\ ocjr ^ m
niliSNrNVINOSHimS S3 I dVd a n"LI B RAR I ES ^SMITHSON IAN ~ I NSTITUTlOi
1 1 | | °
IAR I ES^SMITHSONIAN^ INSTITUTION W NO I XfllllSNI^NVI NOS H1IWS10 S3 I avaai
_ _ co 5 ^ ^ ^ ._ . to
S2^X 5x5 . m Xs^2^>\ w y<^E*T
QC
<
a:
co
Sim:
° _ UllSNI^NVINOSHIIINS S3 I MVd ail^LIBRAR I ES^SMITHSONIAN^INSTITUTI!
*. > p
m X^osv^x1
ARIES $MITHSONIAN~INSTITUTION%OlinXIXSNI~NV!NOSHJJINS S3ldVd9
^ Z t CO 2 CO
~ E i
”j AWWl 2 H »4. 2 H
HI I fJLJI I xW l i
s X^|S 5 •* s ''^SSro^f'' >’P
A' 2 ^ ** 2 CO 2
1XIXSNI_ NVINOSHXIINS S3 I HVH8 11 LIBRARIES SMITHSONIAN INSTITUTI to ™ in
CK
< w
O ^W' 2 5 “ 5
2 oJ 2 awl 2
[ARIES SMITHSONIAN INSTITUTION NOlXflXIXSNI NVINOSHXIWS S3 1 hVUi 2 r- z i” . 2
S5^ 52 'W m 2 n,
CO £ CO
nXIXSNI NVINOSHXIWS S3IBVd9IT LIBRARIES SMITHSONIAN institut
2 ^ 5 X*’ ^ 2 «v
.£ 5 . E £ m&vi
z ApwA -H . a/
_ > w ^
RAR I ES^SMITHSONIAN^ INSTITUTION NOIXOXIXSNI NVINOSHXIWS^SS I d Vd I
to 5 <o __ = co
,w /^S^\ « V> w
nXIXSNI NVINOSHXIWS S3ldVdan LIBRARIES SMITHSONIAN INSTITUT r- , 2 r- z ^
m ^ m
D1IJLSNI NVINOSH1IKIS S3 I d Vd 8 1 1 L I B R AR I ES^SMITHSONIAN ~INSTITUTI( 2 : co z ,y.. co z
~ 5 . ' v ~ < .*<S£L-
o ° i x
! S &&Kr x ^ ' 8
__ > ^ g '\ >
JAR I ES ^SMITHSON IAN INSTITUTION NOlifUllSNI NVINQSHIHMS^SS IdVd 8 co 5 co __ = co
° -\
niliSNI^NVINOSHlIINS S3 I dVd 8 11 LIB RAR I ES^SMITHSONIAN^INSTITUTIC
m ^ XjvAst^X m ^ m
JAR I ESC/5SMITHSONIAN’”lNSTITUTIONt/5NOIinillSNI~NVINOSHlllNS S3 I d Vd 8 CO __ z co e CO
~ ^ ^ 5 <
x '^>\ o X fMWSPi O /Sffl’''%\ x
co 1 co co c o fefe 3<|
#!%#=
> XjVAS^X ^ >, ' **£ xiJ'Vost^' >
2 ^O X CO z
fUllSNI NVIN0SH1IINS S3 1 dVd0 11 LIBRARIES SMITHSONIAN INSTITUTIC - co _ - co
VARIES SMITHSONIAN INSTITUTION NOlifUllSNI NVIN0SH1IWS S3ldVd8
os
70 >
70
IFStSy J2 v5iP*' m M -
.fUliSNI NVINOSHilWS S3 I yVU8n~ LIBRAR I E S ^SMITHSONIAN “ I NSTITUTI z ^ w z w z ,v-
< -«5^x 5 , . ,*1 ..v^S'x ?
hi m -*w:£w v> mm%>. s> i*s? « wk"\
s ^i±LS>^ > iqr- 2 "\ § xyui^qx 2
RAR I ES^SMITHSONIAN INSTITUTION NOU.rUIJ.SNI NVINOSHilWS^Sa I d VB S
CO ~ CO — CO
^as^x 141 ^va^Tx ^ -^Ov .> uj
CO
.ni!lSNl“lNVIN0SHimS2S3 I dVd a ll^LIB RAR I ES^SMITHSONIAN^INSTITUTI
PSYCHE
A Journal of Entomology
Volume 92 1985
Editorial Board
F. M. Carpenter, Editor H. W. Levi
W. L. Brown, Jr. M. D. Bowers
|
E. O. Wilson |
J. M. Carpenter B. K. Holldobler |
Published Quarterly by the Cambridge Entomological Club Editorial Office: Biological Laboratories 16 Divinity Avenue Cambridge, Massachusetts, U.S.A.
