e #4

3L

6(

'n

103

JANUARY 1995

No. 1

Journal

of the

New York

Entomological Society

(ISSN 0028-7199)

Devoted to Entomology in General

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY

Editor: James M. Carpenter, Department of Entomology, American Mu-

seum of Natural History, Central Park West at 79th Street, New York, New York 10024

Book Review Editor: James S. Miller, Department of Entomology, Amer-

ican Museum of Natural History, Central Park West at 79th Street, New York, New York 10024

Publications Committee: James K. Liebherr, Cornell University, Ithaca,

Chairman; David L. Wagner, University of Connecticut, Storrs; Alfred G. Wheeler, Jr., Pennsylvania State Department of Agriculture, Hanis- burg.

The New York Entomological Society Incorporating The Brooklyn Entomological Society

President: Randall T. Schuh, Department of Entomology, American Mu-

seum of Natural History, New York, New York 10024 Vice President: Sule Oygur, Department of Entomology, American Mu-

seum of Natural History, New York, New York 10024 Secretary: Parker Gambino, Bronx High School of Science, Bronx, New

York 10468

Treasurer: Louis Sorkin, Department of Entomology, American Museum

of Natural History, New York, New York 10024 Trustees: Class of 1994 — David Alsop, Queens, New York; Cal Snyder,

New York, New York. Class of 1995 — Paul W. Ehlers, East Northport, New York; Eddie Scheibel, East Northport, New York.

Annual dues are $23.00 for established professionals with journal, $10.00 without journal, $15.00 for students with journal, $5.00 without journal. Sustaining memberships are $53.00 per year, institutional memberships are $125.00 per year, and life memberships are $300.00. Sub- scriptions are $45.00 per year domestic, $50.00 Mexico and Canada, $55.00 all other countries. All payments should be made to the Treasurer. Back issues of the Journal of the New' York Entomological Society, the Bulletin of the Brooklyn Entomological Society, and Entomologica Americana can be purchased from Lubrecht and Cramer, RD 1, Box 244, Forestburgh, New York 12777. The Torre-Bueno Glossary of Entomology can be purchased directly from the society at $45.00 per copy, postage paid.

Meetings of the Society are held on the third Tuesday of each month (except June through September) at 7 p.m. in the American Museum of Natural History, Central Park West at 79th Street, New York, New York.

Mailed January 12, 1996

The Journal oj the New York Entoniological Soeiety (ISSN 0028-7199) is published 4 times per year (January, April, July, October) tor the Society by Allen Press, Inc., 1041 New Hampshire, Lawrence. Kansas 66044. Second class postage paid at New York, New York and at additional mailing office. Postmaster: Send address changes to the New York Entomological Society, '% American Museum of Natural History, Central Park West at 79th Street, New York, New York 10024-6192. Known office of publication: American Museum of Natural History, New York, New York 10024.

Journal of the New York Entomological Society, total copies printed 700, paid circulation 602, mail subscription 602, free distribution by mail 19, total distribution 621, 79 copies left over each quarter.

0 This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper).

J. New York Entomol. Soc. 103(1): 1-14, 1995

• '-AM 2 6

STREPSIPTERA DO NOT SHARE HINDWING VENA^ SYNAPOMORPHIES WITH COLteO^^^J^^fES A REPLY TO KUKALOVA-PECK AND LA^

Michael F. Whiting and Jeyaraney Kathirithamby

Department of Entomology American Museum of Natural History, Central Park West at 79th Street,

New York, New York 10024, U.S.A. and Department of Zoology, South Parks Road, Oxford 0X1 3PS, U.K.

Abstract. — Kukalova-Peck (1991) and Kukalova-Peck and Lawrence (1993) proposed new characters to support a sister-group relationship between Strepsiptera and Coleoptera based on hind wing venation. Through the use of scanning electron microscopy (SEM) and light mi- croscopy, we have examined these putative synapomorphies in multiple strepsipteran taxa and find discrepancies between the author’s presentation of strepsipteran venation and those veins we could observe on the specimens themselves. We find that most of the authors’ putative synapomorphies are defined imprecisely and do not consist of discrete character states. While the authors have expressed their results in cladistic terminology, they have failed to use standard cladistic methodology in character evaluation. We object to the authors’ use of hypothetical groundplans for defining synapomorphy prior to formal cladistic analysis, the heavy reliance on evolutionary scenarios in phylogenetic inference, the lack of adequate outgroup comparison, and the absence of a simultaneous parsimony analysis of the character data. Based on obser- vational discrepancies and methodological improprieties, we conclude that the authors’ putative synapomorphies as currently constituted provide no evidence to support a sister-group relation- ship between Strepsiptera and Coleoptera.

Kukalova-Peck (1991) proposed new synapomorphies for Strepsiptera and Cole- optera based on hind wing venational characters. These characters were later revised, expanded, and presented in more detail in Kukalova-Peck and Lawrence (1993). Previous to this work, there has only been one character supporting a Coleoptera- Strepsiptera sister- group which has survived scrutiny: the ability to power flight with the hind wings (Kinzelbach, 1971, 1990; Kathirithamby, 1989; Kristensen, 1991) Because the phylogenetic position of Strepsiptera has remained one of the most controversial questions in insect ordinal phylogenetics (Kristensen, 1991), and few characters have been found for its placement among the insect orders, these putative synapomorphies are important and deserve closer scrutiny.

The thrust of Kukalova-Peck and Lawrence’s work was towards deciphering the phylogeny of Coleoptera using hind wing venation. They examined 200 specimens from 108 families of Coleoptera and three species of Strepsiptera — Mengenilla sp., Coriophagus rieki, and Lychnocolax sp. — though the actual specimen number was

' This manuscript was originally submitted to the Canadian Entomologist in June, 1994; the journal in which the Kukalova-Peck and Lawrence (1993) paper appeared. Due to unacceptably long delays required to get a final review of this manuscript and await a response from Ku- kalova-Peck prior to publication, it was subsequently withdrawn.

2

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

not mentioned. While the authors’ putative synapomorphies should be carefully eval- uated in the Coleoptera, Strepsiptera, and appropriate outgroups, we will restrict our specific criticisms to their presence and distribution in Strepsiptera. This is because synapomorphies are statements of shared, derived character state distributions. If it can be demonstrated that a certain character is unobservable or undefinable in the Strepsiptera, then regardless of its occurrence in Coleoptera, there is sufficient reason to reject it as evidence supporting a sister-group relationship between Strepsiptera and Coleoptera.

CHARACTER CRITICISMS

We use the following criteria to evaluate Kukalova-Peck and Lawrence’s putative synapomorphies: (1) precision of character and state definition e.g., relative size, position, fine structure; (2) whether the states can be defined discretely; (3) distri- bution of states throughout ingroup taxa; (4) observability of character states in extinct or extant taxa. We agree with Hennig (1966) that only synapomorphy con- stitutes evidence for monophyly and that symplesiomorphy is phylogenetically un- informative. We further concur with Farris (1990) that non-discrete characters are of little use in phylogenetic analysis because states cannot be objectively defined and state transformations cannot be unambiguously specified. We are therefore concerned that every venational character be defined in such a way as to make the determination of states objective when observing veins on specimens. We further insist that the states be observable in the taxa themselves because inferred states based on precon- ceived notions of venational evolution in hypothetical prototypes do not constitute prima facie evidence for phylogenetic inference.

We first provide specific evaluations for each synapomorphy presented by Kuka- lova-Peck (1991) and Kukalova-Peck and Lawrence (1993). Quotations of original character descriptions (in italics) are followed by the authors’ polarity designation. If the same character occurs in both publications, we include both descriptions and polarity designations. Then we evaluate the character and attempt to interpret the states based on the descriptions of the authors. This is followed by specific criticisms of the character and its distribution in Strepsiptera.

Character 1: A sclerotised and shortened ScP entering the pterostigma (Kukalova- Peck, 1991; synapomorphy); ScP ending after entering the pterostigma! radial cell (Kukalova-Peck and Lawrence, 1993; synapomorphy).

Interpretation: Kukalova-Peck and Lawrence treat C and Sc of Kinzelbach (1971) (Figs. 1-4) as two discrete veins: PC+C + ScA and ScP. According to the authors, ScP runs parallel to the anterior margin of RA and ends abruptly beyond the middle of the wing after entering the pterostigma in Strepsiptera and Coleoptera. They di- agrammed PC+C+ScA and ScP as distinct veins visible in the mesothoracic wings of Mengenilla, Coriophagus, and Lychnocolax (Kukalova-Peck and Lawrence, 1993: figs. 69-71). In the authors’ drawings of Mengenilla and Coriophagus, ScP runs into the strepsipteran “pterostigma” (the darkened region between RA,+2 and RA3+4); in the figure of Lychnocolax ScP is present but the pterostigma is absent. According to the authors’ brief description, the two states of this character are “ScP ending before entering the pterostigma” (plesiomorphy) and “ScP ending after entering the pter- ostigma” (apomorphy).

1995

STREPSIPTERA HIND WING VENATION

3

Figs. 1-2. SEM of strepsipteran metathoracic wings. Fig. 1 Coriophagus rieki Kinzelbach (Halictophagidae), wing base; Fig. 2 Lychnocolax drysdalensis Kathirithamby (Myrmecolaci- dae), costal margin and wings base.

4

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

Figs. 3^. SEM of strepsipteran metathoracic wings. Coriophagus rieki Kinzelbach, costal margins. Note that the costal margin apically bears a single vein (Sc), not the two extended veins (PC + C + ScA and ScP) as drawn by Kukalova-Peck and Lawrence (1993) for this species.

1995

STREPSIPTERA HIND WING VENATION

5

Figs. 5-7. Strepsiptera: Corioxenidae. 5, Corioxenos sp. (Mexico) Scale of 0.5 mm.; 6, Loania sp. (Panama) Scale of 0.5 mm.; 7, Dundoxenos sp. (N. Africa) Scale of 0.2 mm. Venational notation in parenthesis after Kinzelbach (1971) and without parenthesis after Kukalova-Peck and Lawrence (1993). Note that the strepsipteran “pterostigma” (the region between RA,+2 and RA3+4) is absent.

Evaluation: This character relies on two distinct morphological features for its def- inition: the presence of ScP as a discrete vein and its position relative to the pter- ostigma. We have examined the same strepsipteran taxa surveyed by Kukalova-Peck and Lawrence as well as other Strepsiptera taxa, including the family Corioxenidae. We failed to observe any vein which could possibly equate to the PC+C + ScA as drawn by the authors in the wings of these taxa. Contrary to these drawings, we could not find this vein using scanning electron microscopy (Figs. 1^) nor using light microscopy (Figs. 8-10). The authors provide no evidence supporting the sup- position that this is a vein, and their interpretation relies on the presence of PC+C + ScA and the compliance this interpretation has with the presumed ancestral state. Since PC+C+ScA cannot be distinguished from ScP, it is incorrect to specify a state for ScP and homologize it with the state in Coleoptera. In the Corioxenidae C+Sc splits to Sc which is a single vein without any darkened region posteriorly (Figs. 5-7). In this family. Sc can be clearly distinguished from C but a subdivision of Sc into ScA and ScP is unobservable and the pterostigma is absent (Kathirithamby and Peck, 1994).

Even if the authors feel justified in equating ScP with Sc of Kinzelbach (1971) (in the absence of observing PC+C + ScA), the distribution of this character is prob- lematic. In Strepsiptera, the posterior margin of Sc commonly does not reach the

6

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

pterostigma, and in many taxa the pterostigma itself is absent (see character 8 below regarding the strepsipteran “pterostigma”).

Character 2: The apical part of the anterior [wing] margin not strengthened by RA (Kukalova-Peck and Lawrence 1993; synapomorphy).

Interpretation: The authors provide no criteria for distinguishing a strengthened an- terior wing margin from one which is not strengthened. Presumably the states of this character are “strengthened” (plesiomorphy) and “not strengthened” (apomorphy). Until a more precise definition is provided, we cannot evaluate this vague character. Character 3: RA and RP diverging abruptly from one another close to the wing base (Kukalova-Peck and Lawrence, 1993; shared autapomorphic trend). Interpretation: We presume the states are “not diverging abruptly” (plesiomorphy) and “diverging abruptly” (apomorphy).

Evaluation: The authors present no criterion for what constitutes an abrupt diver- gence and how it can be distinguished from a non-abrupt divergence. Apparently, there is some angle at which a divergence is abrupt and another at which it is not abrupt, and the two angles do not overlap. If the authors were to argue that the precise angles of divergence (or range of angles) is irrelevant at the ordinal level in insects, then we would likewise counter that this is an irrelevant line of evidence for ordinal level phylogenetic reconstruction. We have observed sufficient variation in the angle of these two veins in Strepsiptera, however, to make us doubt that it can be defined with discrete states (Figs. 5-10). We are further unclear what the authors mean by “shared autapomorphic trend” and how this represents evidence for phylogenetic affinity (discussed below).

Character 4: Plesiomorphous separation ofRA and RP at the wing base (Kukalova- Peck, 1991; symplesiomorphy?).

Interpretation: The states are apparently “RA and RP basally separate” (plesiomor- phy) and “RA and RP basally fused” (apomorphy).

Evaluation: All Strepsiptera and Coleoptera have RA and RP fused basally, as Ku- kalova-Peck and Lawrence (1993) recognized when they formulated character 3. In the author’s “generalized Neopteran wing,” these veins are also fused basally. If indeed the separation of RA and RP at the wing base is “plesiomorphous,” then how does this symplesiomorphy support the monophyly of Strepsiptera and Cole- optera?

Character 5: The radial and medial basivenale not fused together into a large plate (Kukalova-Peck and Lawrence, 1993; shared autapomorphic trend).

Interpretation: The states of this character are presumably “radial and medial basi- venale fused” (plesiomorphy) and “radial and basal venale not fused” (apomorphy). Evaluation: Once again, it is not clear what a shared autapomorphic trend is and how this represents phylogenetic evidence.

Character 6: RP branches supporting folds (Kukalova-Peck, 1991; synapomorphy); the apical field supported by RP branches, which have a somewhat fan-like arrange- ment (Kukalova-Peck and Lawrence, 1993; shared autapomorphic trend). Interpretation: The RP branches are equivalent to Kinzelbach’s (1971) R,-R4. For the first description, the states are apparently “RP branches not supporting folds” (plesiomorphy) and “RP branches supporting folds” (apomorphy). For the second description, the states are “RP branches not supporting apical field, not fan-like”

1995

STREPSIPTERA HIND WING VENATION

7

(plesiomorphy) and “RP branches supporting apical field, somewhat fan-like” (apo- morphy).

Evaluation: The authors need to define clearly what constitutes “supporting.” For instance, must RP be directly contiguous to a fold in order to support it or just in the general vicinity of a fold? How far must RP be from a fold before it is considered no longer supporting? Is it possible for a fold to exist in the anterior portion of the wing without coming into contact with RP? The second description assumes that no other insect groups have RP branches in the apical field. The real question, of course, is what is being homologized here. Is it the fact that Strepsiptera and Coleoptera both have folds in the wings? This seems suspect because the system of folds in Coleoptera is quite different from that of Strepsiptera, as the authors have recognized, and we doubt that the folds themselves are homologous. Is the homology implied that only Coleoptera and Strepsiptera have developed a novel way of supporting their folds through the use of RP? This makes the dubious assumption that no other insect has folds that are supported by RP the same way folds are supported in Coleoptera and Strepsiptera. The authors need to clarify what they mean by support and how this type of support is novel to Strepsiptera and Coleoptera.

Character 7: Reduced CuP (Kukalova-Peck, 1991; synapomorphy); CuP reduced (Kukalova-Peck and Lawrence, 1993; shared autapomorphic trend).

Interpretation: In Kinzelbach’s (1971) drawings of strepsipteran wings CuP is a distinct and often large vein. Kukalova-Peck and Lawrence call this vein AA and treat CuP as either absent or extremely reduced in Strepsiptera. The presumed states of this character are “not reduced” (plesiomorphy) and “reduced” (apomorphy). Evaluation: In the authors’ drawings of Mengenilla sp. and Coriophagus rieki (Ku- kalova-Peck and Lawrence, 1993: figs. 69-70) CuP is absent and in Lynocholax (fig. 71) CuP is present as a small vein basally separate from, and apically fused to AA (CuA2 of Kinzelbach [1971]). In the numerous Lynocholax species we have exam- ined, we have not observed this small vein (Fig. 10), nor have we observed such a vein in any strepsipteran taxa. We see no reason why the large posterior vein in Strepsiptera should be considered homologous to AA rather than CuP, and the au- thors provide no justification for this designation. Hence there is a serious question of homology between the CuP of Coleoptera and what the authors consider the CuP of Strepsiptera.

Character 8: A shortened RA forming a pterostigma between RAj^2 ^s+4 (Ku- kalova-Peck, 1991; synapomorphy).

Interpretation: Kukalova-Peck treats the R, of Kinzelbach (1971) basally as RA and distally as RA,+2 and RA3+4, with the darkened region between these branches as a “pterostigma” homologous to the coleopteran pterostigma. It is not clear whether the synapomorphy is a short RA, the formation of a “pterostigma” between RA,+2 and RA3+4, or both. If both, the states are “RA not shortened, RA,+2 and RA3+4 not forming a pterostigma” (plesiomorphy) and “RA short, RA,+2 and RA3+4 forming a pterostigma” (apomorphy).

Evaluation: Kukalova-Peck provides no criteria for distinguishing a shortened RA from a non-shortened RA. The varying lengths of RA we have observed in Strep- siptera, however, make us doubt that it can be defined discretely. The formation of a pigmented pterostigma between RA,+2 and RA3+4 cannot be considered a syna-

8

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

8

3(R4)

MAi

PC+C+ScA (C)

MAj

Eig. 8. Light microscope photo of strepsipteran metathoracic wing. Note that RA1+2 is absent in Mengenilla and Coriophagiis. Scale: X35.

pomorphy unique to Strepsiptera and Coleoptera since, as the authors correctly rec- ognize, Hymenoptera and Mecoptera possess this state as well.

We have some reservations with Kukalova-Peck’s designation of a pterostigma in Strepsiptera. What she has labelled RA, RA,+2, and RA3+4 is, according to Kinzel- bach, a single vein R,. In some groups of Strepsiptera its distal margins are laterally expanded and the medial region is somewhat sclerotised. In Corioxenidae Sc is distinguishable as a single vein posteriorally (Figs. 5-7) but R, is not laterally ex- panded and this family clearly shows that there is no pterostigma. Furthermore, because Kukalova-Peck and Lawrence (1993) treat the pterostigma as a landmark for homologizing veins (p. 191), they have no criterion for determining whether the pterostigmas themselves are homologous; they simply assume homology.

Character 9: A very long fork of MP (Kukalova-Peck, 1991; synapomorphy). Interpretation: Kukalova-Peck and Lawrence’s drawing (fig. 69) of Mengenilla shows MP forked into MP,+2 (MA, of Kinzelbach) and MP3+4 (MA2 of Kinzelbach). The states of this character are “fork not very long” (plesiomorphy) and “fork very long” (apomorphy).

Evaluation: The authors provide no criterion for distinguishing a “very long” fork from one which is “not very long.” We have examined all the genera of the most basal Strepsiptera, the Mengenillidae {Mengenilla, Eoxenos, and a new genus from N. Africa [Kathirithamby, in prep.]). Contrary to the author’s drawings, in all of these taxa MP is not forked (i.e., MA, and MA2 are not joined basally. Fig. 8).

CRITIQUE OF PHYLOGENETIC METHODOLOGY

While the authors couch their terminology in cladistic parlance, their methodology is pseudo-cladistic and at discord with the theoretical basis of cladistics. The authors

1995

STREPSIPTERA HIND WING VENATION

9

Fig. 9. Light microscopic photo of strepsipteran metathoracic wing. Coriophagus rieki Kin- zelbach (Halictophagidae). Scale: X26.

PC+C+ScA

(C)

AA (CuP)

RP, (R3) (Rs)

3(R4>

RPi (R2)

CuA (CuA)

do not follow cladistic principles in selecting characters, determining character po- larity, appealing unduly to groundplans and evolutionary scenarios, neglecting a formal parsimony analysis prior to their conclusions, and using “shared autapo- morphic trends” to support monophyly.

Rooting and polarity

The authors do not follow the application of character polarity in a cladistic con- text. Character polarity is assessed by outgroup comparison or ontogenetic study and is determined directly by where the root is placed in a branching network (Waltrous and Wheeler, 1981; Farris, 1982; see Nixon and Carpenter [1993] for an excellent discussion). Character states are scored for the ingroup and outgroup taxa, an un- rooted network is generated, the ingroup is rooted to the outgroup, and the polarity of the characters are subsequently obtained by their optimization on the tree. There is no need — and indeed, no clear way — to establish whether a character is “primi- tive” or “derived” prior to cladistic analysis (Nixon and Carpenter, 1993). Neither is there any need for scenarios involving the direction a suite of characters must have evolved in order for the characters to be phylogenetically informative. The distinction between synapomorphy and plesiomorphy is meaningless in the absence of a rooted cladogram, and the assignment of polarity in the absence of a tree is specious.

10

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

1

RPi (R,)

RP, (R3)

RP3(R4)

PC+C+ScA

(C)

ScP (Sc) RA (Ri)

(CuP

CuA (CuAi) ^

Fig. 10. Light microscopic photo of strepsipteran metathoracic wing. Lychnocolax drysda- lensis Kathirithamby (Myrmecolacidae). Scale: X54.

The polarization of the venational characters presented by Kukalova-Peck and Lawrence is not based on any empirical evidence of character state distribution in ingroups and outgroups. Instead, the authors rely on comparisons to the “all-ptery- gote groundplan” and appeals to scenarios of directionality in the evolution of insect veins. How are these groundplans generated? “Venational groundplan is a compi- lation of primitive features [sic] assembled over many years by comparing the prim- itive representatives of all [sic] extinct and extant pterygote orders” (Kukalova-Peck and Lawrence, 1993: 194). The authors’ designations of character polarities are sim- ple statements of concordance with some preconceived notion of primitive wing venation; this is a far cry from polarization as it is commonly practiced in cladistic analysis.

Beyond the sheer subjectivity of this methodology, and the rather dubious claim that all extinct insect orders have left traces for these authors to include in their groundplan, we find a number of problems with this procedure. First of all, organisms are not compilations of distinct features found in different groups (if so, phyloge- netics would be a meaningless pursuit). By the authors’ own admission, there has never been an organism observed which possesses all of these putatively primitive wing features in combination. Then why should the combination of character states observed in extant (or extinct) taxa be polarized by comparing them with a combi-

1995

STREPSIPTERA HIND WING VENATION

11

nation of states which has never been observed in any taxon? How can unobserved states in a hypothetical taxon constitute evidence for polarization? Why should pre- sumed data (which are not really data anyway) take precedence over observable data?

In the authors’ explanation of the groundplan, they suggest that the groundplan represents the primitive states of characters as found in the most recent common ancestor of all the taxa used in the compilation. But how is this combination of presumed ancestral states obtained? It is clearly not obtained by the cladistic practice of optimizing characters on nodes of a cladogram. It is in fact quite possible that the combination of character states as proposed for the “all-pterygote” ancestor cannot co-occur once those states are optimized on a tree. Moreover, why would we expect the common ancestor of those taxa to have all the presumably primitive features of all the taxa used in the compilation? This would seem to suggest that all the taxa share only that ancestor in common rather than sharing a hierarchy of ancestry.

More critically, how do the authors know that these states are indeed primitive? This tautological conclusion stems from the procedure of using “primitive represen- tatives” to infer “primitive features”. How do we know these representatives are primitive? Simple, they have retained primitive features. How do we know they have retained primitive features? Because they are primitive taxa. But of course whether they are a priori considered primitive or not is moot for cladistic analysis. The real question should be how do the authors know, in the absence of a cladogram, that a given character is a synapomorphy? Once more, the authors are confronted with drawing conclusions of polarity without reference to a specific phylogeny.

This confusion over cladistic methodology is best summarized by the authors’ statement that “it is also not possible [to use coleopteran venation in phylogeny] without determining the succession of veinal character states based on the all-pter- ygote groundplan” (p. 194). It is possible and, in fact, is routine to use venational characters for phylogenetic inference by scoring similar vein modifications in mul- tiple taxa and appropriate outgroups without reference to a groundplan. A tree is reconstructed, a rooting selected, and then (if one wishes) the “succession of veinal character states” can be hypothesized by the optimization of these characters on the tree. It is not that we are specifically arguing that this groundplan is incorrect, we are only arguing that the groundplan has not been inferred correctly and that it should not be used as a means to polarize characters. The use of artificial amalgamations of presumed primitive characters as a basis for character polarization is without theoretical and empirical support and is far outside the realm of cladistic analysis.

Outgroups and analysis

We are concerned with the authors’ apparent lack of adequate comparison of their putative synapomorphies with the appropriate outgroups. These characters were not explicitly scored for other holometabolous insect orders, nor were they specifically scored for the Paraneoptera which are the currently accepted outgroup to the Holo- metabola (Kristensen, 1991). Because the authors have presented us with only a two taxon statement, we have no way of knowing the level of generality of these char- acters. It appears that the authors assumed a priori that Strepsiptera and Coleoptera

12

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

are sister-groups, thereby forcing a coleopteran venational scheme upon Strepsiptera, and they then searched for characters which would validate this supposition.

The authors also did not provide a specific cladistic analysis of their character data in the Strepsiptera, Coleoptera, and outgroups. In the absence of this analysis, it is premature for the authors to make any claims regarding whether a character is a synapomorphy or not. Hence, even if the authors’ putative synapomorphies could be defined as discrete characters, they have yet to demonstrate in an analysis that any of these shared similarities are unique to Strepsiptera and Coleoptera and are synapomorphic rather than symplesiomorphic. The conclusion that a particular char- acter is a synapomorphy, with no explicit cladistic analysis to determine polarity, smacks more of authoritarianism than science.

Evolutionary scenarios

The success of cladistics has lain in part with its ability to separate pattern and process: phylogenies represent the pattern from which evolutionary processes are inferred. Thus cladistics attempts to tease apart the evidence for phytogeny from any specific model of evolution (Eldredge and Cracraft, 1980).

The authors have unduly mixed pattern with process into their phylogenetic con- clusions by proceeding under the assumption that veins evolve according to a known set of rules and that character designation and polarity determination can be confi- dently based upon these rules. These rules include the “two major venational prin- ciples”: (1) the loss of primary veins and their main branches is irreversible, and (2) the fusion of two primary veins near the wing base is irreversible. According to the authors, the veracity of these principles is established by the fact that entomol- ogists have been studying veins for over 100 years and that “through this long experience, the sequences of character change have become well established” (Ku- kalova-Peck and Lawrence, 1993: 186). We are not so confident. It is not clear to us why the authors consider these principles well established as they have never been empirically tested on a phytogeny created independent of these principles. As no one (to our knowledge) has used parsimony to optimize the fusion and loss of primary veins on a ordinal phytogeny for the insects and demonstrated evolution according to Dollo parsimony, we would argue that there yet remains no specific test for the veracity of these principles.

Our argument is not that these principles are wrong, only that the authors have placed undue weight on their veracity in drawing phylogenetic conclusions. These principles may be true. But because the authors have needlessly based their phylo- genetic conclusions on these assumptions, they cannot specifically test these prin- ciples using their phytogeny.

Autapomorphic trends

The authors appear confused as to what types of characters constitute evidence for phylogenetic inference — is it synapomorphy, symplesiomorphy (character 4), or “shared autapomorphic trends” (characters 3, 5, 6, & 1)1 The use of the term “shared autapomorphic trend” is not derived from cladistic literature. How is a shared autapomorphic trend indicative of phytogeny? In what sense is a shared au- tapomorphic trend a derived homologous feature in a group of organisms? If a trend

1995

STREPSIPTERA HIND WING VENATION

13

is shared, what is the level of generality of this sharing? How does sharing a trend translate to a unique, derived evolutionary event in the common lineage of these presumed sister-taxa? How does one in practice distinguish a shared autapomorphic trend from a synapomorphy?

