- cq y / 6 2 6 XEffosSgV' Z ' o SZSMITHS0N!AN‘J|NSTITUTI0NZN0linillSNl"JNVIN0SHllWS S3 I dVd 811 z r- ^ ^ 2 r* z t <5 <$ipr m " XOm^X 2 1 •S^SMITHSONIAN^ INSTITUTION^NOIXfUIXSNI NVINOSHXIWs' 'S3 I a Va 8 11. CO -7 ^ ,>K.. 60 w 4 S /^x ^ w _ O _ ^ NI^NVINOSHilWS^SB I ava 8 II^LI B RAR I ESZSMITHSONIAN~INSTITUTION . 03 33 > m g X^sv^ m fjy - „ CO _ c/3 £S SMITHSONIAN INSTITUTION NOIXflXIXSNI NVINOSHXIIMS S3iavaan CO z * C/> 2 < "a 2 — 1 I § t 1 1 Ni^NviNOSHiiws^sa i ava a ii^li b rar i es^'smithsonian institution 5 \ « = w O . ^ z ES^ SMITHSONIAN^ INSTITUTION NOIlfUllSNI NVIN0SH1IWS S3 I d Vd 8 II 2 f" ... 2 ’'iZ O Z X^!2^4X o OT .^C •> o g\ 2 lol > 73 SNI~~NVIN0SH11WS $3 l dVd 8 ll~LI B RAR 3 ES ^SMITHSON IAN INSTITUTION Z . ^ CO 2 Z 'Stf iwKs, q . ,y a* 4 z =j /®5«mS\ z S I I S > ES^SMITHSONIAN ~ INSTITUTION N0I1D1I1SNI NVINOSHill/MS S3 I d Vd 8 I to ^ m UJ ~ Yv CQ .< *>V- . Xk H ''Cl? “ : 5 _ « = m % 1 (ir |l < (fr 'm JJ S * Vf LUSNI^NVINOSHill/NS S 3 J H V H Q 1 1 LIBRARIES SMITHSONIAN INST1TUTM r* . z r~ z m ^ rn ^ w \ R | ES ^SMITHSONIAN^INSTITUTION^ N0IlflJLIJLSNI~~NVIN0SHilINS S3 I HVH8 co 2 * co z c/5 IIISNI^NVINOSHIIINS^SS I 8 VH 8 lV\l B R AR I ES^SMITHSONIAN INSTSTUTH ^ X co __ 5 w 5 ftRIES SMITHSONIAN INSTITUTION NOlinilJLSN! WINOSHlIfeMS S3IHV8E 2: r” 2 n ... z ~~ JjV.m o 03 H /> Ij i*J j2 I? ^ rn. UIJLSNf’NVINOSHJLMS S3 I H V ^3 8 n~U B RAR I ES SMITHS0N1AN~!NST!TUT 2 co z w 2 ■“ /0%S&\ ~ A" — 1 |t *3] CO X ^wc; o co z co V z m ARIES SMITHSONIAN INSTITUTION NOIlfUllSNI NVIN0SH1I1AIS S3IBVBS CO = CO — CO PSYCHE A Journal of Entomology Volume 84 1977 Editorial Board Frank M. Carpenter, Editor P. J. Darlington, Jr. W. L. Brown, Jr. H. W. Levi E. O. Wilson Alfred F. Newton, Jr. B. K. Holldobler R. E. Silberglied Published Quarterly by the Cambridge Entomological Club Editorial Office: Biological Laboratories 16 Divinity Avenue Cambridge, Massachusetts, U.S.A. The numbers of Psyche issued during the past year were mailed on the following dates: Vol. 83, no. 3-4, September-December, 1976: August 29, 1977 Vol. 84, no. 1, March, 1977: November 30, 1977 Vol. 84, no. 2, June, 1977: March 28, 1978 ■>°a y + PSYCHE A JOURNAL OF ENTOMOLOGY founded in 1874 by the Cambridge Entomological Club Vol. 84 March, 1977 No. 1 CONTENTS Attacks on Large or Heavily Defended Prey by Tropical Salticid Spiders. Michael H. Robinson and Carlos E. Valerio 1 The Taxonomic Status and Biogeographic Significance of the Sumatran Formica (Formicidae, Hymenoptera). Andre Francoeur 11 Aerial Dispersal Behavior of Two Orb Weaving Spiders. Wayne W. Tolbert ... 13 Ecology, Zoogeography and Taxonomy of the Lower Rio Grande Valley Mesostenines (Hymenoptera, Ichneumonidae). Charles C. Porter 28 Taxonomy of the United States Leucochrysa (Neuroptera: Chrysopidae) Phillip A. Adams 92 Egg Guarding by Male Assassin Bugs of the Genus Zelus (Hemiptera: Reduviidae). J. Scott Ralston 103 Notice of Reprints of Articles by Professor W. M. Wheeler 107 CAMBRIDGE ENTOMOLOGICAL CLUB Officers for 1976-1977 President Vice-President Secretary Treasurer Executive Committee Paul S. Miliotis Gary D. Alpert Karen S. Vinson Frank M. Carpenter John A. Shetterly Jo B. Winter EDITORIAL BOARD OF PSYCHE F. M. CARPENTER (Editor), Fisher Professor of Natural History, Emeritus, Harvard University J. F. LAWRENCE, Coordinator of Entomological Collections, Harvard University W. L. BROWN, Jr., Professor of Entomology, Cornell University, and Associate in Entomology, Museum of Comparative Zoology P. J. DARLINGTON, JR., Professor of Zoology, Emeritus, Harvard University B. K. HOLLDOBLER, Professor of Biology, Harvard University H. W. LEVI, Alexander Agassiz Professor of Zoology, Harvard University R. E. SlLBERGLIED, Assistant Professor of Biology, Harvard University E. O. WILSON, Baird Professor of Science, Harvard University PSYCHE is published quarterly by the Cambridge Entomological Club, the issues appearing in March, June, September and December. Subscription price, per year, payable in advance: $8.00 for United States and Canada, $9.50 for other countries. Single copies, $2.50. Checks and remittances should be addressed to Treasurer, Cambridge Entomo- logical Club, 16 Divinity Avenue, Cambridge, Mass. 02138. Orders for missing numbers, notices of change of address, etc., should be sent to the Editorial Office of Psyche, 16 Divinity Ave., Cambridge, Mass. 02138. For pre- vious volumes, see notice on inside back cover. IMPORTANT NOTICE TO CONTRIBUTORS Manuscripts intended for publication should be addressed to Professor F. M. Car- penter, Biological Laboratories, Harvard University, Cambridge, Mass. 02138. Authors are expected to bear part of the printing costs, at the rate of $22.50 per printed page. The actual cost of preparing cuts for all illustrations must be borne by contributors: the cost for full page plates from line drawings is ordinarily $ 1 8.00 each, and for full page half-tones, $30.00 each; smaller sizes in proportion. The Sept. -Dec., 1976, Psyche (Vol. 83, No. 3-4) was mailed August 29, 1977 The Lexington Press, Inc. Lexington, Massachusetts PSYCHE Vol. 84 March, 1977 No. 1 ATTACKS ON LARGE OR HEAVILY DEFENDED PREY BY TROPICAL SALTICID SPIDERS By Michael H. Robinson1 and Carlos E. Valerio2 Introduction Spiderlings in the first active instar have severe limitations in prey capture, because of their small size (Valerio, 1975) and par- ticularly in those species that ambush or stalk their prey. The pres- ence of snares or catching webs characteristic of several families expands considerably the range of potential prey items, which is undoubtedly an important pressure in the evolution of such struc- tures. Even web-building spiders have problems with the large heavily-sclerotised prey items (see for instance Robinson & Robin- son 1973a, 57-58). Insects with chemical defenses also prove trouble- some to spiders (Eisner & Dean, 1976). However, the use of silk in the immobilization wrapping of araneid spiders considerably en- hances their ability to subdue large or heavily defended prey (see experimental analyses summarized in Robinson 1975). Salticids, on the other hand, are among the hunting spiders that subdue their prey without the aid of silk. For this reason, it is widely assumed that they are limited, in general, to prey which is smaller than themselves or to soft-bodied defenseless items (Enders 1975, 745 and references). At first sight this assumption seems perfectly reasonable, since the salticid attacking prey larger than itself must contend with a strength (perhaps) superior to its own. The insect under attack would presumably push against the sub- strate and exert sufficient pressure either to escape or to injure ■Smithsonian Tropical Research Institute, P.O. Box 2072, Balboa, Panama Canal Zone. 2Escuela de Biologia, Universidad de Costa Rica, Cuidad Universitaria, Costa Rica. Manuscript received by the editor June 8, 1977. 2 Psyche [March the spider. We here report on field observations (in Panama and Costa Rica) that show that certain tropical salticids do attack and subdue prey considerably larger than themselves. Among these prey are large araneid spiders that are attacked on the web (but not across the web). In making these attacks on large prey the spider may utilize the technique of dropping on its dragline to isolate such prey from the substrate. This action allows the spider to safely attack other types of prey which, although small, are normally protected by social defenses. Observations An adult Phiale was observed in Panama, preying upon a fully sclerotised adult dragonfly that was at least three times as long as the spider. The spider was on the upper surface of a leaf about 1.5 meters above ground level. At the time of discovery the dragon- fly was fluttering spasmodically but the actual capture was not observed. There is little doubt that the dragonfly was attacked after alighting on the leaf. The relative proportions of the spider and its prey are obvious from the photograph (Figure 1.). Observations in the Central Valley, Costa Rica, provide a clue about how small salticids may subdue large prey. An immature Menemerus bivittatus was seen pouncing on a large moth resting on a fence wire. The moth was about half as long again as the spider and perhaps twice as heavy. After the pounce the moth started beating its wings strongly and the spider immediately dropped, on its dragline, until it was well clear of the substrate (figure 2). The spider held the moth with its chelicerae and front legs until the prey was subdued. Clearly this method of “playing” the prey on the end of a line until envenomation occurs or the prey is exhausted, or both, is a strategy that could be applied to any prey item that tried to escape from the spider by jumping, dropping or flying off the substrate. The tensile strength of the dragline silk, in all probability, greatly exceeds the load exerted by the spider and her prey. The tenacity of the spider’s jaw hold may be the critical factor in such attacks. Dropping below the substrate on a dragline also provides the spider with an effective method of dealing with some species of ants that have social defenses. Thus some species of Pseudomyr- mex possess a strong alarm pheromone that directs large numbers 1977] Robinson & Valerio — Salticid Spiders 3 Figure 1. Adult female Phiale sp. feeding on anisopteran dragonfly, Navy Pipe- line Road, Canal Zone, Panama. June 18th 1976. of individuals to the exact place where a member of their colony is in danger. The response to the alarm pheromone is very rapid and may occur within seconds (Janzen 1966). This adaptation could effectively deter salticid predation on the ants were it not for the use of the dragline described above. An unidentified salti- cid (not collected) was observed making effective use of this tech- nique at a lowland site in Guanacaste, Costa Rica. The Pseudo- myrmex were attacked in an Acacia tree. The spider simply pounced on the ant, dropped off the branch and held the ant, suspended on the end of the dragline, and ate it. The ants, attracted by the 4 Psyche [March Figure 2. Immature Menemerus bivittatus feeding on moth in Central Valley, Costa Rica. The spider has dropped on its dragline beneath the fencing wire. 1977 ] Robinson & Valerio — Salticid Spiders 5 alarm pheromone, found the end of the dragline, but were unable to descend the thread. The spider returned to the branch and re- peated the operation several times during the period of observa- tion. (It is worth noting that at night we frequently find salticids and other diurnal non-web-building spiders suspended on their draglines beneath the vegetation. This may provide the safest way of spending the hours of darkness, since they are virtually isolated from the vegetation on which prowl innumerable preda- tory arthropods. Should any of these be capable of descending the dragline, the vibrations thereby induced would presumably alert the resting spider.) Attacks on web-building spiders. There are indications that web-building spiders are preyed upon by an extensive array of predators although the records are scattered throughout the literature and detailed observations are surprisingly few in number. Bristowe (1941; 331-443) deals comprehensively with the enemies of spiders in general and also describes a wide variety of anti-predator adaptations that spiders possess. The defenses of tropical orb-weavers are reviewed by Robinson & Robinson (1970; 649-653) and these authors describe particular defensive structures or behaviors elsewhere (1973a, 1973b). Tolbert (1975) has reviewed some of the available litera- ture on araneid defensive behaviors in conjunction with an experi- mental study of the defensive responses of Argiope aurantia and A. trifasciata. Records of attacks on orb-weavers by other spiders have been few in number. Bristowe (1941; 377-378) lists a number of at- tacks on web-building spiders by hunting spiders, and, in par- ticular, by the salticid Linus fimbiatus. The spiders attacked in- cluded at least one araneid. Bristowe (ibid; 378) implies that the spiders were captured in their webs, “The Linus ... sat in its vic- tim’s web to eat the owner”. Tolbert (1975) mentions attacks on Argiope aurantia and A. trifasciata by salticids and states that attacks in the field can be induced by prodding the Argiope to move (Tolbert, in litt.). Enders (1974) reports attacks on orb- weavers by orb-weavers and (1975; 970) on the “invasion” of the webs of orb-weavers by errant salticids. In three months (May-July, 1976) during extensive census- ing of webs in a number of forest fringe habitats in the Summit 6 Psyche [March and Gamboa areas of the Panama Canal Zone, 14 adult female Argiope argentata were found being consumed by Phiale adults. (On one count Phiale were found consuming 3 out of 64 spiders censused.) The spiders were, in all cases, off the web and resting on nearby vegetation. The araneid is considerably larger than the salticid (figure 3) and at least twice as heavy. A Phiale was also seen feeding on a late instar Nephila clavipes (F. Vollrath, pers comm.). No attacks were seen and it was not clear how the salticid had captured the araneid. To settle this problem, salti- cids were introduced into cages containing adult A. argentata (in webs) and watched. The web-builders were not fed and no attacks or “invasions” of the web were seen during intermittent observa- tions over a period of three days. Feeding one spider immediately gave a clue as to the attack method of the salticid. As the Argiope moved to attack a grasshopper the salticid became active and moved along the walls of the cage to various positions from which it clearly “looked” at the moving Argiope. No attack was made, but when the spider returned to the hub, leaving the wrapped prey at the capture site, the salticid moved to a position on the cage wall almost horizontally opposite the stored prey, and after a number of side to side movements of the cephalothorax, it leapt upon the prey to stand astride it, biting. The Argiope immediately started to make pumping movements at the hub (‘web-flexing’, Tolbert 1975). This movement shook the prey item and shortly after its commencement, the salticid jumped off and regained its former position on the cage wall. Feeding the same spider a second time resulted in a similar response on the part of the salti- cid. This time, after leaping on the completely motionless prey package, it did not provoke the Argiope into pumping, and fed undisturbed on the cricket for over five minutes. At this point the host ran to the stored prey and dragged it closer to the hub, and the salticid leapt off to regain the cage wall. The salticid made one more attack on the prey package and then 16J4 minutes after the start of the activity, attacked the spider at the hub by leaping on it. The Argiope was on the opposite side of the hub to the salticid and immediately dropped to the cage floor. The Phiale then walked on the web to the stored prey and fed upon it. Subse- quent experimentation showed that the salticids could regularly be induced to attack Argiope if the latter were provoked into moving. Attacks on the wrong side of the hub were not successful 1977] Robinson & Valerio — Salticid Spiders 7 Figure 3. Adult female Phiale sp. feeding on adult female Argiope argentata. The salticid is perched close to the upper left hand corner of the araneid’s web. Old Gamboa Road, Canal Zone, Panama. May 19th 1976. 8 Psyche [March but attacks from above the dorsal surface of the araneid were successful in all cases. (Eventually, all four A argentataw/ere killed.) In all cases, the araneids jumped off the hub when the Phiale con- tacted them. At the cage floor they moved about but could not displace the salticid and were eventually pulled up the cage wall to a feeding site. The conditions in the cages probably made the attacks easier than they would be in field conditions. The Argiope web was surrounded by a continuous rigid surface on all sides. In the field the salticid must have to rely on discrete vegetation units for origi- nating its attacks and though it can jump from plant to plant until it finds a suitable site, it may not be able to keep the spider in view continuously. In the cages the salticids looked at the Argiope from the cage floor, the cage walls and even the cage roof, before eventually lining up on the wall to launch an attack. Where it was possible to gauge the point of origin of attacks with some accur- acy, they seemed to occur from a position only slightly above the point horizontally opposite the spider. When launching attacks off the glass sides of cages, the salticid turned around several times before jumping. Subsequent examination showed several silk attachments on the glass in this region. This suggests that the spider may make multiple dragline attachments before long aerial attacks. Take off postures were always head down (i.e. with the cephalothorax lowermost but strongly angled towards the target, and with legs I off the substrate). These observations made on a small sample in simplified con- ditions show that an attack on the dorsal surface of a large prey item can be very successful. Movement seems to be necessary for the initiation of hunting behavior, but attacks were made on subsequently motionless prey. The salticids made accurate dis- tance terminations and traversed horizontal distances measured at greater than 12cm. The failure of attacks made from the ‘wrong side of the hub’ (i.e. with the web between the salticid and the araneid) suggest that the behavior of shuttling (= switching sides of the web, Tolbert 1975), may be an effective defense, as argued by Robinson & Robinson (1970). Dropping from the web clearly did not aid the araneid in the experimental situation but could help in defense against salticids in a more natural one. The araneid might be able to brush off its attacker against the vegetation be- low the web. It can clearly work in other contexts against other predators. 1977] Robinson & Valerio — Salticid Spiders 9 The basic predatory techniques. Dropping on the dragline to isolate a large prey from the sub- strate may be partly fortuitous in some cases. Certainly it depends on the prey moving off the substrate as a result of its own escape movements, since the salticid cannot lift it off. However, the case of the attacks on the ants suggests that it may be part of the normal predatory repertoire for dealing with some types of small prey. Attacking large araneids from above their dorsal surface presumably utilizes a technique that is part of normal prey capture but capitalizes on the araneid’s inability to make strong scraping movements against its upper surface. It is also probable that such attacks benefit from the fact that the spider is not standing on a rigid substrate when attacked. The peculiar defensive posture adopted by Nephila spp. in response to direct tactile stimulation of their dorsal surfaces (Robinson & Robinson 1973a) results in a “barrier” of flexed legs being erected above the spider and could serve to frustrate some dorsal attacks. Summary 1. Some tropical salticids regularly catch prey larger and heavier than themselves. 2. Such salticids may utilize a dorsal attack on the prey followed by dropping on a dragline to effectively isolate the prey from the substrate. 3. This technique could be much more common than we know and definitely extends the size range for the potential prey of these spiders. 4. The drop and hold technique allows the salticids to attack prey that would normally be protected by social defense. 5. Salticids can make aerial attacks on araneid spiders in their webs and the normal defensive dropping responses of these spiders may, in certain circumstances, facilitate the salticid attack. References Bristowe, W. S. 1941. The Comity of Spiders, Volume II. London, Ray Society. Eisner, Thomas and Jeffrey Dean 1976. Ploy and counterploy in predator-prey interactions: Orb-weaving spiders versus bombadier beetles. Proc. Nat. Acad. Sci. USA, 73: 1365-1367. 10 Psyche [March Enders, F. 1974. Vertical stratification in orb-web spiders and a consideration of other methods of coexistence. Ecology 55: 317-328. Enders, F. 1975. The influence of hunting manner on prey size, particularly in spiders with long attack distances (Araneidae, Linyphiidae, and Salticidae). Amer. Natur., 109: 737-763. Janzen, D. H. 1966. Coevolution of mutualism between ants and acacias in Central Ameri- ca. Evolution, 20: 249-275. Robinson, M. H. 1975. The evolution of predatory behaviour in araneid spiders. In Baerends, G., Beer, C., & A. Manning, Eds. Function and Evolution in Behaviour, Clarendon Press, Oxford., 293-312. Robinson, M. H. and B. Robinson 1970. The stabilimentum of the orb web spider, Argiope argentata: an im- probable defence against predators. Canad. Entomol. 102: 641-655. 1973a. Ecology and Behaviour of the Giant Wood Spider Nephila maculata (Fabricius) in New Guinea. Smithsonian Contr. Zool., 149:1-76. 1973b. The stabilimenta of Nephila clavipes and the origins of stabilimentum- building in araneids. Psyche, 80: 277-288. Tolbert, W. W. 1975. Predator avoidance behaviors and web defensive structures in the orb weavers Argiope aurantia and Argiope trifasciata (Araneae, Aranei- dae). Psyche, 82: 29-52. Valerio, C. E. 1975. Population structure in the spider Achaearanea tepidariorum (Araneae, Theridiidae). J. Arachnol., 3: 185-190. THE TAXONOMIC STATUS AND BIOGEOGRAPHIC SIGNIFICANCE OF THE SUMATRAN FORMICA (FORMICIDAE, HYMENOPTERA)* By Andre Francoeur Departement des Sciences Pures, Universite du Quebec a Chicoutimi, Quebec, Canada G7H 2B1 In a paper on the occurrence of Formica fusca in Sumatra, W. M. Wheeler (1927) erected the variety fairchildi for 12 workers collected above Kota Dah at an altitude of 4,000 feet in a pine forest. It has never been found again as far as I know. The ex- amination of 10 of these specimens located in 3 different U.S. museums revealed a surprising similarity with Formica glacialis, a name that I have recently resurrected in a taxonomic revision of the nearctic species belonging to the Formica fusca group (Fran- coeur, 1973). I compared workers of F. fairchildi to F. glacialis types and topotypes collected by me at South Harpswell, Maine, and no significant difference was noted. All the above specimens meet very well my description of the F. glacialis worker. There- fore the formal synonymy is: Formica glacialis Formica fusca var. glacialis Wheeler, 1908, Bull. Amer. Mus. Nat. History 24: 624, worker, female, male. Formica fusca fusca: Wheeler (in part), 1913, Bull. Mus. Comp. Zool. Harvard 53: 494-497. Formica fusca: Creighton (in part), 1950, Bull. Mus. Comp. Zool. Harvard 104: 532. Formica glacialis: Francoeur, 1973, Memoire Soc. Ent. Quebec 3: 152-161. Formica fusca var. fairchildi Wheeler, 1927, Psyche 34: 40-41, worker. Lectotype in MCZ, paratypes in AMNH, MCZ, USNM. new synonymy. This new synonymy eliminates the concept of a distinctive form of Formica in the southern half of the Oriental region. The pres- ence of the genus in northern Sumatra perhaps may now be con- sidered as an unexplained introduction rather than a tropical relict as reinterpreted by Gregg (1969) from Wheeler (1927). However, * Manuscript received by the editor August 11, 1977 11 12 Psyche [March the possibility might be considered that the Kota Dah Formica sample represents a labelling error, or the misplacement of a vial with North American ants in the Fairchild collecting kit. Such explanation seems much more likely than any introduction of For- mica glacialis live into Sumatra. Other examples of this sort of mishap are very common in the Wheeler collection. In papers pub- lished in 1922 and 1927, Wheeler reported and discussed a similar case for the Philippines. With this puzzling case once solved, the natural geographic dis- tribution of the genus Formica appears to be entirely holarctic. Biogeographic boundaries of course do not follow straight lines; rather, they reflect topography and other factors affecting climate. Mountain ranges carry holarctic elements southward toward and into the tropics in both the Old World and the New. Thus the presence of Formica species in the high mountains of Taiwan and Burma is not surprising, since these ranges are nearby outliers or direct continuations of the holarctic uplands of mainland Asia. The range of the genus includes also the high elevations of central Mexico, in North America. Nevertheless, that Formica could at one time have had a much wider or somewhat different distribution can still be supported by its richness, greater than previously recognized, in living species in the southern half of the Holarctic region, and by the presence of fossil Formica among numerous other subtropical and warm tem- perate insects found in the Baltic Amber of Oligocene age. Acknowledgements I am indebted to Mrs. Favreau, American Museum of Natural History, New York, H. E. Evans, Museum of Comparative Zool- ogy, Harvard University, and D. R. Smith, U.S. National Museum, Washington, for loan of specimens. E. O. Wilson and W. L. Brown have my thanks for critically revising the manuscript. Research supported by National Research Council of Canada Grant A6501. References Gregg, R. E. 1969. Geographic distribution of the ant genus Formica (Hymenoptera: For- micidae). Proc. Ent. Soc. Washington 71(1): 38-49. Wheeler, W. M. 1922. Ants of the genus Formica in the Tropics. Psyche 29: 174-177. AERIAL DISPERSAL BEHAVIOR OF TWO ORB WEAVING SPIDERS By Wayne W. Tolbert* Graduate Program in Ecology University of Tennessee Knoxville, Tennessee 37916 Introduction Aerial dispersal, the transport of spiders from place to place by wind and/or convection currents, has been recognized as a feature of spider behavior since the time of Aristotle (Duffey, 1956). Many natural historians and arachnologists have observed and briefly commented on this phenomenon (Emerton, 1908; Bristowe, 1939; Gertsch, 1949; Nishiki, 1966; and Kaston, 1972), and a few studies have been devoted to the environmental conditions associated with the general phenomenon of spider dispersal. Most studies and observations have been made of mass migra- tions of several species of spiders, particularly migrations occur- ring during the winter months (Bristowe, 1939). Duffey (1956) determined that temperature, population density, and stages of the breeding cycle are associated with mass aerial migrations of several species of Linyphiidae. Van Wingerden and Vugts (1974) pro- duced results similar to Duffey’s for one lingphiid species, Erigone artica (White). Richter (1970, 1971) has studied in the laboratory some micro- climatic factors which influence aerial dispersal in eight species of Pardosa wolf spiders. Richter (1970) related the frequency of aero- nautic behavior of each species to the abundance and stability and that species’ preferred habitat. The purposes of this study are to describe the aerial dispersal behavior of Argiope trifasciata (Forskal), to compare behavioral elements of this species with a sympatric population of the con- gener, A. aurantia (Lucas), and to determine, under actual field conditions, the major physical parameters which influence these behaviors. Factors influencing emergence from the egg sac are discussed elsewhere (Tolbert, 1976, in preparation). *present address — Science Applications, Inc. P. O. Box 843 Oak Ridge, Tennessee 37830 Manuscript received by the editor May 27, 1977 13 14 Psyche [March Methods This study was conducted in April and May 1975, in a two hectare overgrown pasture in Loudon County, Tennessee; a complete de- scription of the study area is reported elsewhere (Tolbert, 1976). The study area is 3.2 kilometers (2 miles), west of Glendale com- munity. Spiderlings of two orb weaving spider species, Argiope trifasciata and Argiope aurantia, were observed after emergence from more than 50 egg sacs. The spiderlings from egg sacs were monitored for dispersal and related activities during daylight hours. Spider body temperatures were estimated by the use of a thermo- couple junction and associated cylindrical solder model of the same dimensions as a first instar spiderling (0.5 mm X 1.5 mm). Direct readout of model temperatures at dispersal height were recorded on an Esterline-Angus Continuously Recording Potentiometer. Wind speeds at dispersal height were measured with a Rimco miniature cup anemometer, stall speed 0.25 m/sec. Behavior Prior to Dispersal Unlike Argiope aurantia spiderlings, which emerge from an egg sac over a period of several days or weeks (Tolbert, 1976), Argiope trifasciata emerged in mass during the study period of April and May 1975. All spiderlings emerged from any given egg sac within a single day, usually within a period of one to two hours. This was confirmed by cutting open egg sacs from which spiderlings had recently emerged and checking for spiderlings that were left. An ethogram which summarizes the findings of this study on Argiope spp. dispersal behavior is depicted in Figure 1 . Individuals quickly constructed a communal tangle or communal web, which is a mesh- work of interlocking threads, by laying down draglines. Division of labor was not observed; each spiderling simply contributed a small amount of silk to the tangle. Spiderlings observed in 1975 resided on such tangles for several days (x ± S. E. = 3.5 ± 0.52 days) before dispersing. Valerio (1975) reports that the common house spider Achaeranea tepidorium (C. L. Koch), also spends three to four days in dense clusters on the maternal web before dispersing aerially. Communal tangle formation by Argiope au- rantia was less common. In the four instances where communal tangles were constructed by A. aurantia, the egg sacs had fallen to the substrate. Possibly a differing microclimate near the 1977] Tolbert — Orb Weaving Spiders 15 [spiders in egg sac) EMERGENCE M BEHAVIOR ON THREAD CHANGE WEB SITE 'REBAUOONING' BY WALKING Figure 1 . An ethogram of Argiope spp. dispersal behavior. 