The numbers of Psyche issued during the past year were mailed on the following dates:
Vol. 91, nos. 3-4, April 22, 1985
Vol. 92, no. 1, June 28, 1985
Vol. 92, nos. 3-4, November 28, 1985
ISSN 00
A> -//
^ — #6/ F97^
&d'-
PSYCHE
A JOURNAL OF ENTOMOLOGY
founded in 1874 by the Cambridge Entomological Club
Vol. 92 1985 No. 1
Ub r arses
CONTENTS
Ants of the Dominican amber (Hymenoptera: Formicidae). 1. Two new myr-
micine genera and an aberrant Pheidole. Edward O. Wilson 1
Ants of the Dominican amber (Hymenoptera: Formicidae). 2. The first fossil
army ants. Edward O. Wilson II
Ants of the Dominican amber (Hymenoptera: Formicidae). 3. The subfamily
Dolichoderinae. Edw ard O. Wilson 17
Hostplant choice of Checkerspot larvae: Euphydrvas chalcedona, E. colon,
and hybrids (Lepidoptera: Nymphalidae). M. Deane Bowers 39
Micralymma marinum (Stroem) in North America: biological notes and new distributional records (Coleoptera (Staphylinidae). Margaret K.
Thayer 49
Distribution of inhibitory queen pheromone among virgin queens of an ant, Solenopsis invicta. A. Ann Sorensen, David J. C. Fletcher, and
S. Bradleigh Vinson 57
Spontaneous polydomy in laboratory colonies of the ant, Leptothorax cur-
vispinosus Mayr (Hymenoptera: Formicidae). Robin J. Stuart 71
Synonymy of Pseudoberis Enderlein with Nothontvia Loew, with notes on the genus and a key to the South American species (Diptera: Stratiomyidae).
Norman E. Woodley 83
Temperature preferences of four species of fire ants (Hymenoptera: Formicidae: Solenopsis ). James E. Cokendolpher and Oscar Francke 91
(continued on back cover)
CAMBRIDGE ENTOMOLOGICAL CLUB
Officers for 1984-1985
M. Deane Bowers Wayne Maddison Ardis Johnston Frank M. Carpenter Barbara L. Thorne Norman F. Carlin
EDITORIAL BOARD OF PSYCHE
F. M. CARPENTER, (Editor), Fisher Professor of Natural History, Emeritus, Harvard University
W. L. Brown, Jr., Professor of Entomology, Cornell University and Associate in Entomology, Museum of Comparative Zoology B. K. HOLLDOBLER, Professor of Biology, Harvard University H. W. LEVI, Alexander Agassiz Professor of Zoology, Harvard University M. D. Bowers, Assistant Professor of Biology, Harvard University Alfred F. Newton, Jr., Curatorial Associate in Entomology, Harvard University
E. O. WILSON, Baird Professor of Science, Harvard University J. M. CARPENTER, Assistant Professor of Biology, Harvard University
President
Vice-President
Secretary
Treasurer
Executive Committee
PSYCHE is published quarterly by the Cambridge Entomological Club, the issues appearing in March, June, September and December. Subscription price, per year, payable in advance: $15.00, domestic and foreign. Single copies, $4.00 Checks and remittances should be addressed to Treasurer, Cambridge Entomological Club, 16 Divinity Avenue, Cambridge, Mass. 02138.
Orders for missing numbers, notices of change of address, etc., should be sent to the Editorial Office of Psyche, 16 Divinity Avenue, Cambridge, Mass. 02138. For previous volumes, see notice on inside back cover.
IMPORTANT NOTICE TO CONTRIBUTORS Manuscripts intended for publication should be addressed to Professor F. M. Carpenter, Biological Laboratories, Harvard University, Cambridge, Mass. 02138.
Authors are required to bear part of the printing costs, at the rate of $29.00 per printed page. The actual cost of preparing cuts for all illustrations must be borne by contributors: the cost for full page plates from line drawings is ordinarily $10.00 each, and for full page half-tones, $12.00 each; smaller sizes in proportion. There is ordinarily no additional charge for setting tables of less than six columns; for tables of six or more columns the cost is $25.00 per page.
Psyche, vol. 91, no. 3-4, for 1984, was mailed April 22, 1985
Ihf I exiii^lon I’lvss. Inc.. I cxm{!lon. Massachusetts
PSYCHE
Vol. 92
1985
No. 1
ANTS OF THE DOMINICAN AMBER (HYMENOPTERA: FORMICIDAE).
I. TWO NEW MYRMICINE GENERA AND AN ABERRANT PHEIDOLE
By Edward O. Wilson
Museum of Comparative Zoology, Harvard University, Cambridge, Massachusetts 02138, U.S.A.
Ants rival dipterans as the most abundant fossils in the Domini- can Republic amber. Since they are also phylogenetically compact and relatively easily identified, these insects offer an excellent opportunity to study dispersal and evolution in a Tertiary West Indian fauna.
The age of the Dominican amber has not yet been determined, but combined stratigraphic and foraminiferan analyses of its matrix suggest an origin at least as far back as the early Miocene (Saunders in Baroni Urbani and Saunders, 1982). I am inclined to favor this minimal age (about 20 million years) or at most a late Oligocene origin, for the following reason. In a sample of 596 amber pieces containing an estimated 1,248 ants that I recently examined (439 now deposited in the Museum of Comparative Zoology), I found 36 genera and well-defined subgenera, to which may be added one other, Trachymyrmex, reported earlier by Baroni Urbani (1980a). Of these 37 taxa only three, or 8%, are unknown from the living world fauna (see Table 1). The relative contemporaneity of the Dominican amber ants contrasts with that of the Baltic amber, which is Eocene to early Oligocene in age (Larsson, 1978) and pos- sesses 44% extinct genera; that is, 19 of the 43 genera recorded by Wheeler (1914) are unknown among living ants. The Dominican amber ants also differ to a similar degree from those of the Floris- sant, Colorado, shales, which are upper Oligocene in age and con- tain 8 of 20, or 40%, extinct genera (Carpenter, 1930).