The author’s claim that certain characters constitute autapomorphic trends seems to indicate they think them weaker than a synapomorphy, but still phylogenetically informative. This may explain why characters 6 and 7 were changed (without ex- planation) from synapomorphies in the 1991 paper to shared autapomorphic trends in the 1993 paper. The notion that shared autapomorphic trends are phylogenetically informative, however, is dangerous in that it allows any character distribution to be interpreted as synapomorphic evidence. The concept of shared autapomorphic trends as indicative of phytogeny is foreign to cladistic theory.

CONCLUSIONS

Venational homologies in the highly modified hind wings of Coleoptera have long eluded entomologists. The attempt by Kukalova-Peck and Lawrence is a commend- able effort, but we have some basic criticisms of their character interpretations and phylogenetic methodology.

As detailed above, we are concerned with the author’s misinterpretation of strep- sipteran morphology. The characters they have proposed are in many cases unob- servable, continuous, or of questionable homology. We have further demonstrated discrepancy between the veins the authors have drawn on the three strepsipteran species they examined, and those which we observe in the specimens. Kukalova- Peck (1991) concludes that “quite clearly, Strepsiptera venation can be derived only from a common ancestor with Coleoptera or from a coleopteroid stem group’’ (p. 178). We argue that their current analysis does not support this conclusion.

By failing to score these characters in other holometabolous insect orders and neglecting to score any other venational character which Strepsiptera (or Coleoptera) may share with any other insect order, Kukalova-Peck and Lawrence have biased their results to support their conclusions. We feel that the authors should be more concerned with scoring putatively homologous venational features across multiple ingroup and outgroup taxa, creating a character matrix with these and other char- acters for a formal cladistic analysis, and allow parsimony to arbitrate among pos- sible phylogenetic conclusions and to decide which characters are synapomorphies. The hypothetical ancestral states can then be derived by optimizing the venational states on the tree and any scenario for trends in venational evolution could likewise be derived from the tree. The characters lead to the presumed ancestral states and not the presumed ancestral states to the characters.

We find no evidence from the hind wing venation to support a sister-group rela- tionship between Strepsiptera and Coleoptera.

ACKNOWLEDGMENTS

We thank J. Carpenter, J. Liebherr, W. Wheeler, and D. Grimaldi for helpful suggestions. MEW was supported in part by a NSF dissertation improvement grant. JK wishes to thank the Leverhulme Trust for the Research Fellowship and the Linnean Society for the taxonomic publications grant.

14

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

LITERATURE CITED

Eldredge, N. and J. Cracraft. 1980. Phylogenetic patterns and the evolutionary process. Co- lumbia Univ. Press, New York. 349 pp.

Farris, J. S. 1982. Outgroups and parsimony. Syst Zool. 31:328-334.

Farris, J. S. 1990. Phenetics in camouflage. Cladistics 6:91-100.

Hennig, W. 1966. Phylogenetic systematics (transl. D. D. Davis and R. Zangerl). University of Illinois Press, Urbana.

Kathirithamby, J. 1989. Review of the order Strepsiptera. Syst. Ent. 14:41-92.

Kathirithamby, J. and E. B. Peck. 1994. Strepsiptera of South Florida and the Bahamas with the description of a new genus and species of Corioxenidae. Can. Ent. 126:125-134.

Kinzelbach, R. K. 1971. Morphologische befunds and Eacherfluglern und ihre phylogenetische bedeutung (Insecta: Strepsiptera). Zoologica 1 19(1/2): 1-256.

Kinzelbach, R. K. 1990. The systematic position of Strepsiptera (Insecta). Am. Ent. 36:292- 303.

Kristensen, N. P. 1991. Phylogeny of extant hexapods, in The Insects of Australia: A Textbook for Students and Research Workers, 2nd edition. I. D. Naumann, P. B. Came, J. F. Lawrence, E. S. Nielsen, J. P. Spradberry, R. W. Taylor, M. J. Whitten, and M. J. Little- john (eds.), CSIRO, Melbourne University Press, pp. 125-140.

Kukalova-Peck, J. and J. F. Lawrence. 1993. Evolution of the hind wing in Coleoptera. Can. Ent. 125:181-258.

Kukalova-Peck, J. 1991. Eossil history and the evolution of hexapod structures, in The Insects of Australia: A Textbook for Students and Research Workers, 2nd edition. I. D. Nau- mann, P. B. Came, J. E Lawrence, E. S. Nielsen, J. P. Spradberry, R. W. Taylor, M. J. Whitten, and M. J. Littlejohn (eds.), CSIRO, Melbourne University Press, pp. 141-179.

Nixon, K. C. and J. M. Carpenter. 1993. On outgroups. Cladistics 9:413^26.

Watrous, L. E. and Q. D. Wheeler. 1981. The outgroup comparison method of character anal- ysis. Syst. Zool. 30:1-11.

Received 24 March 1995; accepted 19 May 1995.

J. New York Entomol. Soc. 103(1); 15— 38, 1995

CLADISTIC ANALYSIS OF THE ULMERITUS-ULMERITOIDES GROUP (EPHEMEROPTERA, LEPTOPHLEBIIDAE), WITH DESCRIPTIONS OF FIVE NEW SPECIES OF ULMERITOIDES

Eduardo Dominguez*

Department of Entomology, Cornell University, Ithaca, New York 14853

Abstract. — A cladistic analysis of the species of the genera Ulmeritus and Ulmeritoides is performed. Five new species of Ulmeritoides are described: U. fidalgoi n. sp. from male and female imagos and U. misionensis n. sp. from male and female imagos and nymphs from Argentina, U. spinulipenis n. sp. from male imagos from Uruguay, and U. guanacaste n. sp. and U. tifferae n. sp. from male imagos and nymphs collected in Costa Rica. U. flavopedes is redescribed; U. uruguayensis ( = U. adustus n. syn.), U. luteotinctus and U. patagiatus are discussed. The generic diagnosis of Ulmeritoides is modified to include all known species. Further evidence from the cladistic analysis supports the synonymy of Pseudulmeritus with Ulmeritoides, the monophyly of Ulmeritus and Ulmeritoides, and their status as sister groups.

The Ulmeritus-Ulmeritoides group is a very distinctive complex among the Lep- tophlebiidae. For a long time it was known only from a few species collected from NE Brazil, Uruguay and NW Argentina, with a single record from Suriname. New collections are now available to clarify the taxonomy and phylogeny of the group.

Ulmeritoides was established as a subgenus of Ulmeritus by Traver (1959) for Ulmeritus {Ulmeritoides) uruguayensis Traver, known from imagos of both sexes, and U. (Ulmeritoides) luteotinctus Traver, known only from female imagos. In 1960, Thew described U. (Ulmeritoides) adustus from imagos of both sexes and Ulmeritus patagiatus from subimagos, not assigning the last species to any sub- genus. At that time, the only known nymph of Ulmeritus was that of U. carbonelli Traver.

In 1987 Savage suggested that Ulmeritus might be related to the Hermanella complex, but Flowers and Dominguez (1991) proposed a preliminary cladogram in which Ulmeritus and Ulmeritoides were more closely related to Atopophlebia and Meridialaris. A new species of the subgenus Ulmeritoides was reared for the first time and, based on the characteristics of both sexes, the subgenus was raised to full generic status (Dominguez, 1991).

In this paper I describe five new species of Ulmeritoides and discuss the species U. uruguayensis, U. luteotinctus and U. patagiatus. The species U. adustus is syn- onymized with U. uruguayensis, and U. flavopedes (Spieth) is redescribed. The ge- neric diagnosis of Ulmeritoides is modified to include all known species and a phy- logeny for the Ulmeritus-Ulmeritoides complex is reconstructed and discussed.

' Permanent Address: CONICET-Facultad de Ciencias Naturales, Universidad Nacional de Tucuman; Miguel Lillo 205, 4.000 Tucuman, Argentina.

16

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

MATERIALS AND METHODS

Material from the following institutions was used: Universidad de la Republica, Montevideo, Uruguay (URU); Florida Agricultural and Mechanical University, Tal- lahassee, Florida, USA (FAMU), National Museum of Natural History, Smithsonian Institution, Washington, D.C., USA (NMNH), American Museum of Natural History, New York, USA (AMNH), Illinois Natural History Survey, Illinois, USA (INHS), Instituto Nacional de Biodiversidad, Costa Rica (INBio) and Instituto-Fundacion Miguel Lillo, Tucuman, Argentina (IFML).

The phylogenetic relationships of the group were reconstructed with the aid of the Hennig86 (Farris, 1988) and CLADOS (Nixon, 1992) programs.

Genus Ulmeritoides

Ulmeritus {Ulmeritoides) Traver, 1959:8; Thew, 1960:125.

Ulmeritus {Pseudulmeritus) Traver, 1959:8.

Ulmeritoides', Dominguez 1991:160; Flowers and Dominguez, 1991:52.

Type species: Ulmeritoides uruguayensis (Traver), Original designation, as type spe- cies of Ulmeritus {Ulmeritoides).

Species included: U. luteotinctus (Traver), U. patagiatus (Thew), U. uruguayensis (Traver), U. flavopedes (Spieth), U. spinulipenis new species, U. fidalgoi new spe- cies, U. misionensis new species, U. tijferae new species, U. guanacaste new species. Distribution: From Costa Rica (11°N) to NE Argentina (27°S).

Discussion: This genus was characterized recently by Dominguez (1991) for both stages and sexes. With new material available, some modifications in the generic diagnosis are needed. 1) The development of the median denticle on the antero- median emargination of the labrum is not a constant character for the genus, but differs among species; 2) the generic description of the male genitalia should read: “Penis divided from the base, apex of penis lobes rounded to rather straight, with spines, small projections or with lateral groove”; 3) the posterolateral projections on abdominal segments VIII-IX (Fig. 2A) are of unusually large size for the family, presenting lateral spines. This character is also present in Ulmeritus species (Fig. 2B).

Material certainly belonging to this genus, but not assignable to any described species is given here to make distributional data available for further investigations: Ulmeritoides spp. (2 species): BRAZIL, Para, Rio Xingu, Camp (52°22'W, 3°39'S) ca. 60 km S. Altamira. 8-16/X/1986. P. Spangler and O. S. Flint, Jr., Igarape-Jabuti (2 female imagos, 2 male subimagos) (NMNH); Ulmeritoides sp. GUYANA: Ma- zaruni-Potaro District, Takutu Mountains (16°15'N, 59°5'W), 18/XII/1983, P. J. Spangler, W. E. Steiner and M. L. Levine. Earthwatch Expedition (1 nymph) (NMNH); Ulmeritoides sp. PARAGUAY: Paraguari, Depto Ybycui (25 km SE) in Ybycui National Park, 12-24/IV/1980, P. J. Spangler et al (12 nymphs) (NMNH); Ulmeritoides sp. VENEZUELA: T. F. Amazonas, Puerto Ayacucho (40 km S) To- bogan, 19/11/86. P. J. Spangler, Col. # 1. (1 nymph) (NMNH); Ulmeritoides sp. GUATEMALA: 20 mi SW Puerto Barrios, 16/VIII/1965. P. J. Spangler (1 nymph) (NMNH); Ulmeritoides sp. FRENCH GUIANA: Sinnamary River, Saut Dalles Fleuve, 15/16/VI/92, V. Horeau (10 male subimagos, 3 female imagos, 2 female subimagos, 9 nymphs) (FAMU); Sinnamary River, Saut Maipouri, 24-26/V/93, V. Horeau (1 female imago, 1 male subimago) (FAMU).

1995

CLADISTICS OF ULMERITUS-ULMERITOIDES

17

Fig. 1. Ulmeritoides misionensis nymph, dorsal view.

18

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

Key to the species of the genus Ulmeritoides Male imagos

1. Forewings with costal and subcostal areas tinged with brown (Figs. 8A, 9A) 2

Forewings with costal and subcostal areas hyaline (Figs. 4A, 12A) 4

2. Apex of penis lobes straight, ending in an acute, apical projection (Fig. 8F, G); abdom- inal color pattern as in fig. 8D, E U. misionensis n. sp.

Apex of penis lobes more or less rounded, abdominal color pattern not as above .... 3

3. Longitudinal veins of forewings brownish, cross veins blackish; penis lobes with a shallow small prominence on outer margin, near apex (Eig. 9E, G) . ... U. tijferae n. sp. Veins C, Sc and R1 of forewings brown, remaining longitudinal veins yellowish, cross veins whitish; penis lobes with a small spine on ventral surface of outer corner (Fig.

7B, C) U. fidalgoi n. sp.

4. Apex of penis lobes rounded (Figs. 4F, 12G) 5

Apex of penis lobes not rounded (Figs. 5A, 6D) 6

5. Veins C, Sc and R1 of forewings yellowish; each penis lobe with a lateral groove (Fig.

4E, F) U. flavopedes (Traver)

Veins C, Sc and R1 of forewings yellowish, blackish the stigmatic area; each penis lobe with a small prominence on outer margin (Fig. 12G) U. guanacaste n. sp.

6. Penis lobes with apical margin ending in an acute projection (Fig. 5A)

U. uruguayensis (Traver)

Penis lobes with several small spines on apical margin (Fig. 6D) . . U. spinulipenis n. sp.

Nymphs

1. Tibiae I with two black bands; medial denticle on anteromedian emargination of labrum

much larger than the other four (Fig. 8H) U. misionensis n. sp.

Tibiae I almost completely washed with black, except base and apex lighter; medial denticle on anteromedian emargination of labrum subequal in size to other denticles (Fig. lOA, B) 2

2. Dorsum of femora II and III with numerous short, acute spines; femora II with a median

black spot; abdominal color pattern as in Figure 9D U. tijferae n. sp.

Dorsum of femora II and III with a few short, blunt spines; femora II without a median black spot; abdominal color pattern as in Figure 12D U. guanacaste n. sp.

Ulmeritoides luteotinctus (Traver)

Ulmeritus (Ulmeritoides) luteotinctus Traver, 1959:11.

Ulmeritoides luteotinctus, Dominguez, 1991:162.

Discussion: Traver (1959) described this species based on female imagos and sub- imagos of both sexes. As the original description is adequate, I will not redescribe it. However, due to its different wing coloration, it is possible that the male subima- gos do not belong in the same species. Within the paratypes, there are three that resemble the color pattern of U. uruguayensis.

U. luteotinctus can be separated from the other species of the genus by the fol- lowing combination of characters: 1) forewings hyaline, veins orange-brown, lighter toward posterior margin; 2) abdominal terga and sterna yellowish, slightly tinged with brownish, posterior margin of each segment narrowly darker.

Material: Holotype female imago: URUGUAY, Artigas, Arroyo La Invernada, 21/ 11/54, C. S. Carbonell (URU). Paratypes: 7 female imagos, Timbauba, Arroyo Tres

1995

CLADISTICS OF ULMERITUS-ULMERITOIDES

19

Figs. 2-3. Nymphs. 2, posterolateral projections on abdominal segments VIII-IX: A, Ul- meritoides tijferae\ B, Ulmeritus carbonelli. 3, Detail of tibia III: A, Ulmeritoides tijferae\ B, Ulmeritus carbonelli.

Cruces, 21/11/54, C. S. Carbonell (URU); idem, 1 female imago, 20/11/55; 1 female imago, Sepulturas, Picada del Negro Muerto, Rio Cuareim, 15/XII/57, C. S. Car- bonell (URU). Also studied: 2 female imagos, 2 male subimagos, 1 female subimago, Cerro Largo, Arroyo Quebracho (curso superior), 4-8/III/59, C. S. Carbonell (URU); 1 male subimago, Tacuarembo, Tacuarembo Chico, 20/1/60, C. S. Carbonell (URU).

Ulmeritoides flavopedes (Spieth)

Thraulodes flavopedes Spieth, 1943:1 1 .

Atalophlebioides flavopedes, Traver, 1946:426.

Ulmeritus (Pseudulmeritus) flavopedes, Traver, 1959:8.

Ulmeritoides flavopedes, Dominguez, 1991:162.

Holotype male imago (pinned, one pair of wings and genitalia mounted on slides). Length: body, 6. 6-6. 7 mm; forewings, 8.0-8. 1 mm; hind wings, 1.5-1. 6 mm. Gen- eral coloration bright orange-brown. Wings hyaline. Head light brown. Upper portion of eyes orange-brown, lower portion blackish. Antennae: scape and pedicel brown- ish, flagellum lighter. Thorax: pronotum light brown, with posterior margins black; mesonotum, pleura and sterna bright orange, carinae darker; metanotum light orange, weakly washed with black. Wings (Fig. 4 A, C): membrane of both wings hyaline, wing bases brown. Veins C, Sc and R1 yellowish, remainder hyaline. Legs: Leg I brown, except apex of tibiae and tarsi I yellow, tarsomeres II-V grayish; tarsal claws orange. Legs II-III yellowish except coxae, trochanters, base and median band on femora brownish. Abdomen (Fig. 4D): terga orange-brown with posterior margins blackish. Sterna grayish-orange. Genitalia (Fig. 4E, F): Apex of penis lobes rounded, each with a lateral groove (colors faded).

20

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

Fig. 4. Ulmeritoides flavopedes. Male imago. A, forewing; B, hind wing; C, hind wing enlarged; D, abdominal terga II-VII; E, genitalia (ventral view); F, penis lobes (ventral view, enlarged).

1995

CLADISTICS OF ULMERITUS-ULMERITOIDES

21

Material: Holotype male imago: SURINAM, Moengo, 12/IV/1939, D. C. Geijskes col (AMNH). Paratype: SURINAME, Litani river, Feti Creek, 17/VII/1939, D. C. Geijskes (AMNH).

Discussion: Traver (1959) established the subgenus Pseudulmeritus for this single species. When I elevated Ulmeritoides to the generic level (Dominguez, 1991) I tentatively included U. flavopedes. Based on the phylogenetic relationships of Ul- meritoides species, this placement is confirmed (see discussion under phylogeny).

Ulmeritoides patagiatus (Thew)

Ulmeritus patagiatus Thew, 1960:128.

Ulmeritoides patagiatus, Dominguez, 1991:162.

Discussion: Thew (1960) described this species based on subimagos of both sexes, not assigning it to any subgenus. After studying the type material it is clear that this species belongs in Ulmeritoides (Dominguez, 1991); however, until imagos of this species are obtained it cannot be redescribed or synonymized.

Material: Holotype male imago and Allotype female imago: BRAZIL, Santa Ca- tarina, Nova Teutonia, F. Plaumann, IX/1956 (INHS).

Ulmeritoides uruguayensis (Traver)

Ulmeritus {Ulmeritoides) uruguayensis Traver, 1959:8.

Ulmeritoides uruguayensis, Dominguez, 1991:162.

Ulmeritus {Ulmeritoides) adustus Thew, 1960:126. NEW SYNONYMY.

Discussion: This species was adequately described by Traver (1959). I include an illustration of the penis, because they can be of use for the identification of this species. The abdominal color pattern and the wings are very similar to Ulmeritoides spinulipenis.

Ulmeritus {Ulmeritoides) adustus is a synonym of U. uruguayensis. In his original description, Thew (1960) indicated as diagnostic characters for this species two small spines on the penis lobes, and the abdominal and leg coloration. I had the opportunity to study the types and I did not find any difference between the genitalia of U. adustus and U. uruguayensis. Also, the coloration fits within the intraspecific vari- ation of U. uruguayensis.

Ulmeritoides uruguayensis can be separated from the other species of the genus by the following combination of characters: 1) forewings hyaline, brown spot at base; 2) abdominal color pattern similar to that in Figure 6B; 3) 7-8 cross veins basal to bulla in forewings; 4) penis as in Figure 5A.

Material: Holotype male imago: URUGUAY, Artigas, Arroyo de la Invernada, C. S. Carbonell, 21/11/54 (URU). Paratypes: 2 male imagos, 20 female imagos, 30 male subimagos, 15 female subimagos (idem Holotype); 1 male imago, 15 female imagos, 3 male subimagos, 4 female subimagos, Sepulturas, Rio Cuareim, C. S. Carbonell, 13/1/52 (URU). Other material: 40 male subimagos, 5 female subimagos, Cerro Lar- go, Arroyo Quebracho (Curso Superior), C. S. Carbonell, 4-8/III/59 (URU); 1 female subimago, Tacuarembo, Valle Eden, F. Achaval, IV/80 (URU). Also studied: U. {U.) adustus, Holotype and Allotype: BRAZIL, Santa Catarina, Nova Teutonia, F. Plau- mann, 11/57 (INHS).

22

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

Fig. 5. Ulmeritoides uruguayensis. Male imago. A, genitalia (ventral view); B, hind wing.

Ulmeritoides spinulipenis, new species

Holotype male imago (in alcohol). Length: body, 7. 0-7. 3 mm; forewings, 1 .2-1 .5 mm; hind wings, 1.4-1. 5 mm. General coloration orange-brown, abdomen lighter. Head whitish, diffusely washed with black. Upper portion of eyes reddish-brown, lower portion blackish. Ocelli white, ringed with black at base. Antennae light or- ange, flagellum lighter. Thorax: pronotum light orange-brown with lateral and pos- terior margins and paramedian areas grayish, median area blackish; mesonotum bright orange-brown, carinae darker; metanotum grayish-brown; pleura brownish, with white spots close to wing bases; sterna light-brown, darker in median area. Wings (Fig. 6A): membrane of forewings hyaline, light brown at base, stigmatic area translucent, three small spots, one on bulla and others two in line with it on the next two veins; veins whitish, except apical Vi of C, Sc and R1 brownish; 9-10 cross veins basal to bulla. Membrane of hind wings hyaline (Fig. 6A), brown spot at base; veins C and Sc yellowish, remaining veins whitish. Legs: Leg I orange-brown, washed with black on coxae and trochanters and on external side of femora; narrow black band located at % apical of femora, tibiae blackish except basal part orange- brown and distal part whitish; tarsi yellowish; legs II and III yellowish, washed with black on coxae and trochanters; two narrow black bands on femora, one on apical y3 and the other on apex. Abdomen (Fig. 6B): terga brownish-yellow, terga I-III almost completely washed with black, remaining segments with black markings en- closing a lighter central area, tergum X yellowish; sterna yellowish, diffusely washed with black. Genitalia (Fig. 6C, D): bright orange-brown, forceps orangish, paler toward apex. Penis whitish with small spines as in Figure 6D. Caudal filaments whitish, with black rings at each intersegmental joint, alternating one wide and one narrow.

Female and nymph: Unknown.

1995

CLADISTICS OF ULMERITUS-ULMERITOIDES

23

Fig. 6. Ulmeritoides spinulipenis. Male imago. A, hind wing; B, abdominal terga II-VII; C, genitalia (ventral view); D, penis lobe (ventral view, enlarged).

Material: Holotype male imago: URUGUAY, Tacuarembo, Tacuarembo chico, C. S. Carbonell, 20/1/1960 (URU). Paratypes: 2 male images (idem holotype) (IFML). Other material: 2 male subimagos, Paysandu, Santa Rita, C. S. Carbonell et all. 20/ 1/62 (URU); 1 male subimago, ARGENTINA, Misiones, Puerto Libertad, O. S. Flint, Jr., 24/XI/73 (USNM).

Etymology: L. spinula, meaning small spine, and L. penis, penis.

Discussion: Ulmeritoides spinulipenis can be separated from the other species of the genus by the following combination of characters: 1) forewings hyaline, with three small spots, one on the bulla and the other two in line with it on the next two veins; 2) abdominal color pattern as in Figure 6B; 3) 9-10 cross vein basal to bulla in forewings; 4) penis lobes with small spines as in Figure 6D.

Biology: Unknown.

Ulmeritoides fidalgoi, new species

Holotype male imago (in alcohol, genitalia on slide). Length: body, 6. 5-7. 2 mm; forewings, 8. 0-8. 5 mm; hind wings, 1.6-1. 9 mm. General coloration: orange-brown, abdomen washed with black. Head: whitish, with anterior margin and lines between ocelli black. Upper portion of eyes orange-brown, lower portion blackish. Ocelli white, base black. Antennae: scape and pedicel orange-brown [flagellum broken off and lost]. Thorax: pronotum light-orange-brown, with lateral margins and median and paramedian areas blackish; mesonotum light yellow-brown, margins and carinae darker; metanotum light brown, washed with black; pleura orange-brown, with mar- gins of sclerites washed with black; sterna bright orange-brown, washed with black

24

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

Fig. 7. Ulmeritoides fidalgoi. Male imago. A, abdominal terga II-VII; B, genitalia (ventral view); C, penis lobe (ventral view, enlarged).

in central area. Wings: membrane of forewings hyaline, costal and subcostal areas brown, paler toward apex, wing base brown. Veins C, Sc and R1 and cross-veins between them brownish; remaining longitudinal veins yellowish, cross veins whitish; 8-10 cross veins basal to bulla. Membrane of hind wings hyaline, brown spot at base; longitudinal and cross veins in costal area light brown, darker at base, remain- ing veins yellowish, lighter toward hind margin. Legs: coxae and trochanters orange- brown; femora light brown, femora I and II with medial and apical black spots; femora III with basal half, median band and apex blackish; tibiae yellowish, tarsi I- IV yellowish washed with gray, tarsi V and claws grayish-black. Abdomen (Fig. 7A): terga brown-orange, with black markings as in Figure 7A; markings darker in the first segments, paler toward the last ones; sterna orange-yellowish, washed with brown. Genitalia (Fig. 7B, C): subgenital plate and base of forceps light brown, remainder of forceps and penis yellowish. Each penis lobe with a small spine close to the outer corner (Fig. 7C). [Caudal filaments broken off and lost].

Allotype female imago (In alcohol). Length: body, 7. 3-7. 8 mm; forewings, 10.5- 1 1.0 mm; hind wings, 1.9-2. 2 mm. Similar to holotype except as follows: head light yellow, posterior margin blackish; eyes black; pronotum yellowish; mesonotum bright yellow; abdomen light grayish-brown, pattern paler.

Nymph: Unknown.

Material: Holotype male imago: ARGENTINA, Misiones, Bompland (Camping), Arroyo Martires, 26/XI/86, E. Dominguez col. (IFML); Allotype female imago, same data as holotype. Paratypes: 2 male imagos, 10 female imagos, 1 male subimago, same data as holotype; 2 female imagos, 1 male subimago, same data as holotype, except collected 15/11/85; 1 female subimago: Misiones, Arroyo Pepiri Mini (De-

1995

CLADISTICS OF ULMERITUS-ULMERITOIDES

25

sembocadura Rio Uruguay) l/XII/86, E. Dominguez col. All material deposited in IFML, except 10 female imagos, 5 in (FAMU) and 5 in (NMNH).

Etymology: I name this species after my friend, the chalcidologist A. A. R Fidalgo, with whom I was traveling when I collected this species.

Variation: In some females the mesonotal and abdominal color pattern is much paler than in the allotype.

Discussion: Ulmeritoides fidalgoi can be separated from the other species of the genus by the following combination of characters: 1) forewings hyaline, with costal and subcostal areas tinged with brown; 2) abdominal color pattern similar to the one in Figure 7A; 3) 8-1 1 cross veins basal to bulla of forewings; 4) penis as in Fig. 7B, C.

Biology: Unknown.

Ulmeritoides misionensis, new species Ulmeritoides sp. Dominguez, 1991:166, figures. 18-29.

