A REVISION OF THE GENUS Paratrechina
(HYMENOPTERA: FORMICIDAE) OF THE CONTINENTAL UNITED STATES
JAMES C. TRAGER
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
According to the International Code of Zoological
Nomenclature (Art. 9), dissertations submitted in partial
fulfillment of requirements for doctoral, master's or other
degrees do not constitute publications in the nomenclatural
sense, even when reproduced on microcards, by microfilm
or other similar methods. Therefore, the novel names or
name-combinations proposed in this dissertation are not
entered into the public domain. The information given in
this work is in effect confidential and cannot ethically be
utilized by anyone in any publication (Art. 8 of Code) until
the author of the dissertation has published it in a legit-
imate outlet, unless the author or the author's adviser
gives permission for someone else to do so, or unless after
a reasonable interval neither can be consulted even with
concerted effort (in conformance with the Code of Ethics,
Appendix A of the International Code of Zoological Nomen-
Protection under the articles and appendices of the
Code is hereby claimed for this dissertation.
I dedicate this study to the memory of Dr. William
F. Buren, who first suggested the subject and gave me the
initial nudging necessary to get it underway. It is my
hope that this will constitute a suitable tribute to his
contributions to ant taxonomy.
As in all such studies, this work could never have
been brought to its present state without the aid of many
people. Foremost among these is my wife, Kim, whose unfail-
ing love and devotion and assistance with the minutiae of
curation and preparation of the manuscript have been immense-
ly important and are thoroughly appreciated. I give heart-
felt thanks to the late Dr. William F. Buren and to Dr.
Jerry L. Stimac for their many discussions with me on this
study and related matters, for their friendship and en-
couragement, and not the least of all for arranging finan-
cial support during much of the course of the work. I
thank Dr. Daniel P. Wojcik and Mr. Ruediger W. Klein for
making extra efforts to collect Paratrechina for me. I
wish to thank Dr. William L. Brown, Jr., and Mr. Roy R.
Selling for their indirect guidance through the example
of their work in any taxonomy and for commenting on early
drafts of the work. For typing the earlier versions, I
thank Anne Keene and Renee Lofgren. For assistance with
the computer program and data input, I thank Dr. Robert J.
O'Neil, Lois Wood and Joan Denicola. Finally, I thank all
the institutions and persons listed in Chapter III for their
cooperation in lending or donating specimens, including
valuable types, without which the study could never have
been carried out.
TABLE OF CONTENTS
ACKNOWLEDGMENTS ......................................... iv
LIST OF TABLES ........................................viii
LIST OF FIGURES ......................................... ix
ABSTRACT ............................................. xii
I INTRODUCTION ..................................... 1
II NOMENCLATURAL HISTORY AND AFFINITIES .......... 4
III METHODS, MEASUREMENTS AND ABBREVIATIONS ........ 8
IV SYNONYMY OF THE SUBGENERA OF Paratrechina ...... 13
V GENERIC DESCRIPTION .............................. 17
Worker ......................................... 17
Queen .......................................... 22
Male .......................................... 23
VI IDENTIFICATION OF Paratrechina OF THE UNITED
STATES ......................................... 24
Preface to Keys ................................ 24
Key to Workers ................................. 28
Key to Males ................................... 35
VII FORMAT OF SPECIES DESCRIPTIONS ................ 41
VIII VIVIDULA COMPLEX ................................. 46
Diagnosis of Complex ............................ 46
Paratrechina vividula ....................... 46
Paratrechina terricola ...................... 56
IX PARVULA COMPLEX ................................... 74
Diagnosis of Complex ............................ 74
Paratrechina concinna, New Species ........... 74
Paratrechina faisonensis .. .................. 82
Paratrechina flavipes ....................... 92
Paratrechina parvula ......................... 103
Paratrechina wojciki, New Species ............113
Paratrechina austroccidua, New Species ....... 122
X ARENIVAGA COMPLEX ............................. 134
Diagnosis of Complex ......................... 134
Paratrechina arenivaga ...... ................... 134
Paratrechina phantasma, New Species .......... 146
XI BRUESII COMPLEX ............................... 156
Diagnosis of Complex ........ ................. 156
Paratrechina bruesii ......................... 156
XII HYSTRIX COMPLEX ............................... 166
Diagnosis of Complex ......................... 166
Paratrechina hystrix, New Species ........... 166
XIII GUATEMALENSIS COMPLEX ...... ................... 178
Diagnosis of Complex ......................... 178
Paratrechina guatemalensis .................. 178
XIV FULVA COMPLEX ................................. 187
Diagnosis of Complex ......................... 187
Paratrechina pubens ........ ................. 187
XV BOURBONICA COMPLEX ............................. 194
Diagnosis of Complex ......................... 194
Paratrechina bourbonica ..................... 194
XVI LONGICORNIS COMPLEX ........................... 206
Diagnosis of Complex ......................... 206
Paratrechina longicornis ....................... 206
BIBLIOGRAPHY ................ .......................... 231
A MEASUREMENTS OF Paratrechina SPECIES TREATED
IN THIS REVISION .............................. 237
B SUMMARY STATISTICS OF MEASUREMENTS LISTED IN
APPENDIX A ....................................... 253
BIOGRAPHICAL SKETCH ................................... 263
LIST OF TABLES
1 Reference collections of Paratrechina,
abbreviations for the collections used in
the text, and individuals responsible for
the loan or gift of specimens ................ 8
2 Synopsis of taxonomy and authorship of
U.S. Paratrechina ............................. 27
3 Selected characteristics of members of the
Guatemalensis Complex of Paratrechina ........ 185
LIST OF FIGURES
1 Cephalic microsculpture of longicornis
worker ........................................ 14
2 Cephalic macrochaeta of faisonensis male ..... 18
3 Abdominal macrochaeta and sculpture of
faisonensis male .............................. 21
4 Head of vividula worker, dorsal view ......... 71
5 Head of terricola worker, dorsal view ........ 71
6 Head of terricola male, dorsal view .......... 71
7 Male terricola, lateral view .............. ..... 73
8 Male vividula, lateral view .................. 73
9 Worker of terricola, lateral view ............ 73
10 Worker of vividula, lateral view ............. 73
11 Worker of concinna, lateral view ............. 102
12 Worker of faisonensis, lateral view .......... 102
13 Worker of flavipes, lateral view ............. 102
14 Head of concinna worker, dorsal view ......... 102
15 Head of faisonensis worker, dorsal view ...... 102
16 Head of flavipes worker, dorsal view ......... 102
17 Worker of austroccidua, lateral view ......... 133
18 Worker of wojciki, lateral view .............. 133
19 Worker of parvula, lateral view .............. 133
20 Head of austroccidua worker, dorsal view ..... 133
21 Head of wojciki worker, dorsal view .......... 133
22 Head of parvula worker, dorsal view .......... 133
23 Worker of arenivaga, lateral view ............ 155
24 Worker of phantasma, lateral view ............ 155
25 Head of arenivaga worker, dorsal view ........ 155
26 Head of phantasma worker, dorsal view ........ 155
27 Worker of bruesii, lateral view .............. 175
28 Worker of hystrix, lateral view .............. 175
29 Head of bruesii worker, dorsal view .......... 175
30 Head of hystrix worker, dorsal view .......... 175
31 Worker of bourbonica, lateral view ........... 216
32 Worker of longicornis, lateral view .......... 216
33 Worker of pubens, lateral view ............... 216
34 Worker of guatemalensis, lateral view ........ 216
35 Head of guatemalensis worker, dorsal view .... 218
36 Head of pubens worker, dorsal view ........... 218
37 Head of bourbonica worker, dorsal view ....... 218
38 Head of longicornis worker, dorsal view ...... 218
39 Male of bruesii, lateral view ................ 177
40 Male of hystrix, lateral view ................ 177
41 Male of concinna, lateral view ............... 220
42 Male of longicornis, lateral view ............ 220
43 Male of bourbonica, lateral view ............. 220
44 Male of pubens, lateral view ................ 220
45 Genitalia of faisonensis male, dorsal view,
slide mounted ......... ......................... 222
46 Genitalia of guatemalensis male, dorsal view,
slide mounted .................................. 222
47 Genitalia of arenivaga male, dorsal view,
slide mounted .................................. 222
48 Genitalia of phantasma male, dorsal view,
slide mounted ........................ .......... 222
49 Distribution in North America of vividula
( ), terricola ( o ), guatemalensis (A ) and
pubens ( ) .................................. 224
50 Distribution in North America of concinna
(o ), faisonensis ( ), flavipes ( ), hystrix
(A ) and bruesii ( ) ......................... 226
51 Distribution in North America of parvula (e ),
wojciki (o ) and austroccidua (A ) ........... 228
52 Distribution in North America of arenivaga
(* ), phantasma ( o ) and bourbonica ( n ) ..... 230
Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
A REVISION OF THE GENUS Paratrechina
(HYMENOPTERA: FORMICIDAE) OF THE CONTINENTAL UNITED STATES
JAMES C. TRAGER
Chairman: Jerry L. Stimac
Major Department: Department of Entomology and Nematology
The taxonomy of the ant genus Paratrechina in the con-
tenental United States is revised. Identity of previously
published taxa is determined, and the workers and males are
redescribed. The workers and males of the new species P.
concinna, P. phantasma and P. wojciki from the Southeast and
P. austroccidua and P. hystrix from the Southwest are de-
scribed. P. fulva pubens, P. melanderi arenivaga and P.
arenivaga faisonensis are raised to full species. P. par-
vula grandula is transferred to Conomyrma. The establishment
of P. flavipes of temperate eastern Asia, in Pennsylvania
and New York, and of P. guatemalensis of the Caribbean
region, in southern Florida, is reported for the first time.
The synonomy of P. kincaidi with P. vividula and the sibling
species relationship of vividula to P. terricola are
discussed as are the reasons for resurrecting Buckley's name
terricola for the species described by Wheeler as melanderi.
Biological and distributional data for each species are also
presented. Keys are given for the workers and males of all
Paratrechina is a common though often inconspicuous
element of the ant faunas of almost all continental areas
habitable by ants. Native species are lacking or nearly so
from Europe (Brown, 1973) and colder temperate, Mediterra-
nean climate and desert regions of other continents. The
genus is most diverse in tropical Asia and Australia, which
together have the greatest number of species groups, includ-
ing one-third of the described forms (Emery, 1925). Based
on my acquaintance with the New World forms, I estimate that
perhaps three-fourths of the true number of species has been
described. For the Old World tropics, the proportion may
be much smaller.
A number of species have been transported by shipping
activity to localities well beyond their places of origin.
Some of these species have attained at least minor pest
status through their invasion of nurseries, greenhouses,
laboratories, insectaries, homes, and other buildings (Smith,
1965). Foremost among such species in the New World are
longicornis, vividula, bourbonica, and members of the fulva
group. Other species have been transported by man and become
established in new localities, but the taxonomy of the genus
is so poorly known at present that it would be impossible
to state with certainty which species are involved. Wilson
and Taylor (1967) discussed the taxonomic difficulties in
Paratrechina with special reference to Polynesia.
This study is the first attempt to monograph Paratre-
china and constitutes Part I of a worldwide revision. I
expect that this will lend a degree of clarity to a group
for which the taxonomy "looks very much like a hopeless
muddle" (Creighton, 1950). Apart from the purely academic
interest in unravelling the taxonomic tangle of this diffi-
cult genus, it is hoped that the keys and descriptions will
provide nursery and quarantine inspectors and other applied
entomologists a means to readily identify the species of
this frequently encountered genus.
In the next chapter, I review the nomenclatural history
of Paratrechina and discuss what is known about its rela-
tionships to other ant genera. In the third chapter, I
describe the methods and equipment used in examining the
specimens studied and give a complete listing of abbrevia-
tions used in the text. In Chapter IV, I tentatively synony-
mize the subgenera of Paratrechina and give my reasons for
doing so. In Chapter V, I describe the workers, males, and
queens of Paratrechina at the generic level. Chapter VI
contains a list of the U.S. species of Paratrechina and
those introduced to the region, as well as their authors and
known synonyms. Also in Chapter VI are keys to workers and
males of these species. Chapter VII establishes the format
of the treatment of each species and includes my reasons for
the use of the term "complex" for groups of related species.
The remainder of the chapters contain the descriptions of
the species of each complex. A brief diagnosis of the com-
plex treated is at the beginning of each of these chapters.