2
Psyche
[Vol. 92
Table 1 . List of ant genera and well-defined subgenera known from the Domini- can amber. CBU: recorded by Baroni Urbani (I980a-d) and Baroni Urbani and Saunders (1982). CBIJ/EOW: recorded independently by both Baroni Urbani (I980a-d; and with Saunders, 1982) and E. O. Wilson (hitherto unpublished). Generic names without accompanying initials represent determinations by the author and are recorded here for the first time. (*) unknown in modern faunas.
|
Subfamily Ponerinae |
Oligomyrmex |
|
Anochetus (CBU / EOW) |
* Oxvidris, new genus |
|
Cylindromvrmex |
Paracryptocerus (CBU / EOW) |
|
Gnamptogenys (CBU/ EOW) |
Pheidole |
|
Hvpoponera |
New genus, near Rogeria |
|
Odontomachus |
Smithistruma |
|
Paraponera |
Solenopsis ( Solenopsis ) |
|
Platvthyrea |
Solenopsis ( Diplorhoptrum) |
|
Prionopelta |
Trachymyrmex (CBU) |
|
Trachymesopus |
[Zacrvptocerus: see Paracryptocerus ] |
|
Subfamily Dorylinae |
|
|
Neivanivrmex |
Subfamily Dolichoderinae |
|
Azteca |
|
|
Subfamily Pseudomyrmecinae |
Dolichoderus |
|
Pseudomvrmex (CBU/ EOW) |
Hypoclinea Iridomyrmex |
|
Subfamily Myrmicinae |
[Leptomyrmex = Camponotus'h |
|
Aphaeno^aster |
CBU / EOW] |
|
Crematogaster ( Acrocoelia ) |
Monads (CBU/ EOW) |
|
Cremaiogaster ( Orthocrema ) Cyphomyrmex Erebomyrma |
Tapinoma |
|
* Ilemomyrmex, new genus |
Subfamily Formicinae |
|
Leptothorax ( Macromischa ) |
Camponotus |
|
Leptothorax ( Nesomvrmex) |
Paratrechina |
|
Octostruma |
Prenolepis |
The purpose of this first article of a planned series on the Domini- can fauna is to describe the three most distinctive new species encountered in any collection known to me: two new myrmicine genera ( Ilemomyrmex , Oxvidris ) and a remarkable pheidoline which I have provisionally placed in the genus Pheidole.
Ilemomyrmex, new genus
Diagnosis {worker). Small, eyeless myrmicine possessing the fol- lowing distinctive combination of features: large, flaring frontal lobes that are raised well above the antennal insertions and cover most of the clypeus in full-face view; shallow antennal scrobes with
1985]
Wilson — Dominican amber ants. 1
3
posterior margins curving laterally to embrace the ends of the scapes; paired clypeal carinae close together and projecting beyond the remainder of the anterior clypeal margin to form a short concav- ity between them; narrow, 3- or 4-toothed mandibles (apical region indistinct in the single specimen available); and 12-segmented antennae with 3-jointed clubs.
Queen ( tentative association)". Overall similar to worker, except that frontal lobes extend only part way over clypeus; mandibles are 5-toothed; and eyes and ocelli are well developed. (From Gr. eilema, envelope; and Gr. mvrmex, ant).
Type species: Ilemomyrmex caecus.
Ilemomyrmex caecus, new species (Figs. 1, 2)
Diagnosis [worker). Distinguished from all other known ant species by the combination of traits cited above for Ilemomyrmex. In addition, possessing a robust alitrunk with thick, triangular propodeal spines; and short, thick petiole and postpetiole, the latter with an acute, forward-projecting ventral spine.
Holotvpe worker. Head Width 0.51 mm. Head Length 0.58 mm. Scape Length 0.44 mm. Head coarsely rugoreticulate and completely opaque, the rugae near the rims of the antennal scrobes parallel to one another and following the contours of the rims. Entire alitrunk and waist similarly rugoreticulate and opaque, but the gaster is nearly smooth and is feebly shining to subopaque. Color (which may not have remained true in the fossil state) dark reddish brown.
Queen ( tentative association). Winged. Differing from worker as described in generic diagnosis. Head Width (across and including eyes) 0.52 mm, Head Length 0.54 mm. Eye Length 0.16 mm.
Based on a single (holotype) worker and one alate queen in separ- ate pieces of Dominican amber; no further locality data. Both spec- imens have been deposited in the Museum of Comparative Zoology.
Ilemomyrmex resembles the Old World, principally African gen- era Calyptomvrmex and Dicroaspis in antennal form and the pecul- iar shape of the frontal lobes. However, it differs from them in the following important respects: its mandibles are narrower, with fewer teeth (5 or more in Calyptomvrmex and Dicroaspis ); its antennal scrobe is much shallower; its subpostpetiolar process is better devel- oped; its head is narrower and overall less modified from the primi-
4
Psyche
[Vol. 92
Fig. !. Ilcmoniyrme.x coccus. Holotype worker: A, frontal view of head; B, oblique rear view of head; C, side view of body.
Fig. 2. Ilemomyrme.x coccus. Frontal view of head of queen provisionally placed in this species.
