of Leptothorax du/oticus colonies show that individual
slaves sometimes manifest what might be interpreted as “rebellious
behavior”. If our inferences about polydomy are correct, whole
1983]
Allow ay & Del Rio Pesado — Harpagoxenus
435
slave worker forces may even organize slave-maker colonies in a
way which produces inefficient raiding. Nevertheless, we doubt that
any of these behavioral phenomena are manifestations of evolved
host-species defenses against slave-makers. We suppose that the
behavior of host-species workers has evolved to maximize the
reproductive potential of host-species queens. Slave-maker popula-
tions are so sparse that only a small proportion of host-species
colonies are ever raided. Thus, slavery seems unlikely to exert signif-
icant selection pressure on host-species populations; and we believe
that the facultative polydomy and polygyny found in these host-
species are adaptations to conditions in host-species (not parasite)
colonies.
In these host species, polygyny involves the acceptance of newly
mated young queens in existing colonies. Simultaneously, poly-
domy involves a more or less continual exchange of workers,
queens, and brood among nests; and such commerce requires
workers in one nest to accept workers, queens, and brood from
other nests of the same colony (Alloway et al. 1982). An incidental
effect of these characterstics of host-species colonies is to produce a
worker caste which is vulnerable to enslavement. Of course, a
second effect of polydomy is to produce a worker caste which tends
to organize multiple-nest colonies; and life in multiple nests may be
disadvantageous to slave-makers. In other words, Harpagoxenus
americanus parasitizes the labor of workers which possess a “mixed
bag” of behavioral characteristics. Some of these characteristics
may facilitate enslavement, while others may produce inefficient
slave-maker colonies. However, the assertion that host-species
workers have evolved to be inefficient slaves seems only a little more
likely than the assertion that they have evolved to be slaves at all.
Summary
Colonies of the slave-making ant, Harpagoxenus amerieanus
(Emery), and two of its host species {Leptothorax anihiguus Emery
and L. lohgispinosus Roger) were observed under “seminatural”
conditions, in which the ants lived in artificial nests arranged to
reconstruct the spatial relationships among their natural nests.
Some of the slave-maker and host-species colonies were polydom-
ous.. In some polydomous slave-maker colonies, the slaves carried
all the H. amerieanus workers into one nest before the onset of
raiding. When thus assembled, the slave-makers efficiently captured
436
Psyche
[Vol. 85
brood from nearby host-species colonies. In other polydomous col-
onies where the slave-makers remained in more than one nest, the
parasites conducted unco-ordinated raids and incurred many casu-
alties. Several kinds of slave aggression against the slave-makers are
described. However, slaves ‘’peacefully” augmented the slave worker
forces of some incipient H. americanus colonies.
References
Allow AY, T. M.
1979. Raiding behaviour of two species of slave-making ants, Harpagoxenus
americanus (Emery) and Leptothorax duloticus Wesson. Animal Behav-
iour 27: 202-210.
Alloway, T. M., a Bu.schinger, M. Talbot, R. Stuart, & C. Thomas.
1982. Polygyny and polydomy in three North American species of the ant
genus Leptothorax Mayer (Hymenoptera: Formicidae). Psyche 89:
249-274.
Buschinger, a., & T. M. Alloway.
1977. Population structure and polymorphism in the slave-making ant Harpa-
goxenus americanus (Emery). Psyche 83: 233-242.
Buschinger, A., W. Ehrhardt, and U. Winter.
1980. The organization of slave-raids in dulotic ants; A comparative study
(Hymenoptera; Formicidae). Zeithschrift fur Tierpsychologie 53:
245-264.
Del Rio Pesado, M. G..
1983. Polydomy in the slave-making ant, Harpagoxenus americanus (Emery)
(Hymenoptera; Formicidae). M.Sc. Thesis, University of Toronto.
Del Rio Pesado, M. G. & T. M. Alloway.
1983. Polydomy in the slave-making ant, Harpagoxenus americanus (Emery)
(Hymenoptera; Formicidae). Psyche 90: 151-162.
Gladstone, D. E.
1981. Why there are no ant slave rebellions. American Naturalist 117: 779-781.
Sturtevant, a. H.
1927. The social parasitism of the ant Harpagoxenus americanus. Psyche 34:
1-9.
Wesson, L. G., Jr.
1939. Contribution to the natural history of Harpagoxenus americanus Emery
(Hymenoptera; Formicidae). Transactions of the American Entomolgi-
ca! Society 65: 97-122.
Wilson, E. O.
1971. The Insect Societies. (Cambridge, Mass.,; Belknap Press of Harvard
University Press, x + 548 pp.)