Holotype male imago (in alcohol, one pair of wings and genitalia on slides.) Length: body, 7. 9-8. 2 mm; forewings, 9. 2-9. 6 mm; hind wings, 1. 8-2.0 mm. General col- oration yellow-brown, abdomen washed with black. Head: light yellow, with median line and anterior margins washed with black. Upper portion of eyes light-brown, lower portion blackish. Ocelli white, black basally. Antennae: scape and pedicel orange-brown, flagellum yellowish, washed with black. Thorax: pronotum light brown with lateral, posterolateral and medial and paramedian areas blackish; meso- notum light yellow-brown, margins and carinae darker; metanotum light brown, washed with black; pleura yellow-brown, washed heavily with black; pro- and meta- sternum light brown, heavily washed with black, mesosternum yellowish, tinged with black in median area. Wings (Fig. 8A, C): membrane of forewings (Fig. 8A) hyaline, costal and subcostal areas light brown, stigmatic area much lighter, wing base light brown. Veins C, Sc and R1 and cross veins between them brownish; remaining veins whitish; 10 cross veins basal to bulla. Membrane of hind wings hyaline, brown spot at base (Fig. 8B, C); basal portion of vein Sc brownish, remaining veins whitish. Legs: coxae and trochanters light brown, heavily washed with black; femora I light brown with a black spot at mid-length and washed with black in basal half of apex; tibiae I light brown with a subbasal and a subapical black band; femora II yellowish with a black spot a little apically of median area; femora III yellowish with a black mark in median area and washed with black in basal 1/2 and apex; tibiae II and III yellowish-white; tarsi of all legs light yellow washed with gray; claws grayish-black. Abdomen (Fig. 8D, E): terga yellow-brown, with black markings covering postero- lateral angles, lateral margins and delimiting a circular area lighter, as in Figure 8D; sterna light brown, heavily washed with brown on the anterior segments, lighter posteriorly. Genitalia (Fig. 8F, G): yellowish-brown, washed with black mainly in subgenital plate and apical 4/5 of forceps segment I. Apex of penis lobes straight, ending in an acute apical projection (Fig. 8G). Caudal filaments whitish, with black rings at each intersegmental joint, alternating one wide and one narrow.

Female subimago (in alcohol). Length: body, 7. 0-8.0 mm; forewings, 1 1.0-1 1.5 mm; hind wings, 2.0-2. 2 mm. Coloration as in male imago, except: head yellowish-white, posterior margin blackish; wing membrane translucent, tinged with yellowish-brown.

26

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

8D

8G

Fig. 8. Ulmeritoides misionensis. Male imago (A-G), Nymph (H). A, forewing; B, hind wing; C, hind wing enlarged; D, abdominal terga II-VII; E, abdominal segments V-VII, lateral view; F, genitalia (ventral view); G, penis lobes (ventral view, enlarged); H, detail of antero- median emargination of labrum.

1995

CLADISTICS OF ULMERITUS-ULMERITOIDES

27

except costal and subcostal areas brownish; abdominal markings less marked than in male imago.

Mature nymph (in alcohol) (Fig. 1). Body length, 7.0-8. 8 mm. General coloration: bright orange-brown with black markings. Head: light orange-brown with central area and between ocelli and eyes yellowish. Ocelli whitish with inner margins black. Eyes of male with upper portion reddish-brown, lower portion black. Eyes of female black. Antennae: scape and pedicel light brown, flagellum yellowish, paler toward apex. Mouthparts: clypeus, labrum, basal 2/3 of mandibles, basal 1/2 of maxillae and segment I of palpi and labium brownish, remaining parts lighter. Medial denticle on anteromedian emargination of labrum much larger than others (Fig. 8H). Tusk on inner apical margin of maxillae of medium size, similar to Figure IOC. Thorax: nota bright orange-brown, with black markings as in male imago, medial line yellowish, pleura and sterna yellowish-brown, washed with black as in male imago. Legs: light brown, with coxae washed with black and black markings as in male imago, but smaller; numerous short, pointed spines on dorsum of femora II-III. Claws light brown, apex orangish. Abdomen: terga bright orange-brown, darker posteriorly, black markings as in male imago; sterna yellowish washed with black, darker anteriorly. Gills yellowish, trachea and fimbriae blackish. Caudal filaments brownish lighter toward apex.

Material: Holotype, male imago, ARGENTINA, Misiones, Inta San Vicente, 30/XI/ 86, E. Dominguez col. (IFML); Paratypes: 1 male imago, 1 male subimago, 3 female subimagos, 40 nymphs. All deposited at IFML, except 10 nymphs in (FAMU) and 10 nymphs in (NMNH). The association of the adult and nymphs was made from rearing by E. Dominguez.

Etymology: misionensis, from Misiones Province, Argentina, where this species was collected.

Variation: The male paratype color is lighter than in the holotype, especially in the abdominal pattern.

Discussion: The nymph of U. misionensis was used to characterize Ulmeritoides (Dominguez, 1991) and all illustrations referred to as ''Ulmeritoides sp.” are of this species.

Ulmeritoides misionensis can be separated from the other species of the genus by the following combination of characters. In the imagos: 1) forewings (Fig. 8 A) hy- aline, with costal and subcostal areas light brown, stigmatic area much lighter; 2) abdominal color pattern as in Figure 8D, E; 3) 10 cross veins basal to bulla in forewings; 4) penis as in Figure 8F, G. In the nymph: 1) medial denticle on anter- omedian emargination of labrum much larger than others (Fig. 8H); 2) tibiae I with subbasal and subapical black bands; 3) dorsum of femora II-III with numerous short, pointed spines; 4) femora II with median black spot.

Ulmeritoides tijferae, new species

Holotype male imago (in alcohol). Length: body, 6. 8-7. 2 mm; forewings, 8. 0-8. 2 mm; hind wings, 1. 9-2.0 mm. General coloration: brownish, abdomen slightly washed with black. Head: light orange, heavily washed with black. Upper portion of eyes yellow-brown, lower portion blackish. Ocelli white, heavily washed with brown laterally, blackish basally. Antennae: scape and pedicel light brown, flagellum

28

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

lighter. Thorax: pronotum light brown, heavily tinged with black on lateral and posterior margins, median and paramedian areas blackish; mesonotum light orange- brown, margins and carinae darker; metanotum light brown, washed with black; pleura light brown, with margins of sclerites darker; sterna orange-brown, washed with black. Wings (Fig. 9A-C): membrane of forewings (Fig. 9A) hyaline, costal and subcostal areas light brown, wing base brown; longitudinal veins brownish, cross veins blackish, both lighter toward hind margin; 5 cross veins basal to bulla. Mem- brane of hind wings (Fig. 9B, C) hyaline, tinged with brown in costal area, brown spot at base; longitudinal and cross veins in costal area brownish, darker at base, remaining veins yellowish toward hind margin. Legs: coxae brownish, washed with black; trochanters light brown; femora light brown with apical and medial black spot except in femora III where black spot almost joins black stain in basal half; tibiae I blackish, base lighter; tarsi and claws in fore leg yellowish washed with black; tibiae and tarsi I-IV in legs II and III yellowish, tarsi V and claws yellowish washed with black. Abdomen (Fig. 9D, E): terga light brown washed with black; sterna lighter. Genitalia (Fig. 9F, G): subgenital plate, penis and base of forceps segment I light brown, distal part of forceps segment I and segments II and III washed with black. Apex of penis lobes rounded, each with a shallow small prominence on outer margin (Fig. 9G). Caudal filaments light brown, washed with black.

Female imago: Unknown.

Mature nymph (in alcohol). Body length, 7. 0-8.0 mm. General coloration: light or- ange-brown with areas between ocelli and eyes heavily tinged with black. Ocelli whitish with inner margins black. Eyes of male with upper portion orange-brown, lower portion black. Eyes of female black. Antennae: light yellow, paler toward apex. Mouthparts: clypeus, labrum, basal 2/3 and base of molars of mandibles, basal 1/2 of maxillae and segment I of palpi and labium light brown, molars and incisors of mandibles and setae on galea-lacinia of maxillae orange-brown, remaining parts lighter. Denticles on anteromedian emargination of labrum subequal (Fig. lOA, B). Tusk on inner apical margin of maxillae of medium size (Fig. IOC). Thorax: terga yellow-brown, with irregular black markings especially on the lateral margins, sterna and pleura light yellow. Legs: light brown, with coxae washed with black, apical and median blackish spots on femora, inner margin of tibiae blackish, narrow median band on tarsi dark brown; numerous short, pointed spines on dorsum of femora II- III. Claws light brown, apex orange-brown. Abdomen: terga light orange-brown, with black markings as in male imago; first sterna yellowish, last orangish. Gills whitish, trachea and fimbriae gray-violet. Caudal filaments bright orange-brown.

Female imago: Unknown.

Material; Holotype, Male imago: COSTA RICA, Guanacaste, Quebrada Alcor- noque, Cerro El Hacha, Parque Nacional Guanacaste, 18/VIF1989, C. de La Rosa Col (INBio). Paratypes: 6 Male imagos, same data as holotype, 2 in (INBio), 2 in (FAMU) and 2 in (IFML); 18 nymphs, same data as holotype except data: 26/V/91, R. Tiffer and R. W. Flowers col., 10 in (INBio), 4 in (FAMU) and 4 in (IFML). The association of the adults and nymphs was made by the abdominal color pattern, from material collected at the type locality.

Etymology: This species is dedicated to Miss Ruth Tiffer S., former “Research Coordinator of the Guanacaste Conservation Area,” who was one of the collectors of the nymphs of this species.

1995

CLADISTICS OF ULMERITUS-ULMERITOIDES

29

Fig. 9. Ulmeritoides tijferae. Male imago. A, forewing; B, hind wing; C, hind wing en- larged; D, abdominal terga II-VII; E, abdominal segments V-VII, lateral view; F, genitalia (ventral view); G, penis lobes (ventral view, enlarged).

30

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

Figs. 10-11. Nymphs. 10. Ulmeritoides tijferae. A, labrum (dorsal view); B, labrum (detail anteromedian emargination); C, right maxilla. 11. Ulmeritoides guanacaste, apex of right max- illa.

Variation: One of the paratypes has more extensive black markings on the fore legs; some of the nymphal legs are light yellow and the size of the tibial spot varies slightly.

Discussion: U. tijferae can be differentiated form the other species of the genus by the following combination of characters. In the imagos: 1) forewings hyaline, with costal and subcostal areas tinged with brown (Fig. 9A); 2) abdominal color pattern as in Fig. 9D, E; 3) 5 cross veins basal to bulla in forewings; 4) penis as in Figure 9F, G. In the nymph: I) denticles on anteromedian emargination of labrum subequal (Fig. lOB); 2) tibiae I almost completely tinged with black; 3) dorsum of femora II- III with numerous short, pointed spines; 4) femora II with a median black spot. Biology: Nymphs were collected from leaf packs and wood in a pool area of Que- brada Alcornoque.

Ulmeritoides guanacaste, new species

Holotype male imago (in alcohol, one pair of wings and genitalia on slides). Length: body, 7. 2-7. 5 mm; forewings, 7.6-7. 8 mm; hind wings, 1.2-1. 3 mm. General col- oration: brownish, abdomen tinged with black. Head: orange-brown washed with black on anterior margins. Upper portion of eyes brownish-yellow, lower portion grayish-black. Ocelli white, ringed with black at base. Antennae: scape and pedicel yellowish, flagellum lighter.

1995

CLADISTICS OF ULMERITUS-ULMERITOIDES

31

Thorax: pronotum orange-brown with lateral and posterior margins and median and paramedian areas blackish; mesonotum bright orange-brown, margins and carinae darker; metanotum light brown; pleura light brown, with margins of sclerites darker; sterna orange-brown, washed with black. Wings (Fig. 12A-C); membrane of fore- wings (Fig. 12A) hyaline, light brown at base, stigmatic area translucent; veins C, Sc and R1 yellowish, except blackish in stigmatic area, remaining longitudinal veins whitish, except distal portions of veins Rs and MA tinged with black; cross veins in stigmatic area blackish, remaining cross veins whitish; no cross veins basal to bulla. Membrane of hind wings (Fig. 12B, C) hyaline, brown spot at base, veins C, Sc and cross veins in costal area grayish, remaining veins whitish. Legs: foreleg: coxae and trochanters brown, washed with black; femora light brown with a spot on apical 2/3 and apex blackish; tibiae almost completely black, except little portion at base and apex, brownish; tarsi and claws grayish-brown, washed with black; legs II and III: coxae brown, washed with black; trochanters light brown; femora II light brown with apex blackish, femora III with basal 2/3 washed with black; tibiae and tarsi yellowish washed with black; claws grayish. Abdomen (Fig. 12D, E): terga brownish, washed with black, especially along median-dorsal line as in Figure 12D; sterna marking similar to those of terga but with ganglial areas darker. Genitalia (Fig. 12F, G): subgenital plate, penis and basal 1/5 of segment I and segments II and III of forceps light brown, apical 4/5 of forceps segment I blackish. Apex of penis lobes rounded, with a small prominence on outer margin (Fig. 12G). Caudal filaments gray-yellowish.

Female imago: Unknown.

Mature nymph (in alcohol, mouthparts on slide). Body length, 5. 1-5.2 mm. General coloration: bright orange-brown, with black markings. Head: yellowish-brown, washed with black. Ocelli whitish with inner margins black. Eyes of male with upper portion bright orange-brown, lower portion black. Eyes of female black. Antennae: scape and pedicel light yellow, flagellum whitish. Mouthparts: light brown except wide V shape mark that starts on clypeus and ends close to the anterior margin of labrum, base and inner 2/3 of mandibles, basal 2/3 of maxillae and maxillary palpi and mentum and base of paraglossae and base of labial palpi segment I washed with black, molars and incisors of mandibles and setae on galea-lacinia of maxillae or- ange-brown. Denticles on anteromedian emargination of labrum subequal in size. Tusk on inner apical margin of maxillae very small (Fig. 11). Thorax: nota light- brown, with black markings as in male imago, pleura and sterna yellowish, washed with black. Legs: light brown, black markings as in male imago; claws yellowish, apex orange-brown. Abdomen: terga bright yellow-orange, black markings as in male imago but more marked; sterna yellowish- white in first segments turning to yellow- orangish in last ones, all washed with black. Gills whitish, trachea and fimbriae dark gray. Caudal filaments yellow-orange, paler toward apex.

Material: Holotype, male imago: COSTA RICA, Guanacaste, Arroyo #1; Estacion Pitilla, Sendero La Laguna. 12/VI/ 1989, C. de La Rosa Col (INBio). Paratypes: 1 male imago, same data as holotype (EAMU), 2 nymphs, same data as holotype, except date 16/V/91, R. W. Flowers col. (INBio). The association of the adults and nymphs was made by the abdominal color pattern, from material collected at the type locality.

32

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

Fig. 12. Ulmeritoides guanacaste. Male imago. A, forewing; B, hind wing; C, hind wing enlarged; D, abdominal terga II-VII; E, abdominal segments V-VII, lateral view; E, genitalia (ventral view); G, penis lobes (ventral view, enlarged).

1995

CLADISTICS OF ULMERITUS-ULMERITOIDES

33

Etymology: Guanacaste: from the type locality, Parque Nacional Guanacaste, Costa Rica.

Variation: The male paratype is darker than the holotype, mainly in the abdominal and genital coloration. The younger nymph is light-brown.

Discussion: Ulmehtoides guanacaste can be separated from the other species of the genus by the following combination of characters. In the imagos: 1) forewings (Fig. 12 A) hyaline, with veins C, Sc and R1 blackish at stigmatic area; 2) abdominal color pattern as in Figure 12D, E; 3) no cross veins basal to bulla in forewings (Fig. 12A); 4) penis as in Figure 12F, G. In the nymph; 1) Medial denticle on anteromedian emargination of labrum subequal in size to other denticles; 2) tibiae I almost com- pletely tinged with black; 3) dorsum of femora II-III with few short, blunt spines; 4) femora II without median black markings.

Biology: Nymphs were found at the type locality living in leaf packs that had ac- cumulated under a fallen log. The water current was very slow and at the date of collection (May) the water level in the stream was very low.

PHYLOGENY OF THE ULMERITUS-ULMERITOIDES GROUP

When Traver (1959) established the three subgenera of Ulmeritus, the relationships among the species within the subgenera were unknown, and only the nymph of Ulmeritus carbonelli was described. When analyzing the phylogenetic relationships of the Hermanella complex (Flowers and Dominguez, 1991), including a reared nymph of a new species of Ulmeritoides, it became clear that Ulmeritus and Ul- meritoides each had enough apomorphies as justify placement in different genera (Dominguez, 1991). Ulmeritus (Pseudulmeritus) flavopedes (Spieth) was clearly more closely related to species of Ulmeritoides than to Ulmeritus and for this reason U. flavopedes was transferred to Ulmeritoides.

But the relationships between the monobasic Pseudulmeritus and the other com- ponents of the group remained unknown (Dominguez, 1991). As several new species are described herein, it was necessary to establish their phylogenetic relationships in order to determine the status of Pseudulmeritus. When the first draft of this manu- script was completed, only adult characters were used, since nymphs of only Ul- meritus carbonelli and Ulmeritoides misionensis n. sp. were known.

Traditionally, most of the specific characters in mayflies referred to coloration and genitalic structures; these characters are difficult to polarize and even to homologize. Because the Ulmeritus-Ulmeritoides complex is rather homogeneous, I tried to select characters on which I could establish a transformation series to study the phyloge- netic relationships.

Twelve adult characters (characters 1-12, Appendix I) were used and the polarities determined using two outgroups: Atopophlebia, the sister group of this complex, and Meridialaris which is one of the components of the sister group of Atopophlebia + Ulmeritus/Ulmeritoides (Flowers and Dominguez, 1991) as outgroups. Binary char- acters were coded as 0 (plesiomorphic) and 1 (apomorphic). Multistate characters were assigned different numbers indicating different apomorphies and treated as additive. Character 6 (Appendix I) is polymorphic in Meridialaris and was coded as “missing.” In order to avoid the problems discussed by Nixon and Davis (1991), treating polymorphic characters as “missing entries,” the behaviour of this character

34

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

was studied using alternative codings without any change in the results. Only char- acters separable into discrete states were used. Polarities were determined by out- group comparison (Watrous and Wheeler, 1981) and following previous studies of Flowers and Dominguez (1991) and Pescador and Peters (1978).

The option used with the Hennig86 (Farris, 1988) program was ie*, the option guaranteed to find all shortest trees; only one tree was found, with a length of 29, a Consistency Index (Cl) of 72 and a Retention Index (RI) (Farris, 1989) of 80.

After this analysis was completed, the nymphs of U. tifferae and U. guanacaste were made available to me, by Dr. R. W. Flowers. Using nymphs of Ulmeritus carbonelli and U. balteatus (recently obtained) and Ulmeritoides misionensis, U. tijferae and U. guanacaste, I repeated the analysis, this time using characters of both adult and nymphal stages to test the original results. 14 nymphal characters were coded (characters 13-26, Appendix I), and added to the original matrix. (The nymph- al characters are separated from adult characters by an empty column in the matrix (Appendix II)). Characters 16 and 24 are treated as non-additive.

Using the same program options, the only tree obtained (length = 50; Cl = 82; RI = 85) is identical in branching pattern and adult character distribution to the one based only on adults. This tree was printed with Clados (Nixon, 1992) (Fig. 13).

The analysis indicates that the Ulmeritus-Ulmentoides group is monophyletic, their synapomorphies being: vein ICu2 of forewings attached to Icul and CuP (char- acter 2), presence of basal swelling on segment I of forceps (character 10), basal or medial position of dorsal row of setae on labrum (character 14), dorsal row of setae on labrum entire and sinusoidal or divided (character 15), row of setae at base of outer incisor present (character 20), dorsal row of spines present on segments II and III of labial palpi (character 21), presence of lines of pectinate setae on dorsum of tibiae III (character 24) (Fig. 3A) and posterolateral projections on abdominal seg- ments VIII-IX wide and with spines on margins (character 26). There are two major monophyletic groups, one formed by the three species of Ulmeritus {U. carbonelli, U. balteatus and U. saopaulensis) and the other by the species of Ulmeritoides. Within Ulmeritoides, U. uruguayensis + U. spinulipenis + U. misionensis form a monophyletic group, but it is not possible to establish the relationships among them. Their synapomorphies are: the apex of penis lobes rather straight (character 12) and denticles on anteromedian emargination present, with the median one larger (char- acter 13(2)). With U. flavopedes they share characters 10(3): inner corner of basal swelling acute, which is homoplastic with U. fidalgoi, and as a reversal the absence of ventral projections on penis lobes (character 11). Its sister group is ({U. guana- caste + U. tijferae) + U. fidalgoi). Synapomorphies shared by U. guanacaste + U. tijferae are: cross veins in forewings less than 100 (character 3) and as a reversal vein ICu2 attached to ICul (character 2); U. fidalgoi share with them the synapo- morphy of ventral subapical projections of penis very short (character 11(4)). These two sister groups are linked by the following synapomorphies: in forewings veins C, Sc and RI darker than the rest (character 6(1)), which reverses in U. tijferae to (6); Vein Sc of hind wings less than .8 length of wing (character 8); basal swelling of forceps not rounded (character 10(2)); tusk on inner apical margin of maxillae small to medium size (character 16(1)); spines on margin of glossae few and big (character 23); and homoplastic with Atopophlebia are: cross veins less than 20 in

1995

CLADISTICS OF ULMERITUS-ULMERITOIDES

35

1 11 13 22

Fig. 13. Cladogram of the Ulmeritus-Ulmeritoides group. Black boxes = apomorphies; dark gray boxes = parallelisms; light gray boxes = reversals; empty boxes = plesiomorphies.

hind wings (character 9) and medial row of long setae on dorsum of paraglossae

(22(2)).

The three species of Ulmeritus form a monophyletic group supported by the fol- lowing synapomorphies: spots around cross veins (character 5); basal position of dorsal row of setae on labrum (character 14(2); dorsal row of setae on labrum divided (character 15(2)); tusk on inner apical margin of maxillae big (character 16(2)); palpifer of maxillary palpi enlarged (character 17); ratio segment Il/segment III of maxillary palpi more than 0.9 (character 18); the outer margin of maxillae strongly curved (character 19) and line of pectinate setae on dorsum of tibiae III bifurcated at 1/2 length (character 24(2)) (Fig. 3B). U. balteatus and U. saopaulensis clade is supported by the synapomorphy; digitiform ventral projection of penis medium to short (character 11(2)).

Some characters, such as abdominal color pattern, did give some phylogenetic information. For example U. uruguayensis, U. spinulipenis and U. misionensis all have on the abdominal terga the same pattern of light rounded spot, surrounded by black (Figs. 6B, 8D), not shared by any other component of the group. This character

36

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

was not used because it was not possible to discern discrete states for the other species of the group.

The cladistic analysis supported the synonymy of Pseudulmeritus with Ulmerito- ides, the monophyly of Ulmehtus and Ulmeritoides and their status as sister groups.

It is important to remember that the nymphs of many species in this analysis, including U. flavopedes, are still unknown. As soon as they become available, the distribution of their characters will test the pattern of relationships proposed in this paper.

ACKNOWLEDGMENTS

1 want to thank Dr. L. C. de Zolessi (URU), Dr. W. L. Peters and Dr. R. W. Flowers (FAMU), Drs. O. S. Flint, Jr (NMNH) for the loan of material, and the last three persons and Dr. Q. D. Wheeler and Dr. R. Meier (Cornell University) for comments on the manuscript. This manu- script was prepared while the author was supported by the Short-Term Visitor Program from the National Museum of Natural History, Smithsonian Institution and finished while at the Department of Entomology, Cornell University, with an External Fellowship, from the Argen- tine National Council of Scientific Research (CONICET).

LITERATURE CITED

Dominguez, E. 1988. Ecuaphlebia: a new genus of Atalophlebiinae (Ephemeroptera; Lepto- phlebiidae) from Ecuador. Aquat. Ins. 10(4);227-235.

Dominguez, E. 1991. The status of the genus Ulmeritus (Ephemeroptera: Leptophlebiidae: Atalophlebiinae) and related taxa. Pages 157-168 in: J. Alba-Tercedor and A. Sanchez- Ortega (eds.). Overview and Strategies of Ephemeroptera and Plecoptera. Sandhill Crane Press, Gainesville, Florida.

Dominguez, E. and R. W. Flowers. 1989. A revision of Hennanella and related Genera (Ephemeroptera: Leptophlebiidae: Atalophlebiinae) from Subtropical South America. Ann. Ent. Soc. Am. 82(5):555-573.

Farris, J. S. 1988. Hennig86 version 1.5 manual; software and MSDOS program. Port Jefferson Station, New York.

Farris, J. S. 1989. The retention index and the rescaled consistency index. Cladistics 5:417- 419.

Flowers, R. W. 1980. Atopophlebia fortunensis, a new genus and species from Panama (Lep- tophlebiidae: Ephemeroptera). Fla Ent. 63(1): 162-165.

Flowers, R. W. 1987. New species and life stages of Atopophlebia (Ephemeroptera: Lepto- phlebiidae: Atalophlebiinae). Aquat. Ins. 9(4):203-209.

Flowers, R. W. and E. Dominguez. 1991. Preliminary cladistics of the Hennanella Complex (Ephemeroptera: Leptophlebiidae: Atalophlebiinae). Pages 49-62 in: J. Alba-Tercedor and A. Sanchez-Ortega (eds.). Overview and Strategies of Ephemeroptera and Plecoptera. Sandhill Crane Press, Gainesville, Florida.

Nixon, K. C. and J. I. Davis. 1991. Polymorphic taxa, missing values and cladistic analysis. Cladistics 7:233-241.

Nixon, K. C. 1992. CLADOS version 1.2 manual; software and MSDOS program.

Pescador, M. L. and W. L. Peters. 1987. Revision of the Genera Meridialaris and Massartel- lopsis (Ephemeroptera: Leptophlebiidae: Atalophlebiinae) from South America. Trans. Am. Ent. Soc. 112:147-189.

Savage, H. M. 1987. Biogeographic classification of the Neotropical Leptophlebiidae (Ephem- eroptera) based upon geological centers of ancestral origin and ecology. Stud. Neotrop. Fauna Environ. 22:199-222.

1995

CLADISTICS OF ULMERITUS-ULMERITOIDES

37

Spieth, H. T. 1943. Taxonomic studies on the Ephemeroptera. III. Some interesting ephemerids from Surinam and other Neotropical localities. Am. Mus. Novit. 1244:1-13.

Thew, T. B. 1960. Taxonomic studies on some Neotropical Leptophlebiid mayflies (Ephem- eroptera: Leptophlebiidae). Pan-Pacif. Ent. 36:119-132.

Traver, J. R. 1956. A new genus of Neotropical mayflies. Proc. Ent. Soc. Wash. 58(1): 1-1 2.

Traver, J. R. 1959. Uruguayan mayflies. Eamily Leptophlebiidae: Part I. Rev. Soc. Urug. Ent. 3:1-13.

Watrous, L. E. and Q. E. Wheeler. 1981. The outgroup comparison method of character anal- ysis. Syst. Zool. 30:1-11.

Received 6 April 1995; accepted 19 May 1995.

APPENDIX I. CHARACTERS USED IN CLADISTIC ANALYSIS

FOREWINGS

ADULTS

1. Slanting cross vein above MA fork (< with MA approximately 45°): 0 (absent), 1 (present).

2. Attachment of Vein ICu2: 0 (attached to ICul), 1 (attached to ICul and CuP).

3. Number of cross veins: 0 (more than 110), 1 (less than 100).

4. Cross veins basad to bulla: 0 (present), 1 (absent).

5. Spots around cross veins: 0 (absent), 1 (present, not forming bands), 2 (present, forming bands).

6. Coloration of longitudinal veins posterior to and in relation with C, Sc and Rl: (lighter or same color), 0 (same color), 1 ( lighter).

HIND WINGS

7. Location of apex of costal projection: 0 (in basal 1/2), 1 (beyond basal 1/2).

8. Length of Sc: 0 (>0.8 of wing length), 1 (<0.8 of wing length).

9. Number of cross veins: 0 (25 or more), 1 (less than 20).

MALE GENITALIA

10. Basal swelling on segment I of forceps: 0 (absent), 1 (inner corner rounded), 2 ( quadrangular), 3 (inner corner acute, projecting).

11. Ventral subapical projections on penis lobes: 0 (absent), 1 (long, digitiform), 2 (me- dium-length, digitiform), 3 (short, digitiform), 4 (very short, shallow prominence), 5 (small, spine-like).

12. Shape of apex of penis lobes: 0 (rounded), 1 (rather straight, inner corner obtuse, margin entire), 2 (as in 1, but margin with spines).