16 Psyche [March ground as demonstrated by Geiger (1965) influences communal tangle formation. Spiderlings emerging high in the vegetation ex- perience greater exposure to wind which might stimulate ballooning (discussed below). Argiope aurantia locate egg sacs higher in the vegetation than A. trifasciata. In 1974-1975 and 1975-1976 the means and standard errors of A. aurantia egg sac heights were 0.91 ±0.11 and 1.10 ±0.13 m, respectively (Tolbert, 1976). Argiope trifasciata egg sacs by comparison, were 0.30 ± 0.04 m in 1975-1976 (Tolbert, 1976). Spiderlings on a communal tangle were tolerant of one another at all times and readily accepted conspecifics from different egg sacs. Several such “transplants” were made during the spring of 1975 with spiderlings from one egg sac transferred to the communal tangle of spiderlings from a different egg sac. After a brief flurry of activity produced by the arrival of the transplants, activity de- creased to levels noted before the introductions. At night, early in the morning, and on overcast days Argiope trifasciata spiderlings maintained individual spacing of several times their own body length while on the communal tangle (Figure 2). If disturbed by predators, wind gusts, or rain, spiderlings became agitated but quickly resumed a quiescent attitude if their body temperatures were less than 26° C. On the other hand, if exposed to full sunlight individuals clustered closely together on the com- munal tangle (Figure 3). If cues were not present for dispersal, the spiderlings remained clustered until after nightfall before spacing out. Though not timed, clustering appeared to take less than a minute. A longer period was required for clustering if the group was partially shaded. The exact cue or clues producing clustering are not known, but clustering occurred only after full exposure to bright sunlight. Air temperatures, spiderling body temperatures, light levels, and perhaps other factors which might provide the required stimuli for clustering all change at this time. More experimenta- tion is needed for elucidation of this problem. When Argiope trifasciata spiderlings are clustered and their body temperatures equalled or exceeded 26° C they were susceptible to dispersal. As indicated in Figure 4, dispersal ceased below 26° C and the preferred temperature range for dispersal was between 33° and 38° C. Ninety-two percent of all A. trifasciata spiderlings be- came airborne or “ballooned” when their estimated body tempera- tures were between 33° and 38° C. Argiope aurantia also dispersed 1977] Tolbert — Orb Weaving Spiders 17 Figure 2. An emergence of Argiope trifasciata with spiderlings exhibiting individual spacing. Arrow in the center of the cluster of spiderlings. 18 Psyche [March Figure 3. An emergence of Argiope trifasciata with spiderlings tightly clustered. 1977] Tolbert — Orb Weaving Spiders 19 when their body temperatures were above 26° C, and their preferred range was virtually identical to that of Argiope trifasciata (Figure 5). Seventy-four percent of A. aurantia spiderlings dispersed when their body temperatures were between 33° and 38° C. A disturbance (gust of wind, striking the vegetation to which the communal tan- gle is attached, striking the tangle itself, and fanning the cluster of spiders directly) resulted in spiderling movement on the tangle. The nature of the disturbance seems to be vibrational and subsequent movement by spiderlings on the interlocking mass of silk threads of the communal tangle appeared to reinforce the initial disturb- ance. When a sufficient disturbance occurred and spider body tem- peratures were 26° C or higher, the spiderlings climbed to the top of any available object, usually vegetation. Spiderlings followed one another in mass to the top of a promontory where they pre- pared to disperse. Neither disturbance nor elevated temperature by itself was sufficient to trigger climbing behavior of A. trifasciata (Figure 1). By monitoring spiderling body temperatures on the communal tangle (spiderlings of eight Argiope trifasciata egg sacs in 1975), it was apparent that the spiderlings often experienced tem- peratures within their observed dispersal range (26° -42° C), but did not disperse. This was also evident by the time of residency on communal tangles during sunny periods (Table 1). Striking the vegetation near the tangle or fanning the cluster of spiderlings during times when body temperatures equalled or exceeded 26° C invariably induced climbing. Similar stimulation at body temper- atures of less than 26° C failed to produce climbing behavior. In April 1976, two additional clusters of A. trifasciata spiderlings were tested. Even when body temperatures exceeded 26° C and a vibratory stimulus was applied, these spiderlings could not be in- duced to climb and disperse. While these findings might simply be abberations or due to some genetic differences between populations between the two years, or to some microclimatic variables which were not examined, it seems likely that developmental differences between the spiderlings which emerged in May and those in April may account for the observed difference in behavior. A. trifasciata overwintered as eggs and the active spiderlings did not develop until April (Tolbert, 1976). A difference of as much as one month to six weeks in the developmental age of spiderlings could influence behavior. It is possible that the older spiderlings would be more prone to dispersal behavior than younger spider- 20 Psyche [March SPIDER BODY (MODEL) TEMPERATURE (°C) Figure 4. Number of Argiope trifasciata ballooning as a function of body (model) temperature. lings. Residence time by spiderlings on communal tangles was somewhat longer in early May than in late May (Table 1). Com- puting mean residence time (in days) from the data in Table 1, re- veals 7.5 days for the spiderlings emerging on 10 May, 8 days for 11 May, 5 days for 14 May, 2.3 days for 19 May, and 4 days for 22 May. Richter (1970) has stated that dispersal by spiders normally oc- curs on days that are unusually warm and calm for the time of year. In the same study, he demonstrated that a wolf spider, Par- dosa purbeckenis F. O. B. Cambridge exhibited aeronautic be- havior when laboratory air temperatures were varied between 18° and 34° C. The percentages of aeronautic behavior for each tem- perature range he used (which I calculated from his Table 4) were 6.9% (18-19° C), 34.8% (28-29° C) and 36.4% (33-34°C). In 1975, 20 of the 24 A. trifasciata spiderling masses dispersed on sunny days (see Table 1). Thus, it appears that dispersal in Argiope is influenced by climatogical factors, and in particular temperature. In addition to these findings, Duffey (1966) in monitoring winter 21 1977] Tolbert — Orb Weaving Spiders 20 15 10 5 . 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 SPIDER BODY (MODEL) TEMPERATURE (°C) Figure 5. Number of Argiope aurantia ballooning as a function of body (model) temperature. dispersal of linyphiids, indicated that increasing litter temperatures were partially responsible for increased aerial dispersal by mem- bers of this litter-dwelling spider family. Gypsy moth ( Porthetria dispar L.) larvae ascend trees preparatory to aerial dispersal when ambient air temperatures are between 15.6°C (50° F) and 29.4° C (85° F). These larvae, however, have black dorsal surfaces and their body temperatures when exposed to direct sunlight can easily ex- ceed air temperatures (McManus, 1973). Thus, the aerial dispersal of some spider and insect species is influenced by temperature. Aeronautic Behavior In preparation for dispersal, individuals may adopt either a ’’tip- toe” posture, as defined by Richter (1970) or hang suspended from a dragline from which they become airborne (Figure 1). The tip- toe stance, which is widely employed by lycosid spiders (Richter, 1970), results when the spider depresses the cephalothorax toward the substrate and elevates the abdomen. Silk lines are then exuded from the spinnerets. Multiple lines of ballooning silk were often 22 Psyche [March Table 1 . Argiope trifasciata residency on communal tangles and associated weather conditions during May, 1975. S = sunny, PC = partly cloudy, C = cloudy, X = presence, Y = dispersal on same day as emergence. Dispersal occurred on last day spiderlings were present. May 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Weather S SPCS SCCCS S S S SPCPCS Egg sac number 5 6 11 12 13 14 15 16 17 18 19 20 21 22 23 25 26 27 29 30 31 32 33 34 X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X XXX X X X X X X X X X X X X X X X X X Y X X X X X X X X X X X X Y XXX XXX XXX XXX X X X X X X observed and there appeared to be four to six lines, but the exact number was not determined. Spiderlings also dispersed directly from silk threads. They dropped on a dragline, cut it and while suspended and holding this thread by leg pairs I and II, let out ballooning silk as before, until they became airborne. Spiderlings, as a rule, became airborne during relatively calm periods with a preferred windspeed for dispersal of approximately 0.5 m/sec. (Figure 6). During gusty periods spiderlings either did not exhibit tip-toe postures at all or if in that posture, returned to 1977] Tolbert — Orb Weaving Spiders 23 WIND SPEED (meters/second) Figure 6. Number of Argiope trifasciata ballooning as a function of windspeed. a normal resting attitude. Spiderlings did not disperse at wind- speeds of greater than 0.9 m/sec. (Figure 6). Richter (1971) showed that Pardosa purbeckensis had preferred windspeeds for dispersal which were related to their body sizes; larger spiders preferred greater windspeeds. All size classes, however, preferred wind veloc- ities between 0.35 and 1.70 m/sec. These values exceed those ob- served for Argiope trifasciata, which is a slightly larger spider than Pardosa purbeckensis . From their study of spiders on Frisian Is- land (the Netherlands) Van Wingerden and Vugts (1974) concluded that aeronautic behavior ceased when the wind velocity at 2 m above the substrate exceeded 3.0 m/sec. They also found in the same study that unstable air masses near the ground and at dispersal height served as stimuli to aeronautic behaviour. Disturbances (wind gusts) apparently function in a similar fashion for Argiope trifasciata . After a spiderling becomes airborne it will travel on wind cur- rents until deposited at a potential web location. The spiderling may then build a web, or move a short distance from the deposi- tion site (perhaps searching for a site of better potential for web construction), and then build a web (Figure 1). It might also re- balloon with or without “searching behavior.” Data on the relative 24 Psyche [March frequencies of these behaviors are scarce. Approximately 20 argio- pids were observed to reballoon during the course of the study, some as many as six times with the majority (90%) of these observed reballooning bouts occurring when the spiderlings were in close proximity to the egg sac. It is possible that a minimum time or distance requirement must be achieved before the spiderling has satisfied a “ballooning drive.” It might also reflect investigator error since spiderlings which drop into dense vegetation are diffi- cult to detect and the probability of finding individuals must de- crease with distance from the egg sac (with decreased density of spiderlings). Nevertheless, multiple ballooning bouts probably func- tion to increase emigration from a given area (which would tend to lessen competition for food, web sites or other resources if such resources are in short supply). Since Argiope spp. actively select web sites (Enders, 1972, 1973), it is probable that areas judged un- suitable by the spiderlings can be quickly and easily abandoned via reballooning. This finding is consistent with the predictions of Doyle’s (1975) habitat selection model. Organisms which encounter coarse-grained (patchy) environments, as Argiope do, can improve fitness by selecting habitat types yielding highest survivorship. Riech- ert (1973) demonstrated that a desert agelenid spider, Agelenopsis aperta (Gertsch) improved the quality of its web sites by successive relocation in better web sites. A. aperta walked rather than bal- looned to new web sites, however. Riechert and Tracy (1975) pro- duced a model of reproductive success that demonstrates a thirteen fold advantage in fecundity of spiders living in good versus poor web sites. Habitat selection and differential survival of Argiope spp. is an area worthy of additional research. One or more days may elapse after the spiderling locates a web site before it actually builds a web there. This is based on two sets of observations. First, spiderlings discovered in the field without webs and checked later the same day had not built webs (N = 5). Only spiderlings found early in the morning and thus not recently dispersed are considered here. Two of these individuals built webs one day later; one built two days later. Second, even during the height of Argiope trifasciata aerial dispersal (mid-May) few spider- lings were observed to have webs. Possibly the spiderlings were adjusting from a colonial, passive existence to one of active preda- tion. Whatever the reason(s), considerable mortality is suffered during this period (Tolbert, 1976). Spiderlings would certainly be 1977] Tolbert — Orb Weaving Spiders 25 vulnerable to predation, especially since they do not have the pro- tection a web affords (Tolbert, 1975). Some spiderlings probably starve during this time also (Tolbert, 1976). Spiderlings underwent a radical change in behavior toward con- specifics after dispersing from the communal tangle. As long as the spiderlings were on the communal tangle they were completely tolerant of one another and during hot sunny periods clustered very closely together. However, after a spiderling constructed an orb web it attacked and ate any prey that contacted the web. Litter- mates were invariably attacked and treated as prey items when they encountered or were placed on conspecific’s orb web. However, it is unnecessary for spiderlings of either Argiope species to engage in aerial dispersal before building an orb web. I have successfully reared both species in the laboratory after removing them from egg sacs. After building their first orb web and without the benefit of living on or building a communal tangle or engaging in aerial dispersal, Argiope spiderlings still attacked and killed littermates placed on their orb webs. Although Enders (1972) indicated he has observed unsuccessful ballooning attempts by mid-instar Argiope aurantia, my observa- tions indicate that only the emergent (first instar) disperses aerially. Summary Although both species are capable of all the behaviors in the ethogram (Figure 1), Argiope aurantia, possibly by virtue of their higher, more exposed egg sac locations, generally dispersed shortly after emergence from the egg sac. Argiope trifasciata produced communal tangles and engaged in preaeronautic behaviors on these structures before dispersing. In 1975, both spider species ballooned when their estimated body temperature exceeded 26° C, with most dispersal occurring when body temperatures were between 33° and 38° C. A vibrational stim- ulus occurring when spider body temperatures are above 26° C re- sulted in climbing behavior by the mass spiderlings on the com- munal tangle. Upon reaching the top of a promontory they either became airborne from that point or dropped on a dragline and ballooned from that position. Spiderlings ; ballooned at relatively low wind speeds (0.5 m/s). Multiple ballooning bouts were ob- served for some individuals. It is* hypothesised *1tet aerial dispersal serves to rapidly space spiderlings oveFiavailaWe' Habitats such that 26 Psyche [March overcrowding is minimized. Aerial dispersal would also allow suc- cessional species such as A. trifasciata and A. aurantia to colonize newly opened, ephemeral habitats. Acknowledgments I gratefully acknowledge the assistance of James Tanner, Gordon Burghardt, Charles Pless and David Etnier who ably served on my doctoral committee. I am especially grateful to Susan Riechert, my committee chairperson. I acknowledge the financial support of the National Science Foundation, Grant For Doctoral Disserta- tion Improvement (BMS 74-17602) and the Graduate Program in Ecology of the University of Tennessee. References Bristowe, W. S. 1939. The comity of spiders, Vol. I. Johnson Reprint Corp., New York. Doyle, R. W. 1975. Settlement of planktonic larvae: A theory of habitat selection in vary- ing environments. Amer. Natur. 109: 113-126. Duffey, E. 1956. Aerial dispersal in a known spider population. J. Anim. Ecol. 25: 85-111. Enders, F. 1972. Web-site selection by Argiope aurantia Lucas and other orb weaving spiders (Araneidae). Unpublished Ph.D. dissertation, N. C. State Uni- versity. Enders, F. 1973. Selection of habitat by the spider Argiope aurantia Lucas (Araneidae). Amer. Midi. Natur. 90(1): 47-55. Emerton, J. H. 1908. Autumn flights of spiders. Psyche 15: 121. Geiger, R. 1965. The climate near the ground. Harvard Univ. Press, Cambridge. Gertsch, W. J. 1949. American spiders. D. Van Nostrand Co., Inc., Princeton. Kaston, B. J. 1972. How to know the spiders, 2nd edition. W. C. Brown Co., Dubuque. McManus, M. L. 1973. The role of behavior in the dispersal of newly hatched Gypsy Moth Larvae. USDA Forest Service Research paper NE-267, lOp. Nishiki, S. 1966. On the aerial migration of spiders. Acta Arachnologica 20(1): 24-34 (Japanese with an English Summary). 1977] Tolbert — Orb Weaving Spiders 27 Richter, C. J. J. 1970. Aerial dispersal in relation to habitat in eight wolf spider species ( Par - dosa, Araneae, Lycosidae). Oecologia 5? 200-214. Richter, C. J. J. 1971. Some aspects of aerial dispersal in different populations of Wolf Spiders, with particular reference to Pardosa amentata (Araneae, Lycosidae). Misc. Pap. Landb. Hogesch. Wageningen 8: 77-88. Riechert, S. E. 1976. Web site selection in a desert spider, Agelenopsis aperta (Gertsch). Oikos 27: 311-315. Riechert, S. E. and R. Tracy. 1975. A model relating web-site characteristic to spider reproductive success. Ecology 56: 265-284. Tolbert, W. W. 1975. Predator avoidance behaviors and web defensive structures in the orb weavers Argiope aurantia and Argiope trifasciata (Araneae, Araneidae). Psyche 82: 29-52. Tolbert, W. W. 1976. Population dynamics of the orb weaving spiders Argiope trifasciata and Argiope aurantia (Araneae, Araneidae): Density changes associated with mortality, natality and migrations. Unpubl. Ph.D. dissertation, U. of Tennessee. Valerio, C. E. 1975. A unique case of mutalism. Amer. Natur. 109: 235-238. Van Wingerden, W. K. R. E. and H. F. Vugts. 