NOMENCLATURAL HISTORY AND AFFINITIES
The name Paratrechina was first used in the combina-
tion P. currens Motschulsky (1863). The species name was
not used for over sixty years, until Emery (1925) published
his treatise on the formicine genera. In the interim, the
great majority of species known today were described and
were assigned to Prenolepis Mayr (1862), as was Formica
longicornis Latreille (1802), the senior synonym of
In 1906, Emery published a note, unusual for its time
in that it truly made clear the identity of vividula and
pointed out its probable place of origin and close rela-
tionship to terricola melanderii), shown below to be its
sibling species. More germane to the immediate discussion
is that in his note, Emery erected the subgenera Prenoleois
s.s., Euprenolepis and Nylanderia. The first two of these
are today considered distinct genera. Euprenolepis is not
in the scope of this paper. Prenolepis and Nylanderia are
Prenolepis s.s. was distinguished by Emery in that all
castes lack erect pilosity on the scapes and tibiae and
instead possess long, dense, oblique (i.e., subdecumbent)
pubescence on these segments. In workers, the mesothorax
is extremely constricted, such that it is less than half as
broad as the pronotum in dorsal view. The pronotum and pro-
podeum are broad and globular, yielding in Prenolepis workers
a "dumbell-shaped" mesosoma. Later, Emery (1925) separated
Prenolepis s.s. as a full genus and added to its distin-
guishing features the placement of the eyes posterior to the
middle of the head and the possession of cerci by the males.
In North America, at least, the short scape of the males of
Prenolepis (about equal to or less than length of head) and
the size of the females (10 or more mm vs. 5 mm or less)
further distinguish it from our species of Paratrechina.
Nylanderia was construed by Emery in 1906 to contain
all of what are now called Paratrechina. In 1925, Emery
established the genus Paratrechina with Paratrechina and
Nylanderia as subgenera, containing longicornis and the
remainder of the species, respectively. Wheeler (1936)
elevated Nylanderia to generic level. Though Donisthorpe
(1943) and Kempf (1972) followed this arrangement in major
taxonomic publications, most other authors have used Emery's
(1925) arrangement. Donisthorpe (1947) proposed a third
subgenus, Paraparatrechina, to include his New Guinean
species pallida. Brown (1973) and Snelling (1981) listed
Nylanderia as a synonym of Paratrechina without discussion.
The validity of the subgenera is discussed in Chapter V.
Paratrechina is said to be most closely related to
Prenolepis, Pseudolasius, Lasius and Formica (Wilson,
1975)--to the last two certainly more distantly than to
the first. Unfortunately, our knowledge of Pseudolasius is
insufficient to state the degree of relation to Paratre-
china. Future investigation should be directed toward
reconstruction of the phylogeny of these apparently related
genera and the rest of the Formicinae.
The relationship of Paratrechina and Prenolepis is
clear. Together they form a monophyletic group. The larvae
of these genera share a plump, dolichoderoid appearance,
indistinct segmentation, lack of a "neck," and relatively
few and short head hairs (G.C. and J. Wheeler, 1953, 1968).
These last authors place the two genera in Forel's tribe
Prenolepidini. (The larvae of Brachymyrmex appear related
to the Prenolepidini, but other characters suggest differ-
ently and the larval similarities may be due to conver-
gence.) Both prenolepidine genera have nude pupae, an
unusual trait in Formicinae. In addition, the workers show
similar reduced sculpture and restricted patterns of
thoracic pilosity (and often of pubescence). The single-
toothed male mandibles and 5- or 6-toothed female mandibles
of the two are virtually indistinguishable, as is the
structure of their proventriculi (Eisner, 1957). Workers
of both groups have a reduced mesothorax, a trait much
further developed in Prenolepis. Finally, both groups con-
tain, with well-known exceptions, mostly smallish, soil or
litter-dwelling inhabitants of more or less undisturbed
mesic areas. All known species in this pair of genera can
readily be placed in one genus or the other.
METHODS, MEASUREMENTS AND ABBREVIATIONS
About one-third of the specimens studied were collected
by myself in Florida, primarily in the vicinity of Gaines-
ville and at Archbold Biological Station in Highlands County.
The remainder of the specimens were obtained by loan or gift
from the collections listed below, thanks to the cooperation
of the individuals listed beside each collection (Table 1).
Table 1. Reference collections of Paratrechina, abbrevia-
tions for the collections used in the text, and
individuals responsible for the loan or gift of
American Museum of Natural
British Museum (Natural
California Academy of Science
Frost Entomological Museum
Florida State Collection of
Helsingfors Zoological Museum
& J. Wheeler
Natural History Museum of
Los Angeles County
Harvard Museum of Comparative
Museum d'Histoire Naturelle,
North Carolina State
State Biological Survey of
Kansas and Private Collection
Texas Tech University
(Collection funded by Texas
Department of Agriculture)
United States National Museum
University of Arkansas
USDA Fire Ant Project
Collection and Private
Point-mounted specimens were measured at 50X with an
ocular micrometer. A computer program was used to convert
the measurements from micrometer units to the nearest 0.01 mm
equivalent, to calculate indices of relationship between the
length of certain body parts and to calculate summary sta-
tistics. Only summarized, rounded-off, data are presented
in the text, but the full raw data set has been reproduced
in Appendix A. All measurements are reported in mm.
Measurements are counts taken were as follows:
TL--HL + WL + GL (see below)
HL--Head length in full face view from a line perpen-
dicular to sagittal axis and tangent to the pos-
teriormost parts of the rear border to such a line
tangent to the anterior most parts of the clypeal
HW--Maximum width of head (including eyes) in full face
EL--Maximum diameter of compound eye
SL--Length of scape (chord if curved) from base (exclu-
sive of basal radicle) to its terminus
PW--Maximum width of pronotum
WL--Weber's length of thorax, from anterior edge of
pronotum (exclusive of anterior flange) to posterior
corner of metapleuron
MCL--Length (chord if curved) of largest pronotal macro-
FL--Length of fore femur
GL--Length of gaster in dorsal view (exclusive of cone
of acidopore in worker or of genitalia in males)
SM--Number of standing barbulate macrochaetae on scape
PM--Number of barbulate macrochaetae on pronotum to
one side of sagittal plane
MM--Number of barbulate macrochaetae on mesonotum to
one side of sagittal plane
CI = x 100
01 = E- x 100
SI = |- x 100
FI = x 100
Illustrations were prepared with the aid of a Nikon
SMZ10 stereo dissecting microscope with drawing tube attach-
ment. All are diagrammatic and are modelled on more than
one specimen; proportions and macrochaetal counts were drawn
at or near their mean values for the species. The pubescence
patterns indicated on the head drawings are actual setal maps
of the middorsal portion of the head of a specimen chosen
for its "typical" pubescence.
Slide preparations of genitalia prepared by Dr. Arnold
Van Pelt, loaned by USNM, were consulted in preparation of
the drawings of the genitalia of faisonensis and arenivaga.
The drawings of the genitalia of guatemalensis and phantasma
are based on my own dissections but were positioned as
though based on slide mounts.
SYNONYMY OF THE SUBGENERA OF Paratrechina
As has been shown by Wilson (1955) and Snelling (1976),
subgenera may be useful taxonomic tools when they represent
clearcut phyletic trends comprising major monophyletic group-
ings of species within a genus. It seems clear to me that
the nominate subgenera of Paratrechina do not live up to
this criterion. Indeed, some of the species groups described
in 1925 by Emery such as his caledonica group, and microps
and its relatives in the Caribbean region come closer to
meeting it, but it would be premature to describe subgenera
for such groups without a better knowledge of the world fauna.
I have studied workers of longicornis, pallida and
cisipa to determine their relationship and to determine the
validity of the subgenera. Longicornis is an unusual mem-
ber of the genus in many respects, including its large eyes,
great elongation of the scapes and tibiae, reduction of pil-
osity, unusual microsculpture (see Fig. 1 and description)
and unique biology. Pallida appears related to longicornis.
Pallida, too, has 5-tooth mandibles, long scapes and tibiae
and even longer maxillary palps, further reduced pilosity
(except that there are two pairs of macrochaetae on the
propodeum not found on longicornis), fine very short
Fig. 1. Cephalic microsculpture of
pubescence, and slight bluish reflections. The thorax of
the two species is of virtually identical conformation.
The single type specimen of pallida available to me could
not be subjected to study in the SEM to determine micro-
sculptural details, but the macroscopic appearance of the
integument and the thick, blunt appearance of the macro-
chaetae are similar to those of longicornis.
Smith and Lavigne (1973) described cisipa from Puerto
Rico, noting that its elongated, flattened thorax and long
legs and scapes resembled those of longicornis but that its
6-toothed mandibles and bristly scapes forced them to place
cisipa in Nylanderia. In fact, cisipa, while not related
to longicornis, is not close to any other member of Paratre-
china either. Some cisipa workers I have seen (RJL, RWK),
including one in the type colony, have 5-toothed mandibles.
The bristles on the scapes and tibiae are not the usual
barbulate macrochaetae but are in fact thickened, oblique
pubescence very much like that of Prenolepis. Interestingly,
the eyes of cisipa are placed further back on the head than
usual for a Paratrechina (but not as far back as in Prenol-
epis), and the mesothorax is quite constricted. On the other
hand, the lack of cerci on the male and the pattern of
thoracic pilosity on the worker of cisipa are typical of
Paratrechina. The microsculpture and nature of the barbula-
tion of the macrochaetae are as found in "Nylanderia," but
this species cannot otherwise be to closely linked to that
grouping of species.
Thus, longicornis and pallida appear to form a real
species-group but hardly constitute a major phyletic trend.
Nylanderia is a paraphyletic group including all the re-
maining diversity within the genus, including some species-
groups at least as removed phenetically from the mainstream
of the genus as are longicornis and pallida. I, therefore,
consider the subgenera Paraparatrechina and Nylanderia
synonymous with Paratrechina and recommend that the names
no longer be used. The only feasible alternative is a more
restricted Nylanderia and the erection of several new sub-
genera, but this must await further study.
The males and workers of Paratrechina have been
characterized by M.R. Smith (1943, 1947, 1965). Some
important points from his diagnoses are included below.
In addition, characters not discussed by him and a de-
scription of the female are included.
Small, 1.5-3.5 mm (4 mm in one Australian form). An-
tenna 12-segmented, arising from a fossa placed close to but
not confluent with rear border of clypeus; scape longer than
head, usually with two to four rows of subdecmebent to erect
barbulate macrochaetae (as in Fig. 2) (lacking in Darvula,
opaca, longicornis, etc.); funiculus filiform to very
weakly clavate, segments longer than broad, terminal one
almost twice as long as the others. Eye typically well-
developed, about one/fifth or more as long as head (much
smaller in subterranean microps group of Caribbean region,
caledonica, etc.). Ocelli usually indistinct, or absent;
often only median ocellus is visible. Mandible striate 6-
or 5-toothed; typical dentition as in vividula (Fig. 4)
(Smith, 1939, was in error in stating that myops had four
Fig. 2. Cephalic macrochaeta of faisonensis
mandibular teeth.) Palpal formula 6, 4; palpal segments most
often subequal in length. Clypeus bears 6-20 slender, barbed
macrochaetae of varying lengths and a smoothly rounded median
longitudinal angle. Clypeus emarginate or evenly rounded
anteriorly; in side view with a declinate anterior face.
Frontal carinae very short, extending little beyond antennal
fossae, thus, the frontal area often indicated only by its
possession of four longitudinal rows of regularly spaced
thick, barbulate macrochaetae (the inner two, often, and the
outer rows, occasionally, reduced to a pair or so of
bristles); the rows extending back through the ocellar area
to the rear border. Sides of head in full face view nearly
parallel, or more often, rounded and weakly convergent
anteriad; rarely broader before than behind the eyes. Sides
of head with macrochaetae shorter than those on preocellar
area, subdecumbent and often greatly reduced in number.
Posterior border straight or rounded, often weakly emarginate
medially, with somewhat longer, more curved, erect pilosity
than on sides. Venter of head with fine flexuous pilosity
similar to that on clypeus, or none.
Pronotum convex or subangular. Mesonotum usually nearly
flat in profile, the latter with a very short anterior and
posterior face in many species such that the flat dorsal
surface is slightly raised above pronotum, at least anterior-
ly. Mesonotum somewhat reduced, about 1/2 to 3/5 as broad
as pronotum seen from above. Metanotum a narrow band usually
in a distinct furrow, bordered laterally by mesothoracic
spiracles which are nearly dorsal in position and orienta-
tion of their openings. Pronotum and mesonotum each with
two pairs of large macrochaetae like those on head (Fig. 2)
and usually one or more (rarely up to 16) pairs of smaller
macrochaetae on pronotum and usually one or 2 (up to 8)
such ancillary pairs on mesonotum.