1985]
Wilson — Dominican amber ants. 1
5
tive myrmicine shape; and its hairs are thinner and less uniform and regularly distributed (furthermore, in Calyptomyrmex and Dicro- aspis the hairs are blunt-tipped or, in the case of most species, spatulate or scale-shaped). In addition, Dicroaspis has 11- segmented antennae. 1 am inclined to regard the resemblance in frontal lobe shape between llemomyrmex and the two African gen- era as having arisen by convergent evolution.
Oxyidris, new genus
Diagnosis {worker). A very small myrmicine with closest overall resemblance to the South American genus Oxvepoecus, particularly in the general form of the antenna and waist; but differing in its lack of eyes, its 12-segmented antennae (1 1 in Oxvepoecus ), in its 3 (pos- sibly 4) mandibular teeth (4-5 in Oxvepoecus ), and in its unarmed propodeum (angular or spinous in Oxvepoecus). (From Gr. oxys, sharp, acute; and Gr. idris, wise one; also to note resemblance to Oxvepoecus ).
Type species: Oxyidris antiliana.
Oxyidris antiliana, new species (Fig. 3)
Diagnosis {worker). Distinguished from all known ant species by the combination of traits just described for Oxyidris.
Holotype worker. Head Width 0.36 mm. Head Length 0.45 mm, Scape Length 0.30 mm. Antenna 12-segmented with 3-jointed club. Head densely and evenly rugulo-punctate (rugulae with longitudinal orientation) and opaque. Alitrunk and waist densely and uniformly punctate, and opaque. Gaster shagreened, subopaque. Color (which may be altered in the fossil state) light reddish brown.
Dominican Republic: Palo Quemado Mine, Santiago Province.
Paratvpe workers. Six additional workers, one each in 6 amber pieces from Palo Quemado Mine.
Holotype and paratypes deposited in the Museum of Compara- tive Zoology.
Pheidole tethepa, new species (Figs. 4, 5)
Diagnosis {minor worker). An unusual pheidoline tentatively placed in Pheidole, differing from all known species of that genus by
6
Psyche
[Vol. 92
A
Fig. 3. Oxyiiiris antiliana. Holotype worker: A, frontal view of head; B, side view of body.
the bulging, more-than-hemispherical eyes, and proportionately very large mandibles. Pronotum armed with two well-developed spines, a trait shared with members of the Old World P. quadrispina group but not with any known living New World Pheidole species. (Gr. tethepa, amazed; referring to the eyes).
Holotype minor worker. Head Width exclusive of eyes 0.76 mm, Pronotal Width 0.43 mm. Eyes with approximately 30 ommatidia. Head sparsely rugose to rugoreticulate with predominantly longi- tudinal orientation. Pronotum with several transverse rugae. Rest of dorsal surface of alitrunk evenly shagreened and subopaque.
Paratvpe minor workers. Two individuals poorly preserved but clearly sharing the diagnostic traits of the holotype.
Holotype and paratypes in a single amber piece from La Toca Mine, Dominican Republic.
Discussion
Are llemomyrmex and Oxyidris really extinct? If so, they are extreme exceptions in the generic ranks of the Dominican amber ants. It may be significant that both are small, eyeless, and possess narrow, sharp-toothed mandibles. In addition, llemomyrmex is dis- tinguished by expanded frontal lobes and scrobes that together can mostly cover the antennae. In the living fauna these traits are charac- teristic of cryptobiotic, often scarce myrmicine ants that are among the last to be collected and recognized. Examples of such living genera that have been recently discovered or at least recognized as
1985]
Wilson — Dominican amber ants. 1
7
Fig. 4. Pheiciole tethepa. Holotype worker: A, dorsal view of head and body; B, frontal view of head and pronotum.
Fig. 5. Pheiciole tethepa. Paratype worker from holotype amber piece, showing different view of body and head.
8
Psyche
[Vol. 92
higher taxa in the New World tropics are Creightonidris (1949), Dorisidris (1948), Phalacromvrmex (1960), Protalaridris (1980), and Taiuidris (1967). Hence it is entirely possible that contempo- raneous species of Ilemomvrmex and Oxyidris may yet be disco- vered, although not necessarily in the West Indies.
Pheidole tethepa has been placed in the genus Pheido/e as a pro- visional measure. It may well represent a phyletic line sufficiently divergent from other members of the tribe Pheidolini to justify generic rank. The exophthalmic eyes and proportionately large mandibles are unique within Pheidole on a worldwide basis, while the pronotal spines were almost certainly derived independently from the Old World P. quadrispina group. More material is clearly needed to resolve the matter. In particular, the demonstration of a large-headed major caste (if one exists) would give added reason to retain tethepa in Pheidole.
Acknowledgments
I am grateful to Cesare Baroni Urbani, Barry Bolton, and Wil- liam L. Brown for their advice on the three new species described here. The research was supported by National Science Foundation grant no. BSR 81-19350.
Literature Cited
Baroni Urbani, C.
1980a. First description of fossil gardening ants (Amber Collection Stuttgart and Natural History Museum Basel: Hymenoptera, Formicidae. I: Attini). Stuttgarter Beitr. Naturk., B (Geol. Palaontol.), 54: 1 13.
1980b. Anochetus corayi n. sp., the first fossil Odontomachiti ant (Amber Col- lection Stuttgart: Hymenoptera, Formicidae. II: Odontomachiti). Stutt- garter Beitr. Naturk., B (Geol. Palaontol.), 55: I 6.