1974. Factors influencing aeronautic behavior of spiders, Bull. Brit. Arach. Soc. 3: 6-10. ECOLOGY, ZOOGEOGRAPHY AND TAXONOMY OF THE LOWER RIO GRANDE VALLEY MESOSTENINES (HYMENOPTERA, ICHNEUMONIDAE) By Charles C. Porter1 Department of Biological Sciences, Fordham University Bronx, N.Y. 10458 Introduction This study analyzes results of five years’ fieldwork with net and Malaise Traps on mesostenine ichneumonids of semiarid subtropi- cal scrub and moist gallery woods habitats in the Lower Rio Grande Valley of south Texas. It lists 18 genera and 35 species. The genus Bicristella and the species Trachysphyrus mesorufus, Cryptanura lament aria, and Lymeon leucosoma are recorded for the first time from the United States. Cryptanura vallis Mesostenus opuntiae, Bricristella texana, Diapetimorpha sphenos, D. aspila, and D. pareia are described as new. The zoogeographic relationships, phaenol- ogy, and habitat preferences of each taxon are recorded and con- clusions are adduced as to distributional patterns, annual cycles, habitat selection, and diversity of the entire south Texas mesoste- nine fauna. The south Texas fauna also is compared with meso- stenine communities of other semiarid parts of the Neotropics, such as the Peruvian Coastal Desert and the northwest Argentine Sub- andino, and all these relict or marginal xerophilic faunas are dis- cussed with regard to their origin in wet forest centers of ichneu- monid radiation. Acknowledgements Major support for this research was provided in 1 976— ’77 by a National Science Foundation Grant (DEB 75-22426) and during 1973-75 by grants from the Committee for Research and Explora- tion of the National Geographic Society. Mr. David Riskind and Mr. Sim Oefinger, Jr. of the Texas Parks and Wildlife Department issued permits for insect collecting in the Bentsen Rio Grande Valley State Park. At Bentsen Park, Mr. Reynaldo Ortiz (Superintendent) 'Research Associate, Florida State Collection of Arthropods, Florida Department of Agriculture and Consumer Services, Gainesville, Florida 32602. Manuscript received by the editor, May 24, 1977 28 1977] Porter — Mesostenines 29 and Mr. Antonio Salinas (Park Ranger 3) maintained the Malaise Trap used in my 1976 survey and provided cordial assistance on all my visits to the park. Mrs. Vivian Thacker, as trustee of the Valley Botanical Garden, facilitated collecting in that small but important island of natural vegetation. My father, Mr. Carroll B. Porter, also assisted in the Malaise project and in many other ways. Mr. Charles W. Calmbacher of Fordham University prepared and labeled most of the Malaise samples from Bentsen Park. Finally, Dr. Henry K. Townes of the American Entomological Institute loaned several homotypes which helped resolve crucial taxonomic problems. Materials and Methods Hand collecting with a strong but light net obtained 63% of the 679 specimens captured between May 1973 and March 1977 for use in this study. Sweeping undergrowth yielded numerous mesoste- nines but many others were netted individually in flight from foli- age. Periods annually available for fieldwork included 25 August to 9 September, 18 December to 25 January, 11-21 March (1-8 April in 1975) and 16 May to 10 June. I was in the field 7 days a week and 6-8 hours per day during all visits to south Texas. To obtain a more comprehensive picture of mesostenine diversity than would have been possible by hand collecting alone, I employed two Malaise Traps during this research. The first was installed at the Valley Botanical Garden in a Celtis lindheimeri-C. pallida thicket and functioned from September 1973 until March 1974 but was stolen in April 1974. The second was set up under a large Pithecel- lobium flexicaule in deep woods near a lake at the Bentsen Rio Grande Valley State Park and, having already furnished a complete series of samples for 1976, continues to operate during 1977. In both traps, a pint mason jar filled with 70% isopropyl alcohol (commer- cial rubbing alcohol) was used as the collecting recipient. The trap at the Botanical Garden was changed once a month but I was able to arrange for twice monthly curating of the Bentsen Park trap. Both Malaise Traps were of the “light weight” variety, as perfected by Dr. Henry K. Townes (Townes, 1972, p. 239-247). The Study Area The Lower Rio Grande Valley is an alluvial plain that extends along the Rio Grande River for about 120 km. in Hidalgo and 30 Psyche [March Cameron counties of Texas and the Mexican state of Tamaulipas, beginning on the east at the Gulf of Mexico and ending approx- mately at the level of Mission, Texas on the west. Nowhere is this Valley much more than 25 or 30 km. wide either north or south of the river. It constitutes an island of fertile soil, relatively high humidity, and comparatively lush vegetation surrounded landward on all sides by desert scrub. Because of its latitude (26 degrees N.) and proximity to the Gulf, the Valley experiences an extremely mild temperature regimen. The average yearly maximum at Brownsville is 28 degrees C. and the minimum 18.3 degrees C. Summer highs rarely go above 40 degrees C. and the average daily range for July at Brownsville is 33.6 de- grees C. to 24.2 degrees C. On the other hand, most winters have only two or three frosts during which the temperature normally does not fall below -3 to -4 degrees C., although the record low for Brownsville is -11 degrees C. (registered in February 1899). The average daily range for January at Brownsville is 21.4 degrees C. to 11.2 degrees C. and such temperatures occur quite consistently throughout the Valley in winter, although from November to March warm periods frequently are interrupted by cold fronts that bring 4-10 day stretches of cloudy weather when the temperature stays between about 4 and 10 degrees C. Precipitation in the Valley is rather scant, averaging 669 mm. per year at Brownsville. It occurs in winter as protracted fine drizzle, in spring, summer, and fall as occasional thunderstorms, and sometimes in late summer and early fall as torrential inunda- tions that accompany inland-moving hurricanes. September, with an average of 124.8 mm. is the wettest month while March, with 26 mm., is the driest. Although long-term figures suggest fairly even rainfall distribution, there is actually great variation from month to month and from year to year. Protracted droughts are common but some years may have more than 1000 mm. of rain. Vegetation of the Valley ranges from desert scrub to humid sub- tropical woodlands best developed along the Rio Grande and in the vicinity of water holes. The south Texas flora resembles that which grows in many other semi-arid environments from Mexico to Argentina. Some of the more conspicuous angiosperm genera are Acacia, Bac charts, Bumelia, Celtis, Cercidium, Condalia, Ery- thrina, Opuntia, Parkinsonia, Prosopis, Salix, Tillandsia, and Xan~ thoxylum. This same element occurs also in the ecologically sim- 1977] Porter — Mesostenines 31 ilar Argentine Chaco at the austral extreme of the Neotropics. Almost all my fieldwork on Valley mesostenines was done in the 500 acre Bentsen Park near Mission and the 20 acre Valley Botan- ical Garden at McAllen. Otherwise, except for the Santa Ana Na- tional Wildlife Refuge near Alamo, most natural vegetation has been extirpated from the Valley and replaced by citrus groves, truck farms, sugar-cane fields and other agricultural systems. The Valley Botanical Garden is about 16 km. from the Rio Grande and thus lacks gallery forest and other really humid asso- ciations but offers a sample of scrub communities and moderately humid woods. Here the most abundant or conspicuous larger plants are: Acacia greggii (rare), A.farnesiana (common), Baccharis sp. (common), Bumelia celastrina (rare), Celtis lindheimeri (com- mon), C. pallida (common), Cercidium floridum (rare), Condalia obovata (common), C. obtusifolia (common), Ehretia anacua (com- mon), Forestiera texana (rare), Karwinskia humboldtiana (com- mon), Leucaena pulverulenta (scarce), Morus rubra (rare), Opuntia sp. (common), Parkinsonia aculeata (common), Phaulothamnus spinescens (rare), Pithecellobium flexicaule (rare), Porlieria angus- tifolia (rare), Prosopis juliflora (common), Sabal texana (rare), Salix nigra (common), Serjania sp. (common), and Xanthoxylum fagara (scarce). Within its 20 acres, the Botanical Garden contains several more or less distinct associations. Dry areas are dominated by Prosopis juliflora and Opuntia sp. with Condalia obtusifolia and Parkinsonia aculeata often common also. Moderately dry habitats have at least some of the foregoing species along with Condalia obovata, Celtis pallida, Baccharis sp., and Bumelia cel- astrina. Moderately humid sites usually are dominated by Celtis lindheimeri in the tree stratum, C. pallida in the shrub layer, and by Serjania vines at ground level. They also may contain the small tree Ehretia anacua, the large shrub Xanthoxylum fagara, and the small shrub Karwinskia humboldtiana. The large trees Leucaena pulverulenta and Salix nigra also occur in the Garden but only near ponds and irrigation canals. Finally, some abandoned farm- lands near the Garden support open Acacia farnesiana woods with a monotonous undergrowth of tall grasses. The Bentsen Park is much larger and floristically more varied than the Botanical Garden. Only Sabal texana occurs in the Gar- den but not at Bentsen, while Amyris texana (common), Fraxinus berlandieriana (common), Mimosa berlandieriana (moderately com- 32 Psyche [March mon), Sapindus drummondii (scarce), Tillandsia usneoides (com- mon), and Ulmus crassifolia (common) have been found exclusively in the State Park. In addition to all plant associations described for the Botanical Garden, Bentsen Park supports distinctive gallery forest and water hole communities. Lush woods along the Rio Grande contain Salix nigra, Fraxinus berlandieriana, Celtis lind- heimeri, Mimosa belandieriana, Acacia farnesiana and a profli- gate ground cover of Serjania vines. Even more luxuriant is the flora near a permanent water hole, which includes huge examples of Fraxinus, Ulmus, Leucaena, and Ehretia, some Pithecellobium and Sapindus, numerous Xanthoxylum, some Mimosa, abundant Amyris, and an impressive epiphyton of Tillandsia usneoides on many larger trees. Other dark, damp zones in Bentsen Park are dominated by Pithecellobium flexicaule. Climatically and floristically, the Valley thus emerges as de- cidedly subtropical and the same is true for most of its fauna, from ichneumonid wasps and diurnal Lepidoptera to reptiles and birds. Indeed, south Texas harbors the richest Neotropic biota of any part of the United States. The Tribe Mesostenini Mesostenines are one of the largest groups in the Family Ichneu- monidae and inhabit all continents, having radiated massively in both tropical and temperate regions. They parasitize the pupae of many Lepidoptera as well as of some Coleoptera, Neuroptera, Dip- tera, and certain Hymenoptera. Most species are taxonomically catholic in host selection, each one being attracted to diverse kinds of pupae in a restricted spatial niche (leaf rolls, ground litter, stems, tunnels in tree trunks, etc.) rather than choosing victims from among one particular genus or even family of insects. Like most ichneumonids, mesostenines prefer humid forest habi- tats, so that in the New World they are best represented in the North American Temperate Deciduous Forest and again in various kinds of Latin American subtropical and tropical wet forests. The comparatively dry Lower Rio Grande Valley, thus has a rather de- pauperate mesostenine fauna, whose relations are principally but not exclusively Neotropic. Listed below together with relevant ecological, zoogeographic and taxonomic data are the 18 genera and 35 species of Mesostenini so far recorded from the Lower Rio Grande Valley. 1977] Porter — Mesostenines 33 1 . Gambrus bituminosus Cushman SPECIMENS EXAMINED: 1 female, Bentsen Park, 31 XII 76. HABITAT: Weeds in sandy area at edge of field not far from Rio Grande. DISTRIBUTION: Mass., N.Y., N.J., 111., Minn., Ga., La., Cal., new for Texas. PHAENOLOGY: Summer in north, winter in south. 2. Gambrus ultimus (Cresson) SPECIMENS EXAMINED: 6 females, 2 males: BENTSEN PARK (Net: 2 females, 13 I 76; 1 female, 19 I 76; 1 female, 29 XII 76; Malaise: 1 male, 16 X 76); BOTANICAL GARDEN (Net: 1 female, 12-21 I 76; 1 female, 17-24 III 74; 1 male, 18 III 74). HABITAT: Serjania vine tangles under shade of Celtis lindheimeri and other large trees. DISTRIBUTION: Continental U.S. PHAENOLOGY: Flies in Valley from October to March with peak in January (4 of 8 collections). Active in north from April to October. 3. Trychosis subgracilis (Cresson) SPECIMENS EXAMINED: 1 female, 2 males: bentsen park (Net: 1 female, 23 I 76; Malaise: 2 males, 15-30 IV 76). HABITAT: Serjania vines in gallery woods beneath Celtis lind- heimeri and Salix nigra; entered trap beneath Pithecellobium flexi- caule. DISTRIBUTION: Eastern U.S.; first record for Valley. PHAENOLOGY: January to April in Valley; April to August in northern states. 4. Trachysphyrus mesorufus (Cushman) (Fig. 6) FEMALE: Color: scape black with a broad, nearly percurrent white bar below and brown on dorsal rim; pedicel black; flagel- lum black with a ventrally incomplete white band on segments 5-11; head and mesosoma black with white markings as follows: basal 2/3 of the otherwise somewhat brownish mandibles; blotch covering most of clypeus; most of face except for a large area be- 34 Psyche [March tween and below antennal sockets and a pair of submedian blotches above clypeus that are narrowly confluent mesad and which con- nect laterally with a large black area in anterior 1/4 of malar space; very broad orbital ring interrupted only in malar space and ventro- posteriorly much expanded and almost reaching hypostomal Ca- rina; blotch on apical 1/2 of propleuron; very broad anterior mar- gin of pronotum; very broad humeral margin of pronotum; pair of longitudinal blotches on about median 3/4 of mesoscutum in position of notauli; scutellum; most of postscutellum; tegula; axil- lary sclerites; subalarum; large anterio-median blotch on mesepi- sternum just behind prepectal carina; broad stripe in anterior 2/3 of sternaulus; large blotch in lower hind corner of mesepisternum; mesepimeron pure white on dorsal 1 /4 and more brownish ventrad; most of dorsal metapleuron; large, dorso-posterior blotch on apical 1/2 of lower metapleuron; and a pair of very broad blotches occu- pying all but median 1/3 of hind face of propodeum from cristae to apical margin; first gastric tergite red with a broad white band covering apical 1/2 of postpetiole; second tergite black with red- dish staining baso-laterally and a broad white subapical band; and following tergites black with broad white apical bands; fore and mid legs ferruginous with tarsus duller and fifth tarsomere dusky, trochanter white with brownish above, and coxa white with a small reddish spot above near apex and more broadly marked with dark red to blackish below; hind leg with coxa red except for a small white blotch above at base, trochanter and trochantellus red, femur more ferruginous, tibia dull ferruginous with a slight dusky tinge on base and blackish on about apical 1/10, first tarsomere brown- ish black with white briefly throughout on apex and whitish below on apical 1/2, second tarsomere white, third white with a dusky area above subapically, fourth black with a little whitish on base and fifth black; wings hyaline. Structurally, mesorufus much resembles the Floridian T. weemsi (Porter, 1974, p. 331-335), from which it may be distinguished by most of the characters listed below; Length of fore wing: 6. 1 mm. Pronotum: dorsal margin moder- ately swollen. Mesoscutum: notauli very weak but traceable about 2/3 the length of mesoscutum. Mesopleuron: speculum swollen, smooth and shining with only a few large punctures peripherally; surface otherwise almost uniformly with strong, reticulate wrink- 1977] Porter — Mesostenines 35 ling which obscures its punctures. Wing venation: radial cell 3.3 as long as wide; second abscissa of radius 0.7 as long as first inter- cubitus; disco-cubitus broadly angled with a long and conspicuous ramellus at angulation; upper part of nervellus 3.5 as long as lower. First gastric tergite: post-petiole 1.7 as wide apically as long from spiracle to apex. Second tergite: a little duller and more densely punctate than in weemsi. Ovipositor: sheathed portion 0.34 as long as fore wing; nodus distinct, with a very shallow and broad notch; dorsal valve on tip with a gradual, straight taper between notch and apex; tip 0.17 as high at notch as long from notch to apex. MALE: Unknown. SPECIMENS EXAMINED: 1 female. Botanical Garden, 2 April 1975. DISCUSSION: As noted above, mesorufus closely resembles the Floridian T. weemsi except in color pattern and in some subtle structural characters, whose real value only will be established when more specimens of these elusive ichneumonids are obtained. Townes (1962, p. 256-269) considers all North American repre- sentatives of this group as subspecies of T. planosae. In view of their allopatry and marked differences, however, I prefer to regard them as species, pending proof of intergradation. The Texas specimen was swept from a thorny bush (probably Celtis pallida ) in a dry area of the Botanical Garden dominated by Prosopis juliflora with Condalia obovata and Celtis pallida in the shrub stratum. The above described female is the third known specimen of meso- rufus and the first from the United States. Otherwise, this species inhabits Mexico whence it is recorded by Cushman (1930, p. 2) from Cuernavaca in Morelos state and by Townes (1962, p. 259) from “40 km. southwest of Puebla” in Puebla state. 5. Joppidium brochum Townes SPECIMENS EXAMINED: 1 female, Botanical Garden, 5 I ’76. habitat: Herbaceous undergrowth on shady side of fence row with Celtis pallida, C. lindheimeri, Ehretia anacua and other trees. DISTRIBUTION: Ky., N.C., Ga., to Okla. and Tex. and into Mex- ico at least as far as Veracruz and Mexico City. PHAENOLOGY: Valley record for January; otherwise flies mostly in May and June. 36 Psyche [March 6. Joppidium rubriceps Cresson SPECIMENS EXAMINED: 1 female, Botanical Garden, 1 IV ’75. HABITAT: Herbs, grasses, and pink-flowered verbenas in bright sun. DISTRIBUTION: N. J. to south Texas. PHAENOLOGY: Flies from mid March to early November, appear- ing first and disappearing latest in southern parts of its range. 7. Lanugo picta Townes SPECIMENS EXAMINED: 13 females, 24 males: BOTANICAL GAR- DEN (Net: 4 females, 8 males, 1-15 1 ’75; 4 females, 8 males, 16-26 I ’75; 1 female, 4 IV ’75; 1 female, 1 male, 20-31 XII ’73; 1 female, 5 males, 24-30 XII ’74; Malaise: 1 female. III ’74; 1 female, 2 males, XII ’74). HABITAT: Open and semi-shaded areas; old fields, hedge rows, woods edges; tall grass at edge of thicket dominated by Celtis lindheimeri and C. pallida; a few specimens in Celtis thicket. DISTRIBUTION: South Texas to northern Arizona and as far south in Mexico as Chiapas. PHAENOLOGY: Invernal, Valley records include 3 females and 8 males, for December, 8 females and 16 males for January, 1 fe- male for March and 1 female for April. Flies all summer in moun- tainous parts of west Texas, Arizona and Mexico. Varies in abundance from year to year: 3 specimens in ’73-74, 34 in 74-75, none in 75-76 or so far in 76-77. 8. Compsocryptus texensis Townes SPECIMENS EXAMINED: 26 females, 6 males: bentsen park (Net: 1 male, 12-20 III 77; 2 females, 29-30 XII 76); BOTANICAL GAR- DEN (Net: 6 females, 5-26 I 75; 2 females, 3 males, 28-30 III 75; 5 females, 1 male, 2-5 IV 75; 1 male, 16-30 V 74; 1 female, 19 XII 76; 8 females, 20-28 XII 74; 2 females, 28-30 XII 73). HABITAT: Open, dry areas; fields, hedge rows; short grass and low herbs of incipient secondary succession; herbage of poorly tended orange groves; lawns. DISTRIBUTION: Ka. to Okla. and Tex. south into N. Leon and Tamaulipas of Mexico. 1977] Porter — Mesostenines 37 PHAENOLOGY: Flies from December to May with peak between December and April (13 females in December, 6 males in January, 2 females and 4 males in March, 5 females and 1 male in April, and 1 male in May). Varies in abundance from year to year: 3 specimens in ’73-’74, 25 in ’74-75, none in ’75-76, and 4 so far in ’76-77. Genus Cryptanura The Valley has three Cryptanura, of which one is new and one is here recorded for the first time from the United States. Key to the U. S. Cryptanura (Females only) 1. Second gastric tergite mostly mat; clypeus more or less strongly convex in profile; humeral margin of pronotum not conically produced anteriorly, but often with a carinate elevation above end of epomia; sublateral white stripe of propodeum strongly narrowed basad of crista 2 Second tergite polished; clypeus nasute; humeral margin of pro- notum anteriorly with a prominent subconical to conical ex- pansion; sublateral white stripe of propodeum not narrowed basad of crista 4 2. Epomia not reaching humeral margin of pronotum, the humeral margin not carinate or tuberculate anteriorly; second gastric tergite with a medio-basal white spot C. septentrionalis Cushman Epomia forms a carinate elevation on humeral margin of pro- notum anteriorly; second tergite at most narrowly tinged with whitish medio-basally 3 3. Hind coxa white with conspicuous black markings; femora yel- lowish white with a broad, percurrent dorsal black band; lower metapleuron with coarse oblique wrinkling that be- comes irregular only on about dorsal 1/4 and with at most obscure intercalated punctures; propodeal dorsum behind basal trans-carina coarsely and irregularly wrinkled and puncto-reticulate but without discrete punctures 11. C. lament aria (Cameron). 38 Psyche [March Hind coxa mostly fulvous with whitish above; femora uniformly pale fulvous; lower metapleuron coarsely and densely punc- tate to puncto-reticulate with numerous discrete punctures dorso-anteriad; propodeal dorsum behind basal trans-carina with abundant, mostly discrete coarse punctures, grading into puncto-reticulation only latero-apicad near cristae C. banchiformis (Megerle). 4. Many segments in apical 1/3 of flagellum up to 1.4 as wide as long; first flagellomere 5.1 as long as deep at apex; malar space 0.88 as long as basal width of mandible; frontal horns on a high common base; mesoscutum with abundant large punctures but with no wrinkling along notauli, except api- cad, or on outer margins of lateral lobes; mesoscutum with a large, subcircular median white spot 9. C. compacta (Cresson). Segments in apical 1/3 of flagellum averaging about as wide as long; first flagellomere 7.9 as long as deep at apex; malar space 0.46 as long as basal width of mandible; frontal horns on a very low common base, long and sharp; mesoscutum with moderately numerous medium sized punctures that be- come sparser mesad on lobes and with extensive transverse wrinkling all along notauli and on outer margins of lateral lobes; mesoscutum without a median white spot 10. C. val/is n. sp. 9. Cryptanura compacta (Cresson) SPECIMENS EXAMINED: 1 female, Bentsen Park, 14 III ’77. HABITAT: Clearing with Serjania vines and tall grass on bank of Rio Grande in Salix nigra- Celtis lindheimeri woods. DISTRIBUTION: Southern Texas to Honduras. PHAENOLOGY: Townes (1962, p. 429-30) records a female of compacta from “Cameron County, Texas, 3 August 1928”. 10. Cryptanura vallis n. sp. (Fig. 8) Holotype: female, USA {Texas: Hidalgo County, Bentsen Rio Grande Valley State Park, 27 XII ’76, C. C. Porter). (Townes). 1977] Porter — Mesostenines 39 FEMALE: Color: antenna black with white annulus on flagello- meres 5-12; palpi whitish; mandible white with black on apical 1/3 and narrowly on dorsal and ventral margins except near base; head white with black as follows: broad, irregular mark extending from a little above anterior tentorial pit ventrad to mandibular condyle; face for a short distance below and between antennal sockets; lower and inner margins of antennal sockets; median half of face and ver- tex; occiput much more broadly; most of postocciput; about upper 1/4 of temple very broadly, briefly interrupting white orbital ring; and rest of temple very narrowly along occipital carina; propleuron white with black on most of basal 1/4; pronotum black with a broad white band covering most of front margin, except for lower hind corner and a short break dorso-medially, and with white broadly on the swollen humeral margins; thoracic dorsum black with white on prescutellar ridge, broad anterio-lateral margins and most of apical 1/2 of scutellum, most of postscutellum, and on hind margins of meso and metanotal axillary troughs; tegula white; mesopleuron black on most of prepectus and on most of upper 1 / 3 except for the white subalarum, otherwise wholly white, becoming slightly brownish ventrad, on mesepisternum and mesepimeron; mesosternum black on prepectus but otherwise slightly brownish white; most of dorsal metapleuron white; lower metapleuron white with a pale brownish tinge and briefly stained with darker brown on apex; propodeum white to brownish white with black on a broad, irregular percurrent median longitudinal band which is widest along basal trans-carina and becomes narrower rearward, especially on apical face, as well as at least narrowly blackish throughout along basal carina, blackish around spiracle, and with pale brown staining on much of lateral face between spiracle and crista; first gastric tergite yellowish white with dark brown above on much of apical half of petiole and on most of basal 2/3 of post- petiole as well as a little brownish along ventro-lateral carina; sec- ond tergite black with yellowish white on apical 1/3, on very broad lateral margins, and on entire thyridial areas, between which the ground color becomes brownish yellow basad; third tergite black with yellowish white on apical 1/2 and on very broad lateral mar- gins; fourth and fifth similar but even more broadly yellowish with some darker staining in the yellow zones; sixth and seventh yellow- ish with some darker staining; and eighth dark brown with apex narrowly yellow; fore leg with coxa white, trochanter whitish with 40 Psyche [March Fig. 1. Cryptanura lamentaria, female. Bentsen Park. Side view of hind coxa, trochanters, and femur, showing color pattern. Fig. 2. Mesostenus opuntiae, female holotype. Ovipositor tip. Fig. 3. Lymeon leucosoma, female. Valley Botanical Garden. Fore wing. Fig. 4. Diapetimorpha aspila, male holotype. Dorsal view of propodeum. Fig. 5. Bicristella taxana, female holotype. Dorsal view of propodeum. Fig. 6. Trachysphyrus mesorufus, female. Valley Botanical Garden. Dorsal view of gaster, showing color pattern. Fig. 7. Diapetimorpha pareia, male holotype. Dorsal view of propodeum and first gastric segment. Fig. 8. Cryptanura vallis, fe- male holotype. Dorsal view of propodeum. Fig. 9. Diapetimorpha sphenos, female paratype. Dorsal view of propodeum. 