Propodeum weakly angular with a short, low, dorsal face,
or convex and of varying height relative to thoracic dorsum.
Propodeum without pilosity except in caledonica group, and in
glabra, dugasi (Emery, 1925), microps, hystrix and pallida.
Some caledonica-group species may have only one pair of
macrochaetae on mesonotum. Petiole cuneate with crest vary-
ing from sharp to truncate or rounded in profile; in dorsal
view, crest flat, angular or convex, only rarely notched in
workers (e.g., sakurae). Petiole glabrous or with one or a
few short erect macrochaetae on the crest.
Gaster ovate, acuminate terminally, relatively volumi-
nous, with more or less abundant flexuous pilosity with
less barbs per macrochaetae (Fig. 3) than on those of thorax.
Anterior face of gastral tergite 1 impressed and overhang-
ing petiole, concealing it partially from dorsal view. Bor-
der of impression distinctly angular. Acidopore conspicuous
and conical, with a distinct corona.
Pubescence typically dense on the head and appendages,
shorter and more dilute or lacking on much of thorax and
gaster, the patterns of vestiture forming important diag-
nostic characters of some species or species-groups.
Fig. 3. Abdominal macrochaeta and sculpture
of faisonensis male.
Sculpture usually limited to shallow pubigerous
foveolation on head, delicate shagreening on gaster, and
honeycomb-like reticulation on terminal segments. Some
species in caledonica group are heavily punctate, opaque
forms which contrast strongly with the typical shining
appearance (when not obscured by pubescence) of other
In general, fitting the worker description but dif-
fering as follows. TL 4-7 mm, scape relatively shorter and
with less pilosity than in conspecific workers. Eye a
little larger to much larger in subterranean forms, 1/4 or
more HL. Ocelli fully developed.
Thorax of normal alate formicine type. Pronotum and
other dorsal surfaces invested with barbulate macrochaetae
like those found on workers but not in so regular an array.
Propodeum with relatively shorter dorsal face than in
worker. Petiole typically broader, sharper in profile, and
often with at least a weak median notch.
Gaster much more massive and more thoroughly hiding
petiole in its impressed anterior face.
Pubescence longer, thicker and more dense, usually
covering entire body, except petiole and rear face of
With the important generic characters noted for
females, i.e., barbulate pilosity; the impressed first
gastric tergite and antennal scapes longer than head. In
overall size, approximating that of conspecific worker.
Mandibles well-developed, though smaller and less
heavily sclerotized than in workers; with a single apical
tooth and occasionally a small adjacent cleft, and one or
more inconspicuous subapical denticles. Antennal scapes
with less macrochaetae than conspecific workers or queens.
Antennae 13-segmented, with conformation of segments
generally resembling that of female castes.
Petiole of same conformation as in workers but broader
and more blunt, sometimes notched like that of queen.
External genitalia prominent, varying greatly between
species-groups and interspecifically and, thus, useful in
taxonomy. Cerci (as in Prenolepis) lacking.
Pubescence distributed as in queens, but in length and
density more like that of workers. Thoracic pilosity
usually limited to mesometanotum, rarely more generally
distributed; typically sparse and decumbent on gastral
IDENTIFICATION OF Paratrechina OF THE UNITED STATES
Preface to Keys
In this chapter, I provide what will undoubtedly be the
most frequently read portion of this study, namely the keys
for identification of the workers and males of Paratrechina
in the United States. Also included is a taxonomic synposis
of the species included in the keys and formally treated in
the later chapters (see Table 2).
The key contains a number of species which occur out-
side of the United States but in many cases cannot be used
to identify them outside of this country. There are many
Paratrechina species not treated in this revision, which
makes the use of these keys futile outside of the geographic
region for which they are designed.
The characters which separate Paratrechina species are
more subtle than those used to separate species in some
other insect groups. One will not find here the simple,
one-feature alternatives of older keys to Paratrechina. My
keys assume the user has possession of a stereo dissecting
microscope fitted with a calibrated ocular micrometer and
of a calculator to expedite the calculation of indices.
The reader is urged to read carefully the definitions of
the measurements in Chapter III and the following remarks
on possible difficulties one may encounter with certain
measurements and counts before attempting to use the keys.
Remarks. Certain difficulties were encountered in
taking some of the measurements, and the following comments
may help the user of the keys in selecting the appropriate
alternative at a given couplet (and in reading the descrip-
The posteroventral portion of the head is excised and,
thus, in certain positions obscures the front of the pronotum
making it difficult or impossible to measure WL from the
side. In such cases, WL was taken from a dorsal view. As
a check, the measurement was taken on a number of specimens
in which WL was visible from both views. The two measure-
ments never differed by more than 0.01 mm and usually by
no more than 0.005 mm.
Macrochaeta counts proved to have certain unforeseen pit-
falls. Only barbulate setae of the sort shown in Figs. 2 and
3 were counted. These were usually discerned by their great-
er length and diameter and their different color (usually
darker than pubescence and body color), but in some species
the macrochaetae are nearly the color of the substrate and are
rather short. Compounding the problem in such cases is that
there may occur simple setae of nearly the same macroscopic
appearance. Through observation at high magnification and
checking back at lower power, I learned to discriminate
these setal types. The discrimination turned out to have
a slightly different "Gestalt" for each of the "difficult"
species. In addition, the macrochaetae are all visible only
at certain angles of viewing and illumination or where SM is
large, only by rotating the scape. Though this particular
count is undoubtedly the one for which the most errors were
made, I feel confident that the high end of the ranges re-
ported are close to correct. The count recorded was the
highest repeatable one from among either of the two scapes
resulting in an inherent bias against low counts, whether
natural or due to damage. The scapes of a single specimen
often differed in SM, even on fresh, undamaged specimens.
PM and MM also proved to be asymmetrical in many spec-
imens. Always present (or at least indicated by a socket)
were the 8 largest hairs (the "major thoracic macrochaetae"),
but the number of ancillaries might be different on each
side. The highest repeatable count of hairs and/or unmis-
takable hair-sockets was the one recorded. On occasion,
smaller barbulate setae resembled pubescence, but I learned
to discriminate these as above. The discrimination was
much easier on pale species with dark macrochaetae, and
most counting errors were certainly made on dark species.
I refer loosely to the middle tagma of ants as the
thorax which is the current usage by many hymenopterists
for the homologous body region of ants and other Hymenop-
tera. In practice, there is little need to refer specifi-
cally to the thorax and propodeum collectively, for which
more accurate terms would be alitrunk or mesosoma.
Table 2. Synopsis of taxonomy and authorship of
--kincaidi (Wheeler) syn. n.
terricola (Buckley) stat. rev.
--melanderi (Wheeler) syn. n.
concinna sp. n.
flavipes (F. Smith)
wojciki sp. n.
austroccidua sp. n.
phantasma sp. n.
hystrix sp. n.
pubens (Forel) stat. n.
stat. n. & rev.
Excluded from Paratrechina
parvula grandula (Forel)
--becomes Conomvrma grandula (Forel) comb. n.
Pilosity refers to the large, standing, barbulate setae
(macrochaetae) characteristic of this group (though by no
means exclusive to it). The unbarbed macrochaetae on the
parameres of males are also referred to as pilosity. Simple,
appressed to subdecumbent setae of small diameter are referred
to as Dubescence. Otherwise, terminology used here is that
which has become standard in recent revisions of formicine
genera (Wilson, 1955; Wing, 1968; Bolton, 1973; Snelling,
Key to Workers
la. Species of eastern U.S. and Great
Plains, including pains and woodlands
of east Texas ....................................... 2
lb. Species of Southwest including west
Texas mountains and deserts ....................... 14
2a. Scapes and legs unusually long (SI>165,
FI>115); weakly shining black or gray
with bluish reflections; with sparse,
short, barely visible pubescence; near
human dwellings, Gulf Coast states,
urban areas, sporadic elsewhere .......... longicornis
2b. Scapes and legs of usual proportions
(SI<130, FI<105); colored variously
but never with bluish reflections;
shiny, or if dull, this due to dense
pubescence; habitat various ........................ 3
3a. Yellow to pale whitish; nests in sandy
soil or dunes with entrances surrounded
by conspicuous crater of subsoil in
clearings between vegetation ...................... 14
3b. Uniformly dark-colored or bicolored;
nests inconspicuous; in more mesic
microhabitats under moss, rocks, logs,
in litter; or near man-made structures
(parvula may have crater nests in sandy
areas but is never uniformly yellow) ............... 5
4a. Yellow with gaster infuscated posteri-
orly; thoracic pilosity flexuous and
dark brown (notably darker than body
color); scapes with 5-17 (usually 7-12)
macrochaetae and suberect pubescence;
New Jersey, Florida, Gulf Coast, sand-
hills of Midwest .......................
4b. Yellow or whitish with gaster, at most,
only slightly darker, thoracic pilosity
nearly straight, nearly the same color
or only slightly darker than body color;
scapes with 0-4 (usually 1-3) standing
macrochaetae and short, appressed
pubescence (Florida scrub and dunes) ...
5a. Scapes with not more than 4 standing
5b. Scapes with at least 4 (usually 7 or
more) standing macrochaetae ............
......... ... 7
6a. Typically bicolored, with thorax
yellowish to reddish-brown, head and
gaster darker; middle and hind coxae
pale, much lighter than fore coxae or
rest of legs; scapes with 1-4 suberect
macrochaetae; in full face view eyes
reaching sides of head or failing to do
so by 1 or 2 facet-widths; small (HL
usually 0.51-0.57); (Florida) ................ wojciki
6b. Body typically uniform dark brown (at
most weakly bicolored); appendages
somewhat lighter or even yellowish;
scapes never with standing macrochaetae;
in full face view outer edge of eyes
failing to reach sides of head by 1/4
of eye width; larger (HL>0.57, usually
>0.60) (eastern U.S.) ........................ parvula
7a. Thorax and gaster mostly covered with
pubescence or lacking it only on pleura
and disk of pronotum; dull ......................... 8
7b. Thorax and gaster with greatly reduced
pubescence; shiny ................................. 10
8a. Body light reddish-brown with slender
flexuous light brown pilosity (Fig. 33) ....... pubens
8b. Body dull brown to nearly black or if
faded, pilosity shorter, stouter,
straighter, and dark (Figs. 31 and 34) ............. 9
9a. Uniform dark brown or black, thoracic
macrochaetae thick, often nearly
straight, relatively short and abundant
(PM 5-15); entire body pubescent; large
(HL usually 0.70-0.81); Florida, dis-
turbed habitats, mangrove islands,
occasional along Gulf Coast, in green-
houses, etc., elsewhere .................. bourbonica
9b. Yellowish-brown to dark brown with
middle and hind coxae distinctly
lighter; thoracic macrochaetae more
slender, weakly flexuous and tapering,
longer (at least relatively) and less
abundant (PM 3-7); disk and sometimes
sides of pronotum and pleura glabrous
and shining; smaller (HL usually
0.56-0.66); Homestead, Florida ......... guatemalensis
10a. Uniform dark brown with appendages
somewhat lighter; propodeum with a
dense row of longitudinally aligned
pubescence along anterior edge;
mesonotum and front of pronotum with
at least some dilute pubescence; head
with shallow pubigerous punctae and
dense pubescence which is mostly aligned
with long axis of head; marshes, ditches,
damp pastures, swamp edges; in rotten
wood, cow dung or tussocks; Florida,
Alabama, Georgia, Carolinas ................ concinna
10b. Middle and hind coxae and/or thorax and
legs lighter than gaster and head;
pubescence very sparse or absent from
promesonotum and often from propodeum;
head smooth and shining or irregularly
and weakly punctate beneath pubes-
cence; cephalic pubescence may be very
dilute with some setae markedly not
parallel to long axis of head; not
typically inhabiting marshes or
poorly-drained pastures ........................... 11
lla. Cephalic pubescence dilute; preoccip-
ital area with most spaces between
setae as wide as length of setae or
wider, anterior 1/2 of head (except,
perhaps, frons) lacking pubescence ............... 12
llb. Cephalic pubescence denser; preoccip-
ital area with most spaces between
setae no wider than length of setae,
usually less ................... .................. 13
12a. Eye about 1/4 HL or slightly larger
(OI usually 24-27) inhabiting old fields,
cultivated areas, gardens, disturbed
habitats across southern half of U.S.