1980c. The first fossil species of the Australian ant genus Leptonivrmex in amber from the Dominican Republic (Amber Collection Stuttgart: Hymenoptera, Formicidae. Ill: Leptomyrmicini). Stuttgarter Beitr. Naturk., B (Geol. Palaontol.), 62: I 10.
I980d. The ant genus Gnamptoyenys in Dominican amber (Amber Collection Stuttgart: Hymenoptera, Formicidae. IV: Ectatommini). Stuttgarter Beitr. Naturk., B (Geol. Palaontol.), 67: I 10.
Baroni Urbani, C. and J. B. Sai ndi rs.
1982. The fauna of the Dominican Republic amber: The present status of knowledge. Trans. 9th Caribbean Geol. Conference, Santo Domingo, Dorn. Rep., 1980 (1982), 1: 213 223.
1985]
Wilson — Dominican amber ants. 1
9
Carim nti r, F. M.
1930. The fossil ants of North America. Bull. Mus. Comp. Zool. Harvard, 70: 1 66, I 1 pi. l.ARSSON, S. G.
1978. Entomonograph ( Klampenborg, Denmark), 1 : 1 192.
Wm i i.i r, W. M.
1914. The ants of the Baltic amber. Schrift. Physikal.-Okon. Ges. Konigsberg, 55: I 142.
ANTS OF THE DOMINICAN AMBER (HYMENOPTERA: FORMICIDAE). 2. THE FIRST FOSSIL ARMY ANTS
By Edward O. Wilson
Museum of Comparative Zoology, Harvard University, Cambridge, Massachusetts 02138, U.S.A.
Despite the local abundance of the contemporary army ants, comprising about 105 species in the Old World Dorylinae and 147 species in the New World Ecitoninae (Gotwald, 1979), no fossil remains have hitherto been recovered. In the course of studying a large collection of Dominican amber ants newly assembled in the Museum of Comparative Zoology (see also Wilson, 1985), I encoun- tered two well-preserved workers of an apparently extinct species belonging to the New World genus Neivamvrmex. Their status as the first ecitonine fossils, extending the history of the subfamily back at least as far as the early Miocene, deserves special notice. The discovery also has potential biogeographic significance, because no living species of Neivamvrmex or any other ecitonine is known from the Greater Antilles.
Neivamyrmex ectopus Wilson, new species (Figs. 1, 2)
Diagnosis (based principally on the holotype). A medium-sized species (Head Width 0.4-0. 6 mm) characterized uniquely by the following combination of traits: sides of head parallel or nearly so for most of their length; antennal scapes relatively slender (thicker in the paratype), approaching the occipital angles to within a dis- tance a little less than the maximum scape width; occipital border moderately concave, the lateral angles well defined; dorsal and pos- terior (declivitous) borders of propodeum forming a small, strongly convex but not angulate juncture; petiolar node symmetric, with a well-defined anterior peduncle; the subpetiolar process small, limited to the anterior petiolar border, and projecting forward; body mostly covered with comparatively sparse, semierect pilosity. Head, petiolar node, legs, scapes, and gaster smooth to weakly shagreened and feebly shining. Color dark reddish brown, although this may be an artifact of preservation.
11
12
Psyche
[Vol. 92
Fig. I. Neivumyrme.x ectopus n. sp. holotypc worker: frontal view of head (left) and side view of propodeum, waist, and first gastrie segment (right).
Fig. 2. N. ectopus n. sp. paratype worker: view of entire body. (Photograph courtesy of F. M. Carpenter).
1985]
Wilson — Dominican amber ants. 2
13
Holotype and paratvpe workers. Well-preserved specimens in two separate amber pieces purchased from separate sources in the Dominican Republic. The material enclosing the ants is typical in quality and inclusions of Dominican amber, and the two pieces were in local collections containing other ant species previously estab- lished as typical of the Dominican amber fauna. Although the exact provenance of the specimens is unknown, Mr. J. Brodzinsky of Amberica Inc. states {in litt.) that “90-95%” of the material in which the Neivamyrmex ectopus holotype originated came from the Palo Alto and La Toca mines, near Santiago. The two types have been deposited in the Museum of Comparative Zoology, Harvard Uni- versity, among the contemporary Neivamyrmex species.
Holotype (Fig. 1): Head Width 0.62 mm, Head Length (from the level of the center of the occiput to the level of the center of the anterior clypeal border) 0.72 mm. Scape Length 0.53 mm. Paratvpe (Fig. 2): Head Width 0.42 mm. Scape Length 0.30 mm, Pronotal Width 0.28 mm. The paratype is smaller and has proportionately shorter and thicker scapes, but otherwise corresponds well to the holotype.
Relationships
In order to interpret the status and possible phylogenetic relation- ships of the amber species, I considered all of the 61 contemporary Neivamyrmex species known from the worker caste. Of these, 51 were examined directly from type and reliably determined series, and the remaining 10 were characterized in the essential characters from the reviews by Borgmeier (1955) and Watkins (1976, 1977, 1982).
The following 8 characters visible in N. ectopus were found to vary considerably among the 61 contemporary species: size, shape and length of the scape; shape of the head, especially the occipital border; shape of the alitrunk, especially the profile of the propo- deum; shape of the petiole, including the subpetiolar process; shape of the postpetiole; body sculpturing; and overall pilosity. Six species were found to resemble ectopus closely or identically in 6 of the characters (no species matched in 7 or all 8). These are listed below, along with an indication of the two traits in which they differ from ectopus :
14
Psyche
[Vol. 92
N. agi/is. Propodeum more flattened in side view, the basal and posterior faces merging as a single smoothly convex curve; pilosity denser.