1977] Porter — Mesostenines 41 a brown blotch on basal 2/3 above, trochantellus whitish with some brown staining, femur yellowish white below and mostly brownish above, tibia yellowish white, and tarsus dull yellowish with slight dusky staining toward apex on first segment, more broadly dusky on second and third segments, and blackish almost throughout on fourth and fifth; mid leg similar to fore leg but coxa more dully white with some faint dusky staining, trochanter and trochantellus more broadly brown above, and tarsus mostly dusky to black with yellowish only near base of segments 1-3; hind leg with coxa dull white with a broad but diffuse, nearly percurrent, moderately pale brownish dorsal stripe and with paler brownish staining within and below; trochanter yellowish white with dark brown staining dorso-anteriorly and tinged with paler brown be- hind; trochantellus similar to trochanter but with darker and more extensive brown areas in front and behind; femur dull brownish yellow, becoming darker brown above and brighter yellow behind; tibia yellow with a little dusky on base; and tarsus yellow with dusky only on apical 1/2 of fifth segment; wings hyaline, stigma yellowish with dark brown on broad peripheries. Length of fore wing: 8.7 mm. Flagellum: scarcely flattened be- low on apical 1/3, segments between white annulus and apex av- eraging as wide as long; first segment 7.9 as long as deep at apex. Temple: 0.3 as long as eye at upper 1/3; very strongly and directly receding. Front: horns on a very low common base, large and sharply conical; surface practically without wrinkles between level of horns and anterior ocellus. Clypeus: nasute; apical margin trun- cate. Pronotum: epomia sharp in scrobe but ending above in a large tubercle, so that the otherwise gently swollen dorsal margin of pronotum has a moderately prominent subconical projection at this point. Mesoscutum: shining with numerous medium sized punctures that become sparser centrad on lobes and rather broadly, more or less transversely wrinkled interiorly and exteriorly along notauli and on outer margins of lateral lobes as well as with an area of coarse, irregular wrinkling between notauli toward their terminus; notauli moderately impressed, reaching about 4/5 the length of mesoscutum. Scutellum: weakly convex. Mesopleuron: prepectus with a very short ridge opposite lower hind corner of pronotum; surface between prepectal carina and speculum mostly with strong, nearly regular longitudinal wrinkling which becomes weaker, and mingled with large, obscure punctures, on lower half, 42 Psyche [March where there are also some smooth areas. Lower metapleuron: with uniform, coarsely reticulate wrinkling; juxta-coxal carina not de- fined. Front tibia: moderately inflated. Flind trochantellus: 0.57 as long as its trochanter in dorsal view. Third hind tarsomere: ventrally with about 8-9 strong spines (in addition to the apical group) which are not arranged in regular longitudinal rows. Pro- podeum: spiracle 1.5 as long as wide; cristae large and strongly projecting short ligulate, about 1.1 as long as wide at base, apical carina weakly defined and gently arched forward between them; dorsal face behind basal trans-carina with rather coarse, irregular but not much reticulate wrinkling; the apical face centrally with strong longitudinal wrinkling, becoming smooth sublaterally and then transversely wrinkled laterad. First gastric ter git e: postpetiole 1.4 as wide at apex as long from spiracle to apex; gently arched in profile. Second gastric tergite: smooth and highly polished with a few tiny, very sparse punctures. Ovipositor: sheathed portion 0.43 as long as fore wing; tip 0.20 as high at nodus as long from nodus to apex, weakly sagittate with a direct taper between nodus and apex. MALE: Unknown. TYPE: In collection of Dr. Henry K. Townes, 5950 Warren Rd., Ann Arbor, Michigan, 48105. RELATIONSHIPS: The polished second gastric tergite, nasute cly- peus, and rather strong subconic projection anteriorly on the hu- meral margin of the pronotum suggest affinity with C. compacta, to which vallis runs in Townes’ key to the North American Crypt- anura (1962, p. 427). However, compacta differs from vallis in many chromatic and structural features of which those not already mentioned in my foregoing key are summarized below: Mesopleuron black on lower 2/3 except for a broad, oblique white area extending from prepectal carina to lower hind corner but not invading speculum or approaching mesopleural suture ex- cept far below; propodeum black with a very broad white stripe on each side reaching rearward from basal trans-carina across crista to hind margin; fore and mid femora broadly black above; hind coxa pure white with an almost percurrent broad black stripe dorsally and with other black markings; hind femur black with yellow anteriorly and posteriorly; dorsal margin of pronotum very strongly swollen and produced anteriorly into an exceptionally prominent broadly conical projection; prepectus opposite lower 1977] Porter — Mesostenines 43 hind corner of pronotum with a strong ridge that extends ventrad about 3/4 the distance to prepectal carina; mesopleuron behind prepectal carina with coarser and more oblique wrinkling than in vallis; front tibia a little more strongly inflated than in vallis; pro- podeal spiracle 2.0 as long as wide; propodeal cristae a little more narrowly ligulate than in vallis , 1.4 as long as wide at base; apical trans-carina completely absent between cristae; dorsal face of pro- podeum behind trans-carina with more regularly reticulate wrink- ling than in vallis ; apical face of propodeum with coarse reticulate wrinkling that becomes transversely biased laterad; first gastric tergite more strongly arched in profile than in vallis; postpetiole 1.6 as wide at apex as long from spiracle to apex. FIELD NOTES: Swept from Serjania vines near Rio Grande in shade of Celtis lindheimeri , Salix nigra , Fraxinus berlandieriana and other trees. SPECIFIC NAME: Vallis is the genitive of the Latin noun valles or “valley”. 1 1. Cryptanura lamentaria (Cameron) (Fig. 1) FEMALE: Color: antenna black with a broad white stripe below on scape and a white annulus on flagellomeres 6-11; palpi white with apical segment of each blackish; mandible mostly black with a large white blotch on base; clypeus white with black on median 2/3 and pale brown on lateral 1/3 of apical margin as well as broadly black on lateral and dorso-lateral margins around and mesad of anterior tentorial pits; white on most of face, cheek, and a broad orbital ring which is narrowed and briefly interrupted on upper 1/4 of temple; head otherwise with black on median half of front and vertex and more broadly on occiput and postocciput as well as increasingly narrowly black ventrad along occipital carina to about dorsal 1/2 of temple; mesosoma black with profuse white markings as follows: about apical 3/4 of propleuron; broad front margin of pronotum, ending about 4/5 the distance ventrad to lower hind corner and also enclosing a pair of small, dorso-lateral black spots; all but about median 1/6 of dorsal margin of prono- tum very broadly; large median spot on mesoscutum between apices of notauli; prescutellar ridge; most of scutellum except for an anterio-median black spot; most of post-scutellum; hind rims of 44 Psyche [March meso and metanotal axillary troughs; tegula; subalarum; very broad oblique band across mesopleuron from mid-height of front margin to lower hind corner; about upper 1/3 of mesepimeron; a pair of very large blotches covering much of mesosternum on each side of median groove and confluent posterio-dorsad with white meso- pleural band; most of dorsal metapleuron; about dorso-posterior 3/4 of lower metapleuron; and a pair of broad propodeal stripes which extend dorsad from hind margin to include cristae and then reach forward, a little more narrowly, almost to basal trans-carina; first gastric tergite black with yellowish white dorsally on much of petiole and on about apical 1/2 of postpetiole as well as laterally toward apex of petiole and on most of postpetiole; second tergite black with a brown tinged yellowish white band on base, abruptly widened sublaterally to cover thyridia, as well as broadly yellowish white laterally and on apical 1/3; third and fourth tergites black with very broad apical and lateral yellowish white bands; fifth and sixth mostly yellowish white grading into black dorsad; seventh similar to preceding but more broadly black dorsally; eighth and ninth black with yellowish white laterally and on apical rims; fore and mid coxae white with black on most of apical 1/2 posteriorly; hind coxa yellowish white with a broad, percurrent anterio-dorsal black band and a similar but premedially interrupted posterio- dorsal band as well as blackish on extreme base ventrally; tro- chanters and trochantelli whitish with considerable black above; femora yellowish white with a broad, percurrent dorsal black band; tibiae yellow with some dusky staining posterio-basally on front and mid tibiae and a little more broadly blackish on base of hind tibia; fore and mid tarsi with first segment yellow with a little dusky staining apicad and succeeding segments blackish except narrowly yellow on base of second; hind tarsus with segments 1-4 yellow and 5 mostly black except grading into brownish on base; and wings hyaline with stigma black. Length of fore wing: 10.7 mm. Flagellum: definitely flattened below on apical 1/3, its widest segments 1.5 as wide as long, the first segment 5.7 as long as deep at apex. Malar space: 0.71 as long as basal width of mandible. Temple: 0.4 as long as eye at upper 1/3; strongly and directly receding. Front: horns stout and broad, not on a common base; strongly wrinkled between level of horns and anterior ocellus. Clypeus: strongly and a little asym- metrically convex in profile; apical margin slightly convex. Pro- 1977] Porter — Mesostenines 45 notum: epomia strong in scrobe and reaching far dorsad onto the moderately swollen humeral margin of pronotum, where it forms a carinate elevation. Mesoscutum: shining with abundant coarse, medium-sized punctures which are mostly subadjacent to a little sparser; notauli sharp and narrow, reaching about 2/3 the length of mesoscutum. Scutellum: gently convex. Mesopleuron: prepectus opposite lower hind corner of pronotum with a long ridge that ex- tends 2/3 or more the distance veritrad to prepectal carina; surface between prepectal carina and speculum with strong oblique wrink- ling that on lower 1/2 becomes only gradually a little weaker and mingled with large but mostly obscure punctures; mesopleural su- ture grossly foveolate on lower 1/2. Lower metapleuron: with coarse oblique wrinkling that becomes more irregular dorsad and has only obscure intercalated punctures; juxta-coxal carina trace- able for about basal 0.5 of metapleuron. Front tibia: moderately inflated. Hind trochantellus: 0.36 as long as trochanter in dorsal view. Third hind tarsomere: in addition to apical spines with four longitudinal rows of strong spines numbering about 15 in all. Pro- podeum: spiracle 1.9 as long as wide; cristae strongly projecting ligulate, about 2.0 as long as wide at base, the apical trans-carina absent between them; dorsal face behind basal trans-carina with strong irregular wrinkling and puncto-reticulation but without dis- crete punctures; apical face almost uniformly with strong and quite regular transverse wrinkling. First gastric tergite : postpetiole 1.5 as wide at apex as long from spiracle to apex; in profile pyramidally elevated above spiracle. Second gastric tergite: mostly mat with fine micro-reticulation and on about basal 2/3 with numerous shal- low, medium-sized punctures separated in general by about 1. 0-2.0 their diameters. Ovipositor: sheathed portion 0.46 as long as fore wing; tip 0.2 as high at nodus as long from nodus to apex, its pro- file between nodus and apex only slightly convex. MALE: Unknown. specimens EXAMINED: 2 females, USA {Texas: Hidalgo County, Bentsen Rio Grande Valley State Park, 29 XII ’76, C. C. Porter); PANAMA {Chiriqui: Valley of the Clouds, 17 III ’60, K. W. Brown). (Porter, Townes). DISCUSSION: This is the first record of lamentaria for the United States. The species has been cited previously only from Costa Rica, Guatemala, and Panama. A homotype from Panama, loaned by H. K. Townes, is un- 46 Psyche [March doubtedly conspecific with the Texas specimen, differing only as follows: No white on scape; apical segments of palpi only slightly dusky; white band on anterior margin of pronotum briefly interrupted medially; all of lower metapleuron white; gastric tergites 4-9 more broadly white; posterio-dorsal black band of hind coxa percurrent; first flagellomere 6.3 as long as deep at apex; malar space 0.83 as long as basal width of mandible; ridge on lower prepectus extend- ing about 2/3 the distance ventrad to prepectal carina; hind tro- chantellus 0.47 as long as its trochanter in dorsal view; punctures of second gastric tergite sparser, mostly separated by more than 2.0 their diameters. Among its relatives, lamentaria most resembles C. banchiformis of the eastern United States and the two nearly replace one another geographically since banchiformis ranges down to San Antonio, Texas within less than 500 km. of the Valley. Lamentaria differs from banchiformis mainly in its black and white (instead of mostly fulvous) femora and in having the lower metapleuron and propo- deal dorsum strongly wrinkled but with at most obscure intercalated punctures (instead of with numerous discrete punctures). The two species probably stem from a common ancestor which, during warmer and wetter Tertiary times, ranged uniformly from Mexico up along the Gulf arc into eastern United States but then was frag- mented by Pleistocene glacial maxima into southeastern (Florida) and southwestern (Mexico) isolates, which have practically reestab- lished contact during the present moderately warm interglacial. The unique Texas female was netted as it flew about a tangle of Serjania vines in gallery woods along the Rio Grande. This is the same habitat and same general area where C. compacta and C. vallis also were collected. Genus Mesostenus 12. Mesostenus gracilis Cresson SPECIMENS EXAMINED: 6 females, BENTSEN PARK ( Malaise : 1 fe- male, 1-15 V’76); BOTANICAL GARDEN {Net: 1 female, 16-30 V’74; Malaise: 4 females, III ’74). HABITAT: Herbaceous undergrowth in Celtis lindheimeri-C. pal- lida woods. DISTRIBUTION: U.S. and northern Mexico. 1977] Porter — Mesostenines 47 PHAENOLOGY: March to May in Valley; flies from March to November elsewhere in southern part of its range and from late May to mid-October farther north. 13. Mesostenus opuntiae n. sp. (Fig. 2) Holotype: female, USA (Texas: Hidalgo County, Valley Bo- tanical Garden at McAllen, 10 I ’76, C. C. Porter). (Townes). FEMALE: Color: scape black with a little dull brown below; pedi- cel black with dull brown on apex; flagellum black with a little pale brown on base of first segment and a ventrally interrupted white band on segments 6-11; palpi brownish white; mandible white on basal half with apical half grading through pale brown into black on teeth; head mostly white on clypeus, cheek, face, and on a broad, uninterrupted orbital band, which progressively widens rearward and below to cover much of temple, as well as with pale brownish on mandibular condyles, hypostomal carina, medio-dor- sal margin of clypeus, slightly on apical face and apical margin of clypeus, and on antennal sockets, and with black broadly and ir- regularly along lateral and dorso-lateral margin of clypeus, irregu- larly around antennal sockets below, on about median half of front and vertex, more broadly on occiput and post-occiput, and then increasingly more narrowly ventrad on temple along occipital carina to about its lower 1/5; pronotum black basally with fulvous on apical half that grades into dull white on apical margin; pronotum black with a broad white band on most of front margin and on all but about median 1 / 5 of dorsal margin; mesoscutum black with a median white blotch located between apices of notauli and with white on pre-scutellar ridge; scutellum black, broadly margined with white laterally and behind; postscutellum mostly white; meso and metanotal axillary troughs black with hind rims narrowly whitish; mesosoma otherwise fulvous with white on tegula, subalarum, broadly on about upper 4/5 of front margin of mesepisternum, toward dorsum of mesepimeron, dully in lower hind corner of mesepisternum, and dully on apex of lower metapleuron, as well as with black above, below and behind subalarum, on most of prepectus, on most of apex of metasternum in front of mid coxae, in most of groove at base of propodeum, irregularly ventrad on front margin of lower metapleuron, rather irregularly on submeta- 48 Psyche [March pleural carina and on most of metasternum; gaster fulvous with some faint and diffuse dusky staining; legs fulvous with tibiae and tarsi duller and fifth tarsomeres dusky, front coxa extensively white above and in front with a large dark brown dorsal blotch enclosed by the white, front trochanter white tinged dorso-anteriorly and with slight dusky staining above; mid coxa with a large white blotch on basal half anterio-dorsally, mid trochanter brownish stained above; and with a little dusky on apex of hind trochantellus; wings hyaline with stigma pale brown. Length of fore wing: 6.3 mm. First flagellomere: 5.0 as long as deep at apex. Clvpeus: small, in profile rather strongly and a little asymmetrically convex, its apical margin slightly convex. Malar space: 0.85 as long as basal width of mandible. Temple: at its upper 1/3 about 0.34 as long as eye in lateral view; strongly re- ceding and gently convex. Mesoscutum: smooth and shining with abundant, moderately small, sharp punctures (coarser and denser in thoracicus). Mesopleuron: prepectal carina sharp to about lower 0.2 of hind margin of pronotum and then becoming obsolete in a vertically elliptic, slightly raised white callus. Wing venation: areo- let 1.3 as wide as high at apex; nervulus interstitial. Hind trochan- tellus: 0.20 as long as hind trochanter in dorsal view. Propodeum: elongate and rather strongly sloping rearward with little disconti- nuity between basal and apical face, apical face 0.7 as long as basal; apical trans-carina absent medially, laterally forming very low, broad, weakly oblique subcrescentic cristae; surface strongly and densely punctate with some intercalated wrinkling, especially rear- ward and a little more finely and sparsely punctate basad of basal trans-carina. First gastric segment: postpetiole 0.8 1 as wide apically as long from spiracle to apex. Second gastric tergite: mat with very fine but well developed micro-reticulation and abundant small, shallow punctures separated mostly by a little more or a little less than their diameters. Ovipositor: sheathed portion 0.83 as long as fore wing; tip 0.7 as long from nodus to apex as deep at nodus. MALE: Unknown. TYPE: In collection of Henry K. Townes, 5950 Warren Rd., Ann Arbor, Michigan, 48105. RELATIONSHIPS: This species resembles M. sicarius Townes (1962, p. 448-450), especially because of its short prepectal carina which ends dorsally in a white callus, but may be distinguished by the characters summarized in the following key: 1977] Porter — Mesostenines 49 1. Ovipositor tip 0.7 as long from nodus to apex as deep at nodus; propodeal cristae weakly oblique; first flagellomere 5.0 as long as deep at apex; postpetiole 0.83 as wide apically as long from spiracle to apex; propodeal punctation definitely sparser basad of basal trans-carina; second gastric tergite mat with small punctures separated in general by a little more to some- what less than their diameters M. opuntiae n. sp. Ovipositor tip 1 1.5 as long from nodus to apex as deep at nodus; propodeal cristae strongly oblique; first flagellomere 4.2 as long as deep at apex; postpetiole 0.95 as wide apically as long from spiracle to apex; propodeal punctation hardly sparser basad of basal trans-carina; second gastric tergite more shin- ing with small punctures separated mostly by about 1.5 their diameters M. sicarius Townes. FIELD NOTES: Netted in arid Prosopis-Opuntia association. SPECIFIC NAME: from the genitive singular of Opuntia. 14. Mesostenus longicaudis Cresson SPECIMENS EXAMINED: 5 females, 7 males: BENTSEN PARK (Net: 1 female, 12-20 III ’77; Malaise: 1 male, 16-31 V ’76); BOTANICAL garden (Net: 1 male, 12-20 III ’77; 2 females, 15 III ’76; 1 female, 1 IV ’75; 2 males, 27 VIII ’76; 1 female, 3 males, 3-9 IX ’76). HABITAT: Herbaceous growth in abandoned orange groves; amid pink verbenas on an otherwise well-cut lawn; weedy areas at edge of woods; rarely strays into deep woods; flies in full sunlight. PHAENOLOGY: Peaks in spring and late summer. Seems absent in winter and June-July. Valley records include 4 specimens for March, 1 for April, 1 for May, 2 for August, and 4 for September. Flies from mid-spring to mid-fall in most of south but does not appear before early July in north. DISTRIBUTION: Most of U.S. and Mexico. Genus Bicristella This is the first record of Bicristella from the United States. 15. Bicristella texana n. sp. (Fig. 5) Holotype: female, USA (Texas: Hidalgo County, Bentsen Rio Grande Valley State Park, 29 XII ’76, C. C. Porter). (Townes). 50 Psyche [March FEMALE: Color: antenna black with a white annulus on flagello- meres 5-1 1 and a little brownish toward apex below; maxillary pal- pus white with a little brownish on apical segment; labial palpus white with dusky toward apex of penultimate and on all of apical segment; mandible white with apical 1/4 dark brown; head white with black or blackish markings as follows: narrow apical margin of clypeus; rather broad ventral margin of anterior 1/2 of malar space; weak staining on mandibular condyle; some staining around anterior tentorial pit; narrow median line on about upper 1/3 of face; antennal sockets largely; a little more than median 1/2 of front and vertex; occiput more broadly; most of post-occiput; and temple, increasingly more narrowly, ventrad along occipital carina to about its upper 0.4; propleuron white; pronotum black with front margin very broadly white, except for a narrow dorso-median in- terruption, and all but about apical 1/4 of humeral margin broadly white; thoracic dorsum black with white on a large callus-like area on lateral lobe of mesoscutum above tegula, on prescutellar ridges and broad lateral and very broad apical margins of scutellum, on post-scutellum, and on apical rims of meso and metanotal axillary troughs; tegula white internally grading marginally into blackish; mesosternum and mesopleuron white with weak testaceous suffu- sion on all but about upper 1/3 of mesopleuron, as well as with black above subalarm, obliquely between apex of subalarum and dorsal margin of speculum and in subspecular depression; upper metapleuron almost wholly white; lower metapleuron white with a faint testaceous tinge; part of propodeum basad of basal trans- carina black with slightly testaceous white on much of lateral 1/5, except for black margining spiracle behind, as well as more nar- rowly white all along basal trans-carina and finely whitish on me- dian longitudinal carinae; part of propodeum behind basal trans- carina also slightly testaceous white with an almost percurrent black stripe between median longitudinal carinae and a black stripe along all of pleural carina except near apex; first gastric segment white with a large black area on apex of petiole and about basal 2/3 of postpetiole, narrowly black on apex of post-petiole, and with a broad, almost percurrent black stripe ventro-laterally; second ter- gite black with a broad subapical white band, even more broadly white laterally, and with a transverse white blotch medially at about basal 1/5; third tergite like second but without a sub-basal white blotch; following tergites similar to third but with the subapical 1977] Porter — Mesostenines 51 white band increasingly narrower mesad and broadly interrupted medially on sixth and following; fore leg pale testaceous with coxa white, except for a brownish streak dorso-basally, and tarsomeres 2-5 mostly dusky; mid leg similar to fore leg except for weak testa- ceous staining on coxa; hind leg pale testaceous with coxa grading into white near base and with a broad, percurrent, weakly con- trasting pale brownish stripe on its dorsum and with tarsus yellow except for blackish on the last segment; wings hyaline with stigma whitish grading marginally into pale brown. Length of fore wing: 1 A mm. Flagellum: scarcely flattened be- low toward apex; first segment 6.4 as long as deep at apex. Front: horn large, stout and conical, situated a little below center of front. Clypeus: bluntly nasute; apical margin convex. Malar space: 0.75 as long as basal width of mandible. Occipital earina: fine and sharp, joining the moderately raised hypostomal earina below in a weak depression at a distance above base of mandible equal to about 1/2 basal width of mandible. Temple: strongly and directly receding; 0.21 as long as eye at upper 1/3. Pronotum: humeral margin strongly swollen, evenly rounded and not especially prominent at anterior end; epomia strong throughout in scrobe but not pro- longed dorsad or ventrad; anterior margin bluntly angulate below middle. Mesoscutum: smooth and polished with a few sparse punc- tures; notauli sharply impressed and reaching about 2/3 the length of mesoscutum; lateral lobe opposite tegula with a large, gently raised, nearly circular white callus which is set off internally by a longitudinal impression. Mesopleuron: subalarum swollen; pre- pectal earina reaches dorsad about to upper 0.5 of hind margin of pronotum; prepectus opposite lower hind corner of pronotum with a rather high ridge that extends about 1/3 the distance ventrad to prepectal earina; surface shining with strong longitudinal wrinkling that becomes weaker and partially interrupted on upper 1/2, where there are some scattered large punctures, as well as ventro-posteri- orly, where there are more numerous large punctures. Lower meta- pleuron: with strong oblique wrinkling that grades anterio-dorsad into puncto-reticulation and finally becoming smooth for a short distance near front margin. Wing venation: areolet 1.6 as wide as high at apex; second recurrent a little basad of second intercubitus; nervulus slightly antefurcal; postnervulus broken at lower 0.4. Fore tibia: moderately inflated. F[ind femur: 1.6 as deep at middle as at apex. Flind tibia: inner spur 0.41 as long as basitarsus. Propo- 52 Psyche [March deum: spiracle 1.4 as long as wide; basal trans-carina almost straight medially; apical trans-carina represented sublaterally by strong, broadly cuneate cristae but interrupted on median 1/3 of propo- deum; median longitudinal carinae well defined throughout and enclosing a very narrow, parallel-sided area-basalis and a broader but also long and parallel-sided combined areola and median apical area; pleural carina obsolete; surface shining, apicad of basal trans- carina with more numerous very large puntures that are moderately sparse on area dentipara but which become denser and mingled with longitudinally biased to reticulate wrinkling on areola, median apical area, and latero-apical area. First gastric segment: petiole with a sharply triangular lateral expansion at base; postpetiole 0.6 as wide at apex as long from spiracle to apex. Second gastric ter- gite: smooth and highly polished with scattered tiny punctures emitting short, sparse setae. Succeeding tergites: with denser tiny punctures and setae which in part equal or exceed the length of their interspaces. Ovipositor: sheathed portion 0.51 as long as fore wing; tip 0.15 as high at nodus as long from nodus to apex, dorsal valve with a very long and slightly concave taper between nodus and apex. TYPE: In collection of Henry K. Townes, 5950 Warren Rd., Ann Arbor, Michigan, 48105. RELATIONSHIPS: Texana resembles the Mexican and Guatema- lan B. hwnerosa (Cushman, 1931, p. 51-52, fig. 4 on p. 4) but dif- fers in that the occipital carina reaches the hypostomal carina below (becomes obsolete below in humerosa ), because the propodeum anteriad of the apical trans-carina is punctured on the area denti- para and longitudinally rugose medially (polished before the apical trans-carina in humerosa ), in the well-defined, elongately rectangu- lar areola (in humerosa the apical trans-carina is acutely angled forward medially and a single median longitudinal carina reaches forward from the vertex of the angulation to the basal trans-carina), and in having the apical margins of gastric tergites 2 and 3 narrowly black (no black on apical margins in humerosa ). From the other Mexican Bicriste/la, B. univittata (Cresson), tex- ana differs because the occipital and hypostomal carinae join in a weak declivity (instead of being separated by a broad, deep depres- sion), by its shorter temple (0.21 as long as eye at upper 1/3 vs. 0.38 in univittata ), and by its longer epomia, medially longitudinally wrinkled (instead of mostly smooth and polished) mesopleuron. 