or greenhouses, etc., further north;
also in native scrub vegetation in
southern Texas, New Mexico, northern
Mexico ..................................... vividula
12b. Eye smaller (OI 20-24), under stones,
moss clumps or bark in forest openings
and other open or disturbed habitats
in Texas and Plains states, and at
lower elevations in Southwest mountains,
Arkansas and Tennessee ..................... terricola
NOTE: The workers of terricola and
vividula cannot always be
reliably separated on morpho-
logical grounds. See male key
for separatory characters. See
also descriptions of these two
13a. Thorax, legs, antennae yellow; head
averaging broader (CI usually 85-90)
and with rounded sides (Fig. 16);
wooded areas, Pittsburg, Philadelphia,
Long Island ................ ................. flavipes
13b. Uniform brown or with slightly lighter
head and thorax; middle and hind coxae
usually markedly lighter than fore
coxae; head narrower (CI usually 83-87)
and less convex-sided (Fig. 15); mesic
woodlands, southeastern U.S., north
along the Atlantic coastal plain to
southern New Jersey ...................... faisonensis
14a. Scapes and legs very long and thin
(SI>165, FI>115), Gulf Coast, spora-
dic elsewhere ............................ longicornis
14b. Scapes and legs of more usual length
(SI<130, FI<105) ................................. 15
15a. Cephalic pubescence dense, partially
obscuring the sheen of the integument
beneath; the pubescence arising from
fine punctae; weak bluish reflections
often present on head and pronotum;
pronotum angular, with short, steep
anterior face and longer, flattened
or concave dorsal face (Fig. 17);
mid-elevation woodlands, mountains of
Southwest, Mexico ....................... austroccidua
15b. Cephalic pubescence very short, dilute
or absent, the integument strongly
shining and easily seen between the
setae; pronotum convex in profile, or
if angular only weakly so and otherwise
not fitting above description ..................... 16
16a. SM<10, PM<7, HL<0.65
........... .... terricola or vividula, see couplet 12
16b. SM>13, PM>7, HL>0.65 ............................. 17
17a. Brown to dark brown, or if partly yel-
lowish then with notably darker pleura,
propodeum and gaster; SM 13-19,
PM 7-11; desert washes and riparian
woodlands, Mexico and southern parts of
adjacent Texas, New Mexico and Arizona ....... bruesii
17b. Uniform yellow to yellowish brown with
abundant black pilosity; SM 21-29;
PM usually 12-16; deserts of east-
central California, Nevada and Utah .......... hystrix
Key to Males
la. Parameres broad margined (Figs. 42
and 43) .............................................. 2
lb. Parameres triangular or digitiform
(Figs. 39-41, 44) ................................... 3
2a. Scapes long (SI 168-174), 3/5 or more
of their length protruding beyond
posterior margin of head and com-
pletely lacking standing macrochaetae;
parameres entire-margined (Fig. 42);
urban areas, sporadic elsewhere .......... longicornis
2b. Scapes shorter (SI 128-135) a little
over 1/2 of their length protruding
beyond posterior margin of head, bear-
ing 7-10 suberect macrochaetae; para-
meres emarginate (Fig. 43); Florida,
disturbed habitats, mangrove islands,
occasional in greenhouses, etc.,
elsewhere ................................. bourbonica
3a. Scapes with 0-1 standing macrochaetae;
parameres relatively short, subtrian-
gular (Fig. 48); eastern U.S. ...................... 4
3b. Scapes with 4 or more standing macro-
chaetae; or if with less, then either
a western species (see lug 14b), or
an eastern one with elongate, triangu-
lar parameres and with prominent
cuspides (Fig. 47) .................................. 6
4a. Digiti straight, lying close to
aedeagus ............................................ 5
4b. In posterior view digiti curved
laterad distally (Fig. 48); scrubland
and dunes of Florida ....................... phantasma
5a. Small (TL<1.9); usually with 1-2
standing macrochaetae on scape; middle
hind coxae distinctly paler than fore
coxae; peninsular Florida .................... wojciki
5b. Larger (TL>1.9); scape lacking standing
macrochaetae, middle and hind coxae
usually same color as fore coxae;
Midwest and eastern U.S (except south
Florida) .................................... parvula
6a. Parameres elongate, triangular, cuspides
long and prominent, reaching to about
3/4 or more of the length of the aedea-
gus; sandy areas, Midwest, Gulf and
East Coast states .......................... arenivaga
6b. Parameres variously formed; cuspides
not reaching to 3/4 the length of the
aedeagus, smaller and not prominent ................ 7
7a. Antennal segment IV crooked or curved
(Fig. 6); aedeagus in side view rela-
tively slender and tapering but
spatulate (Figs. 7 and 8) .......................... 8
7b. Antennal segment IV straight, aedeagus
in side view tapering to a point, which
may be rounded but is never spatulate .............. 9
8a. Parameres viewed from the side tapering,
triangular, rounded truncate to subacum-
inate, viewed from behind distinctly
curved mesad, aedeagus in the form of a
weakly spatulate narrow triangle which is
notably shorter than the parameres (Fig.
8); southern half of U.S., disturbed
areas, in greenhouses further north ......... vividula
8b. Parameres viewed from the side tapering
rapidly over their basal 1/2, then dig-
itiform over the distal 1/2, from behind
only weakly curved mesad, or straight;
aedeagus with distal portion drawn out
into a thin blade with a broadened up-
curved terminus, extending well beyond
parameres (Fig. 7); Arizona, New Mexico,
Texas, Louisiana, Arkansas, Tennessee,
Central Plains states ...................... terricola
9a. Dark species with at least gaster
nearly black, head and thorax piceous
brown or darker, and appendages dark
brown; larger spp. (HL>0.60) ...................... 10
9b. Lighter in color; or if nearly black,
then with brownish-yellow antennae and
legs, or pale middle and hind coxae,
and overall size is smaller (HL<0.57) ............ 11
10a. Propodeum and gaster elongate (Fig.
39); scapes usually with 7-10 macro-
chaetae; parameres strongly tapering
basally, then slender and digitiform
over most of their length, notably
longer than volsellae and aedeagus but
not concealing the latter entirely in
side view; Mexican border region, south
to Hidalgo, Nayarit, also Baja Cali-
fornia Sur .................................. bruesii
10b. Body of more usual conformation,gaster
short (Fig. 41); scapes with 5-7 macro-
chaetae, parameres triangular, as long
or a little longer than volsellae and
aedeagus and concealing the latter in
side view; Florida, Georgia, Carolinas ...... concinna
lla. Seen from behind, digiti straight and
lying parallel to aedeagus ........................ 12
llb. Seen from behind, digiti notably
curved laterad distally (Fig. 46);
Homestead, Florida (also Antilles,
Central and South America, Cocos
Islands) ............................... uatemalensis
12a. Scapes with 6 or more (usually 8-10)
standing macrochaetae ............................. 10
12b. Scapes with 6 or less (usually 1-5)
standing macrochaetae ............................. 14
13a. Body densely pubescent; pale reddish-
brown; eye large (I040); parameres
with more than 30 long blondish
macrochaetae forming a dense fringe
(Fig. 44); Miami, Florida, in green-
houses further north (also Antilles) .......... pubens
13b. Body nearly free of pubescence; shiny
yellowish-brown; eye smaller (OI<32);
parameres with about 15 brown, slender
macrochaetae of varying lengths which
do not form a dense fringe (Fig. 40) ......... hystrix
14a. Scapes with 4-7 standing macrochaetae;
eastern states ..................................... 15
14b. Scapes with 2-3 standing macrochaetae;
Southwest and Mexican mountains ......... austroccidua
15a. Uniform dark brown to nearly black (less
often) gaster slightly darker than the
blackish-yellow thorax and head;
appendages yellowish-brown; wooded
areas near Philadelphia and Pittsburg,
Pennsylvania, and Long Island .............. flavipes
15b. Thorax and usually the head notably
lighter brown than the dark brown
gaster; middle and hind coxae pale
relative to fore coxae and remainder
of leg segments; woodlands of south-
eastern U.S. north to New Jersey pine
barrens ................................. faisonensis
FORMAT OF SPECIES DESCRIPTIONS
In the following chapters, I describe the 11 nearctic
Paratrechina species and the five introduced into the region.
I have seen a few series I cannot place, in every case con-
sisting of only a few workers. Thus, it has been possible
for me to identify virtually all nearctic Paratrechina mater-
ial in good condition as belonging to one of the 16 species
treated or, at the very worst, into one of the species com-
plexes (see below). Only workers and males are formally
described. A diagnosis of the queen follows that of the
conspecific male. The queens of most of the native eastern
species are best distinguished by association with workers
It has become increasingly common in ant systematics
to describe holotypes individually, followed by a section
on variation, as is standard in taxonomic works on non-
social insects. This procedure is followed here although
I believe members of a caste from a single colony are
essentially equivalent as name-bearers for ant species in
view of their close genetic relationship and obvious con-
specificity. There is no known case of a mixed colony of
ants comprised of species which are anything less than
clearly distinct from each other. On the other hand, it
often happens that not all syntypes of earlier authors are
concolonial nor even conspecific. I have designated and
described lectotype workers (and allolectotype males
wherever possible) for species described by earlier authors.
Holotypes and allotypes of the 5 species described as new
are in every case progeny of a single queen to the best of
my ability to determine so. Males and workers of the 3 new
eastern species have been deposited at MCZ, AMNH, USMN, LACM
and FSCA. Material of the 2 new western species is deposited
only at MCZ, USNM and LACM.
It has been stated repeatedly that the male genitalia
offer the best characters for separation of Paratrechina
species (Creighton, 1950; and included references), and
indeed, they are useful. The generally accepted corollary
to this is that Paratrechina workers are barely or not at
all distinguishable. In fact, within species complexes,
some of our species are quite similar in the structure of
the male genitalia while the workers are readily distin-
guished. The males of these species are best distinguished
by non-genitalic characters.
All but 2 of our native species form a rather homo-
geneous group. On a world-wide basis, they appear to belong
to the vividula group of Emery (1925). This group includes
all native eastern U.S. species and 2 western species, plus
a number of tropical forms not treated here. Also in the
group are the introduced flavipes and a few other Asian
forms. The workers of these species share reduced thoracic
and abdominal pubescence (but, except in the Vividula Com-
plex itself, fully pubescent heads); reduced thoracic pil-
osity; somewhat shortened legs and scapes with reduced pil-
osity; and fairly uniform size in the medium range for the
genus. The external male genital apparatus typically in-
cludes isoceles-triangular parameres which are curved mesad
and relatively simply constructed volsellae approximating
the structure of those of faisonensis (Fig. 42).
Since only a small part of Paratrechina is treated here,
no attempt was made to reconstruct a phylogeny for the
species. Even a thorough knowledge of the all extant species
would most likely yield only an approximation of the events
leading to the present taxonomic structure of the genus, in
view of the extinctions that have undoubtedly occurred.
All species are grouped in complexes containing groups
of what I believe are ecological or geographical cognate
species. The names of the complexes are simply the names of
their oldest described species. The term "complex" is used
here to avoid confusion with the "species groups" of Emery
(1925) which are larger and more heterogeneous assemblages
within the genus.
Only when I have seen specimens studied by a previous
author or where his description is sufficiently thorough
and clear to make certain the identification have I included
any reference to a Paratrechina species in the synonymies
presented here. Such a conservative approach undoubtedly
will result in omissions but is necessitated by the follow-
ing conditions: (1) Certain taxa, especially vividula and
fulva have been used as catch-alls for new forms which may
not even be closely related and (2) later authors have
rarely bothered to study types or reliably type-compared
material in making identifications. Under such circum-
stances, synonymies based purely on nomenclatural similarity,
without reference to specimens, were considered unreliable.
I have dispensed with the usual practice of listing
the collection data of all specimens studied. Instead, I
have listed states from which specimens studied originated
in the sections on specimens examined, have described the
preferred altitudes and habitats based on my collecting
experience and information gleaned from labels and the
literature in the sections on natural history, and have
provided range maps. In my opinion, these present as much
of the truly useful information as a complete listing of
localities but in a far more readily assimilable form.
Those wishing to collect Paratrechina would do better to
find them in previously uncollected localities than to re-
duplicate past efforts.