NYMPHS

LABRUM

13. Large denticles on anteromedian emargination: 0 (absent), 1 (present, subequal), 2 (present, median one larger).

14. Dorsal row of setae: 0 (apical), 1 (medial), 2 (basal).

15. Dorsal row of setae: 0 (entire, straight), 1 (entire, sinusoidal), 2 (divided).

38

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

MAXILLAE

16. Tusk on inner apical margin: 0 (absent), 1 (present, small to medium size), 2 (present, big).

17. Palpifer size: 0 (normal), 1 (enlarged).

18. Ratio of segment Ill/segment II of palpi: 0 (<0.8), 1 (>0.9).

MANDIBLE

19. Shape of outer margin: 0 (evenly curved), 1 (strongly curved).

20. Row of setae at base of outer incisor: 0 (no), 1 (yes).

LABIUM

21. Dorsal row of long spines on palpi: 0 (on segment III only), 1 (on segments II and III).

22. Row of long setae on paraglossae: 0 (absent), 1 (present, apical), 2 (present, medial).

23. Spines on margin of glossae: 0 (numerous (>10), small), 1 (few (<9))

LEGS

24. Lines of pectinate setae on dorsum of tibia III: 0 (absent), 1 (almost two lines), 2 (one main line, bifurcated at 1/2 length).

25. Spines on dorsum of femora II-III: 0 (numerous), 1 (few).

ABDOMEN

26. Posterolateral projections on abdominal segments VIII-IX: 0 (narrow, with lateral setae), 1 (wide, with lateral spines).

APPENDIX II

Data matrix for the taxa used in this study. Description of character states given in Appendix I. Unknown conditions indicated by polymorphic by “ — Out- groups indicated by *.

Taxon

Character state

*Meridialaris	00000-00000?	00000000000000
^Atopophlebia	100000101010	10000000020000
Ulmeritus balteatus	110020000130	12221111110201
Ulmeritus carbonelli	110010000110	12221111110201
Ulmeritus saopaulensis	110010000120
Ulmeritoides flavopedes	110001011300
Ulmeritoides uruguayensis	110001011301
Ulmeritoides spinulipenis	110001011302	9999999999999'P
Ulmeritoides fidalgoi	110001011350	99999999999999
U Imeritoides misionensis	110001010301	21110001121101
Ulmeritoides tijferae	101000011240	11110001121101
Ulmeritoides guanacaste	101101021240	11110001121111

J. New York Entomol. Soc. 103(1 );39^7, 1995

REVIEW OE THE GENUS COENUS DALLAS, WITH THE DESCRIPTION OF C. EXPLANATUS, NEW SPECIES (HETEROPTERA: PENTATOMIDAE)

D. A. Rider

Department of Entomology, North Dakota State University,

Fargo, North Dakota 58105

Abstract. — Coenus explanatus, new species, is described from Alabama and Georgia. Diag- noses are provided for the genus as well as both previously described species, C. delius (Say) and C. inermis Harris and Johnson. A key to aid in the identification of species of Coenus is given.

While sorting through pentatomid specimens at the National Museum of Natural History, six specimens of an apparently undescribed species of Coenus Dallas were discovered. Credit for this discovery, however, must be given to H. G. Barber as one of the specimens bears the label: ''Coenus n. sp. det. HG Barber.” He apparently never described the species. It is described herein, along with a review of the genus.

The genus Coenus, which lacks a medial spine or tubercle on the third abdominal segment, has been placed in section one of the Pentatomini (sensu Rolston et al., 1980), and is included in a key to section-one genera provided by Rolston and McDonald (1984).

When type label data are cited in the text, each letter in parentheses represents a separate label, with (a) being closest to the specimen on the pin. Type label data are cited as written and placed within quotation marks. All measurements are in milli- meters; measurements in parentheses are of the holotype.

Coenus Dallas, 1851

Coenus Dallas, 1851:194, 230; Stal, 1867:526; Kirkaldy, 1909:72.

Caenus [sic]: Lethierry & Severin, 1893:132.

Type species: Coenus tarsalis Dallas, 1851 [=Pentatoma delta Say, 1831], by mono- typy.

Diagnosis. Third (second visible) abdominal sternite unarmed. Each ostiolar ruga short, auriculate, reaching less than one-fourth distance to lateral metapleural margin; evaporatoria punctured, reaching to about middle of supporting metapleural plate (Fig. 5). Each buccula distinctly lobed posteriorly, reaching to posterior margin of head (Fig. 7). Rostrum reaching to metacoxae; first segment not reaching beyond posterior margins of bucculae. Juga and tylus subequal in length. Inferior surface of each femur often armed with row of widely spaced, small spines; superior surface unarmed distally. Lateral margins of pronotum sharp, but not reflexed; anterior mar- gin not elevated. Width of scutellum at distal ends of frena more than two-thirds basal width; apex broadly rounded, reaching to or beyond apices of coria (Figs. 1, 4, 8). Tarsi 3-segmented; coxae nearly contiguous. Prosternum and mesosternum flat

40

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

Eigs. 1-9. 1-3. Coenus delius. 1. Habitus. 2. Genital plates, caudoventral view. 3. Spermatheca.

4—7. Coenus explanatus. 4. Habitus. 5. Meso- and Metapleural plates. 6. Spermathecal pump. 7. Head, lateral view. 8, 9. Coenus inennis. 8. Habitus. 9. Spermathecal pump. Symbols: bl, buccal lobe; bp, basal plates; ea, evaporative area; gx2, second gonocoxae; ost, ostiolar mga; spb, sper- mathecal bulb; sr, sclerotized rod; slO, tenth stemite; 8pt, eighth paratergites; 9pt, ninth paratergites.

1995

REVIEW OF THE GENUS COENUS

41

posteriorly, each with slight mesial carina anteriorly; metasternum weakly sulcate. Wing venation reticulate.

Parameres F-shaped; proctiger heavily sclerotized; aedeagus with lateral conjunc- tival lobes; median penial plates large; penisfilum long, coiled; dorsoposterior margin of theca tri-lobed, lateral lobes obtuse, medial lobe narrower, nearly acute; with pair of dorsal thecal appendages just dorsad of median penial plates. Mesial margins of basal plates nearly straight; posterior margins sinuous; posteromesial angles not pro- duced caudad. Spermathecal bulb globose, lacking diverticula; spermathecal duct coiled between spermathecal pump and dilation of spermatheca.

Comments. Coenus belongs to a group of genera which includes the well-known genus Euschistus Dallas. All genera within the group possess dorsal thecal append- ages which appear to be unique to this group. Coenus is most closely related to Hymenarcys Amyot and Serville, from which it can be separated by the scutellum not extending beyond the apices of coria.

Key to species of Coenus Dallas

1. Anterolateral pronotal margins concave; anterior angles explanately pro- duced anterolaterad (Fig. 4) (southeastern U.S.) .... explanatus new species Anterolateral pronotal margins straight to convex; neither explanate nor

anteriorly (Figs. 1,8) 2

2(1) Maculations on superior surface of each tibia small, brown, each encircling hair; posterior margin of pygophore with distinct medial tooth (Figs. 24,

25) (east of Rocky Mountains, excluding southeastern U.S.) . . . delius (Say) Maculations on superior surface of each tibia large, irregular, fuscous; pos- terior margin of pygophore entire, lacking medial tooth (Figs. 17, 18) (Lou- isiana, Arkansas, Texas, Oklahoma, Kansas) .... inennis Harris and Johnson

Coenus explanatus, new species Figs. 4-7, 10-16, Map 1

Description. Oval, slightly convex dorsally, strongly convex ventrally. Dorsal sur- face stramineous to pale yellow-brown; punctures brown to fuscous.

Apex of head rounded; lateral jugal margins not reflexed, sinuous, not quite par- allel. Surface of head relatively flat, vertex slightly convex; anterior one-third of head slightly declivant in lateral view; tylus transversely convex, slightly elevated above jugal surfaces. Punctures becoming darker and more crowded near lateral margins, relatively sparse between eye and adjacent ocellus; punctures on tylus smaller and paler than on juga, giving head appearance of having longitudinal, pale, medial band. Antennae rather uniformly red-brown to dark brown, sometimes seg- ment V and distal one-half of segment IV slightly darker.

Anterolateral pronotal margins distinctly concave, anterior angles explanately pro- duced anterolaterad (Fig. 4); punctures slightly more dense near humeral angles and mesially anterior to cicatrices. Pale median line on pronotum and scutellum usually obsolescent. Scutellum and coria uniformly punctured. Connexival punctures darker and more dense near segmental junctures, appearing alternated dark brown and pale yellow. Wing membranes hyaline, veins brown.

Ventral surface pale yellow, punctures brown to fuscous, becoming slightly more

42

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

Eigs. 10-30. 10-16. Coenus explanatus. 10. Posterior margin of pygophore, ventral view.

11, 12. Pygophore. 1 1. Dorsal view. 12. Caudal view. 13, 14. Right paramere. 13. Lateral view. 14. Medial view. 15, 16. Theca and related structures. 15. Lateral view. 16. Dorsal view. 17- 23. Coenus inennis. 17. Posterior margin of pygophore, ventral view. 18, 19. Pygophore. 18. Dorsal view. 19. Caudal view. 20, 21. Right paramere. 20. Lateral view. 21. Medial view. 22, 23. Theca and related structures. 22. Lateral view. 23. Dorsal view. 24-30. Coenus delius. 24.

1995

REVIEW OE THE GENUS COENUS

43

dense and forming vague longitudinal, brown band on each side mesad of spiracles. Head with fuscous vitta just dorsad of each antenniferal tubercle. Rostrum pale brown, segment IV fuscous, reaching between metacoxae. Spiracles piceous. A few punctures slightly darker medially along posterior margin of each abdominal sternite, sometimes with small medial, longitudinal vitta on segment VII; anterolateral angles of each abdominal sternite with large fuscous spot. Females with elongate, trans- verse, pale, impunctate area on each side of median of abdominal segment III. Males with segment III as in females, but also with similar, narrower, areas on segments IV and V. Legs heavily and irregularly maculated with fuscous, including numerous spots on both sides of tibial sulcations; tibial sulcations distinct, often dark brown to black.

Posterior margin of pygophore arcuate, slightly more produced than in C. inermis, lacking medial tooth (Figs. 10, 11); inferior ridge broadly V-shaped, sinuous medially in caudal view (Fig. 12). Penisfilum making nearly two complete revolutions, with two to three coils or twists basally (Fig. 15). Dorsal thecal appendages abruptly curved ventrad near middle (Fig. 15). Parameres rather robust, broadest near middle (Figs. 13, 14).

Measurements. Total length excluding wing membranes 8.97-10.39 (8.97); total width 5.48-5.83 (5.48); medial length of pronotum 2.42-2.64 (2.42). Medial length of scutellum 3.81-4.21 (3.87); basal width 3.57^.00 (3.62); width at distal end of frena 2.58-2.96 (2.58). Length of head from apex to imaginary line drawn through posterior margins of ocelli 1.97-2.04 (2.03); width 2.60-2.73 (2.66); intraocular width 1.81-1.86 (1.82); intraocellar width 1.02-1.13 (1.07); ocellar diameter 0.05- 0.08 (0.07); distance from eye to adjacent ocellus 0.46-0.41 (0.45). Length of seg- ments I-V of antennae 0.52-0.61 (0.53), 0.64-0.80 (0.64), 1.10-1.27 (1.22), 1.01- 1.09 (1.01), and 1.31-1.44 (1.42), respectively. Length of segments II-IV of rostrum 1.58-1.75, 0.70-0.86, 0.69-0.82 (unable to measure holotype), respectively. Holotype. 6, labeled (a) “Adel, Ga 8-11-39 J. D. Beamer” (b) “K. U.” deposited in the National Museum of Natural History (Washington, D.C.).

Paratypes. 2d c3, 3 9 9 . 19 labeled as holotype (USNM); 2S S labeled “Ga.” (DAR, USNM); 19 labeled (a) “Florala, Ala. 1938 L. Henderson” (b) “K. U.” (DAR); and 1 9 labeled (a) “TIFTON GA 6-30-1936” (b) “Collected on cotton” (c) “P. A. Glick Coll.” (d) “Coenus n. sp. det HGBarber” (USNM).

Distribution. Southeastern U.S. (Map 1).

Comments. This species is closely related to C. inermis, but can be separated from it and C. delius by the form of the anterolateral pronotal margins.

All members of the type series were collected in the 1930’s; I have not seen any specimens collected more recently. It appears that this species may either be extinct or at least very rare.

Posterior margin of pygophore, ventral view. 25, 26. Pygophore. 25. Dorsal view. 26. Caudal view. 27, 28. Right paramere. 27. Lateral view. 28. Medial view. 29, 30. Theca and related structures. 29. Lateral view. 30. Dorsal view. Symbols: dtr, dorsal thecal appendages; ir, inferior ridge; mpp, median penial plates; par, parameres; pen, penisfilum; pro, proctiger; th, theca.

44

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

Map 1. United States distribution of Coeuus delius [•], C. explanatus [O], and C. inermis [□].

Etymology. This species is named for the distinctive condition of the anterolateral margins of the pronotum.

Coenus delius (Say)

Figs. 1-3, 24-30, Map 1

Pentatoma delia Say, 1831:8; Say, 1859:320.

Hymenarcys aeruginosa Amyot & Serville, 1843:125. (syn. by Stal, 1872).

Coenus tarsalis Dallas, 1851:230, pi. 8, fig. 6. (syn. by StM, 1867).

Coenus punctatissimus Vollenhoven, 1868:183. (syn. by Stal, 1872).

Coenus delius: Kirkaldy, 1909:72; McPherson, 1982:68-69; Froeschner, 1988:573- 574.

Diagnosis. Antennae red brown to dark brown, usually segment V and distal one- half of segment IV distinctly darker. Anterolateral pronotal margins straight to slight- ly convex; anterior angles not produced or explanate (Fig. 1). Pale median line usually relatively distinct on pronotum and scutellum; pronotal punctures usually relatively dense along anterolateral margins, especially anteriorly. Abdominal seg- ments III-V lacking pale, impunctate areas in both males and females. Maculations on legs rather uniform, relatively sparse; those on superior surface of each tibia limited to very small spot at base of each hair.

Posterior margin of pygophore in dorsal and ventral views truncate with distinct medial tooth (Figs. 24, 25); inferior ridge sinuous in caudal view (Fig. 26). Penis- filum making one complete revolution, not coiled at base (Fig. 29). Dorsal thecal

1995

REVIEW OE THE GENUS COENUS

45

appendages rather uniformly curved ventrad (Fig. 29). Parameres relatively narrow, shaft of each somewhat sinuous (Figs. 27, 28).

Types. Most of Say’s types have been destroyed. The identity of this species is now well established.

Distribution. This species occurs from the Rocky Mountains eastward, but is ap- parently absent from the southeastern U.S. (Map 1).

Specimens examined. CANADA: Ontario: Brantford; Mer Bleu; Toronto. Quebec: Montreal; jet. rtes. 52 & 13. Saskatchewan: Oxbow.

UNITED STATES: Colorado: Boulder: Boulder. Larimer: Ft. Collins. Con- necticut: Fairfield: Norwalk; Stamford; Westport. Litchfield: Lakeville. District of Columbia: Flats; Rock Crk. Idaho: Canyon: Caldwell. Illinois: Champaign: Cham- paign. Du Page. Mason: Havana. Warren: Swan twp. Indiana: Cass: 0.5 m. S. Galveston. Fayette: Connerville. Howard: NW Howard Co. Noble: Sylvan Lake. Starke: Knox. Iowa: Nichole. Allamakee: Harpers Ferry; Waukon. Buena Vista: Sioux Rapids. Dickinson: Lake Okoboji. Emmet: Estherville. Johnson: Hills; Iowa City; Solon. Kossuth: Algona. Lyon: Little Rock. Muscatine: Muscatine. Polk: Des Moines. Sac: Sac City. Story: Ames. Warren: Indianola. Woodbury: Sioux City. Kansas: Ford: Bloom. Johnson. Phillips. Riley: Clearwaters. Rooks. Stafford: Salt Flats. Louisiana: E. Baton Rouge: Pride. Maine: Pretty Marsh. Kennebec: Augusta. Penobscot: Orono. Maryland: Crampton Gap; Meyersville; Plummer’s Isl. Allegany: Cumberland. Anne Arundel: Odenton; 5 km SW Odenton, Ft. Meade. Frederick: Wolfsville. Prince Georges: Beltsville; Silver Hill. Washington: Hagerstown. Mas- sachusetts: Barnstable: Monument Beach; Woods Hole. Essex: Andover; Lynn; Swampscott. Hampshire: Amherst; Cummington; Hadley; Northampton. Middlesex: Sherborn; Tyngsboro. Michigan: Berrien: Warren Dunes. Cheybogan: Douglas Lake. Presque Isle: Ocqueoc Lake. Washtenaw: Ypsilanti. Minnesota: Ereeborn: Albert Lea. Hennepin. Winona: Homer. Missouri: Dameron. Boone: Ashland Wild- life Area; Columbia. Callaway: Tucker Prairie. Howell: West Plains. Jackson: Kan- sas City. Lewis: Canton. Polk: 3.5 m. SE Flemington. St. Charles: St. Charles. St. Louis: Clayton; Manchester; St. Louis. Wayne. Wright: Mountain Grove. Montana: Gallatin. Musselshell: Montana Expt. Stn., Musselshell. Petroleum: 1.5 m. S, 5 mi W Winnett. Nebraska: Douglas: Omaha. Johnson: Sterling. Richardson: Falls City. Sioux: Glen. New Hampshire: Grafton: Pike. Rockingham: Hampton. New Jersey: Beatyestown; Brookside. Camden: Delair. Morris: Madison. Ocean: Lakehurst. Pas- saic: Great Notch. Union: Roselle Park. New Mexico: Bernalillo: Cibola Natl. For- est. New York: Adirondack Mts., Bronx Park; Cascade Lake; Allegany St. Pk.; Flushing; Fort Montgomery; Sangerties; Wading River, Long Island. Columbia: nr. Claverack; Hudson. Dutchess: Fishkill. Erie: Buffalo; Hamburg Genesee: Oakfield. Hamilton: Indian Lake. Monroe: Rochester. Onondaga. Orange: Pine Island; West Point. Suffolk: Cold Spring Harbor, Long Island. Tompkins: Ithaca; McLean Bogs. North Dakota: Tekio. Benson. Billings: Easy Hill; Theodore Roosevelt National Park. Burleigh: Bismark. Cass: Fargo. McHenry. Mckenzie: Theodore Roosevelt National Park. Morton. Nelson. Richland: Hwy 27, 5 mi W Hwy 18. Ransom: Lis- bon; McLeod; 4 mi N McLeod. Slope: Burning Coal Vein; Chalky Buttes. Williams: Williston. Ohio: Ashtabula: Jefferson. Clinton. Geauga: Bainbridge. Highland. Hocking. Licking: Buckeye Lake. Oklahoma: Latimer. Pennsylvania: Philadelphia; Springbrook. Allegheny: Pittsburgh. Cambria: Patton. Eulton: Warfordsburg. Mon-

46

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

roe: Pocono Lake. Perry: 1.3 mi NW New Bloomfield. Wayne: Honesdale. West- morland: Greensburg; Jeannette; nr. Rector. York: 2 mi W Airville. South Dakota: Fall River: Ardmore. Jackson: Weta. Lake: Madison. Minnehaha: Sioux Falls. Tex- as: Brewster: Hwy 90, 1.1 mi W Pecos Co. Brown: Bangs. Tarrant: Ft. Worth Nature Center. Wilbarger: Vernon. Utah: Salt Lake. Utah: Spanish Fork. Weber: Ogden. Virginia: Middle Mt.; Thornton’s Gap. Fairfax: Vienna. Nelson. West Virginia: Preston: Cathedral St. Park. Wisconsin: Dane: Madison. Jackson: Black River Falls. Jefferson: Watertown. Monroe: Warrens. Polk: Osceola. Waukesha: Pewaukee. Wy- oming: Sheridan: Big Horn.

Comments. This species averages slightly smaller than either C. explanatus or C. inermis. It lacks the impunctate areas on the base of the abdomen, the maculations on the tibiae are smaller and more sparse, and the medial tooth on the posterior margin of the pygophore is diagnostic.

Coenus inermis Harris and Johnson Figs. 8, 9, 17-23, Map 1

Coenus inermis Harris & Johnson, 1936:378; McPherson, 1982:68; Froeschner, 1988:574.

Diagnosis. Antennal segments rather uniformly red-brown to dark brown, sometimes segment V slightly darker. Anterolateral margins straight, anterior angles not pro- duced or explanate (Fig. 8). Pale median line on pronotum and scutellum usually obsolescent. Females with elongate, transverse, pale, impunctate area on each side of median of abdominal segment III. Males with segment III as in females, but also with similar, narrower areas on segments IV and V. Maculations on legs relatively large and irregular, including many large maculations on each side of tibial sulca- tions; tibial sulcations relatively shallow.

Posterior margin of pygophore smoothly arcuate in both dorsal and ventral views, lacking medial tooth (Figs. 17, 18); inferior ridge well-developed, V-shaped in caudal view (Fig. 19). Dorsal thecal appendages rather abruptly curved ventrad near distal one-third (Fig. 22); penisfilum making nearly two complete revolutions in opposite direction than C. debus or C. explanatus', not coiled or twisted basally (Fig. 22). Parameres rather robust, broadest near middle (Figs. 20, 21).

Types. This species was described from 1 6 and 3 9 $ specimens from Arkansas and Oklahoma. The holotype was examined, and is conserved in the U.S. National Mu- seum of Natural History, Washington, D.C.

Distribution. Arkansas, Louisiana, Kansas, Missouri, Oklahoma, Texas (Map 1). Specimens examined. Arkansas: Hempstead: Hope. Marion. Kansas: Montgom- ery: Elk City. Louisiana: Union. Oklahoma: Atoka: 1 mi. S Atoka. McCurtain. Texas: Bowie: Maud. Brazos: College Station. Colorado: Alleyton.

Comments. This species is also relatively rare. It is closely related to C. explanatus from which it is easily distinguished by the shape of the anterolateral pronotal mar- gins.

ACKNOWLEDGMENTS

I would like to thank all those individuals at various institutions who generously lent spec- imens pertinent to this study. Special thanks go to T. J. Henry, U.S. National Museum of Natural

1995

REVIEW OE THE GENUS COEN US

41

History, Washington, D.C., who arranged for me to work at the USNM at the time the specimens of C. explanatus were discovered, and the holotype of C. inerrnis was examined. I would also like to thank G. Eauske, and H. J. Meyer, North Dakota State University, J. A. Moore, Eargo, ND, and L. H. Rolston, Louisiana State University, for their comments on an early draft of the manuscript.

LITERATURE CITED

Amyot, C. J. B. and J. G. A. Serville. 1843. Histoire Naturelle des Insectes. Hemipteres. Librairie Encyclopedique de Roret ed., Paris. Ixxvi and 681 pp., 12 pis.

Dallas, W. S. 1851. List of the specimens of hemipterous insects in the collection of the British Museum. Brit. Mus. London, 2 pts., 592 pp., 15 pis.

Eroeschner, R. C. 1988. Eamily Pentatomidae Leach, 1815. The stink bugs. Pages 544-607, ill. in: T. J. Henry and R. C. Eroeschner (eds.). Catalog of the Heteroptera, or true bugs, of Canada and the continental United States. E. J. Brill, Leiden, New York, xix + 958 pp.

Harris, H. M. and H. G. Johnston. 1936. A new genus and species of Podopidae and a new Coenus (Hemiptera: Scutelleroideae [sic]). Iowa St. Coll. J. Sci. 10(4):377-380.

Kirkaldy, G. W. 1909. Catalogue of the Hemiptera (Heteroptera) with biological and anatomical references, lists of foodplants and parasites, etc. Vol. I. Cimicidae. Berlin, xi and 392 pp.

Lethierry, L. and G. Severin. 1893. Catalogue general des Hemipteres. Vol. 1. Heteropteres. Pentatomidae. R. Eriedlander and Eils, Bruxelles, x + 286 pp.

McPherson, J. E. 1982. The Pentatomoidea (Hemiptera) of northeastern North America with emphasis on the fauna of Illinois. S. 111. Univ. Pr., Carbondale and Edwardsville, 240 pp., ill.

Rolston, L. H. and E J. D. McDonald. 1984. A conspectus of Pentatomini of the Western Hemisphere. Part 3 (Hemiptera: Pentatomidae). J. N.Y. Ent. Soc. 92(l):69-86, 54 figs.

Rolston, L. H., E J. D. McDonald and D. B. Thomas, Jr. 1980. A conspectus of Pentatomini genera of the Western Hemisphere. Part I (Hemiptera: Pentatomidae). J. N.Y Ent. Soc. 88(2): 120-1 32, 24 figs.

Say, T. 1831. Descriptions of new species of heteropterous Hemiptera of North America. New Harmony, Indiana, 39 pp.

Say, T. 1859. The complete writings of Thomas Say on the entomology of North America, with a memoir of the author, by George Ord. Ed. by J. L. LeConte. New York, 2 vols. Vol. I: XXIV & 412 pp. & 54 pis., Vol. II: IV & 814 pp. (Reprinted in Philadelphia in 1891.)

StM, C. 1867. Bidrag till Hemipterernas systematik. Conspectus generum Pentatomidum Amer- icae. Ofv. Vet. Ak. Forh. 24(7):522-532.

Stal, C. 1872. Enumeratio Hemipterorum. Bidrag till en forteckning ofver alia hittels kanda Hemiptera, Jemte Systematiska meddelanden. 2. Sv. Vet. Ak. Handl. 10(4): 1-159.

Vollenhoven, S. C. S. van. 1868. Diagnosen van eenige nieuwe soorten van Hemiptera Het- eroptera. Versl. Meded. k. Akad. Wetens Naturk. Amst. (2)2:172-188.

Received 31 March 1994; accepted 10 April 1995.

J. New York Entomol. Soc. 103( 1 ):48-54, 1995

PROTOPOLYBIA BITUBERCULATA, A NEW NEOTROPICAL SOCIAL WASP (HYMENOPTERA: VESPIDAE; POLISTINAE)

Orlando Tobias Silveira and James M. Carpenter

Departamento de Zoologia, Sec. Entomologia, Museu Paraense Emilio Goeldi, Av. Perimetral S/N, Guama, C.P 399, 66040-170 Belem, Para, Brasil, and Department of Entomology, American Museum of Natural History,

Central Park West at 79th Street, New York, New York 10024, U.S.A.

Abstract. — Protopolybia bitubercidata, a new polistine species from the Neotropics, is de- scribed and the nest illustrated.

Protopolybia Ducke is a genus of small neotropical social wasps, belonging to a tribe, Epiponini, the members of which found new colonies by swarms of queens and workers (Carpenter, 1993). The most recent revision (Richards, 1978) recognized 23 species, but with the synonymy of Pseudochcirtergus with Protopolybia (Carpen- ter and Wenzel, 1990) five more species are now included in the genus, for a total of 28. To this total we are adding a new species.

The presently described species has not been properly recognized as a distinct taxon. Ducke (1910:475) misidentified it as a color form of Protopolybia sedula van exigua (Saussure). In the collection of the Goeldi Museum there are specimens la- belled Protopolybia minutissima van sedula (Saussure), probably following the re- vision of the genus by Bequaert (1944). This latter name is a misidentification of Protopolybia exigua (see Richards, 1978). The name sedula is the senior synonym of a different species, called Protopolybia pumila (Saussure) by Bequaert (Richards, 1978).

Protopolybia bituberculata, new species (Eigs. lA, 2 A, 3)

Protopolybia sedula van exigua (de Saussure): Ducke, 1910:474 (in part). Misiden- tification.

Diagnosis: Propodeum with posterior face projecting symmetrically to either side of median furrow, forming two moderately high protuberances (Pig. lA); first meta- somal tergum petiolate, clearly longer than wide at apex. Male genitalia with the medial lobes of the aedeagus pointed laterally and strongly sclerotized (Fig. 2A). Small species; color brown or black and yellow.