1977] Porter — Mesostenines 53 narrow and well-defined area-basalis (broadly and poorly devel- oped in univittata), and black and white (instead of uniformly tes- taceous) gaster. B. bicarinata (Cushman) from Panama may be separated from texana by its color (ferruginous with head black) and by many structural features (clypeus not prominent in profile, occipital Ca- rina separated by a deep groove from hypostomal carina, temple concave in dorsal view, mesopleuron finely and sparsely punctate, subalarum reduced to a carina, and propodeum sparsely punctate with its apical trans-carina forming medially an acute angle from which a single carina extends forward to the basal trans-carina). Cameron’s (1885, p. 236) original diagnosis of B. ehontalensis, the only other described Middle American Bicristella, shows that his species has a short frontal horn (horn is long in texana) and that it deviates chromatically from texana in numerous aspects (mandible black at base, mesoscutum with a yellow line along outer edge of central lobe and without a white callus opposite tegula, lower metapleuron with a black mark over hind coxa, petiole black at base, and fore and mid coxae with a black line at base and a larger black spot at apex, and hind coxa marked with yellow). The Cuban B. tricolor (Brulle) resembles texana because its oc- cipital carina is complete ventrad to the hypostomal carina and the mesopleuron is discally striate but may be distinguished because it has two deep pits at the base of the frontal horn, the anterio-lateral margin of the pronotum sharply angulate below the middle, the scape below and apex of the frontal horn white, the mesoscutum with discal white lines that extend nearly the length of the inner margins of the lateral lobes, and the legs mostly ferruginous. Finally, B. testacea (Taschenberg), the only other described Bi- cristella, ranges over most of South America and differs strikingly from the black, white and testaceous texana in being uniformly ferruginous to testaceous with only the head black. FIELD NOTES: Swept near R. Grande from Serjania vines in Salix-Celtis-Fraxinus gallery woods. SPECIFIC NAME: for the state of Texas. Genus Diapetimorpha The Valley has seven Diapetimorpha, of which three are de- scribed as new. 54 Psyche [March Key to the U.S. Diapetimorpha Females (Females of pareia and aspila unknown) 1. Propodeum black with conspicuous white markings, including at least a broad band on each side that reaches from some- what in front of crista to or nearly to hind margin 2 Propodeum black or ferruginous to fulvous, its pale markings, if any, confined to cristae and often area immediately around cristae 3 2. Malar space 0.50 as long as basal width of mandible; propodeal cristae low, about 0.37 as long as their basal width; propo- deum basad of basal trans-carina with a pair of large white lateral blotches; gastric tergites 2-6 blackish basally with broad fulvous and whitish apical bands 18. D. picta Townes. Malar space 0.70 as long as basal width of mandible; propodeal cristae longer, about 0.9 as long as their basal width; propo- deum wholly black basad of basal trans-carina; gastric tergites 2-6 wholly fulvo-ferruginous D. rufigaster Cushman 3. Pronotum and mesoscutum black, with or without white mark- ings 4 Pronotum and mesoscutum fulvous or ferruginous, with or without white markings 6 4. Malar space 0.80 as long as basal width of mandible; head and mesosoma almost uniformly black and white at most only on propodeal cristae and sometimes very slightly on scutellum; gaster wholly feruginous 21. D. introita (Cresson). Malar space 0.60-0.70 as long as basal width of mandible; head, pronotum, and mesoscutum black with white markings; meso- pleuron, metapleuron and propodeum extensively fulvous to ferruginous; always a large median white blotch on gastric ter- gite 7 5 5. First flagellomere 6.6-7. 1 as long as deep at apex; lower meta- pleuron with coarse, mostly longitudinally biased wrinkling that becomes somewhat reticulate only on dorsal 1/4 or less; dorsal face of propodeum between trans-carinae with strong, more or less longitudinally biased wrinkling; propodeal cristae 1977] Porter — Mesostenines 55 strongly projecting subligulate to ligulate, 0.7-1. 2 as long as wide at base; broad white orbital ring interrupted only at bot- tom of eye; humeral margin of pronotum white throughout; scutellum pure white; first gastric tergite with a white apical band (except in S. Florida populations) 16. D. macula (Cameron). First flagellomere 5. 5-6.0 as long as deep at apex; lower meta- pleuron with strong longitudinal wrinkling* that becomes re- ticulate on dorso-posterior half; dorsal face of propodeum between trans-carinae with strong, complexly reticulate wrink- ling; propodeal cristae prominent, broadly cuneate, 0.4-0. 5 as long as wide at base; white only on frontal orbit; humeral margin of pronotum more than half black; scutellum yellowish ferruginous; no white on apex of first gastric tergite 1 7. D. sphenos n. sp. 6. Lower metapleuron with strong longitudinal wrinkling; whitish at least on pronotal collar and propodeal cristae 7 Lower metapleuron either only punctate or punctate and obliquely wrinkled; no white on mesosoma 8 7. Malar space 0.80 as long as basal width of mandible; mesosoma brownish ferruginous, except for whitish pronotal collar and propodeal cristae; fourth gastric tergite dusky D. brunnea Townes Malar space 0.63-0.73 as long as basal width of mandible; meso- soma fulvous with profuse yellowish white markings includ- ing scutellum, humeral margin of pronotum, and a pair of median stripes on mesoscutum; fourth gastric tergite fulvous. D. alabama Cushman 8. Malar space 0.65-0.75 as long as basal width of mandible .... 22. D. acadia Cushman Malar space 1.05 as long as basal width of mandible D. rugosa Townes Males 1. Gastric tergites 2-7 black or brownish black with broad white apical bands 2 Gastric tergites 2-7 ranging from yellowish fulvous to ferru- ginous, sometimes with blackish basal bands but never with white apical bands 5 56 Psyche [March 2. Tyloides sharp longitudinal carinae on flagellomeres 11 or 12 to 19, those of 12 or 13 to 17 or 18 extending more or less the length of their segments; mesopleuron and lower meta- pleuron largely with medium-sized coarse punctures that mostly are separated by less than 2.0 their diameters or sometimes in part on metapleuron and on mesopleuron in front of speculum with puncto-reticulation to more or less regular longitudinal wrinkling; mesopleuron black with white on subalarum and sometimes one or two whitish blotches on lower 1/4 above sternaulus 21. D. introita (Cresson) Tyloids often very faint and not extending nearly the length of their segments, at most flagellomeres 12 to 15 with sharp, carinate tyloids reaching 0.4-0. 6 their length; mesopleuron and lower metapleuron with small to tiny rather weak punc- tures separated in general by 2.0 or more their diameters and without wrinkling or puncto-reticulation, except some- times above speculum or just along pleural carina; 2/3 or more of mesopleuron white 3 3. Malar space 0.80-0.85 as long as basal width of mandible; scape entirely black; propodeum apicad of basal trans-carina uniformly pale fulvous 20. D. pareia n. sp. Malar space 0.48-0.58 as long as basal width of mandible; scape largely white below; propodeum apicad of basal trans- carina white with black markings 4 4. Temple 0.38 as long as eye at upper 1 /3; pronotal scrobe with- out wrinkles except for the short epomia; juxta-coxal carina absent; propodeal cristae large, strongly projecting, bluntly triangular; propodeum entirely black basad of basal trans- carina 19. D. aspila n. sp. Temple 0.50-0.60 as long as eye at upper 1/3; pronotal scrobe with numerous wrinkles in addition to the scarcely differ- entiated epomia; juxta-coxal carina almost complete but ir- regular at least on apical 1/2; propodeal cristae weakly cu- neate, low, broad, and scarcely projecting; propodeum basad of basal trans-carina black with a large white sublateral blotch 18. D. picta Townes 5. Tyloids sharp carinae extending full length of several segments; lower metapleuron with moderately coarse punctures ... .6 1977] Porter — Mesostenines 57 Tyloids carinate but either obsolescent and almost impossible to see or, if fine and sharp, generally shorter than the length of their segments; lower metapleuron with fine, weak punc- tures 7 6. Malar space 0.75 as long as basal width of mandible; clypeus moderately convex in profile; mesosternum usually entirely fulvous, sometimes partly black and rarely entirely black; epomia rather weak 22. D. acadia Cushman Malar space 0.83 as long as basal width of mandible; clypeus weakly convex in profile; mesosternum largely or entirely black; epomia moderately strong D. rugosa Townes 7. Mesoscutum black with a pair of median white lines or with a median white blotch 8 Mesoscutum mostly or entirely fulvous, with or without a pair of median whitish dashes 10 8. Tyloids sharp longitudinal carinae on 4 segments but in most cases shorter than their segments; flagellum with a white band on about 5 segments; pronotum apicad of basal trans- carina white with a broad black stripe along pleural carina almost to apex and black narrowly along basal trans-carina D. rufigaster Cushman Tyloids hard to see, low, indistinct ridges extending about 0.3 the length of several segments; flagellum at most with a poorly defined postmedian brown section; propodeum api- cad of basal trans-carina dull fulvous to yellowish or whitish and without black or with black only along basal trans- carina 9 9. White orbital ring interrupted on vertex and upper half of temple; pronotum white with a very large black apico-lateral area; mesopleuron black with white on subalarum and spec- ulum and with more or less whitish stained fulvous on most of lower 1/3 of mesepisternum behind prepectal carina as well as throughout on mesepimeron .17. D. sphenos n. sp. White orbital ring complete; pronotum wholly white; meso- pleuron mostly whitish or stramineous to weakly brownish white, in some specimens with black on as much as anterior 1/3-1/ 2 below sub-alarum 16. D. macula (Cameron) 58 Psyche [March 10. Eye about 75% surrounded with yellowish white, the whitish orbit being interrupted on upper part of temple; mesoscutum without a median pair of whitish dashes D. brunnea Townes Eye completely surrounded with yellowish white; mesoscutum with a median pair of whitish dashes D. alabama Cushman 16. Diapetimorpha macula (Cameron) SPECIMENS EXAMINED: 17 females, 4 males: bentsenpark (Net: 1 female, 19 III ’76; 1 male, 9 VI ’76; 12 females, 1 male, 2-10 IX 76; 1 female, 29 XII 76); BOTANICAL garden (Net: 1 female, 16 III 74; 1 female, 19 XII 76; Malaise : 2 males, 10-31 X 73; 1 fe- male, XII 73). HABITAT: Dark, damp woods; herbaceous undergrowth beneath Pithecellobium flexicaule; Celt is lindheimeri-C. pallida associa- tion. DISTRIBUTION: Va. to Fla. west to Tex. and south to Veracruz state in Mexico. GEOGRAPHIC VARIATION: Townes (1962, p. 384-387) divides macula into several subspecies, among which my Valley material agrees most closely with D. m. macula, heretofore not recorded north of Mexico. PHAENOLOGY: Flies almost throughout the year with peak in September. Valley records include 2 females for March, 1 male for June, 12 females and 1 male for September, 2 males for Octo- ber, and 3 females for December. North of the Valley macula is active between March and November. 17. Diapetimorpha sphenos n. sp. (Fig. 9) Holotype: female, USA (Texas: Hidalgo County, Bentsen Rio Grande Valley State Park, 1-15 XI 76, Malaise Trap, C. C. Por- ter). (Townes). Paratypes: 1 female, 2 males, USA (Texas: Hidalgo County, Bentsen Rio Grande Valley State Park, 1-15 V 76, Ma- laise Trap, C. C. Porter; Valley Botanical Garden at McAllen, X 73, Malaise Trap, C. C. Porter). (Gainesville, Porter). FEMALE: Color: antenna black with much dull to pale brown on scape, and sometimes pedicel and base of first flagellomere, and 1977] Porter — Mesostenines 59 with a ventrally interrupted white band on flagellomeres 4 (api- cally) to 9 and sometimes slightly onto base of 10; head black with palpi brownish white, mandible brownish white with dark brown on apical 1/4, brownish white on mandibular condyle, dull brown on apical 1 / 3 of clypeus, at times with a brownish median spot on face somewhat below antennal sockets, and with white broadly on most of frontal orbit; propleuron black; pronotum black with a broad white band on most of front margin except for hind corners, a large, broadly triangular white mark on median 1/3 of dorsal margin which is contiguous medially with white band on front margin and which encloses a more or less well developed central brown area, and with a little pale brown on upper hind corner; thoracic dorsum black with tegula white, a pair of short median white stripes on mesoscutum in apical 1 / 3 of notauli, scutellum shining ferruginous with a yellowish tinge, postscutellum yellow- ish to ferruginous white, and hind rim of meso and metanotal axillary troughs narrowly whitish to ferruginous; mesopleuron and mesosternum black with ferruginous on about ventro-posterior 2/3 of mesepisternum, narrowly on adjacent mesosternum, and on most of mesepimeron as well as with white on much of subalarum and toward dorsum of mesepimeron; mesosoma otherwise uni- formly ferruginous; gaster ferruginous with a trace of faint dusky staining on the more apical tergites and with a very large medio- apical white blotch on tergite 7 and a similar but smaller white area on 8; legs ferruginous to stramineous with blackish anterior and posterior staining on hind trochanter and with 5th tarsomeres dusky; wings hyaline with fore wing faintly brown tinged and stigma rather pale brown. Length of fore wing: 4. 6-4. 8 mm. First flagellomere: 5. 5-6.0 as long as deep at apex. Clypeus: moderately convex in profile. Malar space: 0.62-0.64 as long as basal width of mandible. Temple: 0.19- 0.22 as long as eye at upper 1/3. Pronotum: scrobe extensively wrinkled. Mesoscutum: finely and densely punctate with a broad band of delicate transverse wrinkling along anterior 2/3 of notauli, mostly internally, and with stronger longitudinal wrinkling between notauli on about their apical 1/3. Mesopleuron: in large part with strong wrinkling that is more or less regularly longitudinal above and more irregular below or which sometimes is longitudinally biased throughout. Lower metapleuron: with strong longitudinal wrinkling that becomes very irregular on dorso-posterior 1/2; juxta-coxal 60 Psyche [March carina obsolete, weakly suggested only near base. Propodeum: basal trans-carina rather strongly curved forward medially; apical trans- carina curved far forward and slightly to markedly irregular medi- ally, its cristae broadly and strongly projecting cuneate; surface with strong reticulate wrinkling that is finer basad of basal trans-carina. First gastric segment : postpetiole 1.1 -1.3 as wide apically as long from spiracle to apex. Ovipositor: sheathed portion 0.42-0.48 as long as fore wing; tip 0.23-0.24 as high at nodus as long from nodus to apex. MALE: Color: scape pale brownish to whitish with a little dusky staining dorsad; pedicel dark brown with apex paler; flagellum dark brown, becoming gradually a little paler apicad; palpi white; man- dible white with dark brown on apical 1/3; head black with white on clypeus, face, broad frontal orbits, very broad hind orbit of about 1/2 of eye, and throughout from malar space to hypostomal carina; propleuron white; pronotum white with a very large black area apico-laterally; thoracic dorsum black with tegula white and with a large, transverse, anteriorly emarginate postmedian white blotch on mesoscutum, and with white on scutellum, postscutellum, and narrow hind margins of meso and metanotal axillary troughs; mesopleuron black with white on subalarum and speculum and with more or less whitish stained fulvous on most of lower 1 / 3 of mesepisternum behind prepectal carina as well as throughout on mesepimeron; mesosternum testaceous-stained white except black anteriad of prepectal carina; upper metapleuron white; mesosoma otherwise pale fulvous to yellowish with hind face of propodeum more whitish; gaster pale fulvous with dark brown to black on extreme base of petiole, on basal half of segment 2, and on basal 1 /3 of 3, as well as with paler brown to equally dark brown or black on basal 1/4 of 4-6 and dusky on much of 7; fore leg with coxa, trochanter, and trochantellus white, femur and tibia pallid fulvous, and tarsus whitish with fifth segment dusky and a little dusky above on third and fourth segments; mid leg similar to fore leg but with light fulvous staining dorsad on coxa, trochanter, and trochantellus and with tarsus largely dusky except for whitish narrowly on tips and bases of segments 2-4 and with segment 1 dull fulvous grading into dusky on apical 1/2 with tip whitish; and hind leg pale fulvous with considerable blackish on trochanter, sometimes slightly dusky on apex of trochantellus and base of femur, tibia a little dusky at base and dusky stained on most of apical 2/3 except below, and 1977] Porter — Mesostenines 61 with tarsus blackish with dull white on base of first segment and pure white on apical 1/3 of first segment and on all of segments 2-4; wings hyaline with stigma brownish. Length of fore wing: 3. 5-3. 6 mm. Flagellum: tyloids not clearly defined; first flagellomere 4. 1-4.5 as long as deep at apex. Clypeus: moderately convex. Malar space: 0.55-0.65 as long as basal width of mandible. Temple: 0.47-0.52 as long as eye at upper 1/3. Pro- notum: scrobe smooth, without wrinkles except for short epomia or sometimes with several oblique wrinkles in addition to epomia. Mesoscutum: smooth and polished with numerous, well spaced small punctures that become somewhat larger and denser anteriad. Mesopleuron: smooth and polished with small, scattered punctures and only a little longitudinal wrinkling limited to dorsal region behind subalarum. Lower metapleuron: smooth and polished with scattered small, weak punctures that become only a little larger and denser dorsad toward pleural carina; juxta-coxal carina defined for a short distance near base. Propodeum: basal trans-carina only weakly curved forward medially; apical trans-carina complete, ad- vanced far forward medially, its cristae scarcely raised; surface smooth and polished basad of basal trans-carina, moderately wrinkled apicad of apical trans-carina. First gastric segment: post- petiole 0.84 as wide apically as long from spiracle to apex. TYPES: The holotype is in the collection of Henry K. Townes, 5950 Warren Rd., Ann Arbor, Michigan. One male paratype has been donated to the Florida State Arthropod Collection (Division of Plant Industry, Entomology Bureau, Florida Department of Agriculture, Gainesville, Florida, 32602) and a male and female paratype are retained in the author’s personal collection (301 N. 39th Street, McAllen, Texas, 78501). RELATIONSHIPS: This species is close to D. macula but may be easily distinguished by the structural and chromatic characters given in the key. FIELD NOTES: Sphenos has been collected only by Malaise Trap and appeared at both the Bentsen Park and Botanical Garden col- lecting stations. It is thus a species of shady woods, associated with large trees such as Pithecellobium flexicaule and Celtis lind- heimeri. Sphenos may have separate fall and spring generations, as there are records for October and November and again for May. 62 Psyche [March SPECIFIC NAME: From the genitive of the Greek noun sphen {sphenos) or “wedge”, in reference to the shape of the propodeal cristae. 18. Diapetimorpha picta Townes SPECIMENS EXAMINED: 2 males, BENTSEN PARK {Net: 1 female, 17 III ’76; 1 male, 17 III ’77). HABITAT: Serjania vines in Salix nigra woods along Rio Grande. DISTRIBUTION. Previously recorded only from Florida and south Georgia (Townes, 1962, p. 387). Among material loaned by Townes is a male from Kansas (Clark County, 12 VI ’60, R. L. Fischer). Picta thus probably ranges over the southeastern U.S. and into Mexico. 19. Diapetimorpha aspila n. sp. (Fig. 4) Holotype: female, USA {Texas: Hidalgo County, Valley Botani- cal Garden at McAllen, XII ’73, Malaise Trap, C. C. Porter). FEMALE: unknown. MALE: Color: antenna black with scape broadly white below, a little brown on apex of pedicel and base of first flagellomere, and a white annulus on flagellomeres 9-15; palpi white with a pale brownish tinge; mandible white with apical 1/3 dark brown; head white with black on median 1/2 of front and vertex, more broadly on occiput, on postocciput, and ventrad increasingly more nar- rowly along occipital carina to about its lower 0.1; propleuron white; pronotum black with a very broad white band throughout on anterior margin, a much narrower and apically attenuate white band on most of humeral margin, and a little white on upper hind corner; thoracic dorsum black with a large, more or less rectangu- lar, anteriorly deeply emarginate postmedian white blotch on meso- scutum, and with white on scutellum, postscutellum, and hind rims of meso and metanotal axillary troughs; tegula white; mesopleuron black with white on subalarum and on most of its lower 3/4 apicad of prepectal carina, except for a large, irregular black blotch below speculum and for black in about upper 0.7 of pleural suture; meso- sternum black on prepectus and otherwise white with a little brownish staining in front of mid coxae; upper metapleuron white except for black on its narrow lower 1 / 3; lower metapleuron white 1977] Porter — Mesostenines 63 with black irregularly along its anterior margin; propodeum basad of basal trans-carina entirely black and apicad of basal trans-carina white with a broad anteriorly gradually narrowed black blotch that reaches forward from gastric insertion about 2/3 the distance to basal trans-carina as well as with a broad black stripe along about basal 7/9 of pleural carina; first gastric segment with white on basal 1/2 of petiole above and laterally and on apical 1/2 of postpetiole as well as with brownish yellow on most of petiole ventrally; gastric tergites 2-7 black with very broad white apical bands and tergite 8 and claspers brownish yellow; fore leg with coxa, trochanter, and trochantellus white, femur and tibia dull pale fulvous, and tarsus pale brownish with segments 3-5 blackish; mid leg with coxa white with a yellowish brown tint, trochanter yellowish white with con- siderable brown staining, trochantellus yellowish white, femur and tibia pale fulvous and tarsus blackish grading irregularly into dull fulvous on basal 1/2 and with a little pale brownish on tips of seg- ments 1-4; hind leg bright, deep fulvous on coxa, trochanter, tro- chantellus, femur and tibia and with considerable dark brown on trochanter, a little blackish on base of tibia, and slight dusky stain- ing toward apex of tibia, and tarsus with segment 1 black on basal half and white on apical half, segments 2-4 white, and 5 black; wings hyaline with stigma dark brown. Length of fore wing: 5.0 mm. Flagellum: first segment 4.25 as long as deep at apex; segments 12-15 with tyloides in the form of fine but sharp carinae that extend about 0.4-0. 6 the length of each. Malar space: 0.58 as long as basal width of mandible. Clypeus: moderately strongly and asymmetrically convex, weakly nasute in lateral view. Temple: at upper 1/3 about 0.38 as long as eye in lateral view. Pronotum: scrobe smooth and polished, without wrinkles except for the short epomia. Mesoscutum: smooth and polished with numerous tiny, well spaced punctures. Mesopleuron: mostly smooth and polished with small, scattered punctures. Lower metapleuron: smooth and polished with small, very well spaced punctures that are scarcely denser dorsad; juxta-coxal carina ab- sent. Wing venation: intercubitus about 3.0 as long as width of radial vein. Hind femur: 5.1 as long as deep at middle. Propo- deum: basad of basal trans-carina smooth and shining with numer- ous medium-sized punctures, laterad and behind basal trans-carina rather strongly and irregularly wrinkled; basal trans-carina almost straight; apical trans-carina forming broad and strongly projecting 64 Psyche [March bluntly triangular cristae, between the cristae advanced far forward but, except for a short median segment, practically effaced. First gastric segment: postpetiole 0.91 as wide at apex as long from spiracle to apex; tergite smooth and shining with a few tiny, scat- tered punctures. TYPE: The holotype is in the collection of Henry K. Townes, 5950 Warren Rd., Ann Arbor, Michigan, 48501. RELATIONSHIPS: Aspila closely resembles D. picta but may be distinguished by the characters summarized in the key. FIELD NOTES: Collected by Malaise Trap in a partially shaded Celt is lindheimeri-C. pallida thicket. SPECIFIC NAME: from the Greek adjective aspilos or “unspotted”, in reference to the uniformly black propodeal base of this species. 