The descriptions follow a format that is standard in
systematic literature on ants. Except where otherwise in-
dicated, the descriptions of the head refer to it in full
face view, as in the figures. Descriptions of the thorax,
on the other hand, refer to it in lateral view. Descrip-
tions of the male genitalia, in general, refer to them as
they appear, intact, on the preserved specimen. The vesti-
ture of vividula is described in greatest detail; vestiture
of other species is briefly described in as much as it dif-
fers from that of vividula.
Measurements of the type speciemns are presented in
the following order: TL, HL, HW, SL, EL, PW, MCL, WL,
FL, GL, SM, PM, MM, CI, OI, SI and FL. These are listed
at the start of each worker description. All measurements,
counts, and indices taken are likewise listed at the start
of the description of the type male, but note that PW,
MCL, FL, PM and MM were not recorded for the males. In
the sections on variation and in composite descriptions,
the ranges of all measurements are reported at the head
of the section. For a complete listing of the measurements
and summary statistics, refer to Appendices A and B,
The section on natural history of the species is nec-
essarily rather short. There is little published on any of
these species, including the tramps, and the observations
reported are primarily my own made in Florida. The term
"disturbed" has been used to characterize habitats of some
species. Mild or light disturbance includes grazing activ-
ity, human foot traffic, or selective clearing of vegetation.
Heavy, disturbance refers to more thorough clearing of vege-
tation, tilling of soil, construction of buildings, and
temporary flooding. It is my hope that this revision will
help other investigators to more thoroughly characterize the
lifeways of Paratrechina species.
Diagnosis of Complex
Worker weakly bicolored, very shiny; cephalic pubes-
cence sparse, limited mostly to rear 1/2 of head; scape
pilosity usually subdecumbent to suberect, sometimes incon-
spicuous, SM usually 7 or 8. Nests in open, often disturbed
habitats, under stones, logs, trash, moss clumps, etc.
Male shiny except thoracic dorsum; antennal segment IV
distinctly bent or curved; parameres triangular; digitus
boomerang-shaped; cuspis short; aedeagal lobes triangular
with a spatulate or otherwise broadened tip.
(Figs. 4, 8, 10; Map-Fig. 49)
Formica vividula Nylander, 1846, Acta Soc. Fenn. 2:
900. Type loc., Botanical Garden, Helsinfors, Finland.
Tapinoma vividula (sic): F. Smith, 1858, Cat. Hym.
Brit. Mus. 6: 56 (in part ?)
Prenolepis vividula: Mayr, 1861, Europ. Formicid.,
Prenolepis kincaidi Wheeler, 1906, Bull. Amer. Mus.
Nat. Hist. 22: 350. Fig. 1, e, 2, z. NEW SYNONOMY. Type
Prenolepis vividula: Emery, 1906, Am. Soc. Ent. Belg.
50: 130-134. Figs. 1-4, e, 9, d; Emery, 1910, Deutsch Ent.
Z., p. 131. Figs. 6, 7, e, .
Paratrechina vividula: Emery, 1925, Gen. Insectorum,
Faxc. 183: 223, Creighton, 1950, Bull. Mus. Comp. Zool.
Harvard 104: 409 (in part).
Worker. TL usually 2.0-2.5, OI usually 24-27. A
Southern species of disturbed habitats. Weakly bicolored;
head and gaster yellowish-brown to piceous, thorax and
appendages yellow to dark reddish-brown. Head subquadrate,
sides weakly convex, subparallel. Cephalic pubescence,
mostly limited to rear half of head and often rather varied
in length and spacing.
Male. TL about 1.9-2.25, OI usually 38-40. Body color
dingy yellow to dark brown, thorax usually lighter.
Antennal segment IV curved or bent. Parameres, viewed from
the rear, strongly curved mesad. Aedeagus as short or
shorter than parameres, spatulate, hyaline.
Queen. Not readily distinguishable from those of
terricola, faisonensis or concinna. In contrast to the
latter two species, often bicolored as in worker. Normally
has larger eyes than queen of terricola.
Allolectotype worker. TL 2.15, HL, 0.62, HW 0.53,
SL 0.71, EL 0.15, PW 0.38, MCL 0.16, WL 0.77, FL 0.57,
GL 0.77, SM 8, PM 4, MM 4, CI 85, OI 25, SI 115, FI 92.
In full face view, clypeus arcuate and entire. Dorsum
of clypeus with sides meeting in a rounded, median angle.
Head, exclusive of clypeus, subquadrate, the apparent
squareness accentuated by the anterior 3/4 of the sides
being virtually parallel except where they curve inward to
meet the mandibular insertions. Eyes in full face view
appearing about 1/4 as long as distance from mandibular
insertions to rear border, though in fact their greatest
diameter is longer. Distance between front of eye and
mandibular insertion slightly greater than EL, eyes sep-
arated from sides of head by less than 1/3X eye width.
Scapes curved over basal 1/4 of their length and over short
distal section, otherwise nearly straight. Scapes of about
average length for vividula group. No ocelli visible.
Pronotum in profile weakly angular; the anterior and
dorsal faces flat, the dorsal a little longer than the
anterior. Promesonotal suture clearly visible, though
barely impressed. Mesonotum flat-topped with short anterior
and posterior faces rising steeply to meet the dorsal face
in rounded angles, the posterior face a little longer and
steeper than the anterior. In profile, the propodeum nearly
evenly rounded, its highest point distinctly lower than that
of pronotum, the angle between its anterior and descending
faces weakly obtuse. Legs of about average length for
Petiole sharp-crested, cuneate in profile. In dorsal
view, with straight sides weakly divergent dorsad; crest
comprised of two nearly straight sides meeting in a blunt
Clypeus with 12 forward-projecting macrochaetae, those
closer to the front and center as long as clypeus in dorsal
view, and with a few appressed hairs just to the sides of
the median ridge. Front half of head with a few macro-
chaetae, those on frons somewhat shorter than longest
clypeal macrochaetae and those on infraocular region much
shorter and subdecumbent. Rear half of head with longer
more abundant macrochaetae and pubescence composed of 4 or
5 dozen unevenly spaced appressed simple hairs of varying
length. Scape macrochaetae suberect, less than half as long
as greatest width of scape and of the same color as the
scape; most easily seen in posterolateral view. The eight
major thoracic macrochaetae weakly curved; the remaining
ones half or less as long and straight. Forelegs with 3-6
macrochaetae on flexor surfaces of femora and tibiae and 1-2
on extensor surfaces. Middle and hindlegs with the reverse
pattern, namely several hairs on extensor surfaces and 1 or
2 on flexors. Propodeum with a narrow patch of longitudi-
nally oriented pubescence, along its front edge. Gaster
with sparse, very fine, appressed hairs and the usual com-
plement of macrochaetae.
Smooth and shining from head to first gastral tergite.
Weakly shagreened posterior to this.
Head and gaster light yellowish-brown. Thorax, brown-
ish-yellow. Mandibles light brownish-yellow with dark
reddish-brown teeth. Legs and scapes yellow.
Lectotype male. TL 1.89, HL 0.52, HW 0.48, SL 0.59,
EL 0.19, WL 0.81, GL 0.57, SM 4, CI 94, OI 37, SI 114.
Mandibles apically unidentate, without subapical den-
ticles in this specimen. Clypeus, in dorsal view with a
very shallow median emargination and median raised portion
narrower than in worker, evenly rounded, subumbonate. Sides
of head weakly convex, convergent anteriad, meeting the rear
border through slightly obtuse rounded corners. Limits of
weakly convex rear border lie approximately behind inner
border of eyes. Eyes convex, apparently about 1/3 as long
as head. Median and lateral ocelli separated by about 2X
their diameter. Scapes slenderer than worker's. Antennal
segment IV bent as in Fig. 6.
Petiolar crest more blunt than worker's. Petiole
broader than worker's in posterior view.
Parameres about 2X as long as broad in side view,
nearly triangular, but slightly recurved over distal 1/3
of their length. In rear view, parameres sickle-shaped,
curving mesad distally. Digitus shaped like a short
boomerang, i.e., narrow throughout, bent downward at the
middle through a broadly obtuse angle and tapering gently
to a rounded terminus. Cuspides of the usual shape for
vividula group, i.e., like an inverted bowl of a round
spoon. Aedeagal lobes shorter than parameres, slender,
triangular, weakly spatulate and splayed (as is normal in
preserved males of this species).
Vestiture as in worker but cephalic pubescence more
abundant and evenly distributed over rear half of head and
on frons. Scapes less pubescent than worker's and with a
row of 3, plus 1 offset, macrochaetae; these suberect, short
and pale, inconspicuous as in worker. Parameres with about
25 simple, decurved, yellow macrochaetae on their outer
Head, thorax, front of gaster and appendages uniform
yellowish-brown. Gaster increasingly darkening brown from
middle of first gastral tergite posteriad. Genitalia
yellowish-brown, except aedeagal lobes, which are yellowish-
Workers. TL 1.90-2.64, HL 0.55-0.69, HW 0.44-0.61,
SL 0.61-0.77, EL 0.13-0.17, PW 0.32-0.44, MCL 0.12-0.19,
WL 0.66-0.90, FL 0.46-0.62, GL 0.56-1.17, SM 6-9, PM 3-7,
MM 2-5, CI 80-91, OI 21-27, SI 105-118, FI 83-93, (n=58).
The clypeus may have a shallow concave or flat median
emargination in dorsal view. The sides of the head are
usually subparallel (in eastern specimens) or are wekaly
convex (especially in material from Texas westward). OI is
almost always greater than 24, but two series from separate
localities in the Guadalupe Mountains, Texas, have unusually
small eyes. These may represent a hybrid swarm of vividula
x terricola or simply are aberrant. Mean OI for the species,
including six specimens from Guadalupe Mountains is 25. The
rear face of petiole in lateral view is often convex or sub-
angular, rather than straight as in type, and the crest may
be rounded or with a flat top and sloping sides in rear view.
The pilosity of the scapes is often longer (almost
equal to width of scape) and brown, thus, much more con-
spicuous than in type; this is especially true in Gulf Coast
and Florida material. The latter specimens have more
abundant cephalic pubescence distributed further forward,
especially on frons, and may have some long sparse appressed
hairs on the pronotum and mesonotum. Specimens from Arizona
and California typically have the cephalic pubescence short,
very fine and more evenly spaced but still limited to rear
portion of head.
Western material is generally colored about as type,
and most eastern material is darker. However, I have seen
series from Arizona and Mississippi which are nearly uniform
very dark brown, thus, recognizable only by head shape, eye
size, and pattern of pubescence.
The types and one additional series of kincaidi from
Jamaica, here synonymized with vividula, have color and
vestiture like that of peninsular Florida specimens.
European greenhouse material tends to look more like vivid-
ula from Texas or Northern Mexico. It should be recalled
that Emery (1906) suggested Mexico as the provenience of
Males. TL 1.89-2.25, HL 0.46-0.55, HW 0.45-0.52,
SL 0.54-0.62, EL 0.18-0.21, WL 0.69-0.85, GL 0.57-0.98,
SM 4-6, CI 91-98, OI 33-42, SI 107-119, (n=ll).
The mandibles of the males of vividula vary from simple
apically unidentate to having an apical tooth, a small cleft,
then a subapical denticle or two. The crook in antennal
segment IV is sometimes manifest only as an even curve. Very
few specimens lack the splaying of the aedeagal lobes typi-
cal of preserved males in this species. Most specimens are
significantly darker than the type ranging from a dingy yel-
low to a dark reddish-brown, often with the thorax somewhat
The type material (HM) was collected in a greenhouse;
Helsinki, Finland; 2-25/X/1845. Included are the lecto-
type, allolectotype, three paralectotype workers and a
paralectotype queen (minus gaster). I have deposited
conspecific material from Texas and Florida at HM. In
addition, I have studied many hundreds of specimens from
across the southern U.S. and Europe. The material was
loaned or donated from LACM, USNM, TPN, DPW, and TTU. JCT
contains Florida material. European material and types of
kincaidi (here synonymized with vividula) are at MCZ. The
material from TTU was particularly valuable (in addition
to the suggestions of Emery, 1906), in helping me make a
correct guess about the identity of vividula even before I
saw the types, because of the large number and size of the
series, many of them containing7 associated males. TTU also
contained a similar large amount of terricola, thus helping
to preserve the latter from being synonymized, as was my
This species is very closely related to terricola;
isolated workers cannot always be assigned to one or the
other species with certainty. The larger OI and more
straight-sided head will suffice to separate the great
majority of vividula workers. The ranges of the two species
are different (Fig. 49). I collected one series of concinna,
normally uniform dark brown, which I mistook in the field
for vividula because they were bicolored. The specimens
were in all other characters typical concinna. Males of
vividula are distinguished from terricola by genital char-
acters as described above. Most of the vividula material I
have seen in collections has been identified as melanderii,"
while most of the material identified as vividula previous
to this study does not belong to vividula (see discussion
of faisonensis). I believe this is due to the long standing
dogma stated in Creighton's key (1950, p. 404), "erect hairs
on the antennal scapes [are] abundant and occurring on the
sides as well as the front of the scape" in vividula. In
fact, the macrochaetae on the scape are almost always lim-
ited to two rows on the anterodorsal part of the scape com-
prised of 3 or 4 hairs each. The macrochaetae themselves
are often inconspicuous, giving the casual observer the
impression that there are less of them than, in fact, occur.