Description:

Female: Mean forewing length 4.0 mm. Structure — cuticle finely reticulate, un- punctured except for shallow punctures on posterior margin of metasomal segments; clypeus a little higher than wide, ventral margin forming a rounded median lobe; dorsal pronotal carina low, obtuse, but distinct at sides, gradually sloping posteriad on the humeri; fovea well developed, anterior carina low, obtuse; mesoscutum as long as wide; scutellum slightly convex with an incomplete median line; metanotum

1995

NEW NEOTROPICAL SOCIAL WASP

49

about 1.25X as wide as long, lobe moderately acute apically; propodeum with pos- terior surface projecting to either side of the median furrow, forming two moderately developed protuberances (Fig. lA); first tergum about 1.2 times longer than wide at apex, with a distinct basal petiole; second tergum a little wider than long, about twice as wide as first tergum at apex.

Color — ground color black to brown; margin of mandibular teeth and ventral margin of clypeus, reddish; antennae testaceous brown to black, scape yellow be- neath with flagellum orange; abundant yellow markings, including most of clypeus aside from central inverted bifid mark, bifid frontal mark, ocular loops, most of pronotum except band on anterior surface down to level of foveae and humeral mark, two narrow stripes on mesoscutum, large mesepisternal mark, sometimes connected to spot below scrobe posteriorly, most of tegula yellow except outer margin, most of scutellum, axilla, and metanotum anteriorly, spot on upper part of metapleuron, most of propodeum except two antero-dorsal spots and central stripe, all of forecox- ae, trochanters and femora ventrally, midcoxae anteriorly, posterior margin of first tergum, broad transverse band on second tergum basally, posterior bands on terga II- V, two lateral spots on base of second sternum, posterior bands on sterna II-IV; wings hyaline with dark brown venation.

Vestiture — body covered by short appressed pubescence and more scattered short outstanding bristles, sometimes lacking in poorly preserved specimens; longer hairs on lower margin of clypeus, on propodeum, and distal metasomal segments.

Variation: The color pattern described above is not constant but the degree of variation is nevertheless small, with regard to differences in the proportions of yellow and dark. The yellow on the legs is variable, with often the tibiae and tarsi, and most of the forelegs, yellow. The specimens from Peru and Ecuador tend to be darker, and to have the propodeal protuberances somewhat more pronounced.

Male: General structure and color like the female, aside from the usual sexual di- morphism. Clypeus narrower and ventrally somewhat depressed, extensively covered by silvery hairs; antennae with 13 articles; eyes more swollen below; genae narrow- er; last metasomal sternum is flattened. There is proportionally less yellow, so that on the vertex two small yellow spots are defined behind the posterior ocelli, and the pronotum has just narrow transverse and humeral yellow stripes. The second meta- somal tergum has a pair of basal spots, not a transverse band.

The male genitalia appear to provide excellent diagnostic features. They are quite different from P. bella (von Ihering) (based on dissection of a male from Panama), P. exigua exigua (male from Brasil, Goias), P. holoxantha (Ducke) (specimen from Guyana), P. panamensis (Zavattari) (specimen from Panama), P. sedula (Saussure) (two males from Peru, Loreto), P. weyrauchi (specimen from Peru, Junm) and P. wheeleri Bequaert (specimen from Panama). Most notably, the medial lobes of the aedeagus taper to laterally-directed points and are strongly sclerotized in bituber- culata (Fig. 2A), appearing almost like hooks. In the other species examined, these lobes are apically more or less truncate (although differing in shape. Fig. 2b, c, and with a nipplelike tubercle in panamensis), directed ventrally, and much more weakly sclerotized. In the other species the volsella projects as a distinct medial angle in ventral view, about where the cuspis joins the lamina volsellaris ventral to the base of the digitus, but it is essentially flat in bitubercidata. A definitive judgement on

50

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

1995

NEW NEOTROPICAL SOCIAL WASP

51

A

B

C

Fig. 2. Aedeagus, ventral view. A, Protopolybia bituberculata. B, P. exigua binominata. C, P. weyrauchi. The magnification is 62.5 X.

the utility of these characters, however, must await study of more species of Pro- topolybia, which like other polistines have received little study of male genitalia. Nest: Nine colonies of P. bituberculata, including nests, were collected by JMC and John W. Wenzel in Ecuador and Peru. A further two nests were seen in the Goeldi Museum collection. The nests show typical features of the genus (Fig. 3). The comb is suspended from the supporting leaf by one central and several lateral auxiliary peduncles, and is covered by an envelope with one or more lateral exit-holes. The envelope is brown, usually mottled with whitish streaks which may be spot-like. The envelope ranges in shape from more or less oval (e.g.. Fig. 3 A, 30 X 20 mm) to quite slender and spindle-shaped (e.g.. Fig. 3B, 45 X 15 mm). Six of the nests from Peru and Ecuador, and one from Brasil, were somewhat enveloped by the adjacent leaves, to which the envelope was attached. In three of these nests, 901224-10, 901225-6 and 901227-3, the comb followed the curvature of the supporting leaf, to the extent that the comb almost folded back on itself lengthwise. In one of the Brazilian nests, a second lateral comb was being constructed (Fig. 3B), and in nest 901227-3 a complete second comb was separate and provided with its own envelope. One of the nests, 901217-19, was incipient, with a single naked comb (17 X 15 mm) suspended by a central peduncle. In another nest, 901229-6, the comb was only partly covered by the envelope.

Distribution: Brasil: Amazonas, Para, and Maranhao; Peru: Loreto; Ecuador: Napo. Type material: holotype female Brasil, Para, Vigia, Campo do Palha, 08-xii-1988 (I. S. Gorayeb). Paratypes: Brasil, Amazonas, Tefe, 9-1904, 1 female, 28-9-1904, 1

Fig. 1. Propodeum, oblique posterior view. A, Protopolybia bituberculata. B, P. exigua binominata.

52

JOURNAL OF THE NEW YORK ENTOMOEOGICAL SOCIETY Vol. 103(1)

Eig. 3. Two nests of P. bituherciilata. The scale bar is 1 cm.

male; Maranhao, S. Luiz, 3-6-1907, 1 female; Para, Benevides, 81-911, 3 females; Para, 16-12-1911, 1 female; 1912, 1 female (A. Ducke); Para, Belem, Utinga, 14- xi-1967, 12 females (R. L. Jeanne); Para, S. Francisco, 8-vii-1977, 13 females (W. L. Overal); Para, S. Miguel, 1 l-iv-1979, 24 females (F. F Ramos, W. Franca, and R. B. Neto); Para, Vigia, Campina, 7-xii-1988, 1 female (I. S. Gorayeb); Para, Capitao P090, 25-ii-1978, 1 female (W. Franga); Amazonas, Alvaraes, 17-6-1994, 26 females (O. T. Silveira and I. S. Gorayeb); Peru, Loreto, 80 km NE Iquitos, 22-12-1990, nest 901222-10, 81 females, 2 males (J. M. Carpenter and J. W. Wenzel); Loreto, 80 km NE Iquitos, 22-12-1990, nest 901222-14, 11 females (J. M. Carpenter and J. W. Wenzel); Loreto, Rio Sucusari at Napo, 24-12-1990, nest 901224-10, 240 females, 4 males (J. M. Carpenter and J. W. Wenzel), 4 females emerged 25-12-1990; Loreto, Rio Sucusari at Napo, 25-12-1990, nest 901225-6, 170 females, 31 males (J. M. Carpenter and J. W. Wenzel); Loreto, 80 km NE Iquitos, 27-12-1990, nest 901227- 3, 70 females (J. M. Carpenter and J. W. Wenzel); Loreto, 40 km NE Iquitos, 29- 12-1990, nest 901229-6, 26 females, 1 male (J. M. Carpenter and J. W. Wenzel); Loreto, 40 km NE Iquitos, 29-12-1990, nest 901229-7, 218 females (J. M. Carpenter and J. W. Wenzel); Ecuador, Napo, Tena, 16-12-1990, nest 901216-12, 115 females preserved in ethanol (J. M. Carpenter and J. W. Wenzel); Napo, Tena, 17-12-1990, nest 901217-19, 40 females (J. M. Carpenter and J. W. Wenzel).

Holotype and 93 paratypes deposited in the Goeldi Museum (Museu Paraense Emilio Goeldi, Belem/PA, Brazil). More than 1000 paratype specimens in the Amer- ican Museum of Natural History (New York, USA).

1995

NEW NEOTROPICAL SOCIAL WASP

53

Etymology: the specific name is a reference to the outstanding diagnostic feature of P. bituberculata, the posterolateral propodeal projections.

REMARKS

Protopolybia bituberculata is evidently related to those species of the genus in which the first metasomal segment is petiolate, being longer than broad at the apex, and the propodeal concavity is a narrow furrow. These are probably derived traits, but phylogenetic relationships among the species of Protopolybia are unclear (Car- penter and Wenzel, 1990). In the key of Richards (1978), P. bituberculata runs to couplet 7, which leads either to P. rubrithorax Bequaert or P. exigua (Saussure). P. rubrithorax has a color pattern unique in the genus, with the pronotum and mesos- cutum light reddish and the metasoma mostly black. It also has a very localized distribution, hitherto recorded only from Peru. P. exigua is most similar in color to P. bituberculata but, as shown above, is readily distinguished by the propodeum and the male genitalia. In P. bituberculata the posterior face of the propodeum projects on each side of the median furrow (Fig. lA). In P. exigua that surface is evenly rounded, not projecting (Fig. IB). Males of P. bituberculata differ from P. exigua in having the medial lobes of the aedeagus pointed laterally and strongly sclerotized, appearing hooklike (Fig. 2A) rather than truncate (Fig. 2B).

Ducke (1910:475) treated his specimens of P. bituberculata from Belem and Tefe (Brazil, Amazonia) as variants of P. sedula van exigua, recognizable by a ferruginous ground color. Ducke treated this color form as conesponding to Polybia palmarum Blanchard, described from Guatemala and according to Ducke (1910:474), a syn- onym of P. sedula van exigua. This synonymy was questioned by Bequaert (1944: 1 10), because Bequaert had not seen P. sedula [= exigua] specimens from north of Panama. Bequaert also commented on the very poor description of Polybia palma- rum given by Blanchard saying “The figures of the nest, as well as the size of the wasp, merely allow the conclusion that it was a Protopolybia.'' While noting that Ducke apparently saw specimens of Polybia palmarum at the Paris Museum, Be- quaert concluded as more probable that Blanchard’s species was in fact Protopolybia acutiscutis (Cameron), a species common in Guatemala. The name palmarum was overlooked by Richards (1978), but one of us (JMC) has seen 10 females in the Paris Museum, labelled “Mexique” and marked as types of palmarum. These are evidently syntypes despite the imprecise locality; the specimen bearing the type and determination label is also labelled as exigua by Buysson. Six of the specimens are callows, paler than the other specimens and with crumpled wings. Their color is fundamentally yellow, as stated by Blanchard, not ferruginous as stated by Ducke. However, Ducke’s synonymy with the typical form of exigua is correct, and bitu- berculata is quite distinct from this taxon.

ACKNOWLEDGMENTS

OTS was supported by a grant (300076/91-6 RN) from the Brazilian “Conselho Nacional de Desenvolvimento Cientifico e Tecnologico” (CNPq). Field work by JMC to collect this new species was supported by NSF grant BSR-9006102. The assistance of John W. Wenzel is greatly appreciated. We thank J. Casevitz-Weulersse for a loan of the holotype of exigua and specimens of palmarum from the Museum National d’Histoire Naturelle, Paris.

54

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

LITERATURE CITED

Bequaert, J. 1944. A revision of Protopolybia Ducke, a genus of neotropical social wasps (Hymenoptera, Vespidae). Rev. Ent., Rio de Janeiro 15:97-134.

Carpenter, J. M. 1993. Biogeographic patterns in the Vespidae (Hymenoptera): Two views of Africa and South America. In R Goldblatt (ed.). Biological Relationships between Africa and South America, 139-155. New Haven: Yale Univ. Press.

Carpenter, J. M. and J. W. Wenzel. 1990 (1989). Synonymy of the genera Protopolybia and Pseudochartergus (Hymenoptera: Vespidae; Polistinae). Psyche 96:177-186.

Ducke, A. 1910. Revision des guepes sociales polygames d’Amerique. Ann. Hist. -Nat. Mus. Natl. Hung. 8:449-544.

Richards, O. W. 1978. The social wasps of the Americas excluding the Vespinae. London: Brit. Mus. (Nat. Hist.).

Received 6 November 1994; accepted 27 September 1995.

J. New York Entomol. Soc. 103(1 ):55-68, 1995

TWO NEW SPECIES OF RHAGOVELIA FROM THE PHILIPPINES, WITH A DISCUSSION OF ZOOGEOGRAPHIC RELATIONSHIPS BETWEEN THE PHILIPPINES AND NEW GUINEA (HETEROPTERA: VELHDAE)

Dan a. Polhemus

Department of Natural Sciences, Bishop Museum, P. O. Box 19000- A, Honolulu, Hawaii 96817

Abstract. — Two new species of Rhagovelia are described from the Philippines: R. ridicula from Mindanao, Leyte, and Luzon, and R. phoretica from Negros. R. ridicula belongs to the orientalis group as defined by Polhemus and Polhemus (1988), being the apparent sister species to R. aberrans Andersen, while R. phoretica belongs to a distinctive intrageneric clade defined herein as the caesius group, whose only other members occur on New Guinea. The male paramere of R. werneri Hungerford and Matsuda is figured, and its structure is shown to be correlative to that of a monphyletic group of species occurring on New Caledonia and in the highlands of New Guinea, thus supporting the placement of this species in the novacaledonica group of Polhemus and Polhemus (1988). The evidence of a sister area relationship between the Philippines and New Guinea as indicated by the disjunct distributions of taxa in the caesius and novacaledonica species groups is discussed, and possible tectonic mechanisms that could have led to such a disjunction are evaluated.

Seventeen species of Rhagovelia were described from the Philippines prior to 1993 (Lundblad, 1936, 1937; Drake, 1948; Hungerford and Matsuda, 1961; Andersen, 1965; Polhemus and Reisen, 1976). Of these, R. hoberlandti Hungerford and Mat- suda has proved to be a synonym of R. kawakamii (Matsumura) (Polhemus and Reisen, 1976), R. teretis Drake a synonym of R. luzonica Lundblad, and R. mindan- aoensis a synonym of R. orientalis Lundblad (Polhemus and Polhemus, 1989), leav- ing fourteen valid names. Many additional species are present, including the two new taxa described herein, and a synthetic revision of the Philippine Rhagovelia fauna currently in preparation by Herbert Zettel of the Naturhistorisches Museum Wien, Austria, will raise the total even further. Polhemus and Polhemus (1988) dis- cussed the overall composition of the Philippine Rhagovelia fauna in the context of species groups they had proposed, and commented in passing on the zoogeographic affinities of these taxa. This present paper amplifies on that work, providing descrip- tions of several new species important to the zoogeographical comprehension of the region, and discussing in greater detail the faunal disjunctions between the Philip- pines and New Guinea seen in certain Rhagovelia species groups, and the tectonic mechanisms that may have produced such patterns.

TAXONOMY

Rhagovelia ridicula, new species Figs. 1, 2, 5, 9

Diagnosis: Similar to R. aberrans Andersen, but distinguished by the differing shape and placement of the erect process arising from female abdominal tergite VII (com- pare Figs. 2 and 3), and the shape of the male paramere (compare Figs. 4 and 5).

56

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

Figs. 1-8. 1. Rhagovelia ridicula n. sp. Wingless female, dorsal habitus, legs omitted.

Specimen from Cabigaan River, Leyte. 2. Rhagovelia ridicula n. sp. Lateral view of female terminal abdomen, showing location and structure of vertical process. 3. Rhagovelia aberrans

1995

RHAGOVELIA FROM THE PHILIPPINES

57

Description:

Wingless male: Form narrow and elongate, length 2.88 mm; maximum width (across posterior section of pronotum) 1.08 mm. General dorsal coloration dark grey; ante- rior portion of pronotum with frosty grey transverse band, extending downward onto pleural area; small dark orange spot centrally on anterior portion of pronotum behind head; genital segments black.

Head dark grey, eyes black, frons and vertex bearing scattered stiff dark setae; antennae black, all segments of relatively equal thickness, segment I pale yellow on basal 1/4, all segments covered with very short semi-recumbent dark pubescence, segment I also bearing 6-7 long stiff erect black setae, 1-2 setae of this type also present near middle of segment II; lengths of segments I-IV = 0.76, 0.40, 0.44, 0.44; proepisternum and jugum lacking black denticles.

Pronotum dark grey, anterior 1/4 frosty pruinose grey, length/width = 0.80/0.92, completely covering mesonotum, length of exposed metanotum at midline = 0.10; anterior pronotal lobe bearing small ovate dark orange spot behind head vertex, appearing diffuse due to overlying pruinosity; pronotal surface covered with short recumbent pale setae, these setae longer on posterior margin; disk not distinctly foveate, bearing only scattered, very small, deep pits.

Abdomen dark grey overlain with faint silvery pruinosity, tergite I longer than II (0.16), II-V equal in length (0.12), VI-VIII progressively longer (0.16, 0.20, 0.36 respectively); all tergites and connexival segments covered with short to moderate length semi-recumbent pale setae, connexival margins evenly tapering and conver- gent along their entire lengths.

Legs black, with margins of acetabulae adjoining coxae yellowish brown; all seg- ments thickly covered with short recumbent pale setae; trochanters unarmed; fore, middle and hind femora and tibiae with scattered long erect stout black setae along anterior margins; fore femur with scattered slender erect black setae along posterior margin; fore tibia simple and cylindrical, not bent or expanded; middle femur un- modified, cylindrical, tapering, bearing a line of long slender erect black setae along posterior margin; middle tibia slender, cylindrical; hind femur and tibia with scattered long fine erect brown setae along posterior margins; hind femur very weakly in- crassate, bearing a small black spine on posterior margin near middle, followed by 2-3 much smaller black spines; hind tibia straight, cylindrical, with small tuft of semi-erect black setae along anterior margin on distal 1/3.

Proportions of male legs as follows: fore femur/tibia/tarsal 1 /tarsal 2/tarsal 3 = 0.92/0.88/0.03/0.04/0.20; middle femur/tibia/tarsal 1/tarsal 2/tarsal 3 = 1.60/1.20/ 0.04/0.68/0.64; hind femur/tibia /tarsal 1/tarsal 2/tarsal 3 = 1.28/1.40/0.04/0.04/0.28.

Andersen. Lateral view of female terminal abdomen, showing location and structure of vertical process. 4. Rhagovelia aberrans Andersen. Male paramere. Specimen from Ayala River, Min- danao. 5. Rhagovelia ridicula n. sp. Male paramere. Specimen from Sapa River, Mindanao. 6. Rhagovelia phoretica n. sp. Wingless female, dorsal habitus, legs omitted. Specimen from Amulan, Negros. 7. Rhagovelia phoretica n. sp. Male paramere. Specimen from Amulan, Ne- gros. 8. Rhagovelia werneri Hungerford and Matsuda. Male paramere. Specimen from Mt. Apo, Mindanao.

58

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

Fig. 9. Distribution of Rhagovelia species in the Philippines. Circles = R. ridicula n. sp. Square = R. phoretica n. sp. (exact location of type locality within island uncertain).

1995

RHAGOVELIA FROM THE PHILIPPINES

59

Venter dark grey, weakly overlain with silvery pruinosity; mesosternum, meta- sternum and abdominal ventrites lacking small black denticles; mesosternum with incipient triangular sulcus centrally; metasternum roughly trapezoidal; abdominal ventrite I narrow, lying in semi-vertical orientation, abdominal ventrites II-VI bulg- ing and tumescent, lacking a longitudinal medial carina; ventrites IV-VI thickly set with long semi-erect pale setae directed posteriorly, ventrite VII with a sharp lon- gitudinal medial carina, this carina continuing onto ventrite VIII and widening pos- teriorly; segment VIII (first genital segment) stout, not constricted basally; proctiger and genital capsule retracted into segment VIII, proctiger without basal lobes, distal cone coming to a broad but distinct angle; paramere falciform, shape as shown in Figure 5.

Wingless female: Length 3.36 mm; maximum width (across pronotum) 1.36 mm.; body shape as in Figure 1. Similar to wingless male in general structure and color- ation with the following exceptions: lengths of antennal segments I-IV = 0.80, 0.44, 0.48, 0.48; extreme posterior margin of mesonotum visible behind posterior margin of pronotum; metanotum tumid at anterolateral angles, these swellings covered with numerous long erect black setae; connexival margins evenly converging adjacent to abdominal tergites I-V, meeting over tergite VI, then diverging slightly over tergites VII-VIII, completely covering tergites VI-VII when viewed from above; abdominal segment VII forming an elongate cylinder, with posterolateral angles produced into long projections, bearing numerous long dark setae (Fig. 2), similar setae also present along posterior margins adjacent to genital cavity; posteromedial portion of abdom- inal tergite VII produced to a vertical, finger-like process, projecting upward between connexival margins and bearing a fan-like fringe of setae at the apex (Fig. 2); proc- tiger vertical, gonocoxae recessed into abdominal segment VII, not evident; thoracic and abdominal venter lacking black denticles, abdominal ventrites lacking longitu- dinal medial carinae or long pale setae; hind femur not incrassate, bearing at most a single small black spine near middle.

Winged male: Length 3.16 mm., maximum width (across humeral angles) 1.37 mm. Similar to wingless male in general structure and coloration with following excep- tions: pronotum expanded, length/width = 1.56/1.69, black, with anterior band sil- very pruinose and bearing a small dark orange spot centrally behind head vertex, humeral angles prominent, central section tumid, posterior lobe produced to cover entire metanotum, apex coming to a broadly rounded angle, not elevated or pro- longed; hemelytra exceeding tip of abdomen, bearing 3 closed cells, with 2 long cells originating in basal portion of the plus another cell in the outer portion of the distal half between the radius and the subcosta; hemelytra uniformly dull black, with basolateral cell between subcosta and radius weakly pruinose; dorsal abdominal ca- rinae moderately long, extending to posterior margin of tergite II; hind femur with armature similar to wingless form.

Winged female: Length 3.46 mm., maximum width (across humeral angles) 1.28 mm. Similar to winged male in general structure and coloration with following ex- ceptions: pronotum length/width = 1.75/2.23; abdomen lacking vertical projection formed by tergite VII as seen in wingless form, connexiva more widely separated, revealing all abdominal tergites when viewed from above with wings removed. Discussion: Rhagovelia ridicula is a member of the orientalis group as defined by Polhemus and Polhemus (1988), and is closely allied to Rhagovelia aberrans An-

60

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

dersen. The two species may be separated by the following characters: the postero- lateral connexival angles are prolonged in ridicida and bear long hair tufts, while in aberrans they are unmodified (compare Figs. 2 and 3); the vertical projection on female abdominal tergite VII emerges from between the connexival margins near the middle of segment VII in hdicula, but near the posterior end of the connexiva in aberrans (compare Figs. 2 and 3); the basal portions of the connexival margins in aberrans bear a thick fringe of erect black setae that is absent in ridicula; the posterior margin of abdominal tergite III bears a fringe of scattered long black setae in aberrans that is lacking in ridicula; and the male parameres of the two species are differently shaped (compare Figs. 4 and 5).

Ecological notes: The Sapa River at the type locality was a slow, shaded stream flowing in a muddy bed, and surrounded by numerous habitations and rice fields. This type of habitat may not be obligatory for R. ridicula, however, since it was also taken from swift, rocky streams entering the sea on Luzon and Leyte, and from the clear, rocky Hilusig River, an upland stream on Leyte. All of the localities from which R. ridicula is presently known lie in areas in which the original forest has been heavily disturbed or converted over to agriculture, and it is possible that this species has eluded collection up to now due to its preference for disturbed habitats considered worthless by most collectors.

On Luzon R. ridicula was taken sympatrically with R. luzonica Lundblad, R. niinuta Lundblad, and R. cotabatoensis Hungerford and Matsuda. On Leyte and Mindanao, by contrast, this species was found to be sympatric with R. usingeri Hungerford and Matsuda, and R. orientalis Lundblad.

Etymology: The name “ridicula” refers to the odd and amusing abdominal modi- fications exhibited by this species.

Material examined: Holotype, wingless female: PHILIPPINES, Mindanao, South Cotabato Prov., Sapa River, SE of Koronadal, 550 m, 20 July 1985, CL 1994, D. A. and J. T. Polhemus (BPBM). Paratypes: PHILIPPINES, Mindanao, South Cot- abato Prov.: 1 winged female, 22 wingless males, 47 wingless females, same data as holotype (BPBM, JTPC). Leyte, Leyte Prov.: 3 wingless males, 9 wingless fe- males, Lusig River at Hilusig, 15 July 1985, CL 1979, D. A. and J. T. Polhemus (JTPC); 1 winged female, 10 wingless males, 15 wingless females, Cabigaan River, S. of Ormoc, sea level, 17 July 1985, CL 1987, D. A. and J. T. Polhemus (JTPC). Luzon, Pangasinan Prov.: 1 winged male, 12 wingless males, 7 wingless females, stream 6 km. W. of Sual, sea level, 6 July 1985, CL 1956, D. A. and J. T Polhemus (JTPC).

Rhagovelia aberrans Andersen Pigs. 3, 4

Discussion: Rhagovelia aberrans was originally described from Zamboanga, Min- danao, (Andersen, 1965), and the excellent figures accompanying the description illustrate all the salient morphological features of this species. Despite extensive recent collections in the Philippines this species has not been taken outside the Zamboanga Peninsula, indicating that it may be endemic to this area.

Ecological notes: A series of this species was taken by the author and J. T. Polhemus from the Ayala River, a large, unshaded, rocky river flowing swiftly into the sea

1995

RHAGOVELIA FROM THE PHILIPPINES

61

northwest of Zamboanga City. This habitat was very similar to localities on Luzon and Leyte from which R. ridicula had been taken (see above), and indicates that both of these species may have a preference for swift, open lowland streams. Material examined: PHILIPPINES, Mindanao, Zamboanga del Sur Prov.: 10 wingless males, 23 wingless females, Ayala River at San Ramon, 24 km NW of Zamboanga City, 23 July 1985, CL 1999, D. A. and J. T. Polhemus (JTPC); 1 wingless male, 1 wingless female, Zamboanga, mountain stream, 27 Feb. 1914, T. Mortensen (paratypes, JTPC).

Rhagovelia phoretica, new species Figs. 6, 7, 9

Diagnosis: Immediately recognizable among Philippine Rhagovelia by the broad and ovate form (Fig. 6), sexual dimorphism in body size, with males being much smaller than females, short pronotum with a length at the midline subequal to the length of an eye, and the small, ovate male paramere (Fig. 7).

Description:

Wingless female: Form broadly ovate (Fig. 6), length 3.48 mm; maximum width (across basal abdomen) 1.64 mm. General dorsal coloration dark blackish grey, with basal portions of legs yellowish brown.

Head dark blackish grey, eyes black, frons and vertex bearing scattered stiff black setae; antennae black, segment curved, I thicker than II, all segments covered with very short semi-recumbent dark pubescence, segment I also bearing 3-4 long stiff erect black setae; lengths of segments I-IV = 0.60, 0.36, missing, missing; proepis- ternum and jugum lacking black denticles.

Pronotum dark blackish grey, short, length/width = 1.12/0.24, broadly exposing mesonotum, posterior margin weakly bisinuate; exposed portion of tumid to either side of depressed midline, length along midline = 2.00, posterior margin biconvex posteriorly; metanotum with anterolateral portions moderately tumid, bearing scat- tered erect black setae, length along midline = 0.12; entire thoracic dorsum nearly bare, bearing only widely scattered very short recumbent pale setae, with a few longer stout erect black setae on pleural region.

Abdomen dark blackish grey, tergites I-VI subequal in length (0.16, 0.20, 0.20, 0.24, 0.20, 0.24 respectively), tergite VII longer (0.32), tergite VIII shorter than VII (0.28); tergites nearly bare, with only widely scattered very short recumbent pale setae, connexival segments more thickly covered with moderate length semi-recum- bent pale setae.