20. Diapetimorpha pareia n. sp. (Fig. 7) Holotype: male, USA (Texas: Hidalgo County, Valley Botani- cal Garden at McAllen, 1 I ’76, C. C. Porter). (Townes). Paratype: male, USA (Texas: Hidalgo County, Bentsen Rio Grande Valley State Park, 15-30 IV ’76, Malaise Trap, C. C. Porter). (Porter). FEMALE: unknown. MALE: Color: antenna black with a pale yellow annulus on fla- gellomeres 9-18 or 19; palpi white; head white with black, preceded by a narrow brownish zone, on apical 1/3 of mandible, a small dusky spot just above dorsal corner of clypeus, a little dusky around anterior tentorial pits, and black on median 1/2 of front and vertex, more broadly on occiput, on all but ventral corner of postocciput, and ventrad, increasingly more narrowly, on temple along occipital carina to about its upper 0.5; propleuron white; pronotum black with a broad white band on all but extreme hind corner of front margin, a narrow white band on anterior 2/3 of humeral margin, and with white on upper hind corner, or sometimes white narrowly throughout on humeral margin; thoracic dorsum black with a large more or less rectangular postmedian white spot on mesoscutum, and with white on scutellum, postscutellum, and hind rims of meso and metanotal axillary troughs as well as with some yellowish stain- ing anteriorly in metanotal axillary trough on each side of post- scutellum; tegula white; mesosternum and mesopleuron pale yellow, becoming more nearly white anterio-dorsally on mesopleuron, as 1977] Porter — Mesostenines 65 well as sometimes with a little brownish staining above subalarum, a large black area above speculum, and more or less black on dorsal 1/2 of prepectus; dorsal metapleur'on and groove at base of propodeum yellow and sometimes with a little brownish staining on dorsal metapleuron anteriorly at about upper 1/3; lower meta- pleuron pale yellow testaceous; propodeum pale fulvous with black basad of basal trans-carina in most of region between spiracle and area-basalis, which is dull testaceous or yellow grading into black apically and laterally; first gastric segment yellow with black above on petiole and basal 2/3 of postpetiole, except for a broad median yellow stripe on basal 4/5 of petiole; second gastric tergite black with a broad yellow apical band and pale fulvous on thyridia; suc- ceeding tergites similar to second except that the apical yellow bands reach farther forward laterally; fore leg with coxa, tro- chanter, and trochantellus yellow with a little brown staining above on trochanter, femur and tibia pale testaceous, and tarsus testa- ceous grading into dusky or black on last three segments; mid leg similar to fore leg except that the coxa, trochanter, and trochantel- lus have a faint testaceous suffusion, the tibia sometimes becomes dusky above on apical 1/4, and the tarsus varies from mostly blackish to dusky on segments 2-5 and somewhat paler on 1; hind coxa, trochanter, trochantellus and femur rather intense shining testaceous with a brown streak ventro-anteriorly on base of coxa (absent in paratype) and some brown staining above toward base of trochanter, on apex of trochantellus, and on base of femur; hind tibia yellowish testaceous with a little dusky on base and blackish on apical 1/3 or a little more, and hind tarsus white with black on basal 1/8 of first segment, at least laterally, and on apical 1/4 of last segment; wings hyaline with stigma dull brownish white. Length of fore wing: 7. 1-7.5 mm. Flagellum: first segment 3.9- 4.3 as long as deep at apex; tyloids very weak, sometimes faintly detectable as longitudinal discontinuities on segments 10-16 which extend about 0.5-0. 8 the length of each segment. Malar space: 0.80-0.85 as long as basal width of mandible. Clypeus: weakly convex in profile. Temple: 0.60 as long as eye at upper 1/3. Pro- notum: scrobe smooth and polished and without wrinkles except for the short epomia. Mesoscutum: smooth and polished with abundant tiny punctures separated by 4-5X their diameters. Meso- pleuron: smooth and polished with numerous well spaced, tiny punctures and with a little longitudinal wrinkling above speculum. 66 Psyche [March Lower metapleuron: mostly smooth and shining with many tiny, sparse punctures that become larger and denser, with a little inter- calated wrinkling, only dorsad near the obsolete pleural carina; without juxta-coxal or, in paratype, with juxta-coxal carina de- fined only on basal 1/3. Propodeum: smooth and polished basad of basal trans-carina, otherwise rather coarsely and irregularly wrinkled; basal trans-carina almost straight; apical trans-carina in holotype irregularly traceable between the broad and moderately projecting triangular cristae and in paratype sharp but only slightly curved forward between the lower, more cuneate cristae. First gas- tric segment: postpetiole 0.87 as wide at apex as long from spiracle to apex; its surface smooth and polished with a few tiny, scattered punctures. TYPE: The holotype is in the collection of Henry K. Townes, 5950 Warren Rd., Ann Arbor, Michigan, 48105 and the paratype in the collection of Charles C. Porter, 301 N. 39th St., McAllen, Texas, 78501. RELATIONSHIPS: In habitus and color this species resembles D. picta and D. aspila but may be separated from both by its much longer malar space and by chromatic features such as its entirely black scape and uniformly pale fulvous propodeal base. FIELD NOTES: Taken by sweeping Serjania vines in a partial clear- ing near the edge of a Celtis lindheimeri-C. pallida thicket and in a Malaise Trap located in deep shade beneath a large Pithecellob- ium flexicaule. SPECIFIC NAME: From the Greek noun pareia or “cheek,” in ref- erence to the long malar space. 21. Diapetimorpha introita (Cresson) SPECIMENS EXAMINED: 15 females, 12 males: BENTSEN PARK {Net: 3 females, 1 male, 12-20 III ’77; 1 male, 7 IX ’76); BOTANICAL GARDEN {Net: 1 female, 17 I ’75; 6 females, 4 males, 12-20 III ’77; 1 male, 5 IV 74; 1 male, 16-30 V 74; 1 male, 30 VIII 76; 1 male, 1 IX 73; 1 female, 9 IX 76; 4 females, 20-31 XII 74; Malaise: 1 male, IX 73; 1 male, X 73). HABITAT: Open, sunny places; tall grass at edge of fields and thickets; herbaceous growth in abandoned orange groves; occa- sionally enters dense woods. DISTRIBUTION: N.C. to Tex. and into Nuevo Leon, Mexico. 1977] Porter — Mesostenines 67 PHAENOLOGY: Peaks in spring and fall; possibly absent in sum- mer. Valley records include 1 female for January, 9 females and 5 males for March, 1 male for April, 1 male for May, 1 male for August, 1 female and 3 males for September, 1 male for October, and 4 females for December. North of the Valley introita flies from mid-spring to early fall but disappears by July in most of Texas. 22. Diapetimorpha aeadia Cushman SPECIMENS EXAMINED: 1 female, 4 males: BOTANICAL GARDEN (Net: 1 female, 1 male, 12 III ’77; 1 male, XII ’73; Malaise: 2 fe- males, III 74). HABITAT: Exposed areas to dense woods; weeds in bright sun; partially shaded hedge row; Celtis lindheimeri-C. pallida thicket. DISTRIBUTION: Va. to Tex. and Mexican state of Coahuila. PHAENOLOGY: Valley records are for December and March. North of the Valley it flies from late spring to early fall and peaks between 15 August and 15 September. Genus Listrognathus 23. Listrognathus glomerata Townes Townes (1962, p. 424 -425) gives one undated record of this spe- cies from “Cameron County, Texas”. I have not collected glom- erata personally. Glomerata ranges from New Jersey to south Texas and flies mostly from early spring to late fall. 24. Listrognathus rufitibialis Cushman SPECIMENS EXAMINED: 19 females, 3 males: BENTSEN PARK (Net: 2 females, 12-20 III 77; 2 females, 23-30 XII 76); BOTANICAL GAR- DEN (Net: 6 females, 9-11 I 76; 1 female, 16 I 75; 1 female, 24 I 76; 1 male, 12-20 III 77; 1 male, 16-30 V 74; 1 female, 19 XII 76; 4 females, 22-28 XII 75; 2 females, 1 male, 20-31 XII 73). habitat: Exposed to shaded areas; fields at edge of woods; Ser- jania vines in woodland clearings; in bright sun among small herba- ceous plants on paths through open brushland, DISTRIBUTION: Eastern U.S. from N. J., Ind., and Okla. south- ward. 68 Psyche [March PHAENOLOGY: Valley populations peak in winter, my data in- cluding 8 females for January, 2 females and 1 male for March, 1 male for May, and 9 females and 1 male for December. North of the Valley it flies mostly from June to September. Genus Mallochia 25. Mallochia agenioides Viereck SPECIMENS EXAMINED: 10 males: BENTSEN PARK {Net: 1, 23 I 76; Malaise: 1, 16-31 I 76; 1, 1-15 II 76; 2, 16-28 II 76; 2, 1-15 V 76; 1, 16-31 VIII 76; 2, 1-15 IX 76). HABITAT: Most Valley specimens were collected by a Malaise Trap installed in deep woods under a large Pithecellobium flexi- caule. Farther north, agenioides often is found in grassy, over- grown meadows. DISTRIBUTION: R.I. to Ks. and south to Florida. These are the first records for Texas. PHAENOLOGY: Valley records are for January, February, May, August, and September. Elsewhere agenioides flies mostly from mid-spring to mid-summer. 26. Mallochia frontalis Townes SPECIMEN EXAMINED: 1 male, Bentsen Park, 1-15 IX 76. HABITAT: Caught by Malaise Trap in dense woods under shade of a large Pithecellobium flexicaule. DISTRIBUTION: N.J., Md., Va., N.C., Ks., Tex. PHAENOLOGY: North of the Valley frontalis flies between March and August. Genus Lymeon The Valley has three Lymeon, of which one here is recorded for the first time from the United States. Key to the U.S. Lymeon (Females only) 1. Mesoma black with profuse white markings 2 Mesosoma fulvous or ferruginous, with or without white mark- ings 3 1977] Porter — Mesostenines 69 2. Mesoscutum with a single median white spot; lateral lobe of mesoscutum throughout with medium-sized, sharp, moder- ately dense punctures 27. L. cinctiventris (Cushman) Mesoscutum with a pair of submedian white stripes; lateral lobe of mesoscutum, except on about basal 1/3, sparsely punctate 28. L. orbus (Say). 3. Discoidella completely absent; lateral lobe of mesoscutum pol- ished; second gastric tergite black with a broad white apical band; fore wing without dark cross bands L. bicinctus (Cresson). Discoidella well developed; lateral lobe of mesoscutum mat; second gastric tergite wholly fulvous; fore wing with a median and subapical dark cross band 4 4. Clypeus prolonged ventrally as a conical point; malar space 1.5 as long as basal width of mandible; head mostly fulvous and without black areas; no white on pronotum, subalarum, and propodeal cristae L. nasutus (Pratt). Clypeus strongly convex but not pointed ventrally, the apical margin almost truncate; malar space 0.66 as long as basal width of mandible; head mostly black, sometimes with cly- peus and face fulvous; white on broad anterior margin of pronotum, subalarum, and propodeal cristae 29. L. leucosoma (Cameron). 27. Lymeon cinctiventris (Cushman) SPECIMENS EXAMINED: 1 female, 4 males: BENTSEN PARK (Net: 1 female, 29 XII 76); BOTANICAL garden (Net: 2 males, 16 I 75; 1 male, 1 IV 75; 1 male, 24 XII 74). HABITAT: Damp area in tall grass beneath Salix nigra and Acacia farnesiana near irrigation canal. DISTRIBUTION: Md. to Fla. and west to Tex. PHAENOLOGY: Valley records are for December, January and April. Farther north, cinctiventris flies between April and Septem- ber. 28. Lymeon orbus (Say) SPECIMENS EXAMINED: 3 females: BENTSEN PARK (1, 19 I 76); BOTANICAL GARDEN (1, 1 IV 75; 1, 30 XII 74). 70 Psyche [March HABITAT: Herbaceous undergrowth in dense woods. DISTRIBUTION: N.Y. to Fla. west to Wis., Ks., Tex., and Nuevo Leon state of Mexico. GEOGRAPHIC VARIATION: Specimens from Mexico and the Valley have the female hind coxa white with a black area above, black narrowly throughout on base, and with a large black blotch on basal half anterio-ventrally. In material from eastern North Amer- ica the female hind coxa is fulvous with a more or less well devel- oped whitish dorsal blotch. PHAENOLOGY: Valley records are for December, January and April. Elsewhere it flies mainly from spring to fall. 29. Lymeon leucosoma (Cameron) (Fig. 3) FEMALE: Color: scape pale fulvous; pedicel pale fulvous to dusky; flagellum with first segment pale fulvous to dull brownish grading into blackish on apical 1/2, second and third segments black with a little brownish below, fourth black with a little white on apical 1/3, fifth through eighth white with some pale brownish below, ninth black with white on basal 1/3 above and brownish below, and succeeding segments dull brownish with some dusky staining on the more basal ones and paler apicad; head black with pale brown on mandibular condyles and face and clypeus some- times largely stained with pale fulvous and with mandible mostly white on basal 2/3 but on apical 1/3 grading through pale brown into black; palpi whitish brown; mesosoma rather opaquely pale fulvous with some black basad in mesonotal axillary trough, a little dusky staining in some other areas, and with white broadly on all but about lateral 1/5 of front margin of pronotum, dully anteriad on tegula, on most of subalarum, vaguely toward top of mesepim- eron, sometimes weakly on sides of scutellum and on postscutel- lum, and on propodeal cristae; gaster rather opaquely pale fulvous with slight dusky tinging; legs pale fulvous, duller on tibiae and tarsi, and dusky on fifth tarsomeres; wings hyaline, the fore wing with a median brown crossband occupying basal 1/2 of discocubital cell, extreme base of second discoidal cell, about apical 3/4 of first brachial cell and extending briefly into base of second brachial cell, as well as with a subapical brown cross band that covers apical 1/4 of radial cell, almost median 1/2 of third cubital cell, and about dorso-median 1 / 3 of third discoidal cell; stigma brownish white. 1977] Porter — Mesostenines 71 Length of fore wing: 4. 5-5. 8 mm. First flagellomere: 6. 1-7.5 as long as deep at apex. Clypeus: strongly and asymmetrically convex with apical face distinctly shorter and more strongly declivous than basal. Malar space: 0.66 as long as basal width of mandible. Tem- ple: 0.36 as long as eye at upper 1/3; moderately and directly to a little convexly receding; dully to rather strongly shining with vari- ably developed fine micro-reticulation and more or less numerous small, obscure punctures. Pronotum: humeral margin scarcely swollen; epomia more or less well defined in scrobe among some other variably developed irregular wrinkles. Mesoscutum: notauli weakly defined on about basal 1/3 of mesoscutum and traceable almost to apex as a wide band of oblique rugosities; surface mat with fine micro-reticulation and stronger wrinkling along no- tauli and rearward on central lobe as well as with numerous, me- dium sized, shallow and obscure, mostly subadjacent or sparser punctures. Mesopleuron: mostly mat and rather finely reticulately wrinkled but with some discrete punctures on prepectus and on dorsal 1/3; speculum smooth and polished. Lower metapleuron: with rather strong reticulate wrinkling that sometimes becomes more regularly oblique ventro-posteriad. Wing venation: areolet 0.68-0.76 as high as part of second recurrent above bulla, first intercubitus reclivous; discoidella well developed, upper part of nervellus 1.5- 1.8 as long as lower; brachiella well developed. Fore tibia: only slightly swollen. Propodeum: median part of basal groove with some sharp longitudinal ridges; spiracle 1.2 as long as wide; cristae small but strongly projecting subligulate tubercles, apical carina otherwise lacking. First gastric segment: postpetiole 1.3 as wide apically as long from spiracle to apex, its surface a little dully shining with well defined micro-reticulation; dorso-lateral ca- rina traceable throughout and mostly sharp. Second gastric tergite: thyridium transverse; surface mat and finely reticulate with small, sparse, obscure punctures emitting short, widely spaced setae. Ovi- positor: sheathed portion 0.34 as long as fore wing; tip 0.26 as high at nodus as long from nodus to apex and directly tapering between nodus and apex. MALE: Unknown. SPECIMENS EXAMINED: 4 females, USA {Texas: Hidalgo County, Valley Botanical Garden at McAllen, 12-27 I ’74, C. C. Porter; Bentsen Rio Grande Valley State Park, 19 I ’76, C. C. Porter); MEXICO ( Jalisco : Guadalajara, 17 VII ’51, H. E. Evans). (Porter, Townes). 72 Psyche [March DISCUSSION: These are the first records of leucosoma from the United States and Mexico, the species having been cited previously only from Guatemala (Cameron, 1886, p. 259). A homotype from Guadalajara, Mexico loaned by H. K. Townes differs from my Texas material only in having the tegula a little more broadly white, the sides of the scutellum white, and the post- scutellum white. One of the Valley specimens was collected in the humid “Sunken Garden” at the Valley Botanical Garden, where it was swept from weeds at the edge of a thicket dominated by Ehretia anacua. The other was taken at Bentsen Park along the banks of the Rio Grande in Serjania vines beneath Salix nigra and Celtis lindheimeri. Genus Acerastes 30. Acerastes pertinax (Cress on) SPECIMENS EXAMINED: 87 females, 92 males: BENTSEN PARK (Net: 3 females, 12-13 I ’76; 3 females, 19 I 76; 3 females, 18-19 III 76; 1 male, 17 VI 73; 2 females, 1-13 VI 76; 1 female, 30 VIII 76; 1 female, 7 IX 76; 2 females, 27 XII 75; 6 females, 29-30 XII 76; Malaise (1976): 4 males, 15-30 IV; 5 females, 10 males, 1-15 V; 1 female, 20 males, 16-31 V; 1 male, 1-15 VI; 1 female, 16-30 VI; 1 male, 1-15 VII; 1 male, 16-31 VII; 2 males, 16-31 VIII; 5 males, 1-15 IX; 1 female, 2 males, 1-15 X; 6 males, 16-31 X; 1 female, 1 male, 1-15 XI; 1 female, 1 male, 16-30 XI; 1 male, 16-31 XII); BOTANICAL GARDEN (Net: 1 female, 5 I 75; 1 female, 18 I 75; 1 female, 26 I 76; 2 females, 1 male, 12-21 I 76; 4 females, 1 male, 17-24 III 74; 1 female, 5 IV 75; 1 male, 16-30 V 74; 1 male, 1 VI 75; 5 males, 1 IX 76; 1 female, 18 XII 76; 9 females, 1 male, 20-31 XII 74; Malaise (1973): 3 females, 18 males, X; 26 females, 8 males, XI; 9 females, XII). HABITAT: Open to dense scrub or woods with abundant ground cover of grasses, forbs, or vines; gallery woods; Celtis lindheimeri- C. pallida association; etc. Townes (1962, p. 405) gives the ’’usual habitat” of pertinax as “weedy fields or meadows” but it rarely enters such areas in south Texas. DISTRIBUTION: Md. to Fla. and Tex. south to Brasil; Cuba, Ja- maica. PHAENOLOGY: Valley populations fly almost throughout the year with a spring peak in March-May and a fall maximum from Sep- 1977] Porter — Mesostenines 73 tember to January. Monthy totals include 1 1 females and 1 male for January, 7 females and 1 male for March, 1 female and 4 males for April, 6 females and 31 males for May, 3 females and 3 males for June, 2 males for July, 1 female and 2 males for August, 5 females and 6 males for September, 4 females and 26 males for October, 28 females and 10 males for November, and 25 females and 2 males for December. All but 50 of the 179 specimens were collected by Malaise Traps. Pertinax shows some fluctuation in abundance from year to year. Malaise Trap surveys in the ’73-’74 season and in 1976 have obtained 64 and 65 specimens respec- tively, while hand collecting accounted for 18 specimens in 73-74, 7 in ’74-75, 20 in ’75-76, and 5 so far in ’76-77. In the eastern U. S. pertinax flies mainly during August and September but in subtropical latitudes is more or less active all year. Genus Polycyrtidea 31. Polycyrtidea limitis Cushman SPECIMENS EXAMINED: 17 females, 17 males: BENTSEN PARK (Net: 1 male, 3 I ’76; 1 female, 2 males, 12-13 I ’76; 3 females, 2 males, 19 I ’76; Malaise: 1 male, 1-15 IV ’76; 1 male, 1-15 V ’76; BOTANICAL GARDEN (Net: 1 male, 7 I ’75; 1 female, 2 males, 10-11 I ’76; 1 male, 18 I ’76; 1 female, 1 male, 18-22 I ’76; 2 fe- males, 12-21 I ’74; 1 male, 16-30 V ’74; 2 females, 26 XII ’75; Malaise: 1 female, I ’74; 4 females, 4 males, III ’74; 1 male, XII ’73); SANTA ANA NATIONAL WILDLIFE REFUGE (Net: 1 male, 24 XII ’75). HABITAT: Shaded, weedy places; Serjania vines in gallery woods; in tall grass beneath large Celtis lindheimeri. DISTRIBUTION: Lower Rio Grande Valley to Costa Rica. PHAENOLOGY: Most abundant in January but flies almost throughout the year. My records include 10 females and 9 males for January, 4 females and 4 males for March, 1 male for April, 2 males for May, and 3 females and 1 male for December. Townes (1962, p. 407) gives two additional Valley records: 1 female, 30 July, Hidalgo County and 1 male and 1 female, September, Browns- ville. Limitis may vary in numbers from year to year, since 4 speci- mens were taken in ’73-’74, 10 in HA-15, 18 in ’75-’76, and 2 in ’76-77. 74 Psyche [March Genus Pachysomoides 32. Pachysomoides stupidus (Cresson) SPECIMEN EXAMINED: 1 female, Bentsen Park, 24 XII ’75. HABITAT: Netted from Ulmus crassifolia at edge of forest trail. DISTRIBUTION: N.C. to Fla., Tex., and south to Brasil. PHAENOLOGY: North of the Valley stupidus flies mostly between between August and October. 33. Pachysomoides fulvus (Cresson) SPECIMENS EXAMINED: 5 females: BENTSEN PARK {Net: 1, 12-20 III 77; 1, 29 XII 76); BOTANICAL GARDEN {Net: 26 I 75; 1, 12-20 III 77; 1, 20-31 XII 73). HABITAT: Undergrowth of well shaded woods; dense weeds in untended orange grove. DISTRIBUTION: U.S. and southern Canada to Mexico and Cuba. PHAENOLOGY: North of the Valley fulvus is active mainly from early summer to mid-fall but, in mild years, still may be flying up to December and January as far north as Maryland, Nebraska, and British Columbia. Genus Messatoporus 34. Messatoporus discoidaloides (Cresson) SPECIMENS EXAMINED: 4 females, 5 males: BENTSEN PARK {Net: 2 females, 13-19 I 76; 1 male, 27 XII 75; 1 female, 29 XII 76; Malaise: 1 male, 1-15 IX 76); BOTANICAL GARDEN {Net: 2 males, 12-20 III 77; 1 female, 1 IV 75; 1 female, 19 XII 76). HABITAT: Shaded places; Serjania vines in Celtis-Salix gallery woods; Celtis lindheimeri-C. pallida association. DISTRIBUTION: Quebec and Minn, south to S.C. and Tex. PHAENOLOGY: Most common in the Valley from December to early April. Farther north it flies between May and November. Valley populations vary in abundance from year to year, as my data include no specimens for 73-74, 1 for 74-75, 3 for 75-76, and 5 for 76-77. Genus Agonocryptus 35. Agonocryptus discoidaloides (Viereck) 1977] Porter — Mesostenines 75 SPECIMENS EXAMINED: 97 females, 133 males: BENTSEN PARK (Net: 1 female, 12 I ’76; 3 males, 17-19 III ’76; 2 males, 12-20 III ’77; 2 males, 8-10 IX ’76; 3 males, 29 XII ’76; Malaise (1976): 1 male, 1-15 II; 1 female, 4 males, 15-30 IV; 1 male, 16-31 V; 5 males, 16-31 VII; 7 males, 1-15 VIII; 1 male, 16-31 VIII; 3 males, 1-15 IX; 8 males, 16-30 IX; 7 males, 1-15 X; 9 males, 16-31 X; 1 male, 1-15 XI; 1 female, 2 males, 16-30 XI; 1 female, 1 male, 1-15 XII); botanical GARDEN (Net: 10 females, 7 males, 12-21 I ’74; 2 females, 1 — 15 I ’75; 6 females, 16-21 I ’75; 13 females, 2 males, 1-15 I ’76; 10 females, 2 males, 15-24 III ’74; 2 females, 28 III ’75; 1 male, 15 III ’76; 15 females, 16 males, 12-20 III ’77; 2 females, 1 male, 2-5 IV ’75; 1 female, 1 male, 17-24 V ’74; 1 female, 23 VIII ’73; 1 male, 30 VIII ’75; 1 male, 7 IX ’76; 13 females, 13 males, 20-31 XII ’73; 5 females, 3 males, 20-31 XII ’74; 3 females, 2 males, 23-23 XII ’75; 1 female, 18 XII ’76; Malaise (1973): 3 males, X; 3 females, 15 males, XI; 2 females, 1 male, XII ’73). HABITAT: In or at edges of moderately to deeply shaded woods, especially where there are dead trees, shrubs, or vines; sometimes abundant around fallen Celtis lindheimeri; in winter visits Con- dalia obovata shrubs in bright sun; on Serjania vines or herba- ceous undergrowth in woods; often in orange groves on dead branches of frost-damaged citrus trees. Flies in open during win- ter and keeps to deep shade in hotter months. Males fly much more actively than females. DISTRIBUTION: N.H. and Wis. to Fla. and south Texas. phaenology: Valley populations peak between December and March but show some adult activity throughout the year. Monthly totals include 36 females and 1 1 males for January, 1 male for February, 27 females and 24 males for March, 3 females and 5 males for April, 5 males for July, 1 female and 9 males for August, 14 males for September, 19 males for October, 3 females and 19 males for November, and 26 females and 24 males for December. Of the 230 specimens, 76 (8 females and 68 males) were obtained by Malaise Traps. Discoidaloides fluctuates in numbers from year to year. Malaise Trap surveys in the ’73-’74 season and in 1976 have obtained 24 and 52 specimens respectively, while hand col- lecting yielded 66 specimens in ’73-’74, 19 in ’74-75, 21 in ’75-’76, and 40 in ’76-77. North of the Valley, discoidaloides flies mainly between April and October. 76 Psyche [March Conclusions Zoogeography The 18 Valley mesostenine genera fall into three zoogeographic categories: Neotropic, Sonoran, and Holarctic. Neotropic genera are Latin American taxa with centers in the Brasilian Highlands, the Andean Cloud Forests, and the mountains of Middle America. The Sonoran group includes genera which originated along the Madro-Tertiary geoflora in the southwestern U.S. and northern Mexico. The Holarctic element is circumpolar with maximum de- velopment in Temperate Deciduous Forests. Ten Valley genera are Neotropic: Cryptanura, Bicristella, Diapetimorpha, Mallochia, Ly- me on, Acerastes, Polycyrtidea, Pachysomoides, Messatoporus, and Agonocrvptus; four genera are Sonoran: Joppidium, Lanugo, Compsocryptus, and the Longicaudis group of Mesostenus; and five are Holarctic: Gambrus, Trvchosis, the Transfuga group of Mesostenus, Listrognathus, and Trachysphyrus. The Neotropic group requires special comment. This fauna pre- dominates at the generic level, includes 22 of the 35 species re- ported, and accounts for 543 of the 679 specimens collected. Although the modern Neotropic radiation is centered in Latin American humid forests, only one of the genera cited, Bicristella, reaches its northern limit in the Valley; all others range farther into North America, where they inhabit principally the southeast. This northeastern group seems descended from a larger and more pervasive Middle and North American Tertiary fauna that was pushed south by Pleistocene glaciations. During glacial maxima, a few of these Neotropic elements survived in the southeastern U.S., while many retreated southwest into the more hospitable lower latitudes of Middle America. Interglacials allowed some expansion from Pleistocene refugia but the accompanying aridity in subtropic latitudes has slowed movement of moisture-loving ichneumonids. Thus, southeastern isolates have expanded with the Temperate Deciduous Forest and Southern Pine-Oak Forest as far north as Maryland or New Jersey and southwest into Texas, while some Mexican species have followed subtropical deciduous woods into Texas. However, the semiarid scrub now covering much of south Texas and northeast Mexico has prevented mas- sive interchange between the present-day Middle American and southeast North American Neotropic faunas. 