It should be noted that the scape pilosity of terricola is
of the same character as that of vividula and cannot be used
as a separatory feature for the two species.
The type series of kincaidi (MCZ) unfortunately lacks
the male that originally belonged to it, and he could not
be found in the unit tray in which the series had been
stored by MCZ staff who searched for it upon my request.
Nevertheless, it seems unquestionable to me, based on the
features of the workers and queens in the series and upon
Wheeler's (1906) description and figure of the genitalia
of male kincaidi, that kincaidi can be none other than a
synonym of vividula.
Under conditions of sufficient warmth, alates may be
present at any time of year and may fly any warm day when
humidity is high. In most of the range in the U.S., flights
are restricted to the months May-October. Females are
occasionally attracted to lights at this time of year,
though flights, from the little available evidence, occur
during the day. I observed a group of six males patrolling
a fence post at 1300 hr in Lake Placid, Florida, in January,
and I have alates of both sexes collected from the surface
of a pool in a Juniper-oak woodland in Texas on 7 Septem-
This ant is characteristic of open, usually rather dis-
turbed habitats including beaches, parks and other land-
scaped areas, crop fields, fallow fields, vacant lots,
parking lots, etc. Nuhn and Wright (1979) reported finding
vividula (as Paratrechina sp.) in fields and other open,
disturbed habitats in North Carolina. It commonly estab-
lishes colonies in plant pots or mulch piles and is trans-
ported in these materials to greenhouses and nurseries far
from the areas where it survives outside. It has apparently
been established as a greenhouse ant in Europe for nearly
150 years. This ant occurs in less disturbed rangeland and
other open habitat in southern Texas and northern Mexico,
where it overlaps with and may occur in a mosaic distribu-
tion with respect to its sibling species, terricola.
(Figs. 5, 6, 7; Map-Fig. 49)
Formica (Tapinoma) terricola Buckley, 1966, Proc. Ent.
Soc. Phila. 6: 168. ,, 9, d. Type loc., Austin, TX.
Prenolepis melanderi Wheeler, 1903, Psyche 10: 104.
Fig. 8, t, 9, -. NEW SYNONYMY. Type loc., New Braunfels,
TX (designated by Creighton, 1950).
Prenolepis (Nylanderia) vividula melanderi: Emery,
1906, Ann. Ent. Soc. Belg. 50: 132. Fig. 5, d; Emery, 1925,
Gen. Insectorum, Fasc. 183: 223.
Paratrechina (Nylanderia) melanderi: Creighton, 1950,
Bull. Mus. Comp. Zool. 104: 407; G.C. and J. Wheeler, 1953,
Ann. Ent. Soc. Amer. 46: 144. Pl. II, Figs. 16-20, larva.
Worker. TL usually 1.9-2.4 (southern part of range),
2.2-2.7 (north), OI 20-24. A vividula-like species of open
habitats and disturbed areas of northern Mexico, Arizona
and Louisiana, north to Illinois and South Dakota and in the
foothills of the Smoky Mountains. Particularly abundant
in Texas. Yellowish-brown with darker head and gaster (more
common south) to piceous with yellowish-brown appendages
(more common north). Cephalic pubescence very sparse, head
shiny. Eye small.
Male. TL 2.1-2.4, OI usually 35-37. Piceous brown to
nearly black. Eye relatively small (compared to vividula).
Parameres only weakly curved mesad. Distal portion of
digitus about 2X as broad as that of aedeagus. Aedeagus
longer than parameres, curved upward distally. Crook in
antennal segment IV conspicuous.
Queen. Averages smaller and smaller-eyed than
vividula, often not distinguishable.
Alloneotype worker. TL 1.98, HL 0.61, HW 0.52,
SL 0.68, EL 0.13, PW 0.36, MCL 0.16, WL 0.77, FL 0.54,
GL 0.61, SM 8, PM 4, MM 3, CI 86, OI 22, SI 112, FI 88.
Clypeus lentiform; anterior clypeal border arcuate
with a faint, narrow median concavity; rear border less
strongly arched than front border, also with a small median
concave emargination. Sides of head weakly convex along
most of their length and slightly convergent anteriad.
Rear border of head rounded with a shallow median notch a
little wider than the distance between the frontal carinae.
Head of about average breadth for the species. Eyes smaller
than in vividula, approximately their rear 1/5 overlapping
the midpoint of the postmandibular head length; separated
from sides of head by about 1/2X eye width and from mandibu-
lar insertions by about 1 1/3X EL. Scapes weakly curved
near the ends, straight over most of their length; of about
the mean relative length for the species.
Pronotum broadly angular, the somewhat shorter anter-
ior face flat and the dorsal face weakly convex. Mesonotum
flat on top, sloped downward to the rear, its short anter-
ior face meeting pronotal dorsum face at about the same
angle as that between 2 faces of the pronotum. The poster-
ior mesonotal face shorter and much steeper than the anter-
ior mesonotal face. Propodeum low and arched, its highest
point only about as high as the rear edge of the mesonotal
dorsum. Legs of about average length for the species.
Rear face of petiole obscured by anterior face of
gaster, but petiole apparently cuneate with a blunt rounded
crest and flat front face in profile. From above, petiolar
crest is flat across the middle with rounded corners, and
the sides are straight and weakly divergent dorsad.
Cephalic vestiture resembling that of small vividula
workers but sparser than in most individuals of the latter
species. Cephalic pilosity is rather short, weakly curved,
and sparse, especially on the sides of the head. Cephalic
pubescence sparse, widely spaced, the spaces between the
hairs wider than the length of the hairs in most cases.
The appressed hairs themselves are longer than those of
other U.S. species. Thoracic macrochaetae about half as
long as the greatest width of the scape, of the same color
as the scape, protruding above the finer subdecumbent pubes-
cence which covers the scape. Ancillary thoracic macro-
chaetae about half as long as the shortest major macrochaetae
and of the same dark brown color (pronotum) or somewhat
lighter (mesonotum). Gaster covered with thick, dark brown
rearward-curved macrochaetae of about even length and
spacing throughout; and with some very sparse, long pubes-
Body very shiny and smooth, except gaster which has
some very fine almost indetectable (at 40X) shagreening.
Head piceous brown becoming a little lighter on the
sides and clypeus. Scapes and mandibles brownish-yellow,
the latter with dark brown teeth. Thorax, except mesonotum,
colored like sides of head, mesonotum lighter and yellower.
Legs brownish-yellow with the middle and hind coxae paler.
Gaster piceous brown.
Neotype male. TL 2.27, HL 0.53, HW 0.48, SL 0.59,
EL 0.18, WL 0.74, GL 1.01, SM 6, CI 92, OI 35, SI 112.
Mandible with an apical tooth, a sharp subapical
denticle and a crenulate masticatory border. Anterior
border of clypeus arcuate, rear border with straight sides
and a narrow median concave emargination. Sides of head
convex from posterior corners to just before the eyes,
straight from there to mandibular insertions. Head narrower
than that of vividula. Rear border of head less convex
than sides, about as broad as the distance between the
mandibular insertions. Eyes apparently only about 1/3X HL
(in fact a little longer) but strongly convex, protruding
well beyond sides of head. Ocelli not large but fairly
close set, median and laterals separated by about 2X their
width. Scape weakly sigmoid, a little shorter than average
for the species. Antennal segment IV longer than adjacent
segments and distinctly bent at the middle.
Petiole cuneate with a blunt rounded crest, thicker
than that of worker. In dorsal view, the crest is flat
across the top and about twice as wide as that of the worker
with straight sides weakly divergent dorsad; the anterior
face is concave.
Genitalia distinctive, even from those of closely
related vividula. In side view, paramere is about twice
as long as its basal breadth, but it tapers quickly in the
basal 1/3 of its length and is digitiform over the distal
2/3. Cuspis curved mesad to overlie the digitus in pos-
terior view and is a little larger and more prominent than
in vividula, faisonensis, etc. but much shorter and less
prominent than in arenivaga. Digitus not decurved as in
vividula but with basal edge straight. Distal portion of
digitus about 2X as broad as narrowest portion of aedeagus;
the former subtriangular with angles rounded. Outer face
of digitus convex. Aedeagus distinctly longer than para-
meres; the lobes closely appressed (not splayed, as in
vividula); conspicuously broadened dorsad and truncate
distally, resembling a miniature inverted tomahawk (Fig. 7).
Pilosity finer than worker's, brown to dark brown, of
approximately equal spacing and length over entire dorsal
surface, except on anterior half of gaster. Pilosity
virtually lacking on sides of head and thorax. Parameres
with more than 20 yellowish, weakly decurved hairs. Pubes-
cence on head sparse not obscuring sheen of integument.
Pubescence of thoracic dorsum denser.
Sculpture much as on workers, but shagreening on
gaster consisting of larger, more conspicuous integumental
Eyes grey. Body and appendages uniform piceous,
except articulations and tarsi of legs and post-mandibular
mouthparts, all of which are light yellowish-brown.
Worker. TL 1.87-3.03, HL 0.54-0.73, HW 0.44-0.63,
SL 0.58-0.84, EL 0.11-0.16, PW 0.30-0.44, MCL 0.12-0.19,
WL 0.65-0.93, FL 0.45-0.67, GL 0.61-1.37, SM 5-10, PM 3-6,
MM 2-4, CI 80-92, OI 20-24, SI 106-118, FI 83-93, (n=42).
This species is sympatric (or more precisely, mosa-
ically parapatric) with arenivaga and parvula in the western
half of the latter two species' ranges. Interestingly,
terricola exhibits the same tendency as these other species
in having larger overall body size and a broader head in
the northern and eastern parts of its range. TL of Arizona
specimens averages about 1.9, that of Texas and Oklahoma
specimens about 2.2, that of Kansas and South Dakota speci-
mens about 2.4, and that of Tennessee specimens almost 2.6.
CI of specimens from southern states ranges 80-87, while
CI of Kansas and South Dakota specimens ranges 84-92. Cor-
related with size trends is a greater degree of investiture
of northern material. Four of the five workers with SM 10
and five of the eight workers with SM 9 are from northern
states, as are all individuals with PM>5 and MM>3. Greater
density of cephalic pubescence is normal on larger specimens
wherever they occur. Specimens from Arkansas (UARK), pre-
dictably, span the gap between the large, hairy, Tennessee
material and the smaller, nearly glabrous Texas material.
Male. TL 2.12-2.40, HL 0.49-0.59, HW 0.46-0.54,
SL 0.57-0.69, EL 0.17-0.21, WL 0.73-0.93, GL 0.85-1.05,
SM 4-6, CI 90-94, OI 31-37, SI 110-117, (n=9).
These numbers describe fairly well the metric varia-
tion of terricola males. Most males have CI 35-37. The
two specimens with CI 31 and 33 are unusually small-eyed.
I have seen no terricola males from Arizona but would
expect them to have small eyes as do workers from that
The paramere is generally narrower and straighter
(i.e., more digitiform) than in vividula males. While it
is always less curved, the shape may closely resemble the
weakly recurved triangular form of the vividula paramere.
The aedeagus of terricola males is normally considerably
longer than the parameres, and the lobes of the aedeagus
are closely approximated in preserved material. In side
view, in the intact animals, the aedeagus normally projects
below the parameres. Occasionally, especially in Tennessee
material, the aedeagus more closely resembles that of vivid-
ula in being shorter than the parameres and having its halves
splayed and projecting above the parameres. In such cases,
the structure of the volsella is that typical of terricola,
and the specimens have in each case belonged to series in
which other males have the aedeagal structure more typical
The neotype male and alloneotype worker were desig-
nated from a nest series (TTU) labeled Texas, Real Co. 12
mi NE Leakey. Among small rocks. 19 Mar 1979. O. Francke
et al. #3725. The pins bearing the neotype and alloneotype
are deposited at LACM, and the remaining three pins from
the series are deposited at MCZ. Other series, containing
at least males and workers, from nearby Texas localities
(all from TTU) are deposited at these and the other usual
Material from North Dakota and Oklahoma was from GJW.