N. californicus. Petiolar node more flattened in side view; body pilosity denser.
N. emersoni. Basal and posterior faces of propodeum meeting at only slightly more than right angle; pilosity denser.
N. fallax. Antennal scapes thicker and shorter relative to head capsule; subpetiolar process slightly thicker and pointing straight down (instead of forward as in ectopus).
N. manni. Occipital border flat; pilosity denser.
N. melanocephalus. Occipital border distinctly less concave; pilosity denser.
Of these 6 species, no fewer than 5 have ranges predominantly in Mexico and the southern United States, the northern portion of the range of Neivamyrmex. Only one ( emersoni ) is limited to the south- ern part of the total generic range, in this case South America and Trinidad. Yet only 21 of the 61 Neivamyrmex species occur in the northern portion; the remaining are so far as known limited to South and Central America. Put another way, ectopus closely approaches about 20% of the known northern species but only about 2.5% of the southern species.
Discussion
The exact age of the Dominican amber is unknown, but most writers agree that it is either late Oligocene or early Miocene. The material originates from various mines in two principal areas: in the north of the Dominican Republic between Santiago and Puerto Plata, and in the east between Santo Domingo and the Bahia de Samana. The age of the many deposits in which insects occur no doubt varies. Foraminiferal fossils associated with amber from mines near Palo Alto are characteristic of the lower Miocene (Baroni-Urbani and Saunders, 1982). I would guess a relative youth for most of the ant fossils I have seen, because the great majority belong to modern genera.
Regardless of its exact age, the discovery of a Neivamyrmex in Dominican amber has considerable significance for the interpreta-
1985]
Wilson — Dominican amber ants. 2
15
tion of West Indian biogeography. Army ants are among the least vagile of all social insects. Among the Ecitoninae of the New World, Neivamyrmex nigrescens has been recorded from the Islas Marias, 100 km off the Mexican Pacific Coast, while A. klugi occurs on St. Vincent in the Lesser Antilles. No ecitonine is known farther away from the mainland, either from the northern arc of the Lesser Antilles or any of the Greater Antilles. Similarly, the Old World Dorylinae, represented by Aenictus, extends only as far east as the Philippine Islands, New Guinea, and Queensland (Wilson, 1964). It is wholly unknown from those portions of Micronesia and Polyne- sia that support a native ant fauna (Wilson and Taylor, 1967). The farthest outlier in the western part of the range is a population of A. fergusoni on Great Nicobar Island, 160 km from Sumatra. The existence of A. ectopus is therefore consistent with the common view based upon both geological and paleobotanical studies (Gra- ham and Jarzen, 1969) that the ancestral Greater Antilles were larger and extended closer to the Mexican mainland during the middle and late Tertiary than is now the case.
Furthermore, the overall closer similarity of A. ectopus to con- temporary Mexican and United States species, although far from conclusive, is consistent with a closer approach of the Greater Antilles to Mexico than to the northern coast of South America during the Tertiary Period.
Summary
The first fossil army ant, Neivamyrmex ectopus, new species, is described from the Dominican amber, generally considered to be late Oligocene or early Miocene in age. Because of the extremely limited vagility of the Ecitoninae, and their apparent absence today from the West Indies north of St. Vincent, the presence of A. ecto- pus suggests a closer proximity of the Greater Antilles to the main- land during Tertiary times. Also, A. ectopus is phenotypically closer to species now living in Mexico and the southern United States than to the much richer Central and South American faunas.
Acknowledgments
I am grateful to F. M. Carpenter for making the photograph of the paratype worker and for advising me on techniques of preparing
16
Psyche
[Vol. 92
and studying amber inclusions, and to William L. Brown, William H. Gotwald, Jr., and Julian F. Watkins II for critical readings of the manuscript. The research was supported by grant no. BSR 81- 19350 from the National Science Foundation.
REFERENCES ClTED
Baroni-Urbani, C. and J. B. Saunders
1982. The fauna of the Dominican amber: the present status of knowledge. Trans. Ninth Caribbean Geol. Conf., Santo Domingo, Dom. Rep., 1: 213-223.
Borgmf.ier, T.
1955. Die Wanderameise der Neotropische Region. Studia Entomol., 3: 1-716.
Gotwald, W. H., Jr.
1979. Phylogenetic implications of army ant zoogeography (Hymenoptera: Formicidae). Ann. Ent. Soc. Amer., 72: 462-467.
Graham, A. and D. M. Jarzen
1969. Studies in Neotropical paleobotany. I. The Oligocene communities of Puerto Rico. Ann. Missouri Bot. Garden, 56: 308-357.
Watkins, J. F. II.
1976. The identification and distribution of New World army ants. Baylor Univ. Press, Waco, Texas, x + 102 pp.
1977. Neivamvrmex nvensis, n. sp. (Formicidae: Dorylinae) from Nye County, Nevada, U.S.A. Southwestern Naturalist, 22(4): 421-425.
1982. The army ants of Mexico (Hymenoptera: Formicidae: Ecitoninae). J. Kansas Ent. Soc., 55(2): 197-247.
Wilson, E. O.