1975. Leptothorax duloticus and the beginnings of slavery in ants. Evolution
29: 108-119.
PSYCHE
INDEX TO VOLUME 90, 1983
INDEX TO AUTHORS
Alloway, Thomas M. and Maria Guadalupe Del Rio Pesado. Behavior of the
Slave-Making Ant, Harpagoxenus americanus (Emery) and Its Host Species
under “Seminatural” Laboratory Conditions. 425
Alloway, Thomas M. See Del Rio Pesado, M.G.
Betz, B. W. The Biology of Trichadenotecnum alexanderae Sommerman (Psocop-
tera: Psocidae). III. Analysis of Mating Behavior. 97
Blondheim, Syril A. and Eliezer Frankenherg. ‘Protest’ Sounds of a Grasshopper:
Predator-Deterrent Signal? 387
Brow n, William L, Jr. See Willey, Robert B.
Burnham, Laurie. Studies on Upper Carboniferous insects: 1. The Geraridae
(Order Protorthoptera). 1
Buschinger, Alfred and Andre Francoeur. The Guest Ant, Symmyrmica chamher-
lini. Rediscovered near Salt Lake City, Utah (Hymenoptera, Formicidae). 297
Buschinger, Alfred, Ursula Winter, and Walter Faber. The Biology of Myrmoxen-
us gordiagini Ruzsky, a Slave-Making Ant. 335
Calabi, Prassede, James F.A. Traniello, and Michael H. Werner. Age Polyethism:
Its Occurrence in the Ant, Pheidole hortensis, and Some General Con-
siderations. 395
Carpenter, Frank M. The Structure and Relationships of Eubleptus danielsi
(Palaeodictyoptera). 81
Carpenter, Frank M. Dedication: Philip J. Darlington, Jr. 333
Chandler, Donald S. Larvae of Wrack Coleoptera in the families Corylophidae,
Rhizophagidae, and Lathridiidae. 287
Conner, Jeffrey, and Thomas Eisner. Capture of Bombardier Beetles by Ant-Lion
Larvae. 175
Del Rio Pesado, Maria Guadalupe and Thomas M. Alloway. Polydomy in the
Slave-Making Ant, Harpagoxenus americanus (Emery). 1 5 1
Del Rio Pesado, Maria Guadalupe. See Alloway, Thomas M.
Douglas, Matthew M. Defense of Bracken Fern by Arthropods Attracted to Axil-
lary Nectaries. 313
Eisner, Thomas. See Conner, Jeffrey.
Eisner, Thomas. See Nowicki, Steven.
Faber, Walt her. See Buschinger, Alfred.
439
Frolich, D.R. and F.D. Parker. Nest Building Behavior and Development of the
Sunflower Leafcutter Bee: Eumegachile (Sayapis) pugnata (Say). 193
Frankenherg, Eliexer. See Blondheim, Syril A.
Gordon, Deborah M. Daily Rhythms in Social Activities of the Harvester Ant,
Pogonomyrmex hadius. 4 1 3
Haskins, Caryl P. and Edna F. Haskins. Situation and Location-Specific Factors
in the Compatibility Response in Rhytidoponera mefallica (Hymenoptera:
Formicinae: Ponerinae). 163
Haskins, Edna F. See Haskins, Caryl P.
Henry, Charles S. Temperature Induced Changes in the Calls of the Green Lace-
wing, Chrysoperla plorabunda. 343
Herbers, Joan M. Social Organization in Leptothorax Ants: Within- and Between-
Species Patterns. 361
Johnson, Leslie K. Reproductive Behavior of Claeoderes bivittata (Coleoptera:
Brentidae). 135
Kwait, Ellen C. and Howard Topoff. Emigration Raids by Slave-Making Ants: a
Rapid Transit System for Colony Relocation. 307
Messina, Frank J. See Root, Richard B.
Nowicki, Steven and Thomas Eisner. Predatory Behavior of Bombardier Beetles by
a Tabanid Fly Larva. 1 19
Orgren, M.C. Ferreira. See Strassman, J.E.
Parker, F.D. See Frolich, D.R.
Rissing, Steven W. Natural History of the Workerless Inquiline Ant, Pogonomyr-
mex colei (Uymenopieva, Formicidae). 321
Root, Richard B. and Frank J. Messina. Defensive Adaptations of a Case-Bearing
Beetle, Exema canadensis (Coleoptera: Chrysomelidae). 67
Ross, Kenneth G. and P. Kirk Visscher. Reproductive Plasticity in Yellowjacket
Wasps: a Polygynous, Perennial Colony of Vespula maculifrons. 179.