Kansas material was from JCT and SBSK. A large amount of
Texas material and smaller samples from other states is at
LACM and USNM. Wheeler's types of melanderi (here synony-
mized with terricola) are at MCZ and AMNH. A large sample
of workers from Arkansas (UARK) seems to fill in the middle
portion of the size and hairiness dines between Texas and
Tennessee and upper midwestern specimens.
The resurrection of Buckley's (1866) name terricola
for the ant which has been known for 80 years as melanderi
is based on several considerations. The first is that
despite an adequate description and fine drawings accompany-
ing Wheeler's (1903) description, the name melanderi has
been applied recklessly to almost every native species of
nearctic Paratrechina. Most references to melanderii" in
the literature apply either to other species or to a mix
of species which may or may not include the melanderi of
Wheeler's description. In its modern usage, the name has
become at worst meaningless and at best highly ambiguous.
Of course, this is true for many names in this genus and
would not alone constitute sufficient reason for bringing
into use a name of apparently uncertain identity in place
Another problem concerning melanderi is Creighton's
designation of New Braunfels, Texas, as its type locality.
Among the syntypes of melanderi at AMNH are seven workers
and parts of two others from New Braunfels, but no males.
MCZ contains no syntype material from New Braunfels. The
only male in the entire syntype series, as presently con-
strued, is a rubbed, headless male from Austin.
The importance of the lack of suitable male specimens
among the melanderi syntypes becomes clear when we consider
that vividula and terricola are separable with certainty
only in the males. It is for this reason that I have desig-
nated males as lectotype and neotype, respectively, for
these two species.
There is considerable overlap in the phenology of
workers of terricola and vividula, so much so that lacking
associated males, worker specimens often cannot be identi-
fied. The differences in the males, contained in the de-
scriptions and in Figs. 7 and 8, need no elaboration. What
follows is a listing of hints for identifying workers of
these two species in the absence of males. The list is, of
course, based on the assumption that all other possibilities
besides terricola or vividula have been eliminated.
1. Specimens collected outdoors from north of the
Mason-Dixon line, from undisturbed or moderately disturbed
habitats are almost certainly terricola. Specimens col-
lected in Gulf or Atlantic coast states east of Louisiana
are almost certainly vividula.
2. In Mexico and Arizona, terricola averages about
2.0 mm in length while vividula from the same area averages
a little larger, about 2.25 mm.
3. Specimens with OI 24 or less are usually terricola.
Workers with OI 25 or more are certainly vividula.
4. Specimens with very sparse cephalic pubescence are
usually terricola, though small vividula specimens may have
the pubescence on the head quite dilute. Specimens with
denser (though very fine) pubescence on the rear portion
of the head are probably vividula.
Finally, we consider Buckley's description itself and
enumerate the reasons why it must concern the ant Wheeler
redescribed as melanderi.
1. The name Formica (Tapinoma) terricola--Creighton
(1950, p. 24) points out that Buckley was strongly influ-
enced by the British hymenopterist Frederick Smith. Smith
called the Paratrechinas he described Tapinoma (e.g.,
flavipes). It is possible that Buckley would have used the
name in the same way.
2. The size of the castes--The size reported for the
queen is certainly too large for a Paratrechina (6.86 mm).
The sizes for the worker and male (2.0 mm) are appropriate.
(Buckley gave the queen measurement as 0.27 inch which may
be a misprint for 0.17 inch which equals about 4.32 mm,
approximately the size of a Paratrechina queen).
3. Coloration--The color of the queen is described as
piceous, and the males are said to be "like the female."
The worker is described as dark brown above, lighter below
and on the petiole and legs. Fully sclerotized Texas ma-
terial of this species fits these descriptions.
4. Head shape--The head of the queen is described as
"small, triangular," i.e., strongly convergent anteriad.
The worker's head is said to be "little wider than the
thorax." While the latter is an understatement of the true
width, it does suggest that he observed the relative narrow-
ness of the head of many workers from Texas.
5. Vestiture--The queen is described as "thickly
sprinkled with short grey hairs," probably referring to her
dense pubescence. No such pubescence is mentioned for
workers or males, as is appropriate.
6. Venation--Buckley says the wings have one marginal
and two submarginal cells (i.e., a submarginal and a median
cell) and that the discoidal cells are obsolete--as in
7. Petiole--In Buckley's words, pediclee short, in-
serted in the base of the abdomen . inclined forward,
wedge-shaped." This describes the petiole of Paratrechina
8. Ecology--Terricola is described as "very active"
(i.e., fast-moving) "with shallow nests in soil. Winged
males and females captured in March" and as "rare," which
reflects the great difficulty in encountering nests except
in early spring. Among ants that could fit the rest of
Buckley's description, only Paratrechina has alates in the
nest in March. The color is inappropriate for arenivaga,
the locality is inappropriate for bruesii, and the alate
season is inappropriate for vividula. Melanderi is the only
suitable Texas Paratrechina for the description.
Any one or a few of these points, except perhaps number
8, would leave us in doubt, but considered together, I
believe they describe, however poorly, Wheeler's melanderi
which must, therefore, be relegated to synonym. It is
probable that Buckley's Formica picea (1866, p. 163) is in
fact a description of terricola minims or other unusually
small workers. Lacking sufficient evidence, I do not include
picea in the synonymy. While I realize that resurrecting
terricola from the ranks of Buckley's numerous nomina nuda
constitutes a virtual myrmecological apostasy, I trust that
the evidence presented shows this to be the best course for
eliminating confusion and stabilizing the taxonomy of the
Alates of terricola overwinter in the nest and, as
indicated in both Buckley's and Wheeler's descriptions, are
commonly found in March in Texas. Texas alate records range
from January through April. A Kansas record is from April,
and I collected a colony with male pupae from Lawrence,
Kansas, on 2 August. Series from Tennessee and Missouri
with alates are dated March and April, respectively. Col-
lections of alates in May from Tennessee and Arkansas are
both from mountainous areas.
Mitchell and Pierce (1912) reported "large numbers of
winged forms were collected running, flying and mating,
March 2, 1909." This description recalls to me my observa-
tion of a mating swarm of vividula in which males ran about
on the substrate, periodically flying up then quickly cir-
cling back into the swarm and running about on the substrate
with wings vibrating.
Buckley (1866), Wheeler (1903), Mitchell and Pierce
(1912) and Van Pelt (1983) state that this species nests
in soil, usually beneath a stone, log or cow dung. Records
from the TTU collection indicate they are found in most
soil types, except very sandy soils, and in vegetation
varying from open disturbed areas and pasture to mesquite
and post-oak woodland and occasionally in denser more mesic
woods. Specimens from South Dakota (GJW) were from Juniper--
Rhus trilobata--Chrysothamnus woodland. Kansas collections
are from woodland edges and openings, pastures and meadows.
Male terricola, lateral view.
Male vividula, lateral view.
Worker of terricola, lateral view.
Worker of vividula, lateral view.
Diagnosis of Complex
Worker brown or bicolored, shiny; head with dense pubes-
cence, this arising from very fine to rather conspicuous
punctae; scape pilosity reduced in some species, suberect
to erect. Nests in a variety of undisturbed or lightly dis-
turbed, moist habitats in leaf litter, rotten wood, clump-
ing plants or in soil.
Male shiny, but thoracic and cephalic dorsum usually
dulled by pubescence; parameres triangular; digitus boomer-
anged-shaped; cuspis short; aedeagus triangular, acuminate.
Paratrechina concinna, New Species
(Figs. 11, 14, 38; Map-Fig. 50)
Worker. TL usually 2.3-2.9, CI usually 84-89,
OI 24-28. A species found from Florida to North Carolina,
usually found in marshes, wet grasslands and pastures, and
swamp edges. Uniform brown or with thorax and legs slightly
lighter than gaster and head. Scapes and middle and hind
coxae brown (never tan or whitish). Overall size large
among native eastern U.S. species. Cephalic pubescence
arising from small foveolae giving punctate appearance.
Pubescence of pronotum and proprodeum usually conspicuous,
that on front of pronotum subdecumbent.
Male. TL 2.25-2.5, OI usually 34-37. Closely resem-
bling faisonensis but darker in color, has larger body and
eyes and cephalic sculpture as in workers.
Queen. Resembles vividula and faisonensis. Differs
in having uniform dark color, cephalic punctation, and
slightly larger size.
Holotype worker. TL 2.72, HL 0.72, HW 0.65, SL 0.81,
EL 0.19, PW 0.46, MCL 0.20, WL 0.91, FL 0.65, GL 1.09,
SM 12, PM 7, MM 4, CI 90, OI 27, SI 113, FI 90.
In full face view, clypeus lentiform, anterior border
arcuate with a shallow concave emargination across the mid-
dle 1/4; rear border also arcuate. Dorsum of clypeus with
sides meeting in a rounded median angle. Sides of head
rounded, convergent anteriad; widest just behind the eyes.
Rear border of head straight with rounded corners; the
corners about as far apart as the width of the clypeus.
Head subquadrate, broader than usual for the species (CI 90
versus mean CI 86 for the species). Eyes longer and some-
what broader than in faisonensis, slightly more convex.
Eyes separated from sides of head by less than 1/3 their
width; from mandibular insertions by slightly over IX EL.
Scapes a little shorter than average for the species, which
has SI slightly but probably insignificantly longer than
average for vividula group. Scape very slightly but evenly
curved over its entire length, distal portion tapering giv-
ing the illusion of greater curvature. Small round markings
at the positions of the ocelli probably represent vestiges
of those organs.
Pronotal profile evenly curved, somewhat more convex
than in faisonensis. Anterior face of mesonotum rising from
promesonotal suture to dorsal face through about a 50% slope;
posterior declivity steeper and longer than usual in the
faisonensis; mesonotal dorsum slightly sloped to the rear.
Propodeum low and evenly rounded, only a little higher in
profile than metathoracic spiracles, the angle between its
anterior and declivious faces slightly obtuse.
Petiole sharp-crested, create in profile; in dorsal
view broader and with more rounded sides and crest than
Cephalic pilosity as in faisonensis. Cephalic pubes-
cence very dense, reminiscent of queens of other species,
and neatly aligned, giving a combed appearance. Thoracic
macrochaetae conspicuous like those of faisonensis. Thorax
appears quite bristly because the macrochaetae are less
curved than those of faisonensis, and the eight major hairs
are subtended by ancillary hairs about 3/4X as long and
these in turn (on the pronotum) by smaller hairs about 1/2
length of the latter. Pubescence of thorax is diagnostic:
subdecumbent and unusually abundant (for the vividula
group) on the anterior portion of the pronotum, scattered
appressed hairs on the metanotum and metapleuron, and
abundant suppressed hairs on the anterior portion of the
propodeum. Gaster with a few appressed hairs, unlike fais-
onensis. Petiole with two small macrochaetae on the crest.
Thorax and gaster strongly shining, the former very
smooth, the latter delicately shagreened. Head and
appendages punctate, feebly shining, the sheen reduced by
the denser pubescence which arises from tiny foveolae.
Uniform castaneous brown with only the articulations
of the appendages and the blade of the mandibles a little
Allotype male. TL 2.46, HL 0.65, HW 0.61, SL 0.73,
EL 0.22, WL 0.97, GL 0.85, SM 6, CI 94, OI 34, SI 112.
Head large compared to all other males in vividula
group. Head shape, including that of clypeus, strikingly
like that of worker. Though CI indicates head is propor-
tionately wider than worker, this is because the head width
is taken across the eyes which are much larger than workers'
and protrude beyond the sides of the head. Head is, in
fact, a little narrower than workers' minus the eyes.
Mandibles strongly sclerotized, dark colored with an apical
tooth, cleft and a subapical denticle, the identical larger
than in any specimen of vividula having it. Scape a little
shorter and with fewer macrochaetae than workers but pilose
for a vividula-group species. Eyes larger and more
prominent than faisonensis. Median ocellus notably larger
Upper portion rear face of petiole concave. Petiole
otherwise as in worker but broader. Gaster contracted,
shorter than thorax. Genitalia closely resembling those
of faisonensis but with cuspis a little more convex and
generally heavier and darker colored.
Vestiture as in workers with the usual differences.