1964. The true army ants of the Indo-Australian area (Hymenoptera: Formici- dae: Dorylinae). Pac. Insects, 6(3): 427-483.
1985. Ants of the Dominican amber (Hymenoptera: Formicidae). 1. Two new myrmicine genera and an aberrant Pheidole. Psyche, 92(1): 1-9.
Wilson, E. O. and R. W. Taylor
1967. The ants of Polynesia. Pac. Insects Monogr., 14: 1-109.
ANTS OF THE DOMINICAN AMBER (HYMENOPTERA: FORMICIDAE).
3. THE SUBFAMILY DOLICHODERINAE
By Edward O. Wilson
Museum of Comparative Zoology, Harvard University, Cambridge, Massachusetts 02138, U.S.A.
The Dolichoderinae, in particular the species of Azteca and Iri- domyrmex, are by a wide margin the most abundantly represented ants in the Dominican amber. They are also exceptional in the number of genera occurring in the amber but not in the modern fauna of the Greater Antilles. In other words the Dolichoderinae appear to have undergone a considerable amount of retreat entail- ing local extinction since early Miocene times, and for this reason alone the fossil species deserve close attention. In the present review I have relied principally on the large collections I have assembled for the Museum of Comparative Zoology during the past 10 years, supplemented substantially by material loaned by Dr. Robert E. Woodruff. All of the holotypes (except Dolichoderus dibolia) and most of the paratypes have been deposited in the Museum of Com- parative Zoology. The holotype of D. dibolia and a few paratypes have been placed in the Florida State Collection of Arthropods, Gainesville, Florida. Earlier parts of this series have dealt with extinct myrmicine genera and the ecitonine army ants respectively (Wilson, 1985a, b).
Dolichoderus
This genus of large, slender ants is composed today of 8 species limited to the moist tropical forests of South America. The colonies are mostly or exclusively arboreal. The species to be described below is clearly a member of the distinctive attelaboides group, which ranges from the Amazonian region of Bolivia north to Trinidad.
17
18
Psyche
[Vol. 92
Dolichoderus dibolia, new species (Fig. 1)
Diagnosis {worker). Similar to the living species D. attelaboides and D. imhecillus, particularly in its attenuated occipital “neck” and overall body shape (including especially the alitrunk and propodeal spines) but differs from both forms in its somewhat larger size; proportionately thicker mesothorax; lower, more rounded petiolar node (quadrate in side view in the case of attelaboides and imbecil- lus ), and much longer anterior petiolar peduncle.
The name dibolia, Greek for lance, refers to the strongly devel- oped propodeal spines.
Holotype worker. Length of alitrunk 3.7 mm. With moderately abundant, coarse, erect hairs, especially on the antennal scape. Sculpturing not determinable due to the obscuring of much of the body surface.
The genus Monads may eventually be fused with other blocks of the Dolichoderini (see Brown, 1973, who synonymizes it under Dolichoderus ), but until a thoroughgoing revision of the tribe is completed I regard it as both prudent and convenient to treat this genus as a separate entity. For the moment all the New World species placed in Monads (Kempf, 1959) are distinguished by their possession of angulate or spiny pronotal humeri and laterally mar-
Monacis
1 mm
Fig. I. Dolichoderus iliholia worker holotype, side view of body.
1985]
Wilson — Dominican amber ants. 3
19
ginate mesonotum. Moreover, all recognized Monads are Neotrop- ical. Living members of the genus are almost wholly limited to the mainland from southern South America to southern Mexico, and all evidently nest arboreally. M. bispinosa, the most widespread species, also occurs on Trinidad and has been recorded, quite anomalously, from St. Thomas in the Virgin Islands (Mayr, 1862). The latter record was based on a single queen and is almost certainly an error — or at most represents a population introduced by human commerce. No other Monads has been found in the remainder of the West Indies north of Trinidad, despite the fact that the workers of most of the species, including M. bispinosa, are large, conspicuous insects.
The two species described below from the Dominican amber are typical Monads not much different from two species ( rufescens and laminatus respectively) in the living fauna. They are of exceptional significance because of the current absence of the genus from the Greater Antilles.
Monads earibbaea, new species (Fig. 2)
Diagnosis {worker). A relatively small (Pronotal Width includ- ing humeral spines 0.49-0.66 mm), slender species resembling the modern M. rufescens of the Brazilian Amazon, differing from rufes- cens in its very sparse body pilosity and the more sharply angulate “shelf” separating the dorsal and declivitous (posterior) faces of the propodeum. M. earibbaea is also light reddish brown as opposed to light reddish yellow in rufescens, although its color might well have been altered during fossilization.
Holotype worker. Head Width 0.80 mm, Pronotal Width 0.51 mm. Alitrunk and petiole densely, finely, and evenly punctate, opaque; head still more finely punctate, grading to shagreened, and feebly shining. Body almost devoid of standing pilosity. Body mostly dark reddish brown, legs and parts of petiole and gaster light reddish brown.
Paratype workers. Eleven specimens in as many amber pieces: 2 from Palo Quemado (Pronotal Width of first 0.47 mm, second not measured), one from Bayaguana (no measurement), and 8 with no further locality within the Dominican Republic (Pronotal Width 0.49-0.66 mm).
20
Psyche
[Vol. 92
Fig. 2. Monads caribhaea worker holotype: A, side view; B, dorsal view.