Shapiro, Arthur M. Testing Visual Recognition in Precis (Lepidoptera: Nymphali-
dae) using a Cold-Shock Phenocopy. 59
Shelly. Todd E. Prey Selection by the Neotropical Spider, Alpaida tuonado, with
Notes on Web-Site Tenacity. 123
Strassman, J.E. and M.C. Ferreira Orgren. Nest Architecture and Brood Devel-
opment Times in the Paper Wasp, Polistes exclamans (Hymenoptera: Vespi-
dae). 237
Topoff, Howard. See Kwait, Ellen C.
Traniello, James F.A. See Calabi, Prassede.
Visscher, P. Kirk. See Ross, Kenneth G.
440
Werner. Floyd D. Anthicidae of the Greater Antilles and a New Species from
Venezuela (Coleoptera). 211
Werner, Michael H. See Calabi, Prassede.
Willey, Robert B. and William L. Brown, Jr. New Species of the Ant Genus
(Hymenoptera; Formicidae; Ponerinae)
Winter, Ursula. See Buschinger, Alfred.
INDEX TO SUBJECTS
All new genera, new species and new names are printed in capitai type
Age polyethism in Pheidole, 395 Case-bearing beetle, Exema canadensis.
A Ipaido tuonabo, 1 23
Anepitedius, 44
Ant-lion larvae, 175
Anthicidae of Greater Antilles, 21 1
Anthicus antii i forum, 218
Anthicus Bi ackwei DERI, 219
Anthicus dari ingtoni, 224
Anthicus hispanioi af, 224
Anthicus macgillavryi, 226
Anthicus margaritae, 219
Anthicus soledad, 224
Anthicus subtilis, 222
Anthicus russoi, 230
Arthropods attracted to axillary necta-
ries, 313
Athymodictya, 84
Behavior of Harpagoxenus americanus,
425
Biology of Myrmoxenus gordiagini. 335
Biology of Trichadenotecnum alexande-
rae, 97
Bombardier beetles, 119, 175
Brachinus, 1 1 9
Capture of bombardier beetles by ant-
lion larvae, 175
67
Claeoderes bivittata, 1 35
Corticaria valida, 293
Daily rhythms in social activities of Po-
gonomyrmex, 413
Dedication: P.J. Darlington, Jr., 333
Defense of bracken fern by arthropods,
313
Defensive adaptations and natural ene-
mies of Exema, 67
Emigration raids by slave-making ants,
307
Eubleptus, 81
Eumegachile pugnata, 193
Genentomum, 36
Geraridae, 1
Gerarulus, 42
Gerarus, 12
Green lacewing, 343
Guest ant, Symmyrmica chamberlini,
297
Harpagoxenus americanus, 1 5 1
Larvae of wrack Coleoptera, 287
Leptothorax, social organization, 361
Mating behavior of Trichadenotecnum,
97
441
Myopias c hapmani, 264
Myopias 158
Myopias dfh a. 281
Myopias nrNSFS i ic iA, 268
Myopias gioas, 251
Myopias ,mi ivora, 254
Myopias i obosa, 211
Myopias mfdia. 257
Myopias NOPS. 279
Myopias RUI HAF, 274
Myopias tasnianiensis, 270
Myopias tenuis, 270
Myrmo.xenus yorciiayini, 335
Naeekoniia, 40
Natural history of the workerless inqui-
line ant, Poyononiyrmex colei, 321
Nest architecture and brood devel-
opment times in Polistes exelanians,
237
Nest-building behavior and devel-
opment of Eumegaehile, 193
New species of the ant genus Myopias,
249
Onhoperus seutellaris, 288
Pareuprepocneniis syriaea, 387
Pheidole honensis, 395
Pogononiynnex, 413
Pogonomyrmex colei, 321
Pogonomyrniex anierieanus, 425
Polistes exelanians, 237
Polydomy in Harpagoxenus anieriean-
us, 151
Polyergus lucidus, 307
Polygynous colony of Vespula, 1 79
Precis, 59
Predatory capture of bombardier bee-
tles, 1 19
Prey selection by Alpaida tuonaho, 123
Progenentomum, 39
‘Protest’ sounds of a grasshopper, 387
Protorthoptera, 1
Reproductive behavior of Claeoderes hi-
vittata, 135
Reproductive plasticity in yellowjacket
wasps, 179
Rhytidoponera metallica, 1 63
Slave-making ants, 425, 451
Social organization in Leptothorax, 361
Structure and relationships of Euhlep-
tus, 81
Studies on North American Carbonifer-
ous insects. 7
Studies on Upper Carboniferous insects,
81
Symmyrmica chamherlini, 297
Tahanus punctifer, 119
Temperature-induced changes in the calls
of Chrysoperla, 343
Testing visual species recognition in
Precis, 59
Trichadenoteenum, biology, 97
Upper Carboniferous insects, 1,81
Vespula macuUfrons, 1 79
Visual species recognition in Precis, 59
442
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.