Parameres with decurved pilosity as in faisonensis but
this less abundant, about 20 hairs.
Not as shiny as workers, due to occurrence of pubi-
gerous punctation on thoracic dorsum and coxae. Gaster also
appearing somewhat roughened, this due to a slightly coarser
version of the usual shagreening.
Darker in color than all other vividula group species,
expect flavines, from which concinna differs by larger size
and having dark colored appendages. Head, thorax, and ap-
pendages dark brown, gaster piceous.
Worker. TL 2.13-3.10, HL 0.61-0.76, HW 0.52-0.65,
SL 0.71-0.85, EL 0.15-0.20, PW 0.37-0.49, MCL 0.16-0.20,
WL 0.77-0.93, FL 0.57-0.71, GL 0.65-1.41, SM 7-16, PM 4-8,
MM 2-6, CI 83-90, OI 24-28, SI 110-125, FI 88-97, (n=41).
This species seems to be quite uniform in general
appearance throughout its range. Head shape somewhat vari-
able as indicated by the large range of CI. This seems to
have no relationship to overall size, as even large workers
may have heads in the narrow end of the range. The abun-
dance and degree of appression of the pronotal pubescence,
which occasionally consists of only a few appressed hairs
is also somewhat variable. The macrochaetae on the petiolar
crest vary from 0-2. The color of the thorax is not uncom-
monly a little lighter and yellower than that of the head
and gaster. This is probably the normal condition of
Male. TL 2.24-2.46, HL 0.59-0.67, HW 0.55-0.63,
SL 0.67-0.74, EL 0.20-0.24, WL 0.87-1.01, GL 0.69-0.85,
SM 5-7, CI 92-97, OI 33-38, SI 108-117, (n=8).
The large size, dark color, distinctly short gaster
and worker-like aspect of the head of concinna males show
little variation in the specimens from Florida and North
Carolina I have seen. Most have the ocelli more equal in
size than those of type.
Holotype and allotype and a large series of paratypes,
including queens, were collected 15 April 1980, 1.5 miles
ESE of Gainesville, Florida, under a plant pot on a damp
lawn at the edge of a sweetgum swamp. Holotype and allo-
type are deposited at FSCA. Paratypes are deposited in the
Most of the material studied was collected by myself
in Alachua County and Highlands County, Florida. A number
of collections made by Van Pelt (1957) are stored at FSCA.
Additional small series were found in LACM, USNM, AMHN,
DPW (USDA) and TPN, these from Florida, Georgia, and North
The name concinna (Latin--neat, orderly) refers to the
neatly aligned appearance of the cephalic pubescence.
This species falls into Creighton's concept of vivid-
ula, but I do not believe he ever saw it and, thus, have not
included his vividula treatment in the synonomy. There are
only a few specimens of concinna in the major collections
due to its preference for habitats rarely collected by ant
hunters. My collection contains many more concinna than
all other collections together except FSCA, which reflects
both the poor representation of Florida ants in these col-
lections and the unlikelihood of collectors'entering marshy
areas to find ants.
Concinna could be confused in the field with vividula,
faisonensis or bourbonica. Fiasonensis lacks thoracic
pubescence and is smaller, while bourbonica is more uni-
formly covered with dense grey pubescence and is larger.
Large, dark specimens of vividula can be recognized by
their smooth heads, less abundant cephalic and thoracic
pubescence and lower SM.
Based on the overall similarity of the workers and
of the genitalia of the males, I believe concinna is most
closely related to faisonensis.
Like vividula, this ant may, under appropriate condi-
tions, produce alates at any time of year. I have records
of alates and alate pupae in nest series from south Florida
in December (Highlands County) and May (Dade County), and
I have alates from North Carolina collected in April and
September. In the Gainesville, Florida, area, these ants
seem not to raise alates during the cool months; the peak
rearing of sexual brood occurs March-July.
I believe, based on the collections of Van Pelt (FSCA)
and myself in relatively pristine habitats, that concinna
was originally a species of marshy areas and of openings
in flatwoods and swamp. In such habitats, it nests in
grass tussocks and in pieces of rotten wood that have
fallen from nearby wooded areas. Today, it probably is
more readily found in pastures and lawns of high water-table
areas and along ditches and drainage canals where it nests
in tussocks, cow dung and discarded wood. I found one
colony at a laboratory complex in Gainesville, nesting in
leaf litter beneath shrubbery. Vividula is abundant at
this site, which is adjacent to a low pasture. A collec-
tion from xericc mixed woods" is Seminole County, Georgia
(LACM), is unusual and possibly mislabeled.
(Figs. 1, 2, 12, 15, 42; Map-Fig. 50)
Prenolepis arenivaga var. faisonensis Forel, 1922.
Rev. Suisse Zool. 30: 98. >. Type loc. Faison, NC.
Paratrechina vividula: Creighton, 1950. Bull. Mus.
Comp. Zool. Harvard 104: 408 (in part?); Nuhn and Wright,
1979, Amer. Mid. Nat. 102: 353-362; Thompson et al., 1979,
Psyche 86: 321-325.
Paratrechina melanderi: Lynch, Balinsky and Vail,
1980, Ecol. Ent. 5: 353-371; Lynch, 1981, Oikos 37: 183-198.
Worker. TL usually 2.0-2.5, CI usually 83-87, OI
usually 20-23. A deciduous woodland species of the south-
east U.S. north to the New Jersey Pine Barrens. Brown with
tan to whitish middle and hind coxae. Head narrow, CI
usually 83-87. Eye small, weakly convex. Cephalic pubes-
cence dense, partly obscuring sheen of head. Erect macro-
chaetae on scape conspicuous, almost as long as width of
scape. Thorax and propodeum normally completely lacking
pubescence, very shiny. SM 8-13.
Male. TL usually 2.0-2.25, OI usually 31-33. Color
as in worker. Head narrow due to small, weakly convex eyes.
CI usually around 90 (94+ in sympatric species). Scape
macrochaetae conspicuous, MS 4-7.
Queen. Small eyes, light coxae and very shiny in-
tegument of head and thorax (visible even through dense
pubescence when specimens are clean) distinguish this
species from vividula, concinna and arenivaga.
Lectotype worker. TL 2.10, HL 0.59, HW 0.48, SL 0.69,
EL 0.13, PW 0.35, MCL 0.16, WL 0.71, FL 0.53, GL 0.81, SM 9,
PM 3, MM 2, CI 83, OI 22, SI 117, FI 90.
In full face view, clypeus subtrapezoidal, anterior
border truncate, its straight portion about half as wide as
its sinuate rear border. Sides of head rounded, slightly
convergent anteriad, widest just behind eyes. Middle half
of rear border of head nearly straight except for narrow
median concavity. Head, exclusive of clypeus, longer than
broad, narrower than that of related species. Eyes small,
weakly convex separated from the sides by more than 1/2 the
eyes' width, and separated from mandibular insertions by
about 1 1/2X EL. Scapes a little longer than average for
the species, evenly and very slightly curved for about the
basal 85% of their length, a little more strongly curved
distally. No ocelli visible.
Pronotal profile an even, very weakly convex upward
slope toward the rear. Promesonotal suture feebly impressed.
Mesonotal profile almost flat, parallel to long axis of
thorax and lacking an anterior face, thus yielding a hemi-
parabolic profile together with the pronotum.
Posterior portion of mesonotal profile bearing a small
setigerous swelling, followed by a posterior declivity.
Proprodeum partly crushed on the left side but in profile
appears evenly rounded like that of vividula. The angle
between its base and its rearmost portion near the petiole
weakly obtuse. Legs of usual length for vividula group.
Petiole partly imbedded in glue, appears cuneate, but
more rounded than in vividula. In posterior view, crest
is an even curve.
Cephalic pilosity as in vividula; the pubescence fine,
more densely and more uniformly distributed over entire head
but difficult to see in this callow specimen. Scape macro-
chaetae brown, suberect, a little longer and stouter than
those of vividula and quite visible. Thoracic macrochaetae
dark brown, gently and evenly curved mesad. Pilosity of
gaster and legs just as in vividula. Thorax and abdomen
lacking appressed pubescence.
Head and thorax shiny (in the former case even through
the pubescence). Gaster delicately shagreened.
Head and gaster light brown, thorax, legs and scapes
a shade lighter, middle and hind coxae very pale yellow.
Worker. TL 1.83-3.00, HL 0.53-0.69, HW 0.44-0.61,
SL 0.63-0.80, EL 0.12-0.16, PW 0.29-0.44, MCL 0.14-0.20,
WL 0.65-0.89, FL 0.47-0.65, GL 0.57-1.49, SM 8-13, PM 2-6,
MM 2-5, CI 81-90, OI 20-24, SI 112-119, FI 87-97, (n=40).
TL is typically less than 2.5; the 3.0 specimen's length
largely due to its very distended gaster. CI is usually less
than 87; only one specimen had CI 90.
The clypeal shape described in lectotype is common,
but the anterior border may range from arcuate and virtually
entire to having two straight sides with a narrow concave
emargination. The eyes of some specimens are larger and
more convex, approximating the condition of related species
(see discussion). A vestigial median ocellus is rarely
present. The rear border of the head of larger specimens
in full face view usually has a broader emargination than
the type (and most other smaller specimens), yielding a
weakly subcordate appearance. The widest portion of head
is directly across eyes in many larger specimens. The
promesonotal profile is usually hemiparabolic as described
in type, but often the parabola is broken by a short, steep
anterior mesonotal face. Rarely, the pronotum is very
obtusely subangular. The sides of the petiole appear to
diverge a little more toward the crest than in vividula
(the latter feature not visible in lectotype).
Pilosity is often a little more abundant than on the
type, particularly on the clypeus and promesonotum. In the
latter case, the eight major macrochaetae are 2X or more as
long as the six or eight adjacent ancillary hairs, and a
few still shorter hairs may be present. Some large speci-
mens may have a small amount of pubescence along the anter-
ior edge of the propodeum, or rarely on the anteromedian
portion of the pronotum.
The color of fully sclerotized individuals varies from
nearly uniform dark brown to dark yellowish-brown. The
middle and hind coxae are less contrastingly pale in the
lighter specimens. Occasionally, the thorax and, less often,
the head are a little lighter than the gaster. The color de-
scribed for the type is typical of allows. Many darker
specimens may have some faint bluish reflections on the
Description and Variation
Male. (Composite description, no males in syntype
series. Two males from my collection have been deposited
at MHNG. I have not formally designated them as allopara-
types, since they are not from the type locality.)
TL 1.96-2.24, HL 0.52-0.57, HW 0.46-0.51, SL 0.61-0.65,
EL 0.16-0.18, WL 0.73-0.83, GL 0.65-0.93, SM 4-7, CI 87-92,
01 31-35, SI 113-123, (n=8).
Anterior border of clypeus arcuate, often with a shal-
low median concavity or truncation. Rear border of clypeus
more arcuate than worker's, yielding a lenticular general
form to the clypeus in full face view. Median portion of
clypeus rounded, subumbonate. In full face view, rear bor-
der of head straight with rounded corners. Mandibles with
smooth masticatory border, apical tooth often only indis-
tinctly set off from this border. Eyes smaller and less con-
vex than any other sympatric Paratrechina in full face view.
Eyes less than 1/3 the length of the head, and in some spec-
imens, barely protruding beyond the sides of the head. Head
narrower than in sympatric species.
Petiole shaped as in worker but broader. Genitalia
(Fig. 45) have the form that I consider most generalized
among nearctic species. Parameres in side view triangular,
about 2X as long as broad. In rear view, parameres weakly
curved mesad over their distal half. Volsella weakly
sclerotized,of simple structure; the digitus resembling
the boomerang shape of that of vividula but shorter, less
strongly bent and less tapered distally; cuspis thumb-like
in appearance and not convex in dorsal aspect, appearing
more truncate than rounded distally. Aedeagus resembling
paramere in shape but narrower at the base and less taper-
ing, less sclerotized and slightly decurved near the tip
(Fig. 45 does not show this clearly because it is a more
dorsal view). In rear view, the aedeagal lobes are nor-
mally closely appressed in preserved specimens and are
about equal in length to the parameres.
Vestiture, color and sculpture almost like those of
the workers with the usual differences in the vestiture of
the scapes and thorax. In colonies with fully sclerotized
workers at the lighter end of the color range, the males
tend to be lighter, as well. Parameres with about 25 de-
curved brown hairs, like those of vividula, except for their
Syntype material (MHNG) consists of nine workers on
three pins collected in a forest at Faison, North Carolina,