1977] Porter — Mesostenines 77 The Valley is a comparatively humid refugium surrounded by more arid habitats and located near the southwest limit of inter- glacial expansion for North American Neotropic species and close to the northeast limit for most Middle American species. Its Neo- tropic mesostenines are thus of complex distributional affinities. Four species, Cryptanura compaeta, C. lamentaria, Lymeon leu- cosoma, and Polycyrtidea limitis, are Middle American and range from the tip of Texas to Central America. Cryptanura vallis, Bi- cristella texana, Diapetimorpha sphenos, D. aspila, and D. pareia are apparently endemic to the Valley but almost certainly will be found also in Mexico when that poorly known fauna has been better collected. Acerastes pertinax and Pachysomoides stupidus extend all the way from Brasil to the southeastern U.S. On the other hand, Diapetimorpha picta, Mallochia agenioides, M. fron- talis, Lymeon cinctiventris, Messatoporus discoidalis, and Agono- cryptus discoidaloides are mainly eastern North American and reach their southern limit in the Valley. Diapetimorpha macula, D. introita, D. acadia, and Lymeon orbus likewise are centered in the Atlantic and Gulf states but range south a variable distance into Mexico and Pachysomoides fulvus extends over the entire U.S. and well into Mexico. Obviously, Pleistocene alteration of glacials and interglacials, as well as wetter and drier epochs within interglacials, have produced in the Valley a multiple overlap of northern and southern Neotropic mesostenines. Valley mesostenine genera show extremely wide affinities. Crypt- anura, Diapetimorpha, Mallochia, Lymeon, Acerastes, Pachysom- oides, Messatoporus, and Agonocryptus range from the northeast- ern U.S. to subtropical Argentina, Polycyrtidea from Texas and Florida to Argentina, and Bicristella from Texas to Argentina. The 11 Neotropic mesostenine genera represented in the eastern U.S. all reach Argentina. Of the 34 genera found in Middle Amer- ica, 33 cover at least a large part of South America also, while only one is endemic. About 20 more Neotropic mesostenine genera are endemic to South America. The Neotropic mesostenines thus seem to have evolved in South America during the Tertiary and to have spread northward massively in those warmer and wetter times. The modern Middle American fauna, generically, is a vast sample of the South American and the North American fauna a decimated vestige of the same stock. In the Tertiary, when humid forests covered much of North America below 40 degrees N. Lat. and 78 Psyche [March when temperatures there, if scarcely tropical, were milder than at present, most of the Middle American mesostenine genera prob- ably ranged far up the Gulf and Atlantic coasts. All of these moisture rather than temperature-controlled taxa tolerate winters with repeated frosts (as shown by their present-day altitudinal and latitudinal distribution in Mexico) and, in fact, there are few, if any. Neotropic mesostenines which are “tropical” in the sense of requiring frost-free winters. This is why a fair number of Neo- tropic mesostenines survived glaciation in the southeastern U.S. This is also why the Valley should be viewed not only as a present- day northern outpost of the main Neotropic fauna but also in historical perspective as an area once situated deep within the vaster Tertiary Neotropics. Comparison with other Neotropic Deserts Recent hand collecting surveys in the Peruvian Coastal Desert (Porter, 1975b) and net and Malaise sampling in the northeast Argentine Subandino (Porter, 1975a) allow comparison of Valley mesostenines with those of two ecologically somewhat similar areas in remote parts of the Neotropics. The Peruvian Desert reaches from south Ecuador to north Chile on the Pacific Coastal Plain and adjacent Andean foothills of west- ern South America. Its rainfall varies from 80 mm. per year in northern Peru to 0.5 mm. at Arica, Chile but, near the coast, massive fogs supply additional humidity. This desert is frost-free, at least on the Coastal Plain, but the cold Humboldt Current run- ning just off shore keeps temperatures moderate (at Arica, Chile the summer day-night range is about 20-30 degrees C. and the winter range approximately 10-22 degrees C.). Thus we confront the paradox of a humid but almost rainless, cool but frost-free tropical desert. Hand collecting during June-July of 1 974— ’76, mainly around Chiclayo, Trujillo, and Lima, Peru and Arica, Chile, has obtained from this desert in fertile valleys between sealevel and 2000 m. a mesostenine fauna of 12 genera and 31 species. There are nine Neotropic genera ( Biconus , Cyclaulus, Diapetimorpha, Basileucus, Lyme on, Acerastes, Polycyrtidea, Messatoporus and Agonocryp- tus ), one Pantropic genus ( Baltazaria ), two Sonoran genera ( Comp - socryptus, Mesostenus of the Longicaudis group), and two Holarc- tic genera ( Trachysphyrus , Mesostenus of the Transfuga group). 1977] Porter — Mesostenines 79 Note that all these genera except Biconus, Cyclaulus, Basileucus, and Baltazaria are shared with the Valley. A simple index of af- finity (Odum, 1971, p. 144) thus gives 0.606 (out of a possible 1.000) as the degree of similarity between the Valley and Coastal Desert mesostenine faunas. Actually, the real value probably ap- proaches 0.777, since four more Valley genera ( Cryptanura , Bi- cristella, Mallochia, and Pachysomoides ) occur also in the Ecua- dorian rainforests and Andean cloud forests, so that they very likely penetrate the northern fringes of the desert. The northwest Argentine Subandino of Salta, Tucuman, Cata- marca, and La Rioja provinces resembles the Valley because it is in subtropical latitudes where many Neotropic taxa approach their distributional limits. However, the Subandino differs from the Valley by its inland location and relatively high altitude (study sites between 900 and 2000 m.). Thus it has a much cooler tem- perature regimen that includes frequent winter frosts (annual aver- age temperature 13-15.5 degrees C. vs. 23.4 degrees C. for the Valley). Moreover, the Subandino is a genuine semidesert with only 80-250 mm. of rain per year (vs. 669 mm. for the Valley). Three years’ use of 10 Malaise Traps and frequent collecting trips at all seasons between 1966 and 1972 obtained from the northwest Subandino a mesostenine fauna of 10 genera and 33 species, including seven Neotropic taxa ( Dotocryptus , Diapeti- morpha, Basileucus, Polycyrtidea, Polycyrtus, Messatoporus, and Agonocryptus ), two Sonoran genera ( Compsocryptus and Meso- stenus of the Longicaudis group), and two Holarctic genera (Tra- chysphyrus and Mesostenus of the Transfuga group). Considering Polycyrtus almost certainly present in the Valley (it is in the east- ern U.S. and northeast Mexico), we get for the Valley and the Subandino a similarity index of 0.580, one surprisingly high for two localities separated by about 42 degrees of latitude. In fact, since the Valley genera Cryptanura, Bicristella, Mallochia, Pachy- somoides, and Joppidium reach at least the borders of the Suban- dino in northwest Argentina, the index of similarity eventually could prove as high as 0.8000. With regard to the Neotropic element, the above comparison of south Texas, coastal Peruvian, and Subandean mesostenines, only reemphasizes the vast South and Middle American distributions of so many of these genera. Species, of course, differ widely from place to place but at the generic level we find remarkable similarity 80 Psyche [March between faunas as distant as the south Texan and the northwest Argentine. This correspondence between Texan, Peruvian, and Argentine arid-adapted mesostenine communities also shows that the same versatile genera tend to survive anyplace in Middle and South America where local conditions become dry enough to elim- inate the wet forests which are the optimum habitat for most Neotropic mesostenines. Almost equally, it suggests a correlation between dry and cold tolerance, since nine of the ten Neotropic genera that reach well north in the eastern U.S. also occur in one or more of the dry areas studied (Crypt anura, Polycyrtus, Diapeti- morpha, Lymeon, Mai lochia, Acerastes, Pachysomoides, Messa- toporus, and Agonocryptus). Sonoran mesostenines also show much affinity among the three study areas. These genera are centered in Mexico and the south- western U.S. and evolved there in response to increased aridity and orogeny which affected that region in the latter half of the Tertiary. The Valley has four Sonoran taxa ( Joppidium , Lanugo, Compso- cryptus, and Mesostenus of the Longicaudis group) and two of these, Compsocryptus and the Longicaudis group of Mesostenus, are the only Sonoran mesostenines known from the Peruvian Des- ert and the Subandino. However, Lanugo reaches the Peruvian Andes and so might be found on the coast and Joppidium possibly enters both deserts, since it is known from Ecuador and turns up again in the Argentine Chaco. Most of these genera favor semiarid habitats, from thorn scrub to subtropical deciduous forests, and doubtless extended their distributions in the driest parts of the Tertiary and during interglacial xerothermic episodes. The present moderately wet interglacial has produced some notable Sonoran disjunctions, such as the above-mentioned case of Joppidium and that of Compsocryptus, which has many species in the western U.S. and Mexico, a single representative in Florida and Cuba, an isolated species in the Peruvian Coastal Desert, and another dis- junct species in northern Argentina. The Holarctic element, consisting mainly of genera adapted to temperate forests, shows more discontinuities among Middle and South American arid zones than do the Sonoran and Neotropic faunas. Gambrus, Trychosis, and Listrognathus reach only as far south as the Valley or northern Mexico. On the other hand, Trachysphyrus and the Transfuga group of Mesostenus occur in all three study areas. The Transfuga group is mainly Holarctic 1977] Porter — Mesostenines 81 and Andean, reaching its austral limit in northern Argentina and including only three or four South American species. Trachysphy- rus, in contrast, has a big Holarctic fauna plus a massive endemic South American radiation of more than 150 species centered in mountainous and/or semiarid to arid parts of the southern half of the continent. Thus the Subandino has 21 Trachysphyrus, the Coastal Desert 12, but the Valley only one. Interestingly, the south Texan species, T. mesorufus, is a member of the Planosae group, the only Trachysphyrus stock of South American origin to have invaded Middle and North America. The heterogeneous distribu- tional patterns of Holarctic genera represented in the Valley show that this element has penetrated Middle and South America in different expansions at widely separated times. Trachysphyrus, because of its huge endemic South American fauna, probably had reached the southern continent by the middle or late Tertiary (Raven and Axelrod, 1975, p. 422 point out that even by the late Cretaceous the “northern Andes” were “beginning to approach their modern configuration” and thus could have served all through the Tertiary as a suitable habitat for temperate-adapted genera invading from the north.). The Transfuga group of Mesostenus, mainly Holarctic and in South America practically confined to the Andean region excluding Chile, doubtless moved south with Pleis- tocene glaciations. Finally, Gambrus, Trychosis, and Listrognathus, which only reach northern Mexico perhaps were pushed as far as the Valley only by the most recent and severest Wisconsin glacia- tion. Phaenology Hand collecting and Malaise Trap records of Lower Rio Grande Valley mesostenines obtained between June 1973 and March 1977 are summarized in Table 1. Since hand collecting was possible only between 25 August-9 September, 18 December-25 January, 11-12 March (1-8 April in 1975) and 16 May-10 June, the data furnished in Table 1 show strong bias toward those periods. Fieldwork during February, October, and November, in particular, doubtless would have shown that these months have much larger mesostenine faunas than sug- gested in the table. Nonetheless, the phaenology emerges as uni- modal and invernal with an impressive peak that extends from December to March and includes 390 of the 679 specimens and Table 1. Phaenology of Valley Mesostenines 82 Psyche [March I I I ■ ~ ^ i I I I I I 1IU iii i ■ I oa I I I I I I I i n i i i i i i i i H * aj ^3 I I 1 I I I I I 1 1 1 I S Ov I I - I I I “ - - I I I & I w | I I I I vO — j | ''t | m | CM I I I I I i i i i M C o •” 3 C fi x I C e D £ W 3 c « I I 3 .52 53 o 3 S3 1 5 s, .52 3 3 C 3 .52 .3 5 3 3 .5 -C) o 2 3 -3 O .3 £ -3 3 .3 3 8{ 3 a, 3 (-S .52 3 5 3 5 3 3 ~3 -3 3 5, >3 -2 bo O S s 0 u U _• 77 « _ ^ ■'t — I 9 2 ^ rrt OO OO goo 3 co 03 03 00 3 O •H 3 3 6 3 0) rH > r3 O O 43 -H 3 4-) 3 3 3 3 •H 6 3 3 •H %H 3 N 03 3 3 3 4-1 3 tO 3 3 3 3 3 O 4-> 03 43 43 3 O O 3 3 3 3 3 i—l 3 3 CD 3 3 O O > •H M— 1 43 2 3 4-1 o O r-l •H 3 43 03 3 3 O 3 3 •H O O CD 3 43 3 O 3 £ 3 3 3 03 3 43 & 3 •r-l 92 4-) 3 -H 4-1 i 3 3 6 3 4= 3 O 3 O 3 o 3 >4-1 3 O O -3 t>9 3 3 3 3 m 3 O 3 3 03 •H 3 > 3 3 3 3 3 43 3 3 4-4 3 4J •H 4-1 3 o3 > ■H 43 4-1 3 3 03 O 3 3 03 •— ' 3 3 4-4 3 3 4-1 O e 3 3 3 3 i-H tH 3 O to 3 O 00 4-4 3 bC l 3 3 •H O 3 43 -3 3 3 3 3 3 3 0 3 3 6 33 [June 1977] Kanz — Monarch Butterfly Orientation 129 The orientation of summer laboratory-reared, non-migrant Mon- archs in transparent periphery cages was random with or without the sun visible (Table II), and was comparable to the random orientation shown by opaque periphery cage non-migrants and migrants in the absence of sun cues. Therefore, the presence of terrestrial cues over-rides the sun-orienting escape response of non-migrant Monarchs. It is possible that using terrestrial instead of sun cues each non-migrant butterfly persisted in its orientation cue for escape. These experiments demonstrate that the majority of non-migrant Monarchs chose terrestrial rather than sun cues for orientation when both were available. Fall migrants, exposed to both terrestrial and sun cues, continued to orient to the sun’s azimuth (Fig. 4, Table II) with an orientation closer to the sun’s azimuth from 1000-1400 hrs, EST, than either before or after this period (| 0-sun azimuth | value being 35°, 52° and 77°, respectively). An exception was seen in one sun-visible test, in which 0’s from most significantly grouped animals main- tained an approximately 240° heading (southwest). However, this occurred only once and one cannot determine whether terrestrial or celestial cues were used. Random orientation resulted with terrestrial cues present but sun clues absent (Table II). The per- sistent sun orientation of fall migrants, when the sun and terrestrial cues were visible, is additional support for the hypothesis that the sun orientation of fall migrants is a migratory response and not merely an escape response. Flight Directions Following Release Monarchs were released after each experiment and an azimuth reading taken on the vanishing direction of each butterfly using a Silva compass compensated for declination angle. Only those but- terflies that flew to the horizon were used in the analysis of Monarch vanishing directions. All Monarchs exhibited speed flight (Urqu- hart, 1960) immediately upon release. Occasionally Monarchs showed a feeding flight pattern with short randomly directional flights between flowers. When feeding flight took a butterfly to the horizon of the test field within 2 minutes, its azimuth at the periphery was included in the analysis; if feeding flight persisted, the vanishing azimuths of such animals were excluded from com- putations. Flight altitudes of released Monarchs were evenly di- vided between those above and below approximately 8 meters. 130 Psyche [June sun azimuth line TIME OF DAY IN HOURS (EST) Fig. 4. Orientation displayed by fall migrants in a transparent periphery cage on a clear day. In contrast to summer non-migrants, fall migrants continue to sun-orient in the presence of both sun and terrestrial cues. Readings beginning at 1345 hrs (EST) were an artifact of a sudden ambient temperature drop. The results of summer Monarch releases with the sun visible showed that most vanishing azimuths appeared to be down-wind rather than toward the sun (Fig. 5). Only the releases of (c), (f) and (h) were significantly grouped about their respective mean orientation direction. Vanishing azimuths for summer non-mi- grants on overcast days were down-wind (Fig. 6). The distribution of releases in all but (c) were significant about their respective 6' s. The releases of (f), with only a light surface wind, appear to be in the direction of the sun’s azimuth even though the sun’s position was obscured by clouds. However, most of these butterflies flew at an altitude ^ 20 m and, therefore, likely encountered stronger winds. The vanishing directions of summer Monarchs, therefore, appeared more influenced by wind than by sun. Vanishing azimuths were more scattered with light winds (^ 5 mph or 8 km/ hr). Low flight enabled Monarchs to fly against head winds that exceeded 10 mph (16 km/ hr). Thus, Monarch flight direction was greatly influenced by the wind but was not competely determined by wind direction. Figure 7 shows the patterns of vanishing azimuths for fall mi- grants when the sun was visible. All 20 distributions were signifi- cant about their respective 0’s. The vanishing azimuths generally corresponded to the direction in which the wind was blowing with the exception of (a), (m) and (t). The releases of (a) occurred from a site that was surrounded by water except to the west; Urquhart 1977] Kanz — Monarch Butterfly Orientation 131 g h i Fig. 5. Release orientation for summer monarchs on clear days. For each circle: Dot indicates the vanishing direction of a butterfly; star indicates sun position at the time of release; radial arrow indicates direction in which wind was blowing and its velocity in miles per hour; N indicates true north. Most vanishing azimuths are seen to be down-wind. Fig. 6. Release orientation for summer monarchs on overcast days. Symbols are as in Fig. 5. Most vanishing directions were down-wind. (1960) has observed that Monarchs tend to avoid flying over water when possible. Most flights in (m) were low to the ground, as was true for the majority of flights showing vanishing azimuth into the wind. The releases of (t) occurred at 1600 hrs (EST) when migrants return to roosting trees for the night. The roosting trees for mi- grants at Gloucester were west and northwest of the release and feeding sites. Thus, the migrants of (t) were possibly returning to their roosting trees; however, it is unclear why the butterflies in (c) chose a similar direction at 1300 hrs (EST) with the same wind velocity as in (t). Vanishing directions appeared to be down-wind, particularly when winds exceeded 10 mph (16 km/ hr) unless the butterfly flew low to the ground. Two-way analysis of variance between the mean angular deviation of 22° for (h), (i), (k), (1), (n) and (o) and the mean angular deviation of 42° for the remaining fourteen release distributions indicate significant differences (F = 9.5, df = 16, P< .01). In the former case, the wind was blowing in the same direction (southwest) as the sun’s azimuth. In the latter case, the wind and sun azimuth directions did not coincide. Analysis of variance of 1977] Kanz — Monarch Butterfly Orientation 133 Fig. 7. Vanishing directions of fall migrants on clear days. Symbols are as in Fig. 5. Most releases flew down-wind but distributions for north and northeast winds are tighter than those for southerly winds. See text for discussion. 134 Psyche [June N N Fig. 8. Vanishing directions for fall migrants released on overcast days. Sym- bols are the same as in Fig. 5. Monarchs flew generally down-wind. angular deviations for winds of 13 mph (21 km/ hr) shows that the mean AD of 17° for distributions (h), (k) and (n) with the sun and wind in the same direction was significantly smaller than the mean AD of 28° for distributions (g), (j) and (p) with sun and wind di- rections dissimilar (F = 4.9, df = 5, P< .1). The same was true for winds of 16 mph (26 km/ hr) where the mean AD (18°) for (i), (1) and (o) with sun and wind to the southwest was significantly smaller than the mean AD (46°) for (d), (e), (f), (m), (r) and (s) with the sun and wind directions dissimilar (F = 30.8, df = 8, P < 0.005). Therefore, these data suggest that fall migrants were not just flying with prevailing winds but were also orienting toward the sun. The vanishing azimuths for fall migrants released on overcast days were predominantly down-wind and each distribution was significant about its 6 (Fig. 8). Releases (c) and (d) occurred at 1630 hrs and 1700 hrs (EST), respectively, and could be examples of the roosting orientation described for Fig. 7 (t). Thus, although the sun is an important cue in oriented flight, fall migrants utilize favorable winds to facilitate migration, but dis- play oriented flight without the aid of the wind. 1977] Kanz — Monarch Butterfly Orientation 135 Discussion The experiments reported here were designed to delineate the role of the sun in Monarch orientation. The sun was selected as the most probable cue in the orientation of Monarch butterflies for several reasons: (1) the Monarch is a diurnal animal and the sun is a prominent cue in its environment; (2) a positive phototaxis has been reported in a number of Lepidoptera (Brandt, 1934; Col- lins, 1935; Dolley, 1916; Jander, 1963; and Kelsheimer, 1935); (3) the sun has been shown to be important in the orientation of a number of animals (Hasler, 1967; Schmidt-Koenig, 1961; Taylor and Ferguson, 1969; and von Frisch, 1967), including migrating European butterflies (Baker, 1968a,b). Non-migrant Monarchs (laboratory-reared) demonstrated a sun orientation when tested in an opaque periphery orientation cage with the sun visible. This response was termed an escape response. A sun-orienting escape response for Monarch butterflies is appro- priate for three reasons. First, the most prominent orientation cue available to Monarchs in opaque periphery cages on a clear day is the sun. Second, when fast escape flight is warranted, Monarch escape would be linear and, therefore, fastest when the animals use a constant cue, such as the sun, for orientation. Third, when fol- lowed by a predator, a sunward escape response would put the sun in the predator’s line of sight thus making it more difficult for the Monarch to be detected. The orientation of summer non-migrants was random with both terrestrial and celestial cues present. If Monarchs were attempting to escape, the sun did not appear to be their orientation cue. This random orientation was believed to be indicative of the orientation of uncaged, non-migrant, Monarchs during the summer, since sum- mer animals are known to wander randomly (Urquhart, 1960). Verheijen (1958) has criticized phototaxis experiments on the basis that the test situations eliminated scattered and reflected light, therefore making the illumination of the animal’s environment unnatural. Illumination conditions (as well as conditions in general) during Monarch testing in transparent periphery cages, more closely approximated a natural field situation for Monarchs than the con- ditions encountered with opaque periphery cages. Thus the re- sponses of transparent periphery cage butterflies might be expected 136 Psyche [June to more accurately represent Monarch orientation. It has been assumed that Lepidoptera displaying positive photo- taxis (Johnson, 1969) were flying directly toward a light source. Hsiao (1973), however, has found that the corn earworm moth, Heliothis zea, flies toward a dark band surrounding the light source. Hsiao suggested that this Mach band explanation (Graham, 1966) could explain the attraction of night-flying moths to ultraviolet light sources: moths seek darkness characteristic of their diurnal behavior, although they appear attracted to ultraviolet light. Hsiao’s results raise the possibility that sun orientation of Monarch butter- flies is an orientation to either (1) a Mach band surrounding the sun, or (2) a Mach band perceived between the sun and the darker horizon. However, the Monarch is a diurnal butterfly, not a noc- turnal moth, and butterflies generally seek sun-light areas instead of shaded areas (Klots, 1961). Data from experiments with caged butterflies might represent landing orientation. However, when observations were made (at times ranging from 1-15 min after cage rotation), few, if any, Monarchs were in flight or landing. In general, the butterflies were either stationary or walking. The location and shape of the horizon in opaque vs transparent periphery cages could also have affected Monarch orientation. Orientation to mountain tops by many Coccinellidae (Hagen, 1962), and to tree tops by the Scolytid beetle, Conophthorus coniperda (Henson, 1966), is presumably based on horizon-orientation. Never- theless, sun orientation by fall migrant Monarchs persisted in spite of horizon differences between opaque and transparent periphery orientation cages. It seems unlikely, therefore, that the cage ori- entation of these insects was significantly affected by horizon dif- ferences. The orientation displayed by summer non-migrants was different in the two types of cages. This difference might reflect an horizon influence, but other factors, such as the presence or absence of terrestrial cues, are equally likely. When the sun was visible, fall migrant Monarchs oriented to the sun’s azimuth regardless of the type of orientation cage. In con- trast, non-migrant orientation was sunward only in opaque periph- ery cages. One explanation for this difference in orientation re- sponse could be that a sun-orienting escape response is stronger in fall migrants than in summer non-migrants. Tables