Legs black, with all coxae, fore and hind trochanters, and basal half of fore femur yellowish brown; all segments thickly covered with short recumbent pale setae; tro- chanters unarmed; fore, middle and hind femora and tibiae with scattered long erect stout black setae along anterior margins; fore femur with scattered slender erect black setae along posterior margin; fore tibia cylindrical, weakly curved and slightly more setiferous distally; middle femur unmodified, cylindrical, tapering, bearing a few long slender erect black setae along posterior margin; middle tibia slender, cylindrical; hind femur with a few scattered long fine erect dark setae along posterior margin, not incrassate, unarmed; hind tibia straight, cylindrical, unarmed.

Proportions of female legs as follows: fore femur/tibia/tarsal 1 /tarsal 2/tarsal 3 =

62

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

0.98/0.96/0.04/0.04/0.28; middle femur/tibia/tarsal 1/tarsal 2/tarsal 3/ = 1.40/1.20/ 0.04/0.52/0.84; hind femur/tibia/tarsal 1/tarsal 2/tarsal 3 = 1.00/1.40/0.04/0.05/0.40.

Venter grey; mesosternum, metasternum and abdominal ventrites lacking small black denticles; mesosternum without incipient triangular sulcus centrally; metaster- num trapezoidal, unmodified; abdominal ventrite I flat, lying in horizontal orienta- tion, abdominal ventrites II-VII lacking long setiferation or medial carinae; gono- coxae broadly exposed; proctiger lying in a semi-horizontal orientation.

Wingless male: Length 2.48 mm.; maximum width (across pronotum) 1.16 mm. Similar to wingless female in general structure and coloration with the following exceptions: body size much smaller, not as broadly ovate; dorsum of thorax and abdomen thickly covered with short, semi-recumbent dark setae; lengths of antennal segments I-IV = 0.40, 0.32, missing, missing; abdominal tergites I-III equal in length (0.12), IV and V equal in length but longer than preceding (0.14), VI and VII increasingly longer (0.16, 0.24 respectively), VIII shorter than VII (0.16); fore femur and tibia moderately bowed, modified for phoresy; proctiger with basal section bearing weakly produced lateral lobes, distal section with incipient lateral lobes but otherwise unmodified; paramere small, roughly ovate (Fig. 7).

Winged forms: Unknown.

Discussion: Rhagovelia phoretica n.sp. belongs to a distinctive intrageneric species group, referred to subsequently as the caesius group, that also contains R. caesius Lansbury from eastern New Guinea, and is defined by the following characters: strong sexual dimorphism, with males much smaller than females; a broad and ovate overall shape, with silvery grey ground coloration; a short and posteriorly sinuate pronotum in wingless forms; a tumid and broadly exposed mesonotum in wingless forms, with the posterior margin generally bilobate; a small, ovate male paramere; a simple and unmodified proctiger; and slender male forelegs, with the fore tibia curved and modified for grasping. Although the two above taxa are the only members of this assemblage presently described, a large number of additional undescribed species in this clade are present on New Guinea, and will be treated in a monograph on the Melanesian fauna that is currently in progress. The sister area relationship between the Philippines and New Guinea indicated by the distribution of species in the caesius group is consistent with the narrative zoogeographical analysis presented by Polhemus and Polhemus (1988) in their monograph of the eastern Indonesian Rhagovelia fauna.

Intermixed among the type series from Negros are three specimens (two males and one female) bearing the data “PHILIPPINES, Todaya, Is. Mindanao, July 30, 1970, M. Sato.” It seems possible that these specimens were mislabelled, and they arc thus not included as paratypes. The Mindanao record for this species, if in fact valid, needs to be reconfirmed.

Ecological notes: The type series was taken in company with Rhagovelia usingeri Hungerford and Matsuda, but no information is available regarding the specific hab- itat.

Etymology: The name “phoretica” refers to phoretic modifications of the males for riding on top of the females.

Material examined: Holotype, wingless female: PHILIPPINES, Negros, Prov. un- certain, Amulan, 16-18 July 1970, M. Sato (BPBM). Paratypes: PHILIPPINES,

1995

RH AGO V ELIA FROM THE PHILIPPINES

63

Negros: 11 wingless males, 12 wingless females, same data as holoytype (BPBM, JTPC).

Rhagovelia wenieri Hungerford and Matsuda Fig. 8

Discussion: This species was originally described from high elevation on Mt. Apo, behind Davao, and to date no additional series have been taken. R. werneri is of interest in a zoogeographical context because of the structure of the male paramere, which possesses a distinctive, distally spatulate shape typical of the novacaledonica group as defined by Polhemus and Polhemus (1988) (Fig. 8). These authors had in fact suggested that R. werneri might be a member of this species group, but did not include it as such, since their primary diagnostic character for inclusion involved wing venation, and no winged specimens of werneri were available to them. Sub- sequent analysis of paramere shapes within the Melanesian and Philippine Rhago- velia faunas has shown that the spatulate paramere is a strong symapomorphy uniting the species of the novacaledonica group, and argues for the placement of werneri within it. In addition to R. werneri, this paramere shape is seen in all species from New Caledonia, and in a diverse complex of species (including R. thysanotos Lans- bury, R. crinita Lansbury, R. aureospicata Lansbury, R. herzogensis Lansbury and many additional undescribed taxa) that has radiated in the highlands of New Guinea. Polhemus and Polhemus (1988) proposed a pattern of faunal relationship in which New Caledonia was an ancient continental source area for Rhagovelia species oc- curring on the younger islands of New Guinea and the Philippines, and the distri- bution of the novacaledonica group as now understood supports this hypothesis (Fig. 10), although in a somewhat revised geological context (see following discussion).

The suggestion by Polhemus and Polhemus (1988) that a second Philippine spe- cies, R. hoogstraali Hungerford and Matsuda, might also be assignable to the no- vacaledonica group, is not supported by the current analysis of paramere shapes. The wing venation in this species is still unknown, however, so its group affinities remain uncertain.

Material examined: PHILIPPINES, Mindanao, Davao Prov.: 6 wingless males, 2 wingless females, Sibulan River, Mt. Apo, 7,000-8,000 ft, 21 Sept., C. S. Clagg (JTPC); 1 wingless male, 1 wingless female, E. slope Mt. Apo, stream through original forest, 6,000 ft, November 1946, H. Hoogstraal and E G. Werner, CHNM- Philippine Zool. Exp. (paratypes, JTPC).

ZOOGEOGRAPHY

The Philippine islands comprise a complex amalgamation of island arcs whose geological history is still poorly understood. Although certain of these arcs, such as the Sulu, Palawan, and Sangihe, may be identified as discrete systems where they emerge from the tectonic knot in the center of the archipelago, others, such as the Negros arc and the accreted terranes of the Pacific Cordillera along the east coast of Mindanao, are much more difficult to interpret. At least six accreted arc systems are present in this region (Hamilton, 1979, 1989; see figs. 79, 99, 1 1 1 in the former work), and much additional onshore geology remains to be done before the history of the area can be properly comprehended.

Fig. 10. Distribution of the Rhagovelia novacaledonica group.

The complex geology of the Philippine archipelago is reflected in the multicentric origins of its Rhagovelia biota, which shows alliances to disjunct groups of species occuring on Borneo, Celebes, and New Guinea. The patterns involving the former two areas were discussed in detail by Polhemus and Polhemus (1988) in their mono- graph on the Rhagovelia of eastern Indonesia, while the linkages to New Guinea were alluded to but not treated in detail. The present study illustrates much more conclusively the presence of distinctively Papuan Rhagovelia species groups in the southern and central Philippines, and raises the question of how such patterns were established.

Similar sister area relationships between the Philippines and New Guinea are also seen in other families of aquatic Heteroptera. Polhemus and Polhemus (1987) noted that sagocorine Naucoridae were found in both regions, but not in the intervening northern Moluccas or Celebes, a pattern apparently identical to that displayed by the caesius and novacaledonica species groups in Rhagovelia. In both this and the pre- viously cited study the authors concluded that the Philippines and New Guinea rep- resented sister areas for certain groups of aquatic Heteroptera, and postulated that a tectonic mechanism was responsible for this faunal disjunction. No simple geological explanation was evident, however, and at the time the authors had insufficient knowl- edge of regional tectonics to suggest any mechanistic hypothesis.

The present study further clarifies the linkages between certain elements of the Papuan and Philippine Rhagovelia biotas. In addition, a more thorough knowledge of regional geology now allows a mechanistic hypothesis to be advanced that would explain this pattern. It is proposed herein that a subsequently displaced, eastward

1995

RHAGOVELIA FROM THE PHILIPPINES

65

Eig. 11. Map showing selected tectonic elements in the southern Philippine and Molucca Sea regions, including island arcs discussed in the text (after Hamilton, 1979).

migrating arc system once contributed faunal elements to both the southern Philip- pines and New Guinea. This arc was probably not the system that was accreted to the northern margin of New Guinea in the Miocene, forming the present north coast ranges, because this Miocene arc harbored certain Asian aquatic Heteroptera groups, such as the genus Ptilomera, which are absent in the Philippines outside the Palawan arc. Nor was it part of the Sangihe or Halmahera arcs that were initiated in the Eocene and are presently colliding in the Molucca Sea (Fig. 11). Instead, this arc was probably part of a long, pre-Eocene, northwest-southeast trending system that once extended from New Zealand and New Caledonia through the Solomon region, past the section of northern Australia that would later become New Guinea, and onward to what is now Mindanao.

Portions of this pre-Eocene arc may be represented along the northern flank of the New Guinea central highlands in the Jimi Terrane. This anomalous crustal block contains Mesozoic volcanic rocks that appear to be the result of arc magmatism (Hamilton, 1979), but its tectonic histories are poorly understood. This terrane could

66

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

have attained its present juxtaposition by arc collision with the stable northern Aus- tralian continental margin in the Mesozoic, or could have formed offshore and have been incorporated into an advancing arc that subsequently collided with New Guinea in the Miocene (for CD-ROM video reconstruction of tectonic events in eastern New Guinea see Yan and Kroenke, 1994). Hamilton (1979) postulated that this terrane might be correlative with the eastern Australian tectonic and magmatic belts of Cre- taceous age, which formed about the same time that New Caledonia migrated away from the eastern Australian continental margin (Kronke, 1984); Hamilton’s sugges- tion is thus computable with the zoogeographical pattern displayed by the novaca- ledonica species group of Rhagovelia (Fig. 10).

Arc-related materials of Cretaceous age are also present in southern Mindanao (Hamilton, 1989), and their existence is consistent with the above pre-Eocene arc hypothesis, but little is known regarding the history of these Mesozoic arc systems or whether the terranes derived from them are correlative to the Cretaceous emplace- ments in central New Guinea. The Pacific Cordillera of Mindanao also contains ophiolites and other remnants of more recently collided arc systems, which have been interpreted by Hamilton (1979) and subsequent workers to represent the on- shore remnants of the convergent Halmahera and Sangihe arcs, which sutured along their northern sectors in the middle Tertiary (Hawkins et al., 1985), and have been subsequently overprinted by products from the more recently initiated Mindanao Trench system (for geographical locations of these tectonic structures see fig. 10). If either of these latter two arcs had provided a means for faunal interchange between New Guinea and the Philippines, either as disperal corridors or tectonic rafts, then one would also expect to see remant taxa from some of the groups involved repre- sented in the faunas of the northern Moluccas or northern Sulawesi as well. Several aquatic Heteroptera surveys in these latter islands over the last ten years, however, including one by the author and J. T. Polhemus in 1985, have produced no collections of the groups involved in the New Guinea-Philippine disjunction. Rather than pro- viding a linkage to New Guinea, the Sangihe Arc is currently allowing faunal inter- change between Mindanao and Celebes, as demonstrated by the distribution of the orientalis group (Polhemus and Polhemus, 1988, and examples herein). This arc has moved eastward with time (Hamilton, 1989), and members of the orientalis group distributed along it have not yet reached the Moluccas or New Guinea. I hypothesize that Sangihe arc is in fact a relatively recent faunal conduit, and that the direction of Rhagovelia migration along it has been from the Philippines southward into Cel- ebes, because the former area contains a diverse assemblage of species in the orien- talis group, while the latter island has so far yielded only three (Polhemus and Polhemus, 1988; Nieser and Chen, 1993).

If a Paleocene arc did in fact formerly link the Philippines and New Guinea, then the central sector of this system must have either migrated or been displaced prior to the the development of the convergent Sangihe and Halmahera arc systems, the former of which is apparently of Eocene age (Weissel, 1980). The displacement of this older arc could have occurred along the Sorong Fault, a major left-lateral system that cuts across the extreme western tip of northern New Guinea, and has apparently transported continental fragments from the Vogelkop Peninsula eastward to near Celebes, but this is not supported by biological evidence. Conversely, fragments of this older arc could have been carried northwestward along strike-slip faults bound-

1995

RHAGOVELIA FROM THE PHILIPPINES

67

ing the western margin of the Pacific Plate, in a manner similar to that seen in the vicinity of present day New Ireland. Present plate motions and the apparent youth of the Mindanao Trench (Hamilton, 1979; Kroenke, 1984) are consistent with such a hypothesis of previous left-lateral shear in this region. Further collections of aquatic Heteroptera on the arc islands between Celebes, the Philippines and New Guinea, combined with additional geological investigations in the southern Philippines, will be necessary before these hypotheses can be more critically evaluated.

ACKNOWLEDGMENTS

I am pleased to acknowledge the invaluable assistance of Dr. Victor Gapud, University of the Philippines, Los Banos, and Burt Barrion, International Rice Research Institute, who pro- vided logistical assistance and aid with field collections in the Philippines. I also gratefully acknowledge the generosity of Dr. M. Sato, of Nagoya Women’s College, Nagoya, Japan, for his kind donation of Rhagovelia specimens collected in the Philippines. J. T. Polhemus was a tireless co-worker in the field, and read through prepublication drafts of this manuscript, offering many useful suggestions. Special thanks are also due to Herbert Zettel, Naturhistorisches Mu- seum Wien, who also reviewed the prepublication draft and graciously allowed me to describe the new species treated herein prior to the publication of his monograph on Philippine Rha- govelia, and to Dr. Warren Hamilton of the U.S. Geological Survey, Denver, Colorado and Dr. Loren W. Kroenke of the University of Hawaii, Manoa, who provided vital insights regarding southwest Pacific tectonics and reviewed the geological interpretations presented herein.

Holotypes of all new species described herein are deposited in the Bishop Museum, Honolulu (BPBM); paratypes are held in the J. T. Polhemus collection, Englewood, Colorado (JTPC), and the U.S. National Museum of Natural History, Washington, D.C. (USNM).

This research was sponsored by a series of grants from the National Geographic Society, Washington, D.C. (3053-85, 3398-86, 4092-89), and by grant BSR-9020442 from the National Science Eoundation, Washington, D.C. I thank these organizations for their continued support on research into the systematics and zoogeography of aquatic Heteroptera.

LITERATURE CITED

Andersen, N. M. 1965. A remarkable new species of Rhagovelia Mayr from the Philippines (Heteroptera, Veliidae). Ent. Med. 34:111-117.

Drake, C. J. 1948. Notes on Philippine Rhagovelia (Hemiptera, Veliidae). Proc. Ent. Soc. Washington 50:61-62.

Hawkins, J. W., G. E Moore, R. Villamor, C. Evans and E. Wright. 1985. Geology of the composite terranes of east and central Mindanao. Circum-Pacific Council for Energy and Mineral Resources, Earth Science Ser. 1:437-463.

Hamilton, W 1979. Tectonics of the Indonesian region. U.S. Geological Survey Prof. Paper 1078. Government Printing Office, Washington, D.C. 345 pp.

Hamilton, W. 1989. Convergent tectonics as viewed from the Indonesian region. Pages 655- 698 in: A. M. C. Sengor (ed.). Tectonic Evolution of the Tethyan Region Reidel, Am- sterdam.

Hungerford, H. B. and R. Matsuda. 1961. Some new species of Rhagovelia from the Philip- pines (Veliidae, Heteroptera). Univ. Kansas Sci. Bull. 42:257-279.

Kroenke, L. W 1984. Cenozoic Development of the Southwest Pacific. U. N. ESCAP, CCOP/ SOPAC Tech Bull. 6. 122 pp.

Lansbury, I. 1993. Rhagovelia of Papua New Guinea, Solomon Islands and Australia (Hemip- tera-Veliidae). Tijds. Ent. 136:23-54.

Lundblad, O. 1936. Dit altweltlichen arten der Veliidengattungen Rhagovelia und Tetraripis. Archiv Zool. 28A(21):l-63.

68

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

Lundblad, O. 1937. Einige neue oder wenig bekannte ostasiatische Rhagovelia- Arien. Ent. Tidsk. 58:1-9.

Nieser, N. and R P. Chen. 1993. The Rhagovelia (Heteroptera: Veliidae) of Sulawesi (Indo- nesia). Tijds. Ent. 136:259-281.

Polhemus, D. A. and J. T. Polhemus. 1987. A new genus of Naucoridae (Hemiptera) from the Philippines, with comments on zoogeography. Pan-Pac. Ent. 63:265-269.

Polhemus, J. T. and D. A. Polhemus. 1988. Zoogeography, ecology and systematics of the genus Rhagovelia Mayr (Heteroptera: Veliidae) in Borneo, Celebes and the Moluccas. Insecta Mundi 2:161-230.

Polhemus, J. T. and W. K. Reisen. 1976. Aquatic Hemiptera of the Philippines. Kalikasan, Philiipine J. Biol. 5:259-294.

Weissel, J. K. 1980. Evidence for Eocene oceanic crust in the Celebes Basin. Amer. Geophys. Union Geophys. Mono. 23:37^8.

Yan, C. Y. and L. W. Kroenke. 1994. A plate tectonic reconstruction of the Southwest Pacific, 0-100 Ma (CD-ROM). Proc. Ocean Drilling Prog., Sci. Res., 130, Chap. 43.

Received 17 January 1995; accepted 19 May 1995.

J. New York Entomol. Soc. 103(1):69— 72, 1995

A NEW SPECIES OF AGELAIA LEPELETIER FROM BRAZILIAN AMAZONIA (HYMENOPTERA: VESPIDAE; POLISTINAE)

Orlando Tobias Silveira and James M. Carpenter

Departamento de Zoologia, Sec. Entomologia, Museu Paraense Emilio Goeldi, Av. Perimetral S/N, Guama, C.P 399, 66040-170 Belem, Para, Brasil, and Department of Entomology, American Museum of Natural History,

Central Park West at 79th Street, New York, New York 10024, U.S.A.

Abstract. — A new species of the polistine genus Agelaia Lepeletier is described, and its relationships to other species of the genus are discussed.

The genus Agelaia Lepeletier {=Stelopolybia Ducke) is a representative of the swarming genera of the subfamily Polistinae, with essentially neotropical distribu- tion. These genera constitute the tribe Epiponini (Carpenter, 1993). Agelaia has had a tangled history of name changes, documented in Araujo (1946), Richards (1978) and Carpenter and Day (1988). In the most recent taxonomic revision by Richards (1978), the genus included 22 described species.

While studying the collection of Polistinae of the Goeldi Museum, OTS discovered an undescribed species of Agelaia from the state of Acre. This species is close to A. lobipleura (Richards), and like that species shares some features with A. cajen- nensis (E) and A. brevistigma (Richards). Together, these genera evidently constitute a monophyletic group.

Agelaia acreana, new species (Fig. 1)

Diagnosis: Almost entirely yellow species; eyes distinctly hairy; clypeus with lateral margins little sinuate, diverging slightly dorsoventrally at mid-height, and markedly separated from eyes; malar space with a shining area lacking hairs; subantennal sulci evanescent (Fig. lA); anterior margin of pronotum, at the ventral corner, with a well developed transparent lamella, but not produced very far forward (Fig. ID); forewing with prestigma about one and a half times as long as wide, tip rounded. Description:

Female: mean forewing length 1 1 mm. Structure — cuticle dull, very finely reticulate; clypeus broader than high, lateral margins little sinuate, diverging dorsoventrally at mid-height and markedly separated from eyes (Fig. lA); malar space long, distance between eye and upper margin of clypeal lateral lobe almost as high as antennal socket, malar space with a slightly concave shining area lacking hairs; interantennal prominence weakly produced, with a gently convex profile as seen from above (Fig. IB); frons with a feeble, frequently interrupted, impressed median line; subantennal sulci evanescent; posterior ocelli separated by a distance about equal to their diam- eter; POOL:OOL about Va; head narrowed proportionally in lateral view, upper part of gena slightly narrower than eye in profile (Fig. 1C), occipital carina weakly de- veloped up to middle of the swollen portion of the eye; dorsal pronotal carina very

70

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

Fig. 1. Agelaia acreana, n. sp. A — head in frontal view; B — head in dorsal view; C — head in lateral view; D — pronotum and mesepisternum in lateral view; E — first metasomal tergum in dorsal view.

weak and obtuse but recognizable laterally; frontal margin of pronotum with a mod- erately developed transparent lamella widening above; fovea well developed, cir- cular, in center of a wide concavity (Fig. ID); anterior pronotal carina obtuse in front of fovea, lower part swollen and round; dorsal mesepisternal plate unusually wide, its height less than two times its width (Fig. ID); impressed line on scutellum sometimes weak but distinct in front; propodeal furrow wide, not very shallow and moderately well defined; propodeal valves with a broad hyaline border; first meta- somal tergum with the sides diverging gradually, spiracles not projecting (Fig. IE);

1995

A NEW SPECIES OE AGELAIA

71

forewing with prestigma rather long, about one and a half times as long as broad, tip rounded.

Color — yellow, but the extensive reddish pubescence gives a darker shade to the head and mesoscutum; antennal flagellum orange-yellow; ventral margin of lower metapleural plate, posterior margin and median furrow of propodeum, mesal portion of propodeal valves around the suspensory ligament, black or dark brown; most of dorsal aspect of metasomal tergum I, broad posterior bands on terga II to V, brown or light brown; wing yellow-brown, tegula yellow.

Vestiture — abundant outstanding bristle-like hairs all over the body, less conspic- uous on meso- and metapleura; eyes with numerous strong, moderately long hairs. Male: Unknown.

Type material: holotype female, Brazil: Acre, Rio Branco (25/x-8/xi-1991) F. Ra- mos, A. Henriques, I. Gorayeb, N. Bittencourt. Paratypes: 11 females. Acre, Rio Branco (25/x-8/xi-1991) F. Ramos, A. Henriques, I. Gorayeb, N. Bittencourt.

Holotype and 9 paratypes deposited in the collection of the Museu Paraense Emilio Goeldi, Belem, Brazil. Two paratypes in the American Museum of Natural History. Distribution: Brazil, state of Acre.

Etymology: The specific name A. acreana is a reference to the Brazilian state of Acre.

REMARKS

Agelaia acreana has some uncommon features which are shared with A. lobipleu- ra, A. cajennensis and A. brevistigma. One of these features, previously noted to occur only in the latter two species (Richards, 1978), is the presence in the malar space of a slightly concave shining area lacking hairs. Richards (1978:237, 241) characterized this area as indistinct in A. cajennensis, but the condition varies, and we have seen many specimens with a distinct, shining pale area, like that stated to characterize A. brevistigma. In A. lobipleura this area may have the same color as the adjacent parts of gena (it is differently colored in one paratype from Mato Grosso we have seen), but it is nevertheless distinguishable by the shining aspect and the absence of hairs. In A. acreana this area is as distinct as in many A. cajennensis, being slightly different in color from adjacent parts.

In the key of Richards (1978), an important character separating A. cajennensis and A. brevistigma from the remaining species of Agelaia is the shape of the pres- tigma of the forewing, the prestigma being relatively longer, at least one and a half times as long as broad, with the end rounded. In most Agelaia species it is short, not or little longer than broad, with the end truncate or pointed. The length of the prestigma actually varies in both A. cajennensis and A. brevistigma, and in A. acreana the prestigma has precisely the same shape as in A. brevistigma. Although Richards (1978) stated that in A. lobipleura the prestigma is short with the end truncate, we have noted that in this species (seven specimens were examined) it indeed has the end rather truncate, but it is longer than usual and never wedge- shaped as in most Agelaia species.

Another character shared by the above mentioned four species is the tendency shown by the lateral margins of the clypeus being divergent dorsoventrally in the middle portion, and little or not sinuate (Fig. lA). This is accompanied by reduced

72 JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

contact between the clypeus and the eyes in A. lobipleura and A. cajennesis, and complete separation in A. acreana and A. brevistigma (and a few specimens of A. cajennesis). These features are approached somewhat in a few other species such as A. constructor, A. flavipennis and A. areata.

The species-level phylogeny of Agelaia has not been studied, nor is there even a formal infrageneric classification (Richards, 1978). However, from the morphological evidence discussed above, it seems clear that A. lobipleura, A. acreana, A. cajen- nensis and A. brevistigma together form a monophyletic group. The presence of a concave shining area in the malar space is unique in these species, and an elongate prestigma is very probably a derived character within Agelaia. Besides the restricted distribution within this genus, in the closely related genus Angiopolybia Araujo, the prestigma is short and truncate in A. paraensis and A. obidensis. In A. pallens, it is sometimes longer with the end rounded, but there is much intraspecific variation. Furthermore, if the very weak dorsal pronotal carina of A. lobipleura and A. acreana is considered to be a transitional state towards its complete absence in A. cajennensis and A. brevistigma, then this provides additional support for the monophyly of a group comprising these four species. In Angiopolybia, the dorsal pronotal carina is also residual, but is different from the condition observed in A. lobipleura and A. acreana. In Angiopolybia, the carina is completely reduced at sides, and weak but nevertheless acutely produced at the center.

Among these four species, as discussed above, A. acreana shares some derived features with A. brevistigma: the shape of the prestigma, and the separation between the clypeus and eyes. But as also noted above, both of these features vary somewhat in A. cajennensis. On the other hand, A. cajennensis and A. brevistigma share the derived character of loss of the dorsal pronotal carina, and A. acreana and A. lobi- pleura both have the subantennal sulci pale and very weakly impressed, and the head narrowed proportionally in lateral view, with the upper part of the gena being slightly narrower than the eye (Fig. 1C). This distribution of apomorphies supports the inference of a sister-group relationship between A. acreana and A. lobipleura on the one hand, and A. cajennensis and A. brevistigma on the other, and we consider that to be a better established hypothesis on present evidence.

ACKNOWLEDGMENTS

OTS was supported by a grant, number 300076/91-6 (RN), from the Brazilian “Conselho Nacional de Desenvolvimento Cientifico e Tecnologico,” CNPq.

LITERATURE CITED

Araujo, R. L. 1946. Angiopolybia nom. n., para o conceito revalidado de '"Stelopolybia Ducke, 1914” (Hym.-Vespidae-Polybiinae). Papeis Avuls. Zool., Sao Paulo, 7:165-170. Carpenter, J. M. 1993. Biogeographic patterns in the Vespidae (Hymenoptera): Two views of Africa and South America. In P. Goldblatt (ed.). Biological Relationships between Africa and South America, 139-155. New Haven: Yale Univ. Press.

Carpenter, J. M. and M. C. Day. 1988. Nomenclatural notes on Polistinae (Hymenoptera: Vespidae). Proc. Ent. Soc. Washington 90:323-328.

Richards, O. W. 1978. The Social Wasps of the Americas excluding the Vespidae. London: Brit. Mus. (Nat. Hist.).

Received 6 November 1994; accepted 27 September 1995.

J. New York Entomol. Soc. 103(l):73-77, 1995

A NEW SPECIES OE URANOCORIS WALKER FROM NEW GUINEA (HEMIPTERA: HETEROPTERA: COREIDAE: HOMOEOCERINI)

Harry Brailovsky

Institute de Biologia, UNAM, Departamento de Zoologia,

Apdo. Postal # 70153, Mexico 04510, D.F., Mexico

Abstract. — Uranocoris maculatus is described as a new species from New Guinea. Compar- ison is made with U. suavis Walker, which is the only previously known species of the genus. Habitus view illustrations and drawings of the male genitalia are provided to help distinguish these taxa.