FOR SALE
Republication of Frederick Valentine Melsheimer’s 1806 “A Catalogue of Insects of
Pennsylvania”, the first separate work devoted to American insects. The facsimile
lists more than 1300 species of Coleoptera (other orders were not completed), and
includes a short biography of Melshcimer. Price: U.S.. $5.00 (overseas, airmail
S6.50). Checks payable to Entomological Society of Pennsylvania, c o Entomology
Dept.. Pennsylvania State University. University Park. PA 16802. U.S.A.
.m.
.m,,
■
V-IJ
' y ■
■ «
rt ■' i
y-i
mt.-
■-y.
t
V
r
I
_ , — o
^ -j z —
RAR I ES^SMITHSONIAN^INSTITUTION NOliniliSNI NVINOSHilWS S3ldVd
r 4 ro ~ DO 9
00
?0
> •’!
i-^4|JK^/ —
inillSNI NVINOSHillAIS S3 I W Vd 8 n~L I B R AR I ES^SMITHSONIAN "INSTITU"
Z CO Z CO z V
~ / ^''1 i
i NSSiocJ^?' I i I N^^/ t
RAR I Es‘”sMITHSONIAN INSTITUTION NOlifUliSNI NVINOSHilWs‘”s3 I a Vil
(O <o ~ CO
iruiiSNi NviNOSHims SBidVdan libraries Smithsonian institu
“ V Z r- Z ^
m 2 m y^'21^ 2 m ><r 0
RARIES SMITHS0N!AN“'lNSTITUT10N‘^N0liniliSNI~NVIN0SHiltMS S3ldVi]
CO Z \ CO Z CO
§ ~ 2
X o Mf X
2 '“' 5> ' 2 5> '
ifUllSNI^NVIN0SHilWS^S3 I d Vd 8 IT^LI B RAR I ES^SMITHSONIAN INSTITU'
CO := CO z;
v''
-J V?-
o i: o x^osv^j
z ■' J z _J 2
RARIES SMITHSONIAN INSTITUTION NOliniliSNI NVINOSHil^S S3ldVi
^ X ^ “ /^^\ i " ° '
3 (. J^ ? If jl ?
//\ fc A.
m
lifUliSNI NVINOSHilWS S3 I a Va 3 n~L I B R AR I Es'^SMITHSONIAN INSTITU
2 to z V ^ z
< s ^ „. s I
I \l U \
NviNUbHJLmb lidkmkic.^ bivmiibum/\iN mdiiiuii
“ CO = CO
R ARIES SMITHSONIAN INSTITUTION NOliniliSNI NVINOSHillNS S3ldVyi
z z r- _ . z
__ m
inillSNI NVINOSHillMS S3 1 HVd 8 n”u B RAR I ES^^SMITHSONIAN INSTITUT
z w z w 2
° (fi 1^) - i -
2 ^ > xiiulisjjx :§
R AR I ES^^SMITHSONIAN INSTITUTION NOliniliSNI NVINOSHillMs‘^S3 I d VB 1
CO 5 CO ^ V ^
^ tJj ^ UJ
CO
O — N^uixs^ O
iniliSNl"^NVIN0SHilHS^S3 I bVa a n"’u B RAR I es^smithsonian“’institut
r~ V z f“ Z «”
m <' ^ rn Xi^osvi^ ^ m x^osv^
CO — CO _ CO
RARIES SMITHSONIAN INSTITUTION NOliniliSNI NVlNOSHillNS S3ldVd
^ — z r ^ ^
iniliSNl NVIN0SHilWs"’s3 I avaa H^LI B RAR I ES^SMITHSONIAN^INSTITUT
“ CO ~ CO “
tr\ aJ>*v C*J
, , ^ 1/
C>y ~ rrt Z]
Z ' j J z
RARIES SMITHSONIAN INSTITUTION NOliniliSNI NVINOSHillMS S3ldVd
d ^ ^ ^ .V 21
O . ™ O ><^v)v?Vv _ V/ O
= ^ = (sf
• § ^
CO
CO
m w m
iniliSNI NVINOSHillNS S3 I d Vd 0 n“”L I B R AR I ES SMITHSONIAN INSTITUT
Z CO z ,...• CO z
J D A D I CC QMITUQHMIAM
I
CO
O
z
CO * Z CO
MQTiTiiTinN MniiniiiQMi Mvmn?:w I ILMQ <^’^iMVd