The pronotal humeri of the caribbaea types vary considerably, from well defined spines as in the holotype to simple acute angles approaching the condition that characterizes the second fossil spe- cies, pro/aminata.
Monads prolaminata, new species (Fig. 3)
Diagnosis (worker). A relatively large, robust species closely resembling the living M. laminata of northern South America and
1985]
Wilson — Dominican amber ants. 3
21
Fig. 3. Monads prolaminata worker holotype: A, side view; B, dorsal view.
Central America, differing in the concave dorsal border of the petio- lar scale viewed in full face (as opposed to strongly convex in laminata).
Holotype worker. Head Width 1.07 mm, Pronotal Width 0.93 mm. Dorsum of alitrunk shallowly rugoreticulate, the rugae longi- tudinally oriented on the mesonotum; spaces within the rugoreticu- lar cells densely punctate; the whole surface entirely opaque but with reflections from the numerous evenly packed punctures giving the alitrunk a silvery, “sparkling” appearance. Head, petiole, and abdomen variably punctulate to shagreened and feebly shining. Body almost wholly devoid of standing pilosity. Body uniformly
22
Psyche
[Vol. 92
blackish brown, appendages light to medium brown. Dominican Republic amber: no further locality.
Paratvpe workers. Five specimens in 4 amber pieces: one from Palo Quemado (Pronotal Width 1.11 mm), 4 others from Domini- can amber with no further locality, 2 of which are in one piece (Pronotal Width 0.90-1.30 mm).
Hypoclinea
The modern genus Hypoclinea as presently defined is nearly cos- mopolitan (absent in Africa) and highly diverse in anatomy and behavior. At least 15 species compose the living Neotropical fauna, almost all confined to the mainland of Mexico, Central America, and South America. The most widespread form, H. lutosa, occurs from Brazil to Trinidad and southern Mexico. It has also been recorded from St. Vincent in the southern part of the Lesser Antilles, where it is quite rare (Forel, 1893). Hypoclinea is appar- ently absent in the remainder of the West Indies, including all of the Greater Antilles, a circumstance giving the Dominican amber spe- cies to be described below more than ordinary biogeographic importance.
Hypoclinea primitiva, new species (Fig. 4)
Diagnosis (worker). A medium-sized (Head Width 0.6-0. 7 mm), slender species with several primitive traits for Hypoclinea overall, including a relatively unmodified alitrunk, smoothly rounded propodeum, simple petiolar scale, and a “generalized” head shape with reference to the Dolichoderini in general. Closest in appearance to H. germaini of South America among living species; germaini differs from primitiva, however, in having a blunt, trans- verse ridge that separates the dorsal and declivitous faces of the propodeum, as well as more flattened pro- and mesonota (both of these traits are reasonably interpreted as having been derived in evolution).
The name primitiva alludes both to the antiquity of the species and to the set of traits just cited that are provisionally interpreted to be primitive within the genus and perhaps even in the Dolichoderi- nae as a whole.
1985]
Wilson — Dominican amber ants. 3
23
Fig. 4. Hypodinea primitive! worker holotype: A, side view oi body; B, dorsal view of body; C, head.
Holotype worker. Head Width 0.68 mm. Head Length 0.93 mm, Pronotal Width 0.56 mm. Scape Length 0.87 mm. Eye Length 0.26 mm. Posterior half of head bearing scattered foveae with the interspaces weakly shagreened and feebly shining. Rest of the body uniformly and lightly shagreened and feebly shining. Body uni- formly light reddish brown. Dominican Republic amber: no further locality.
Paratype workers. Four workers: 2 in one amber piece from La Toca, associated with Azteca alpha workers (Head Width of one 0.64 mm, other not measured); two others in separate pieces origi- nating from unknown localities (one with Head Width 0.66 mm, other not measured).
Azteca and Iridomyrmex
The large and complex genera Azteca and Iridomyrmex have proved difficult to separate in their entirety by means of external anatomy (a striking difference in the proventriculus makes the cleavage easier when this internal organ can be examined). The classification of fossil material is therefore difficult, since only external structures are usually preserved.
24
Psyche
[Vol. 92
The problem is further complicated by the separation of workers, queens, and males into different amber pieces, so that most linkages across castes and sexes can only be guessed. In the case of the Dominican amber, however, I was able to solve the problem in part by the discovery of three miniature “Rosetta stones”: 3 amber pieces with closely intermingled workers and males of the commonest spe- cies of Azteca (A. alpha). One of the pieces also contains a male pupa, further supporting the interpretation that the associated workers and males belonged to the same colony and were trapped during a colony emigration.
With the worker-male connection secured, I made comparisons with material representing 48 contemporary species of Azteca and 50 of Iridomyrmex (11 Neotropical, 39 Indo-Australian) in the Museum of Comparative Zoology collection, many with workers and males from the same nests. The following characters appear to hold with consistency:
1 . In Azteca workers the mesonotum is usually moderately con- vex in side view and hence does not form a smooth line with the pronotum; in New World Iridomyrtnex workers the mesonotum is only weakly convex in side view and forms a continuous line with the pronotum.
2. In Azteca workers the dorsal face of the propodeum is much less convex in side view than in Iridomyrmex workers.
3. In Azteca workers the petiolar node is longer than high (the reverse is true in Iridomyrmex) and inclined more strongly forward than in Iridomyrmex.
4. The occiput of Azteca workers is usually more deeply con- cave than in Iridomyrmex workers, although extreme species within