This magnificent genus, of green blue metallic iridescense with shining pink and purple beams over a pale yellow surface, is known only from New Guinea. Walker (1871) described the genus Uranocoris, including the species U. suavis, which was placed in the family Anisoscelidae. For over a century this genus was kept in this classification until Osuna (1984), reviewing the generic concepts of the tribe Ani- soscelidini, concluded that this tribe is exclusively from the Western Hemisphere, and transfered Uranocoris to the tribe Homoeocerini which has an Old World dis- tribution.

Recently the author had the opportunity to check the type material deposited in BMNH, and to assemble a significant lot of specimens, allowing him to confirm the latest taxonomic treatment, describe the second species of the genus and provide the first specific locality of U. suavis in New Guinea.

Among the diagnostic characters that separate the tribes Anisoscelidini and Hom- oeocerini, the following are relevant: Anisoscelidini are distributed exclusively in the new world, presenting a well developed neck, armed anterior and middle femora, and foliate posterior tibiae on one or both faces; Homoeocerini are exclusively from the Old World and have a short or absent neck, unarmed anterior and middle femora, and cylindrical and sulcate posterior tibiae; all these characters are shared with Ur- anocoris.

The following abbreviations are used in the text: Bernice R Bishop Museum, Honolulu, Hawaii (BPBM); the Natural History Museum, England (BMNH); Insti- tute de Biologia de la Universidad Nacional Autonoma de Mexico (IBUNAM); Queensland Museum, Brisbane, Australia (QMBA).

All measurements are in millimeters.

Uranocoris suavis Walker

Uranocoris suavis Walker, 1871. Cat. Hem. Het: IV: 153 Figs. 4-6

This species is recognized by the shape of the parameres, with the posterior lobe longer and incurved (Figs. 4-5), the genital capsule decidedly larger (Figs. 3, 6),

74

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

Eigs. 1—6. 1—3. Uranocoris rnaculatus, new species. 1, 2. Parameres. 3. Caudal view of the

male genital capsule. 4—6. Uranocoris suavis Walker. 4, 5. Parameres. 6. Caudal view ot the male genital capsule.

1995

NEW URANOCORIS FROM NEW GUINEA

75

and by the lack of a yellow spot on the inner third of the apical margin of the corium.

Distribution

Originally described from New Guinea without any particular record.

Material examined. — One male: WEST NEW GUINEA: Star Mts. (Sibil Val), 18.X-8.XI.61. S. & L. Quate. Deposited in BPBM.

Uranocohs maculatus, new species Eigs. 1-3, 7

Male. Dorsal coloration. Head including antennal segments I to III bright yellow; ocelli red; external border of antenniferous tubercle brown; antennal segment IV red brown with anterior third bright orange; pronotum, scutellum, clavus and corium dark brown with following areas metallic blue green: punctures, anterior margin, anterolateral margins, humeral angles, and posterior margin of pronotum, as well as costal margin and apical angle of corial disc; following areas ochre yellow to pale yellow: callar region, apex of scutellum, and clearly discoidal spot located on the inner third of apical margin of corium; hemelytral membrane ambarine with basal angle pale hazel; connexival segments I to VI yellow, and VII bright orange, with upper margin yellow; abdominal segments I to VI with a longitudinal median stripe yellow, and laterally bright orange; segment VII with anterior half yellow, and pos- terior half yellow brown. Ventral coloration. Including rostral segments I to IV (apex of IV brown), legs, anterior and posterior lobe of metathoracic peritreme and adjacent areas yellow with three patches metallic blue green covering the pleural region of prothorax, mesothorax and metathorax; genital capsule yellow with diffusse hazel marks. Structures. Rostrum reaching posterior third of mesosterno. Genitalia. Genital capsule. Posteroventral edge entire, with small and opened “U” concavity (Fig. 3). Parameres. Figures 1-2.

Measurements. Length head: 1.12; width across eyes: 1.80; interocular space: 0.76; interocellar space: 0.18; length antennal segments: I, 4.92; II, 3.68; III, 2.72; IV, 4.96. Pronotal length: 2.44; width across frontal angles: 1.28; width across humeral angles: 3.24. Scutellar length: 1.60; width: 1.28. Total body length: 12.46.

Female. Color. Similar to male. Connexival segments VIII and IX bright orange; abdominal segments VIII and IX dark red with bright orange reflections; genital segments yellow. Measurements. Length head: 1.24; width across eyes: 2.00; inter- ocular space: 0.84; interocellar space: 0.24; length antennal segments: I, 5.28; II, 4.00; III, 2.96; IV, 4.80. Pronotal length: 2.84; width across frontal angles: 1.44; width acro^'^ '^"*^eral angles: 4.12. Scutellar length: 1.84; width: 1.72. Total body length: 1^..^.

Variation. 1 — Head dorsally bright yellow with hazel diffused marks. 2 — Antennal segments I to III bright orange. 3 — Abdominal sterna yellow with a median longi- tudinal stripe bright hazel, running from the III to VII sterna. 4 — Genital plates of female yellow with posterior margin hazel.

Holotype male. PAPUA NEW GUINEA: GOILALA (Loloipa): Owen Stanley Range, 1-15.II.58. W.W. Brandt. Deposited in BPBM.

Paratypes. PAPUA NEW GUINEA: GOILALA (Loloipa): Owen Stanley Range, 1-

76

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

Fig. 7. Uranocoris maculatus, new species.

1995

NEW URANOCORIS FROM NEW GUINEA

77

15.11.58. W.W. Brandt. One female. Deposited in BPBM. PAPUA NEW GUINEA: Wau, 16.1.?. One female. Deposited in BPBM. PAPUA NEW GUINEA: NE: Port Moresby to Brown River (30 mts.) 29.X-1.XI.65. J. Sedlacek. One female. Deposited in BPBM. PAPUA NEW GUINEA: GOILALA (Tapini): Owen Stanley Range (975 mts.), 16-25. XI. 57. W.W. Brandt. Two males, one female. Deposited in BPBM and IBUNAM. PAPUA NEW GUINEA: SE: Brown River (5 mts.), 23.X.60. J.L. Gres- sitt. Two males, one female. Deposited in BPBM and IBUNAM. PAPUA NEW GUINEA: NE: Bulolo (700 mts.), 26.XI.69. Two females. Deposited in QMBA and IBUNAM. PAPUA NEW GUINEA: NE: Bulolo (850 mts.), 29.VIII.65. One female. Deposited in QMBA.

Discussion. This is a medium-sized species, similar in color and habitus to U. suavis Walker, the only previously known species of the genus. U. maculatus is easily distinguished by the pale yellow discoidal spot located on the inner third of the apical margin of the corium (absent in U. suavis), the shape of the parameres (Figs. 1-2, 4-5), and the general view of the posteroventral edge of the male genital capsule (Figs. 3, 6).

Etymology: The specific epithet of the species refers to the pale yellow discoidal spot of the corium.

Distribution. Known only from the type locality. New Guinea.

ACKNOWLEDGMENTS

I thank the following colleagues and institutions for the loan of specimens and other assis- tance relevant to this study: Gordon M. Nishida (BPBM); Janet Margerison-Knight (BMNH); G. B. Monteith (QMBA); Cristina Urbina for the preparation of the dorsal view illustration, and to the Consejo Nacional de Ciencia y Tecnologia, Mexico (CONACyT) for financial as- sistance to visit the Bernice P. Bishop Museum in Hawaii, as well as the Queensland Museum in Brisbane, Australia.

LITERATURE CITED

Osuna, E. 1984. Monografia de la Tribu Anisoscelidini (Hemiptera, Heteroptera, Coreidae) I.

Revision Generica. Bol. Ent. Venez. N. S. 3(5-8):77-148.

Walker, F. 1871. Catalogue of the specimens of Hemiptera Heteroptera in the collection of the British Museum Part IV. London : 1-211.

Received 23 March 1995; accepted 19 June 1995.

J. New York Entomol. Soc. 103( 1 ):78-82, 1995

AN UNUSUAL NEW SPECIES OF BARYSCAPUS FORSTER (HYMENOPTERA: EULOPHIDAE: TETRASTICHINAE) FROM NORTH AMERICA

K. SUREKHA AND JOHN LaSaLLE

Indian Agricultural Research Institute, New Delhi, 110012, India; and International Institute of Entomology,

(An Institute of CAB INTERNATIONAL),

56 Queen’s Gate, London, SW7 5JR, UK

Abstract. — Baryscapus megos, a new species of tetrastichine Eulophidae, is described and illustrated. This species is unusual in having a very large mouth opening. A key is presented to separate it from other species of North American Tetrastichinae with such a large mouth.

LaSalle (1994) provided a key to North American genera of Tetrastichinae which included a single genus (and species) with an enlarged mouth opening, Exastichus odontos LaSalle. An additional species of Tetrastichinae with a large mouth opening, Kocaagizus pirireisi Doganlar was described from North America by Doganlar (1993). This genus was not included in LaSalle’s key, however it was briefly dis- cussed in a note added while in press (LaSalle, 1994).

Baryscapus megos n. sp. is the third species of North American Tetrastichinae known to have an enlarged mouth opening, and the first species of the genus Bar- yscapus. This description is necessary because current generic keys for North Amer- ican Tetrastichinae will have to be slightly modified to accommodate this new spe- cies.

Morphological terminology follows Graham (1987) and LaSalle (1994). Abbre- viations for Museums are as follows: BMNH — The Natural History Museum, Lon- don, UK; CNC — Canadian National Collection of Insects and Arachnids, Ottawa, Ontario, Canada; lARI — National Pusa Collection of Insects, Indian Agricultural Research Institute, New Delhi, 110012, INDIA; UCR — University of California, Riv- erside, California, USA; USNM — The U.S. National Museum of Natural History, Smithsonian Institution, Washington, D.C., USA.

Baryscapus megos, new species (Figs. 2-7)

Diagnosis. Mouth opening broad, 3. 0-3. 5 times wider than malar space (Fig. 2); clypeus produced into two large lobes; mandible large. Malar groove curved. Me- soscutum with 6-12 adnotaular setae, in more than a single row. Median groove on mesoscutum distinct. Callus with 4 setae. Submarginal vein with 2 setae.

Female. Length to apex of last tergite 1.4-1.75 mm. Head, mesosoma and gaster ranging from dull metallic blue to brown with metallic blue shine, the mesosoma dorsal ly more distinctly metallic than other parts. Lower face with yellow streak below torulus; ventral margin of face and clypeus yellow to brown; mandible brown- ish yellow; antenna yellowish brown. Coxae and femora brown; apex of femora.

1995

NEW BARYSCAPUS FROM NORTH AMERICA

79

Figs. 1-4. 1. Exastichus odontos LaSalle $, face. 2-4. Baryscapus megos n. sp. $: 2, head,

frontal view; 3, head, side view; 4, mesosoma, dorsal view.

tibiae and tarsi pale yellow; tibiae may be slightly darkened. Wing veins pale yellow, setae brown. Setae on face, antenna and body yellow.

Head (Figs. 2-3) 1.05-1.15 times broader than high in facial view. Malar groove curved, malar space about 0.38-0.45 eye height. Distance between posterior ocelli about twice as long as distance from posterior ocellus to eye margin. Clypeus dis- tinctly produced, incised medially, with two large lobes. Mouth opening 3. 0-3. 5 times as wide as malar space. Antennal toruli at lower level of eye margin. Side of face with many minute punctures. Mandible large, bidentate, the ventral tooth dis- tinctly longer than dorsal tooth.

80

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

Antenna (Fig. 6) with scape 4. 5-5. 5 times as long as wide. Antenna with one anellus. Funicular segments subequal in width; FI 1.3-1. 4 times, F2 1.1-1.25 times, F3 1.1-1. 2 times as long as wide. Club 2. 0-2. 2 times as long as than wide

Mesosoma (Fig. 4) finely sculptured. Midlobe of mesoscutum with 6-12 adnotau- lar setae, these somewhat scattered and forming more than a single, straight row. Median groove on mesoscutum distinct. Scutellum 1.1 -1.2 times as wide as long. Dorsellum slightly convex. Propodeum with median carina mostly eliminated; with faint paraspiracular carina. Callus with four setae.

Forewing (Fig. 5) with 2 setae on dorsal surface of submarginal vein. Marginal vein 0.70-0.80 times as long as costal cell, 1.75-1.85 times as long as stigmal vein.

Caster 1.25-1.35 times as long as mesosoma, and 1.4- 1.6 times as long as wide. Lateral sides of all tergites pubescent. Ovipositor sheaths short.

Male. Length 1.15-1.6 mm. Similar to female in coloration and morphological fea-

1995

NEW BARYSCAPUS FROM NORTH AMERICA

81

tures except in sexual characters. Scape (Fig. 7) with ventral sensory plaque 0.40-

0. 45. total length of scape, situated near center of scape. FI and F2 1. 1-1.2 times; F3 and F4 1.1-1.25 times as long as wide. Club 3.0-3. 1 times as long as wide. Funicular segments with short, sparse setae; without a distinct basal whorl of long dark setae.

Material examined. Holotype 9, USA, Arizona, Pima, Co., 4 mi. S. Robles Junction, 26.viii.1979, C. W. Melton (USNM).

59 9, 6S paratypes. Same data of holotype (or collected by J. LaSalle) (209, 26 USNM; 1091(3; UCR, BMNH, CNC; 59 M LaSalle; 49 lARI).

Discussion

Baryscapus is the second largest genus of Tetrastichinae in the Holarctic region, with 43 North America species (LaSalle, 1994) and 57 European species (Graham, 1991). It can generally be distinguished from other tetrastichine genera by the fol- lowing characters: submarginal vein with 2 to many setae on dorsal surface; at least the longest two of the cereal setae subequal in length, relatively short and often not conspicuous in their difference from the setae on the surrounding tergites of the metasoma; propodeal spiracle with entire rim exposed; body dark or metallic, without pale or yellow markings; mesosternum generally convex in front of trochantinal lobe, without a distinct precoxal suture; malar sulcus often strongly curved. Baryscapus is included in keys to Tetrastichinae genera for Europe (Graham, 1987, as Eutetras- tichus\ 1991) and North America (LaSalle, 1994).

Within Baryscapus, B. megos is easily distinguished by the presence of a mouth opening that is three or more times as wide as the malar space. All other Holarctic species have the mouth opening less (usually distinctly less) than twice as wide as the malar space.

There are now three species of North American Tetrastichinae which have en- larged mouth openings (mouth opening three or more times as wide as malar space). Only one of these genera, Exastichus, was included by LaSalle (1994) in his key to North American genera. These species may be separated using the following key couplets. These would best be included in LaSalle’s (1994) key at couplet 44, where Exastichus is distinguished.

Key to North American Tetrastichinae with enlarged mouth openings

1. Mandibles exodont (curving outwards), and not meeting medially (Fig. 1)

Exastichus odontos LaSalle

Mandibles normal, curving inwards and when closed meeting or overlapping medially (Fig. 2) 2

2. Propodeum with a raised lobe of the callus which partially overhangs spiracle. Malar sulcus straight or only slightly curved. Female antenna with 4 funicular segments (the

first smaller than the second) Kocaagizus pirireisi Doganlar

Propodeum with entire spiracular rim visible (Fig. 4). Malar sulcus distinctly curved (Fig. 3). Female antenna with 3 funicular segments (Fig. 6) ... Baryscapus megos n. sp.

ACKNOWLEDGMENTS

KS’s visit to the UK as a Darwin Fellow was funded by the Department of the Environment, Darwin Initiative for the Survival of Species, grant no. 162/3/53. Space and facilities during

82

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

this study were kindly provided by the Department of Entomology, The Natural History Mu- seum, London; teehnieal assistance from the SEM and photography units of the BMNH is also gratefully acknowledged. This publication was partially funded through NSF grant BSR- 9020206 (JL).

LITERATURE CITED

Doganlar, M. 1993. A new genus of Tetrastichinae from North America. Entomofauna 14(9): 187-191.

Graham, M. W. R. de V. 1987. A reclassification of the European Tetrastichinae (Hymenoptera; Eulophidae), with a revision of certain genera. Bull. Brit. Mus. (Nat. Hist.), Ent. Sen 55(1): 1-392.

Graham, M. W. R. de V. 1991. A reclassification of the European Tetrastichinae (Hymenoptera: Eulophidae): revision of the remaining genera. Mem. Am. Ent. Ins. 49:1-322.

LaSalle, J. 1994. North American genera of Tetrastichinae (Hymenoptera: Eulophidae). J. Nat. Hist. 28:109-236.

Received 30 October 1994; accepted 14 July 1995.

J. New York Entomol. Soc. 103( 1 ):83-90, 1995

NEW NAMES AND OTHER TAXONOMIC CHANGES IN THE ORIENTAL MEMBRACIDAE (HOMOPTERA: MEMBRACOIDEA)

Ai-Ping Liang and Stuart H. McKamey

Roosevelt Research Fellow, Department of Entomology,

American Museum of Natural History,

Central Park West at 79th Street, New York, New York 10024, USA and Zoologisk Museum, University of Copenhagen, Universitetsparken 15, DK-2100 Copenhagen, Denmark

Abstract. — Nine new replacement names are proposed for species in the genera Tricentrus, Oxyrhachis and Gargara. T. elongatus Kato, T. minuUus Jacobi, and Centrotypus perakensis Distant are reinstated as valid names. T. mckameyei Ahmad is emended as T. mckameyi, T. striimpeli Ahmad and Yasmeen as T. struempeli, and Gargara donitzae Matsumura as G. doen- itzi. Kotogargara alini yunnanensis Yuan and Chou is revised as Gargara {Kotogargara) par- vula yunnanensis (Yuan and Chou), new combination. Fifteen new or reinstated species com- binations in Tricentrus and Gargara are established to follow previous generic changes.

During the preparation of an updated taxonomic catalogue of world Membracoidea (by SHM) and a checklist of the AuchenoiThyncha found in China and adjacent regions (by APL), it became evident that a number of nomenclatorial changes in the family Membracidae were necessary. In this paper, we offer new replacement names for 9 homonyms, reinstate 3 valid specific names, emend 3 incorrect original specific spellings, revise the status of 1 subspecies, and propose 15 new or reinstated species combinations.

REPLACEMENT NAMES

Tricentrus dubitatus Liang and McKamey, New Name

Tricentrus dubius Ananthasubramanian, 1980a:25, pi. 24, figs. 1-6 (preoccupied by Tricentrus dubius Kato, 1960:348, fig. la-i).

Ananthasubramanian (1980a) described Tricentrus dubius from two males and three females collected in West Bengal, India. Because the name T. dubius is pre- occupied, the replacement name dubitatus is here proposed.

Tricentrus taiwanensis Liang and McKamey, New Name

Arisangargara nitida Kato, 1928:48, pi. 1, fig. 8 (preoccupied by Tricentrus nitidus Funkhouser, 1927a:2, figs. 1-3).

Kato (1928) described Arisangargara nitida from an unspecified number of female specimens from Taiwan. Funkhouser (1943) transferred nitida from Arisangargara to Centrotoscelus, which he treated as synonyms. When Ahmad and Yasmeen (1974) synonymized both Arisangargara and Centrotoscelus with Tricentrus, Arisangar- gara nitida Kato 1928 implicitly became a junior secondary homonym of Tricentrus

84

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

nitidus Funkhouser 1927. We accept the placement of nitidus Kato in Tricentrus and propose the replacement name T. taiwanensis for it.

Tricentrus kodaikanalensis Liang and McKamey, New Name

Tricentrus purpureus Ananthasubramanian and Ananthakrishnan, 1975:230, fig. 26- 1-9 (preoccupied by Tricentrus purpureus Funkhouser, 1942:62, pi. 7, fig. 2).

Ananthasubramanian and Ananthakrishnan (1975) described Tricentrus purpureus from nine males, sixteen females, and eight nymphs collected in Kodaikanal, Madras, India. Because the name T. purpureus is preoccupied, the replacement name kodai- kanalensis, derived from the type locality of Ananthasubramanian and Ananthak- rishnan’s species, is here proposed.

Tricentrus nigritus Liang and McKamey, New Name

Tricentrus nigra [sic] Ahmad and Yasmeen, 1979b:265, figs. 19-21 (preoccupied by Arisangargara nigra Kato, 1928:48).

Ahmad and Yasmeen (1979b) described Tricentrus niger (as T. nigra) from a single female collected at Rangamati of Bengal. Upon synonymizing both Arisan- gargara and Centrotoscelus with Tricentrus (Ahmad and Yasmeen, 1974:183), Tri- centrus niger Ahmad and Yasmeen implicitly became a junior secondary homonym of Arisangargara nigra Kato 1928. We accept the placement of T. niger (Kato), new combination, and propose the replacement name nigritus for T. niger Ahmad and Yasmeen.

Tricentrus yasnieeni Liang and McKamey, New Name

Tricentrus matsumurai Yasmeen and Ahmad, 1976:108, figs. 24-26 (preoccupied by Arisangargara matsumurai Kato, 1928:50. Replacement name for Centrotus (Gar- gara) variegatus Matsumura, 1912:21 nec Centrotus variegatus Signoret in Fair- maire and Signoret, 1858:336).

Yasmeen and Ahmad (1976) described Tricentrus matsumurai from two females collected at Radjendra Gang of East Bengal. When Ahmad and Yasmeen (1974) synonymized Arisangargara and Centrotoscelus with Tricentrus, they implicitly transferred A. matsumurai Kato 1928 into the latter genus. We accept the placement of T. matsumurai (Kato), new combination, and propose the replacement name T. yasnieeni for T. matsumurai Yasmeen and Ahmad 1976.

Oxyrhachis ampliata Liang and McKamey, New Name

Oxyrhacliis grandis Ananthasubramanian, 1980a:7, pi. 7 (preoccupied by Oxyrhachis grandis Capener, 1962:97, pi. 22, fig. 50).

The new replacement name is Latin for enlarged or widened, to resemble the meaning connoted by the former name.

Oxyrhachis bulla Liang and McKamey, New Name

Oxyrhachis tuberculata Ananthasubramanian, 1980b: 113, figs. 1-3 (preoccupied by Oxyrhachis tuberculata Walker, 1858:109).

1995

ORIENTAL MEMBRACIDAE

85

Ananthasubramanian’s name is a primary homonym of Walker’s, which was re- turned to the genus Oxyrhachis, from Xiphistes, by Capener (1962:33). The new name, similar in meaning to the former, is Latin for a knob, to describe the distinct tubercle on the pronotum.

Oxyrhachis monochroma Liang and McKamey, New Name

Oxyrhachis unicolor Ananthasubramanian, 1980a: 8, pi. 8 (preoccupied by Oxyrha- chis wn/co/or Walker, 1851:509).

Ananthasubramanian’s name is a primary homonym of Walker’s, which was, like the previous species, formerly in the genus Xiphistes, but returned to Oxyrhachis by Capener (1962:161) when he synonymized these genera. The new name for the junior homonym is Greek but was chosen to preserve the meaning of the Latin name it replaces.

Gargara aliquantula Liang and McKamey, New Name

Gargara minuscula Ananthasubramanian, 1978:291 [preoccupied by G. minuscula (Walker, 1870:191)].

Ananthasubramanian’s name is a primary homonym of Walker’s, which was for- merly placed in the genus Centrotus, but transferred to Gargara by Distant (1915: 492). The new name for the junior homonym is Latin for little, to resemble the meaning connoted by the former name.

OTHER TAXONOMIC CHANGES Tricentrus elongatus Kato, Reinstated Name Tricentrus elongatus Kato, 1929:540, pi. 16, fig. 3a, b.

Tricentrus gracilis Kato, 1930:287; unjustified emendation of T. elongatus Kato, 1929.

Although Kato’s (1930) correction of his original spelling for the species was accepted by Metcalf and Wade (1965), the correction contravenes ICZN Art. 32b. Because the change was demonstrably intentional, the subsequent spelling is an emendation with its own availability (Art. 33b, International Commission on Zoo- logical Nomenclature 1985 [ICZN]). Nevertheless, the original spelling cannot be taken as incorrect (Art. 32c) and the emendation is therefore unjustified. The emen- dation implicitly became a secondary junior homonym of Arisangargara gracilis Kato, 1928, when Ahmad and Yasmeen (1974) synonymized Arisangargara with Tricentrus, but the homonym is not replaced because it is rejected as a junior syn- onym (Art. 60a).

Tricentrus minullus Jacobi, Reinstated Name

Tricentrus minullus Jacobi, 1944:33 [new name for Tricentrus minor Lindberg in Lindberg and Zachwatkin, 1936:4, figs. Ic, 2c (preoccupied by Otaris minor Schmidt, 1911:243)].

Jacobi (1944) proposed Tricentrus minullus to replace T. minor Lindberg 1936,

86

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

which was a secondary junior homonym of Otaris minor Schmidt 1911. However, Metcalf and Wade (1965:401) reinstated T. minor Lindberg as a valid name and listed T. minuUus Jacobi as an unnecessary replacement name for (and objective synonym of) T. minor Lindberg. Because the secondary junior homonym T. minor Lindberg is permanently invalid (ICZN, Art. 59b), T. minullus Jacobi is here rein- stated as a valid name. T. minor Lindberg is now again a junior homonym, but is not replaced because it is rejected as a junior synonym.

Centrotypus perakensis Distant, Reinstated Name

Centrotypus perakensis Distant, 1916:318 [new name for C. alatus Buckton, 1903: 237, pi. 54, figs. 2, 2a (preoccupied by Hemiptycha alata Fairmaire, 1846:317, pi. 6, hg. 24)].

Distant (1916) proposed Centrotypus perakensis to replace C. alatus Buckton 1903, which was a secondary junior homonym of Hemiptycha alata Fairmaire 1846. This was accepted by Funkhouser (1927b:377, 1951:255). However, Coding (1930, 1950) treated C. alatus as a valid species and listed C. perakensis Distant first (1930: 40) as a synonym and later (1950:127) as a nomen nudum under alatus Buckton. Metcalf and Wade (1965:195) followed Coding (1950), reinstated C. alatus Buckton as a valid name and listed C. perakensis Distant as an unnecessary replacement name for (and objective synonym oO C. alatus Buckton. Because the secondary junior homonym C. alatus Buckton is permanently invalid (ICZN, Art. 59b), C. perakensis Distant is here reinstated as the valid name for the species.

Tricentrus mckameyi Ahmad, Emendation

Tricentrus mckameyei Ahmad, 1992:209 [new name for Tricentrus planicornis Yas- meen and Ahmad, 1976:112, figs. 37-39 (preoccupied by Tricentrus planicornis Jacobi, 1944:33)].

Yasmeen and Ahmad (1976) described Tricentrus planicornis from a single female collected at Rangamati of East Bengal. Because the name T. planicornis is preoc- cupied, Ahmad ( 1 992) proposed the replacement name mckameyei, said to be derived from the name of the junior author of this paper. Because Ahmad’s (1992) original spelling mckameyei must therefore be taken as inconect [ICZN, Art. 32c(ii)], it is here emended as mckameyi.

Tricentrus struempeli Ahmad and Yasmeen, Emendation

Tricentrus striimpeli Ahmad and Yasmeen, 1979a:547, figs. 66-72, 550, 555 [also spelled strumpeli],

Ahmad and Yasmeen (1979a) described T. strumpeli from five males taken at Punjab, Pakistan. We apply the principle of hrst revisor to fix the original spelling as strumpeli, which we emend to struempeli in accordance with IZCN Arts. 32c(iv, vi) and 32d(i) (2).

1995

ORIENTAL MEMBRACIDAE

87

Gargara doenitzi Matsumura, Emendation Centrotus (Gargara) donitzae Matsumura, 1912:23.

Matsumura (1912) described Centrotus (Gargara) donitzae from one male and one female from Honshu, Japan. The species was named after Herrn Donitz. Since Matsumura’s (1912) donitzae can not be taken as a correct original spelling [ICZN, Art. 32c(ii, vi)], it is here emended as doenitzi [Art. 32d(i) (2)].

Gargara (Kotogargara) parvula yunnanensis (Yuan and Chou),

New Placement, New Combination

Kotogargara alini yunnanensis Yuan and Chou in Yuan et ah, 1992:210 (Chinese),

219 (English), fig. 19.

Yuan and Chou (see Yuan et al., 1992) described yunnanensis as a new subspecies of Kotogargara alini (Funkhouser) from a single female collected at Lushui of Yun- nan, southwestern China. Kotogargara Matsumura was reduced to a subgenus of Gargara Amyot and Serville and K. alini (Funkhouser) as a junior synonym of Gargara (Kotogargara) parvula Jacobi by Anufriev (1981:167). Here the subspecies yunnanensis Yuan and Chou is revised as a subspecies of Gargara (Kotogargara) parvula.

OTHER NEW OR REINSTATED COMBINATIONS

Ahmad and Yasmeen (1974) synonymized both Arisangargara and Centrotoscelus with Tricentrus and reinstated the synonymy of Otaris with Tricentrus. Anufriev (1981) reduced Kotogargara Matsumura to a subgenus of Gargara Amyot and Ser- ville. However, the implied resultant combinations for many species in those junior synonymic genera have never been published. We accept the generic changes pro- posed by Ahmad and Yasmeen (1974) and Anufriev (1981). To improve nomencla- tural and taxonomic stability, we here propose the following new or reinstated com- binations inferred from those generic changes.

Tricentrus jiavus (Kato), New Combination Arisangargara jiava Kato, 1928:50.

Tricentrus gracilis (Kato), New Combination Arisangargara gracilis Kato, 1928:47, pi. 1, fig. 7.

Tricentrus handschini (Funkhouser), New Combination Centrotoscelus handschini Funkhouser, 1936:196, fig. 6.

Tricentrus intermedius (Schmidt), Reinstated Combination Otaris intermedius Schmidt, 1911:242.

Tricentrus luteus (Funkhouser), New Combination Centrotoscelus luteus Funkhouser, 19 18a: 30.

Tricentrus maculipennis (Funkhouser), New Combination Centrotoscelus maculipennis Funkhouser, 1933:579, fig. 1.

88

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

Tricentrus marginatus (Kato), New Combination Arisangargara marginata Kato, 1928:49.

Tricentrus matsumurai (Kato), New Combination Arisangargara matsumurai Kato, 1928:50.

Tricentrus minor (Schmidt), Reinstated Combination Otaris minor Schmidt, 1911:243.

Tricentrus niger (Kato), New Combination Arisangargara nigra Kato, 1928:48.

Tricentrus nigrifrons (Kato), New Combination Arisangargara nigrifrons Kato, 1928:49.

Tricentrus porrectus Funkhouser, Reinstated Combination Tricentrus porrectus Funkhouser, 1929:118, pi. 1, fig. 10.

Otaris porrecta (Funkhouser); Metcalf and Wade, 1965:423.

Tricentrus shinchikunus (Kato), New Combination Gargara shinchikuna Kato, 1928:44.

Tricentrus truncaticornis Funkhouser, Reinstated Combination Tricentrus truncaticornis Funkhouser, 1918b:8.

Otaris truncaticornis (Funkhouser); Metcalf and Wade, 1965:423.

Gargara (Kotogargara) botelensis (Matsumura), New Combination Kotogargara botelensis Matsumura, 1938:153.

ACKNOWLEDGMENTS

We thank Lewis L. Deitz and Robert L. Blinn of the Department of Entomology, North Carolina State University, Raleigh, North Carolina, for all their help and generosity and Randall T Schuh and Lee Herman of the Department of Entomology, American Museum of Natural History, New York for several discussions. We would also like to thank Lewis L. Deitz and Randall T Schuh for their critical reviews of the manuscript.

LITERATURE CITED

Ahmad, I. 1992. Cladistic analysis of Fairmairei [sic] group of centrotine treehopper’s genus Tricentrus Stal (Homoptera: Auchenrrhyncha [sic]: Membracidae) with new name for T. plcmicornis Yasmeen and Ahmad. Proc. Pakistan Congr. Zool. 12:207-214.

Ahmad, I. and N. Yasmeen. 1974. A new tribe of the subfamily Centrotinae Amyot et Serville (Homoptera: Membracidae) with comments on its phylogeny. Mitt. Hamburg. Zool. Mus. Inst. 71:175-191.

Ahmad, I. and N. Yasmeen. 1979a [dated 1978]. An account of Gibbosulus [sic] group of Tricentrus Stal (Homoptera: Membracidae: Tricentrini). Oriental Ins. 12:531-556.

Ahmad, I. and N. Yasmeen. 1979b. New species and records of the Tricentrus projectus group (Homoptera: Membracidae: Tricentrini) from Pakistan, Azad Kashmir and Bangladesh, with phylogenetic considerations. Pacific Ins. 20:257-278.

Ananthasubramanian, K. S. 1978. Taxonomic notes on a new species of Gargara Amyot & Serville (Membracidae: Homoptera) and its immature stages. Entomon 3(2):29 1-294.

Ananthasubramanian, K. S. 1980a. Descriptions of a new genus and some new species of

1995

ORIENTAL MEMBRACIDAE

89

Membracidae (Homoptera) in the collections of the Zoological Survey of India. Rec. Zool. Surv. India Misc. Publ., Occas. 16:1-36, pis. 1-33.

Ananthasubramanian, K. S. 1980b. Taxonomic studies on Indian Membracidae (Insecta: Ho- moptera). Entomon 5(2): 1 13-128.

Ananthasubramanian, K. S. and T. N. Ananthakrishnan. 1975. Taxonomic, biological and eco- logical studies of some Indian membracids (Insecta: Homoptera) Part I. Rec. Zool. Surv. India 68:161-272.

Anufriev, G. A. 1981. Homopterological reports 1-3 (Homoptera, Auchenorrhyncha). Rei- chenbachia 19(28): 159-173.

Buckton, G. B. 1903. A monograph of the Membracidae. 296 pp., 60 pis.

Capener, A. L. 1962. The taxonomy of the African Membracidae. Part 1. The Oxyrhachinae.

Entomol. Mem. Dep. Agric. S. Afr., Pretoria 6:1-164, 42 pis.

Distant, W. L. 1915. Rhynchotal notes. — LVII. Ann. Mag. Nat. Hist. (8)16:489-496.

Distant, W. L. 1916. Rhynchotal notes.— LIX. Ann. Mag. Nat. Hist. (8)17:313-330. Fairmaire, L. M. H. 1846. Revue de la tribu des Membracides. Ann. Soc. Entomol. Fr. (2)4: 235-320.

Fairmaire, L. M. H. and V. Signoret. 1858. Order Hemipteres. Deuxieme partie. Hemipteres Homopteres Latrielle. In Voyage au Gabon. Histoire naturelle des insectes et des ar- achnides recueillis pendant un voyage fait au Gabon en 1856 et en 1857 par M. Henry C. Deyrolle sous les auspices de MM. le comte Mniszech et James Thomson precedee de I’histoire du voyage par M. James Thomson. Archives entomologiques ou recueil contenant des illustrations d’ insectes nouveaux ou rares par M. James Thomson. 2:330- 343, pi. 40.

Funkhouser, W. D. 1918a. Notes on the Philippine Membracidae. Philippine J. Sci. 13(1 ):21- 39, 1 pi.

Funkhouser, W. D. 1918b. Malayan Membracidae. J. Straits Br. R. Asiatic Soc. 79:1-14. Funkhouser, W. D. 1927a. Fauna sumatrensis. (Beitrag Nr. 30). Membracidae (Homoptera). Suppl. Entomol. 15:1-22.

Funkhouser, W. D. 1927b. General Catalogue of the Hemiptera. Fasc. 1. Membracidae. Smith College, Northampton, Mass., USA.

Funkhouser, W. D. 1929. New Archipelagic Membracidae. Philippine J. Sci. 40:111-131, pis. 1-2.

Funkhouser, W. D. 1933. A new Malayan membracida. J. Fed. Malay States Mus. 17:579- 580.

Funkhouser, W. D. 1936. New Membracidae in the Handschin Collection. Rev. Suisse Zool. 43:189-198.

Funkhouser, W. D. 1943. Synonymy of the Membracidae of Formosa. J. New York Ent. Soc. 51:265-275.

Funkhouser, W. D. 1951. Homoptera family Membracidae. Gen. Ins. 208:1-383.

Goding, F. W. 1930. Synonymical notes on Membracidae. J. New York Ent. Soc. 38:39-42. Goding, F. W. 1950. The Old World Membracidae. J. New York Ent. Soc. 58:117-129. International Commission on Zoological Nomenclature. 1985. International Code of Zoological Nomenclature. 3rd ed. International Trust for Zoological Nomenclature, London, Uni- versity of California Press, Berkeley, xx + 338 pp.

Jacobi, A. 1944. Die Zikadenfauna der Provinz Fukien in Siidchina und ihre tiergeographischen Beziehungen. Mitt. Miinchen. Ent. Gesell. 34:5-66.

Kato, M. 1928. Notes on Formosan Membracidae, with descriptions of one new genus and some new species. Insect World 32(1, 2):2-15, 37-50. [In Japanese.]

Kato, M. 1929. Descriptions of some new Formosan Homoptera. Trans. Formosa Nat. Hist. Soc. 19:540-551, 1 pi. [In Japanese.]

90

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

Kato, M. 1930. [The Japanese Membracidae.] Dobutsugaku Zasshi, Zool. Soc. Japan, Tokyo 42:281-306, pi. 3. [In Japanese with English summary.]

Kato, M. 1960. Homoptera: Membracidae. Ins. Micronesia 6(5):345-35 1 .

Lindberg, H. and A. Zachwatkin. 1936. Schwedisch-chinesische wissenschaftliche Expedition nach den nordwestlichen Provinzen Chinas, unter Leitung von Dr. Sven Hedin und Prof. Sii Ping-chang Insekten gesammelt vom schwedischen Arzt der Expedition Dr. David Hummel 1927-1930. 59. Hemiptera. 3. Homoptera Cicadina. Arkiv Zool. 29A(4):1-18.

Matsumura, S. 1912. Die Cicadinen Japans II. [Tokyo] Annot. Zool. Jap. 8:15-51.

Matsumura, S. 1938. Homopterous insects collected by Mr. Tadao Kano at Kotosho, Formosa. Ins. Matsum. 12:147-153.

Metcalf, Z. P. and V. Wade. 1965. General Catalogue of the Homoptera. A Supplement to Fascicle I — Membracidae of the General Catalogue of Hemiptera. Membracoidea. In two sections. North Carolina State University, Raleigh, vi + 1552 pp.

Schmidt, E. 1911. Beitrag zur Kenntnis der Membraciden. Zool. Anz. 38:233-243.

Walker, F. 1851. List of the specimens of homopterous insects in the collection of the British Museum 2:261-636, pis. 3-A.

Walker, F. 1858. Homoptera. Insecta Saundersiana: or characters of undescribed insects in the collection of William Wilson Saunders, Esq. 1858:1-117.

Walker, E 1870. Catalogue of the homopterous insects collected in the Indian Archipelago by Mr. A. R. Wallace, with descriptions of new species. Linn. Soc. London, Jour. Proc. 10: 82-193, pi. 3.

Yasmeen, N. and I. Ahmad. 1976. New species of Tricentrus Stal from Pakistan, Azad Kashmir and East Bengal with phylogenetic considerations (Membracidae, Centrotinae, Tricentri- ni). Mushi 49(10):95-125.

Yuan, E, I. Chou and Z. X. Cui. 1992. Homoptera: Membracoidea. In S. X. Chen (ed.): Insects of the Hengduan Mountains Region 1:192-219. [In Chinese, English abstract p. 212.]

Received 2 May 1995; accepted 19 May 1995.

J. New York Entomol. Soc. 103( 1 ):91— 99, 1995

EFFECTS OF DIET QUALITY AND QUEEN NUMBER ON GROWTH IN LEPTOTHORACINE ANT COLONIES (HYMENOPTERA: FORMICIDAE)

Jay D. Evans' and Naomi E. Pierce^

Department of Biology, Princeton University, Princeton, New Jersey 08544-1003

Abstract. — Laboratory experiments manipulating the diet of colonies of the facultatively po- lygynous ant, Leptothorax curvispinosus (Mayr), demonstrated that carbohydrates and protein have synergistic effects on egg numbers and brood production in colonies of this ant. Colonies receiving insect prey and sucrose grew significantly faster than colonies reared on unlimited supplies of either of these food types alone. This study also measured the effect of queen number on colony growth rates. Because the occurrence of multiple queens might affect colony growth only under certain nutritional conditions, polygynous colonies were reared in each of the three diet treatments. Queen number did not affect colony worker production in any of the three diet treatments; thus, individual queens in polygynous colonies produced far fewer workers per queen than did queens in monogynous colonies. There were no interaction effects between queen number and diet on colony growth. Several colonies which lacked morphologically dis- tinct queens produced workers over the course of the experiment. Using artificially established colonies of unmated workers, we found no evidence for parthenogenetic (thelytokous) repro- duction in these colonies.

Key Words. — Formicidae, Leptothorax, diet, polygyny, brood development.

Production in social insect colonies generally consists of both a vegetative stage, during which new workers are added to colonies, and a sexual stage involving the release of males and females with the potential to mate. The rate of vegetative growth is determined mainly by workers that have foregone their own reproduction to devote energy to the growth of their siblings (Oster and Wilson, 1978). To date, only a few studies have examined the effect of diet on colony growth during this vegetative phase (Brian, 1973; Buschinger and Pfeifer, 1988; Porter, 1989). Buschinger and Pfeifer (1988) found increased larval and pupal production in colonies of Lepto- thorax acervorum fed insect prey and honey, when compared to those fed an artificial diet. They also found behavioral changes in these ants and in the slave-making species Harpagoxenus sublaevis. Porter (1989) demonstrated that a diet of insect prey and sucrose was optimal for colony growth in the imported fire ant, Solenopsis invicta. Colonies fed insect prey alone fared significantly better than those reared on sucrose alone. Colonies of Myrmica rubra reared on insect prey and a sucrose so- lution also were more productive than those not given insects (Brian, 1973).

The facultatively polygynous ant L. curvispinosus is an excellent subject for mea- suring colony growth and development under controlled circumstances, due to its

' Current address: Department of Biology, University of Utah, Salt Lake City, Utah 84112.

^ Current address: Museum of Comparative Zoology, Harvard University, Cambridge, MA 02138.

92

JOURNAL OF THE NEW YORK ENTOMOEOGICAL SOCIETY Vol. 103(1)

small colony size and the ease with which it accepts artificial nests. Natural popu- lations of these ants are composed of at least three kinds of nests: those having a single queen, those with two or more queens and those containing workers but no queens (Talbot, 1957; Herbers, 1984, 1986 for L. longispinosus). Therefore, one can measure the effects of queen number on growth in unmanipulated colonies.

This study had two initial goals. The first was to examine the effect of diet on the population dynamics of monogynous and polygynous colonies of L. curvispi- nosus by rearing them for 14 weeks on three different diets. Colonies were fed either insect prey alone (high protein diet), a sucrose solution (high carbohydrate diet), or both. The second was to observe whether the effect of queen number was the same in colonies reared on the different diets. Eggs, larvae, and workers were censused weekly to determine diet and queen effects on all life stages.

A substantial fraction of nests in the field contained no distinct queen. Eggs and larvae were present in these nests, suggesting either worker reproduction or egg laying by queens in neighboring nests of the same colony. We reared eight colonies which lacked a morphologically distinct queen on the high protein-high carbohydrate diet. Over a period of four months, members of three of these colonies laid eggs that developed into new workers. To determine whether these workers arose via parthenogenesis (thelytoky) from unmated workers, we established and monitored nests of newly eclosed (unmated) workers.

METHODS

Colonies of L. cunnspinosus were collected from acorns {Quercus rubra) in a wooded area on the northeast corner of the Princeton University campus, Princeton, New Jersey. Whole, intact acorns are impenetrable for the ants and they usually rely on acorn weevils (Curculionidae) to make the nuts habitable (Talbot, 1957). These weevils develop within fallen acorns, feeding on the fruit and finally drilling a small exit hole after eclosion. Acorns with holes indicating the emergence of curculionid beetles were taken in dark canisters to the laboratory, where they were examined for ants. Acorns having these characteristic openings usually contained colonies of L. cun’ispinosiis.

Ants were placed into individual arenas consisting of clear plastic boxes (11 X 11X4 cm) with lids. Fluon (Northern Chemical Co., Woonsocket, R.L) was applied to the container walls to prevent ants from escaping. Artificial nests were placed into each arena. Nests consisted of two glass microscope slides (2.5 X 7.5 cm) separated by a 1.0 mm wide piece of cardboard cut to form a rectangular cavity between the slides. A single entrance hole was cut along the walls of this cardboard spacer, and a layer of red cellophane was placed between the cardboard and the top slide to reduce light in the nest cavity, while allowing colony observation. Colonies usually moved into the artificial nests within ten hours of being placed into the arena. They were maintained at a temperature of 25°C, on a 16L:8D light cycle.

Following collection, queenright colonies were divided by queen number. Thirty six monogynous colonies were divided into three groups of twelve. Similarly, 15 two-queened colonies were divided into three groups of five. For each queen class, one group received the high protein diet, another received the high carbohydrate diet and a third group was given both. Colonies of similar size were randomized before

1995 DIET AND QUEEN NUMBER EFFECTS ON ANT COLONY GROWTH

93

Table 1. Egg, larva and worker numbers from initial census, October 30, 1987.

	No queen (N = 12)	One queen (N = 32)	Polygynous (N = 19)
X	(SE)		(SE)	A	(SE)
Eggs	10.33	(5.08)	21.84	(5.59)	32.68	(5.22)
Larvae	25.75	(5.91)	25.84	(3.57)	38.00	(6.48)
Workers	43.58	(7.66)	46.38	(5.38)	63.95	(10.64)

being placed into the three diet treatments, so that mean colony size was equal across treatments.

Colonies receiving the carbohydrate diet were fed a solution of 10% sucrose. This was dispensed through a capped 1.5 ml. micro-centrifuge tube with a 3^ cm long (0.25 mm diameter) capillary tube inserted through the side wall. To prevent pressure locks, small holes were made in the caps of the microcentrifuge. Colonies not re- ceiving sucrose were given identical feeders filled with distilled water. All colonies were also provided water via micro-centrifuge tubes capped at one end with cotton. Sucrose solutions and water were replaced throughout the experiment as needed. In the high protein diets, adult fruit flies {Drosophila melanogaster) were given to the colonies ad libitum. All flies were killed by freezing before feeding them to the ants, as a moving stimulus was not necessary to induce attack by Leptothorax workers. Uneaten flies were replaced with fresh flies every two days.

Production of brood and workers within each colony was estimated by taking weekly standing counts. For the censuses, nests were removed from their arenas, and eggs, larvae and workers were counted using a binocular microscope.

We compared the final counts of eggs and larvae using a two-way analysis of covariance (ANCOVA), with diet and queen number as factors and colony size (worker number) as the covariate. Change in worker number was also compared using an ANCOVA, with queen and diet classes as factors and initial worker number as the covariate. In both cases, interactions between queen number and diet effects were also examined. Six of the smaller monogynous colonies went extinct during the experiment and one colony lost its queen in the second week. These seven colonies were excluded from the analysis of worker production.

Colonies of unmated workers were established by periodically harvesting pupae from several productive queenright colonies. When these pupae eclosed, colonies were reared on a diet of Drosophila and a 10% sucrose solution. They were main- tained at constant temperature (25°C) for five months, then were kept at approxi- mately 4°C for three months, before being returned to room temperature.

RESULTS

Field collections. Table 1 presents the population structure for 63 queenless, mo- nogynous and polygynous colonies. Twelve (19%) of the acorns had no morpholog- ically distinct queen. Thirty-two (51%) had one queen, while nineteen (30%) were polygynous. Three additional colonies collected with this sample consisted of L. curvispinosus workers enslaved by queens and workers of the slave-making ant, Protomognathus americanus. These colonies were excluded from the experiment.

In the first census, egg presence appeared to be positively correlated with queen

94

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

number, although these differences are not significant when confounding effects of colony size are removed (ANCOVA, P = 0.244). Eight of the twelve queenless colonies lacked eggs, as did eleven of the thirty-two single-queen colonies. All po- lygynous colonies had eggs present.

All colonies, including those without queens, contained larvae. These larvae were predominantly early instars and no pre-pupae nor pupae were present. No male or female sexuals were found in any of the colonies censused. From their colony com- position, we assume that these colonies were preparing to diapause for the winter. Typically, colonies of L. cur\>ispinosus require a period of vernalization before re- commencing healthy brood production. However, in the following growth experi- ments, colonies resumed growth in the laboratory without overwintering. This sug- gests that, at least in certain parts of its range, L. curvispinosus does not have an obligate vernalization requirement.

Standing egg counts. Beginning by the fourth week, colony egg numbers showed the effects of their diets, as shown in Figure la. Colonies receiving flies only and those fed sucrose only had similar increases in egg numbers during the experiment, while those receiving both flies and sugar showed a much greater increase. By the end of the experiment, the effects of these three diet treatments were significant (Table 2). Monogynous and polygynous colonies had similar egg numbers throughout the experiment. When analyzed individually, none of the diet treatments showed any effect of multiple queens on egg numbers.

Standing larva counts. As was the case for eggs, larval numbers were highest in the combined fly and sugar treatment. Interestingly, larval numbers were higher in the sugar-only than in the flies-only treatment (ANCOVA, P < 0.01). As with the data from egg numbers, queen number did not significantly affect the count totals, and there were no interaction effects between queen number and diet. As shown in Table 2, final larval numbers depended significantly on worker number from the first census. Production of workers. Worker numbers diverged between the three diet treatments beginning around week 10 (Fig. lb). These differences occurred six weeks after the egg numbers had begun to show significant differences, giving a rough representation of the development time for workers under these diet and temperature conditions. Production of workers was significantly higher in colonies receiving both flies and sugar than in those receiving flies or sugar alone (Fig. lb. Table 2). Queen number did not influence worker production in any of the three diet groups (Table 2). Initial colony size was also a poor predictor of colony growth.

Production in queenless colonies. Of the eight apparently queenless field colonies, six increased in worker number over the course of the experiment (Table 3). In three of these colonies, the increase in worker number was greater than the combined total of eggs and larvae present initially. In 1 1 artificial colonies consisting of unmated workers, no new workers were produced. Males were produced in each of these colonies, suggesting that laboratory conditions were sufficient for brood develop- ment.

DISCUSSION

Diet effects

Diet quality had a strong effect on both egg and larva numbers and overall worker production of these Leptothorax colonies. Sucrose (carbohydrate) and insect prey

1995 DIET AND QUEEN NUMBER EEEECTS ON ANT COLONY GROWTH

95

a)

Week of Census

b)

Week of Census

Eig. 1. a) Weekly egg numbers during the experiment for monogynous and polygynous colonies given three different diets, b) Weekly worker numbers during the experiment for monogynous and polygynous colonies given three different diets.

96

JOURNAL OF THE NEW YORK ENTOMOLOGICAL SOCIETY Vol. 103(1)

Table 2. ANCOVA results for brood and worker production in experimental colonies. Work- er number was used as a covariate against diet and queen number. Egg and larva counts, and worker production, were strongly dependent on diet class. Larva numbers were also positively associated with worker number. * p < 0.05. ** p < 0.01. *** p < 0.001.

Source of effect	Nparm	DF		F ratio (Prob > F)
Eggs	Larvae	Worker increase
Diet	2	2	6.00 (0.005)**	9.22 (0.0005)***	7.15 (0.0022)**
Queen number	1	1	0.154 (0.697)	0.0046 (0.946)	0.031 (0.860)
Diet X queen number	1	1	0.058 (0.943)	0.554 (0.579)	1.396 (0.244)
Worker number	2	2	0.021 (0.885)	6.58 (0.014)*	0.237 (0.790)

(largely protein) clearly have synergistic effects on egg numbers and subsequent colony growth, as colonies given both grew far more quickly than those raised on an unlimited supply of one or the other. Because a diet of insect prey provides all the nutrients needed for growth, it is surprising that colonies given insect prey alone did as poorly as those fed only sucrose. In contrast. Porter (1989) found that colonies of Solenopsis invicta reared on crickets alone grew substantially faster than those given sucrose alone. Other studies (e.g., Buschinger and Pfeifer, 1988; Tschinkel, 1988) also suggest that protein availability is the main determinant of colony growth. It is possible that these colonies of L. cunnspinosus rely to a larger extent on energy derived from carbohydrates, than do other species.

The data on egg numbers do not provide information about actual egg-laying rates: rather they are a record of the fraction of eggs at each census that has survived oophagy, and it is possible that oophagy varied among the treatments. Brian (1957) found high rates of oophagy by first instar larvae in colonies of Myrmica rubra. Assuming that egg hatching time was constant for all treatments, these counts pro- vide an accurate measure of the relative number of potential offspring among the treatments. Accordingly, differences in egg numbers were correlated with the actual worker production of colonies raised on the three diets (corr. coefficient, r = 0.336, P < 0.05).

Table 3. Census results for eight queenless eolonies, taken over 18 weeks. * = Colony produeed more workers than the number of initial brood. > = Colony worker numbers inereased at some point during experiment.

11/20/1987 1/2/1988 4/2/1988

Colony	Eggs	Larvae	Workers	Total	Eggs	Larvae	Workers	Total	Eggs	Larvae	Workers	Total
1*	3	1 1	48	62	91	31	59	181	97	115	79	291
2*	17	6	31	54	30	41	31	102	—	62	68	130
3*	21	9	23	53	—	104	23	127	33	60	64	157
4>	—	4	4	8	—	4	4	8	—	3	5	8
5>	—	24	37	61	—	20	40	60	—	18	36	54
6>	22	4	21	47	7	15	26	48	1	6	34	41
7	17	9	4	30	2	21	2	25	—	7	1	8
8	28	12	54	94	94	29	51	174	7	29	38	74

1995 DIET AND QUEEN NUMBER EEEECTS ON ANT COLONY GROWTH

97

Queen effects

The laboratory experiment suggests that queen number does not strongly affect colony growth. If we assume that each queen in a polygynous nest lays roughly the same number of eggs as every other queen in that nest, then the individual fecundity of queens in terms of worker production appears to be substantially reduced by polygyny. These results are consistent with those collected by Wilson (1974) for L. curvispinosus colonies monitored over the course of one month. Elmes (1973) found that, while the standing crop of eggs in field colonies of Myrmica rubra increased with queen number, the number of larvae, hence colony growth, varied only with the number of workers. He predicted that colony growth in M. rubra reflects mainly the amount of food taken in, not the number of queens in a nest.

Further behavioral studies are needed to determine whether the reduction in per- queen productivity is due to inhibition among queens (as shown for Solenopsis in- victa by Vargo, 1992) or to limits imposed by the size of the worker force. Bourke (1993) found no evidence for such inhibition in Leptothorax acervorum. Behavioral and genetic analyses also would help to determine whether the “costs” of polygyny are shared equally within the nest. For example, in polygynous colonies of the im- ported fire ant, Solenopsis invicta, particular queens appear to dominate sexual pro- duction (Ross, 1988).

Various colony-level advantages from polygyny are possible. Members of polyg- ynous colonies might benefit from a higher survivorship rate of these colonies as opposed to monogynous colonies. In fact, in the course of this experiment, 7 of the monogynous colonies went extinct, while all of the polygynous colonies survived. While this result may be confounded by differences in worker number, there was overlap in colony size between some of the monogynous colonies that went extinct and the smallest polygynous colonies.

Another explanation for polygyny in ants is an increased ability to exploit newly available resources. For example, polygyny is present in some species of Pseudo- myrmex inhabiting the hollow thorns of tropical Acacia trees. Janzen (1973) sug- gested that polygynous colonies are more efficient at providing eggs for their ex- ponentially increasing Acacia nest sites. Although this has not been examined ex- perimentally, these colonies may use polygyny as a method of displacing other, competing, colonies. Perhaps the similar tendency of Leptothorax colonies to bud off into new nests (Stuart, 1985; Herbers, 1986) has selected for