I and II show that the sun orientation of fall migrants (opaque or transparent 1977] Kanz — Monarch Butterfly Orientation 137 periphery cages) was not as conclusive, statistically, as the sun ori- entation of summer non-migrants in opaque periphery cages. Fur- thermore, a stronger escape response could be expected to be posi- tively correlated with a greater overall level of activity within a cage. It Was found, however, that cage activity during fall tests was less than cage activity during summer tests. The tendency of fall migrants to orient more closely to the sun’s azimuth during the observed fall migratory period (1000-1400 hrs, EST) was another feature distinguishing fall migrant sun orienta- tion from escape response sun orientation. A sun orientation re- stricted to the period 1000-1400 hrs would offer several advantages to Monarchs migrating south: (1) A restricted sun orientation en- compasses an arc of 60° to 70° compared with an arc of 160° to 180° resulting from all-day sun orientation and the 60°-70° arc rarely deviates from the desired south to southwest migratory di- rection. (2) Consequently, the distance traveled, and time and energy expended, would be less. Tunmore (1960) has suggested a similar scheme for bird navigation. (3) A restricted sun orientation also obviates the necessity of sun-compass orientation to explain the precision of the Monarch’s long-distance fall migration. The data suggest that the restricted sun orientation was independent of temperature (for ambient temperatures greater than 1 3° C). (4) Since the highest autumn temperatures generally occur between 1000 hrs and 1400 hrs (EST), fall migrants would be migrating during the warmest part of the day. If fall migrants use sun orientation, then spring migrants return- ing north might use a negative sun orientation. Reversed orienta- tion by insects between leaving and returning to a site is well known (Geir, 1960; Johnson, 1969; Kennedy and Booth, 1963; Pickens, 1934; and Shephard, 1966). Monarch migrations are undoubtedly affected by winds. How- ever, while Monarch migrations appear to be aided by prevailing winds, they are not as dependent on them as locusts (Waloff, 1946, 1958) and aphids (Johnson, 1954, 1969). Figure 7 showed that the distributions of release azimuths were tighter when migratory and down-wind directions coincided than when the two directions dif- fered. Furthermore, several instances were recorded (Kanz, 1973) where fall and spring migrants were engaged in directed migratory flights with little or no wind. The prevailing surface wind patterns for up to 500 m altitude (Prevailing Direction, Mean Speed and 138 Psyche [June Fastest Mile of Wind, U.S. Weather Bureau) for September and October would facilitate a southwestward movement of Atlantic Coast migrants in the fall, and for March and April would facili- tate a north and northeastward movement of migrants passing north through Mexico and Texas. Prevailing wind patterns for March and April suggest a possible explanation for why the pop- ulation of Monarch butterflies in the United States (excluding the West Coast population) is proportionally greater east than west of the Mississippi River (Urquhart, personal communication). Strong March and April winds from the north, west, and northwest in northern Texas and Nebraska, could force spring Monarchs, ori- enting by a negative sun-orientation, to the east and northeast. Therefore, even a broad northerly orientation for spring migrants might still result in biasing the summer population toward the east- ern half of the United States. Thus, it might not be necessary for Monarchs to possess a restricted negative sun orientation in order to assure a northeasterly movement in the spring. Summary Non-migrant and fall migrant male and female Monarch butter- flies, Danaus plexippus L., orient toward the azimuth of the sun when confined in circular orientation cages with only celestial cues present. When both terrestrial and celestial cues are present, non- migrants exhibit random directionality similar to the flight of free- flying summer non-migrants while fall migrants orient to the sun’s azimuth. Both fall migrants and non-migrants exhibit a random cage distribution under overcast sky with or without terrestrial cues. The sun orientation of fall migrants is believed to be a mi- gratory response resulting in a southward movement. Such ori- entation differed from the sun orientation of non-migrants which appears to be an escape response. Upon release, migrants and non-migrants tend to fly with the wind. No conclusive indication of sun-compass or polarized light orientation in migrants or non- migrants was evident. No sex differences in orientational responses were observed. Acknowledgments I wish to thank Dr. E. S. Hodgson for his advice and support during this study. In addition, I am indebted to Dr. F. A. Urquhart of the University of Toronto and Dr. L. P. Brower of Amherst 1977] Kanz — Monarch Butterfly Orientation 139 College for their help in establishing a laboratory colony of Mon- arch butterflies. I would also like to thank Ms. Mary Kanz for her assistance throughout the field studies. This research was sup- ported, in part, by funds derived from an NIH Career Develop- ment Award to Dr. K. D. Roeder, Tufts University. References Baker, R. R. (1968a). A possible method of evolution of the migratory habit in butterflies. Phil. Trans. R. Soc. (B), 253, 309 341. Baker, R. R. (1968b). Sun orientation during migration in some British butterflies. Proc. R. ent. Soc. (A), 43, 89-95. Batschelet, E. (1965). Statistical methods for the analysis of problems in animal orientation and certain biological rhythms. A1BS Monograph, Washington, DC. Brandt, H. (1934). Die Lichtorientierung der Mehlmottle Ephestia Kuehniella Zeller. Z. vergl. Physiol., 20, 646-673. Collins, D. L. (1935). Comments upon phototropism in the codling moth with reference to the physiology of the compound eyes. J. econ. Ent., 28, 103-106. Dolley, W. L. (1916). Reactions to light in Vanessa antiopa, special reference to circus movements. J. exp. Zool., 20, 357-420. Geir, P. W. (1960). Physiological age of codling moth females ( Cydia pomonella (L)) caught in bait and light traps. Nature, Lond., 185, 709. Graham, C. H. (1966). Vision and Visual Perception. New York: John Wiley and Sons. Greenwood, J. & Durand, D. (1955). The distribution of length and components of the sum of n random unit vectors. Ann. math. Statist., 26, 233-246. Hagen, K. S. (1962). Biology and ecology of predaceous Coccinellidae. A. Rev. Ent., 7, 289-326. Hasler, A. (1967). Underwater guideposts for migrating fishes. In: Animal Ori- entation and Navigation, pp. 1-20. Corvallis, Oregon: Oregon State University Press. Henson, W. R. (1966). The analysis of dispersal mechanisms in Conophthorus coniperda Sz. Biometeorology 2. Proc. Ill Int. Congr. Biomet. (Pau, 1963), pp. 541-549. Hsiao, H. S. (1973). Flight paths of night-flying moths to light. J. Insect. Physiol., 19, 1971-1976. Jander, R. (1963). Insect Orientation. A. Rev. Ent., 8, 95-114. Johnson, C. G. (1954). Aphid migration in relation to weather. Biol. Rev., 29, 87-118. Johnson, C. G. (1969). Migration and Dispersal of Insects by Flight. London: Methenun and Co., Ltd. Kammer, A. E. (1968). Motor patterns during flight and warm-up in Lepidoptera. J. exp. Biol., 48, 89-109. Kammer, A. E. (1970). Thoracic temperature, shivering, and flight in the monarch butterfly Danaus plexippus (L). Z. vergl. Physiol., 68, 334-344. 140 Psyche [June Kanz, J. E. (1973). The orientation of non-migrant and migrant monarch butter- flies ( Danaus plexippus). Ph.D. Dissertation. Medford, Massachusetts: Tufts University. Kelsheimer, E. G. (1935). Response of European corn borer moths to colored lights. Ohio J. Sci., 35, 17 28. Kennedy, J. S. & Booth, C. O. (1963). Free flight of aphids in the laboratory. J. exp . Biol., 40, 67-85. Klots, A. B. (1951). A Field Guide to the Butterflies. Boston: Houghton Mifflin Co. (The Riverside Press). Nielsen, A. & Nielsen, E. T. (1952). Migration of the pieride butterfly Ascia monuste L. in Florida. Ent. Meddr., 26, 386-391. Pickens, A. L. (1934). Termites and Termite Control {Ed. by C. A. Kofoid). Los Angeles: University of California Press. Prevailing Direction, Mean Speed, and Fastest Mile of Wind; from National Atlas of the United States. U.S. Weather Bureau, U.S. Department of Commerce, Washington, D.C. Schmidt-Koenig, K. (1961). Sun navigation in birds? Nature, Lond., 190, 1 025— 1026. Shephard, R. F. (1966). Factors influencing the orientation and rates of acitivity of Dendroctonus ponderosae Hopkins (Coleoptera: Scolytidae). Can. Ent., 98, 507-518. Stockhammer, K. (1959). Die Orientierung nach der Schwingungsrichtunglinear polarisierten Lichtes und ihre sinnesphysiologischen Grundlagen. Ergebn. Biol., 21, 23-56. Summarized in: The Dance Language and Orientation of Bees by Karl von Frisch (1967). Cambridge, Mass.: Harvard University Press. Tables of Computed Altitude and Azimuth, Latitudes 40°-49° N. Inclusive (1962). U.S. Navy Hydrographic Office, Publication No. 214, Volume V. Washington, D.C.: U.S. Government Printing Office. Taylor, D. H. & Ferguson, D. E. (1969). Solar cues and shoreline learning in the Southern Cricket Frog, Acris gryllus. Herpetologica, 25, 147-149. The National Almanac, for Years 1970-1973. Issued by the U.S. Naval Observa- tory. Washington, D.C.: U.S. Government Printing Office. Tilden, J. W. (1962). General characteristics of the movement of Vanessa cardui (L). J. Res. Lepidop., I, 43-49. Tunmore, B. G. (1960). A contribution to the theory of bird navigation. Proc. XII Int. Ornith. Cong., Helsinki, pp. 718-723. Urquhart, F. A. (1960). The Monarch Butterfly. Toronto, Canada: University of Toronto Press. Urquhart, F. A. (1976). Found at last: The Monarch’s winter home. National Geographic, 750(12): 161-173. Urquhart, F. A. & Stegner, R. W. (1966). Laboratory techniques for maintain- ing cultures of the monarch butterfly. J Res. Lepidop., 5, 129-136. Verheijen, F. J. (1958). The mechanisms of the trapping effect of artificial light sources upon animals. Archs. neerl. Zoo/., 13, 1 107. von Frisch, K. ( 1967). The Dance Language and Orientation of Bees. Cambridge, Mass.: Harvard University Press. 1977] Kanz — Monarch Butterfly Orientation 141 Walker, J. J. (1914). The geographical distribution of Danaida plexippus L. ( Danais archippus F.) with especial reference to its recent migration. Ento- mologist’s mon. Mag., 50, 181-193, 224-237. Waloff, Z. (1946). A long-range migration of the Desert Locust from Southern Morocco to Portugal, with an analysis of concurrent weather conditions. Proc. R. ent. Soc. (A), 21, 81-84. Waloff, Z. (1958). The behaviour of locusts in migrating swarms. Proc, X Int. Congr. Ent., Montreal, 2, 567-570. Williams, C. B. (1951). Seasonal changes in flight direction of migrant butterflies in the British Isles. J. Anim. Ecol., 20, 180-190. Williams, C. B. (1958). Insect Migration. London: Collins Press. Williams, C. B., G. F. Cockbill, M. E. Gibbs, J. A. Downes (1942). Studies in the migrations of Lepidoptera. Trans. R. ent. Soc., Lond., 92, 101-283. REDESCRIPTION OF XENICOPODA MOORE AND LEGNER (COLEOPTERA: STAPHYLINIDAE, OMALIINAE), WITH SUPPLEMENTARY NOTES* By Margaret K. Thayer Museum of Comparative Zoology, Harvard University, Cambridge, Mass. 02138 Introduction Moore and Legner (1971) described a new genus and species of Omaliinae, Xenicopoda helenae, on the basis of a unique specimen of undetermined sex collected on Mt. Wilson, California. The ma- jor distinguishing character they noted for this genus (op. cit. and Moore and Legner, 1974) was the bizarrely modified protarsi, quite unlike any previously described in the Omaliinae. Since then I discovered a series of six X. helenae in the H. C. Fall Collection at the Museum of Comparative Zoology. Exam- ination of these specimens showed that the unusual front tarsi found on the type are present only in males. Females have normal female omaliine protarsi, slender and with equal tarsal claws. This distinct difference between the sexes and the availabil- ity of additional specimens of Xenicopoda made it seem worth- while to publish an amplified description of the genus. Detailed study of the type and the six other specimens also led to the dis- covery of a number of discrepancies between the specimens and the original description. The proportions as given in the description and as shown in the original habitus drawing are rather distorted, owing partly, perhaps, to the fact that the type is somewhat curled up. Proportions based on careful measurements of the seven avail- able specimens will be mentioned below. The head shape in the original drawing is rather distorted as well, and a new figure is given. * Manuscript received by the editor November 3, 1977 142 1977] Thayer — Redescription of Xenicopoda 143 Redescription Length: mean 2.5 mm (2. 0-2. 7 mm), measured as is (slightly curled); estimated actual length 3.0 mm (2. 5-3. 6 mm). Head about seven-tenths as long as wide (from clypeal apex to nuchal constriction), epistomal sulcus absent (fig. 3). Vertex with well-impressed dorsal tentorial pits and small, rather obscure pale ocelli. (The dorsal tentorial pits are presumably what Moore and Legner (1971) meant by anterior tentorial pits. The latter are ac- tually very slight depressions antero-medial to the antennal inser- tions, indicated by short lines there in figure 3.) Weak nuchal constriction across dorsal surface just behind ocelli. Faint micro- sculpture on dorsal and ventral surfaces of head. Labrum nearly rectangular with rounded anterior corners, about 2.2 times as wide as long; margins entire. Mandibles as illustrated (figs. 1, 2) with well-developed molar areas composed of many small, sharp, buc- cally directed teeth. Maxillary palp four-segmented, more or less filiform, with first segment small; second and third segments larger, subequal to each other, more or less obconical; fourth segment at least twice as long as wide (2.1-3.3x, mean 2.5x), tapering toward apex, the whole segment twice as long as the third, but narrower than its apex. Labial palp three-segmented, each segment slightly longer and narrower than the preceding; third segment two to three times as long as wide. Gular sutures distinctly separate, closest at a level just before the hind margins of the eyes and diverging an- terior and posterior to this. Antenna filiform, basal segments dis- tinctly longer than wide, more distal ones becoming successively shorter and broader up to tenth segment, which is very slightly wider than long. First five or six antennal segments glabrous ex- cept for sparsely scattered long setae; segments six or seven to eleven with shorter setae in addition. The shorter setae become progressively denser on the more distal segments, while the longer setae diminish in number and become increasingly restricted to the apical area of each segment. Pronotum about seven-tenths as long as wide, about half as long as elytra; fairly evenly convex except for a small median basal de- pression; lateral margins evenly arcuate, slightly explanate in basal half; reticulate microsculpture on dorsal surface. In ventral view, postcoxal process of pronotum extends about halfway from lateral 144 Psyche [June 1977] Thayer — Redescription of Xenicopoda 145 pronotal margin to midline of prothorax. Procoxa and protro- chantin very strongly and sharply carinate externally. Prosternum with short acute process barely extending between procoxae ex- ternally. Mesosternum not longitudinally carinate, having a fairly even surface with no distinct depressions fitting against procoxae; with a short rounded process between mesocoxae. Mesocoxae slightly separated, but meso- and metasternum not touching exter- nally. Metasternum without external process between mesocoxae, with a short pair of processes between metacoxae. Metacoxa tri- angular in ventral view, with posterior surface slightly excavate. Apex of coxa slightly explanate laterally, overlapping part of the trochanterofemoral joint when the leg is in a retracted position. All tibiae with vague row of spines along part or all of outer face. Tibiae and tarsi slender except male prolegs as described below. Metatarsus about three-fourths as long as metatibia, first four seg- ments subequal in length, fifth about twice as long as each of first four. Pair of empodial setae between claws on all tarsi, generally about half as long as claws. Elytra together about 1.1 times as long as wide; without micro- sculpture between punctures; probably extending to about the apex of tergite 4 in life. Elytral epipleuron delimited by a distinct lateral keel. Wings fully developed, with a typical omaliine folding pat- tern (my unpublished data). Abdomen with fine reticulate microsculpture on dorsal and ven- tral surfaces; intersegmental membranes with brick-wall pattern (with occasional irregularities) typical of Omaliinae (see Hammond, 1971); sternites of segments 2 and 3 apparently without a keel be- tween metacoxae; tergites 2 and 3 somewhat sclerotized, 4 and fol- lowing more so; tergites 4 and 5 each with a pair of small patches of medially-directed microtrichia (“pruinose” or “tomentose” spots of authors; see fig. 4); only segments 3 and 7 bearing paratergites (“margined segments”), segment 2 with sternites extending onto dorsal surface, other segments with narrow membranous joint di- Figs. 1-8. Xenicopoda helenae Moore and Legner. 1-2. Mandibles; 1., ventral; 2., dorsal view. 3. Head, dorsal view (large circles = ocelli, small circles = dorsal tentorial pits). 4. Abdomen, segments 2-8 and (male) genital segment, dorsal view. 5. Eighth abdominal sternite and external female genitalia, ventral view. 6-8. Male genitalia; 6., Genital segment, ventral view; 7., Aedeagus (as positioned within ab- domen), dorsal view; 8., Genital segment, dorsal view. Membranous areas of geni- talia stippled. Scale line = 0.5 mm. 146 Psyche [June rectly between tergite and sternite; sternite 8 with median basal process as illustrated in figure 5. Male: Protibia abruptly broadened just beyond base, its maxi- mum width about twice that of a mesotibia; apical half of outer face with an irregular row of spines intermixed with a few setae; rounded notch at apex on outside of tibia. Protarsus with first four tarsomeres expanded: the first a pedunculate triangle, next three roughly triangular with their anterior apical corners succes- sively more prolonged; ventral surfaces of first four segments with large strap-like setae (except medially); fifth tarsomere distinctly curved ventrally, apex twice as wide as base. Anterior protarsal claw much longer and thicker than all other tarsal claws, about four-fifths as long as protarsus. Posterior protarsal claw normal. Empodial setae on protarsus shorter than usual, about one-fourth as long as posterior claw. Peg setae (see Hammond, 1972) appear to be absent from legs. Genital segment and aedeagus as in figures 6-8. Aedeagus with parameres dorsal within abdomen, internal sac with dense armature. Female: Protibia slender, similar to meso- and metatibia, spinose along entire outer face. Protarsus narrow (as meso- and metatar- sus), fifth tarsomere only slightly wider at apex than at base; nor- mal slender setae on ventral surface of tarsomeres 1-4. All tarsal claws similar in size and shape. Genitalia as in figure 5, sclerotized spermatheca apparently absent. Material examined: CALIFORNIA: Los Angeles Co.: Mt. Wilson, 6— III— 46, G. P. Mackenzie (Holotype, male) [California Academy of Sciences]; Pasadena, Echo Mt., 18-III-16, 3500 ft., (1 male, 1 female) [Museum of Comparative Zoology]; (Los Angeles Co.?) Pomona Mts., 11-22 (2 males, 1 female) [MCZ]. Santa Barbara Co.: Santa Barbara, 8— II— 9 1 (1 male) [MCZ]. Discussion The distinctive protarsi of Xenicopoda males may be modified to facilitate grasping females during copulation. There seem to be no corresponding special structures in females, but both the pro- notum and the elytra have fairly sharp lateral margins. Assuming that a male mounts a female dorsally (I have collected Eusphalerum mating this way), he might use his protarsi to grasp her pronotum or elytra in either of two ways: 1) with his tarsi dorsal and tarsal claws ventral to the lateral edge of the body; 2) with the anterior 1977] Thayer — Redescription of Xenicopoda 147 side of the tarsus (including anterior protarsal claws) dorsal and posterior side ventral to the lateral edge of the body. The bifurcate nature of the second to fourth protarsal segments of the male and the lack of setae along the midline of the tarsus lend some credence to the latter hypothesis, but of course only direct observation of mating can confirm or deny any of this speculation. Why only one genus, out of all known Omaliinae, has these tarsi remains a mys- tery. The large strap-like setae on male Xenicopoda protarsi also may be an aid to grasping females in copulation. The presence of modified protarsal setae in at least the males is characteristic of nearly all omaliine genera I have seen (approximately 40, of which at most 5 lack these setae entirely). The form of the modified setae varies: some, like those of Xenicopoda, are strap-like, while others are spatulate, more or less like those of Xanthonomus Bernhauer, as illustrated by Steel (1955). Those of other genera form a con- tinuum between these two types. Rarely, females also have modi- fied setae on the protarsi, and in Eusphalerum, Amphichroum, and Pelecomalium, modified setae are found on all tarsi of both sexes, although in all these cases the setae of the males seem to be broader than those of the females. Males of several genera have modified setae on their mesotarsi as well as on their protarsi. Most Omaliinae have a pair of paratergites on the second through seventh abdominal segments. In their description of Xenicopoda, Moore and Legner (1971) stated that paratergites are present on the fourth and fifth “visible abdominal segments” (=sixth and sev- enth segments), although their figure seems to show paratergites on the seventh and eighth segments. Examination of a cleared Xeni- copoda specimen reveals that only the third and seventh segments bear paratergites. The second segment appears at first to have paratergites, but closer examination reveals that the sternite ex- tends continuously onto the dorsal surface, whereas there is a mem- branous articulation between paratergites 3 and 7 and their respec- tive sternites. Xanthonomus appears to have a similar abdomen, but has paratergites present only on the seventh segment (Steel, 1955, description and figs. 1-2; also I have examined specimens of an apparently undescribed Xanthonomus sp. in the Bernhauer Col- lection). I do not intend to imply, however, that these two genera are related because of their similarity in abdominal structure. To include females of Xenicopoda in Moore and Legner’s (1974) key to North American omaliine genera, couplet 29 should be 148 Psyche [June replaced by the following: 29(28) Abdomen lacking paratergites except on segments 3 and 7; male with large unequal protarsal claws, anterior one much longer and thicker than posterior, sometimes nearly as long as tarsus Xenicopoda Moore & Legner — Abdomen with paratergites on segments 3 through 7; protarsal claws equal in both sexes, same size as those on meso- and metatarsi 30 There are no ecological data on any of the specimens seen. Exam- ination of one cleared specimen, however, revealed the gut to be packed with pollen grains, as in Eusphalerum spp., Amphichroum spp., Pelecomalium spp., and some Elonium spp. which are found on flowers. The presence of a mandibular mola composed of small sharp teeth is fairly restricted within the Omaliinae, but all of the above-named genera except Elonium share this character with Xeni- copoda. ( Brathinus spp. and Olophrum spp., which are not flori- colous, also have molar surfaces composed of separate teeth, but the teeth differ in size, shape, and orientation from those of Amphi- chroum, Eusphalerum, Pelecomalium, and Xenicopoda.) Pollen- feeding in Elonium may well be a secondary development, as most species of this genus seem not to be found on flowers; this possi- bility makes the lack of a toothed mola in the flower-dwelling Elonium species less surprising. The other genera mentioned are apparently entirely floricolous as adults. This evidence and the collection dates on the seven known specimens of Xenicopoda sug- gest that an intensive search in February and March on flowers in the fairly restricted area where the genus has been collected might turn up additional specimens of this interesting beetle. The genus Xenicopoda was placed in the tribe Anthophagini by its authors. For the time being it may remain there, pending badly- needed further study of the higher classification of the Omaliinae. Acknowledgements I would like to thank D. H. Kavanaugh, California Academy of Sciences, for the prompt loan of the type of Xenicopoda helenae; H. S. Dybas and E. H. Smith, Field Museum of Natural History, for the opportunity to study the type of Xanthonomus toxopeanus 1977] Thayer — Redescription of Xenicopoda 149 (Bernhauer) and to borrow specimens of Xanthonomus sp.; Nancy Hinnebusch for typing the manuscript; and especially my husband, A. F. Newton, Jr., for continual encouragement and advice in the preparation of this paper. References Cited Hammond, P. M. 1971. The systematic postition of Brathinus LeConte and Camioleum Lewis (Coleoptera:Staphylinidae). J. Ent. (B) 40(1): 63-70. 1972. The micro-structure, distribution and possible function of peg-like setae in male Coleoptera. Ent. Scand. 3: 40-54. Moore, I. and E. F. Legner. 1971. A new genus and species of rove beetle from California (Coleoptera: Staphylinidae). Coleopt. Bull. 25: 51-53. 1974. Bibliography (1758 to 1972) to the Staphylinidae of America north of Mexico (Coleoptera). and Keys to the genera of the Staphylinidae of America north of Mexico exclusive of the Aleocharinae (Coleoptera: Staphylinidae). Hilgardia 42(16): 511-563. Steel, W. O. 1955. Notes on the Omaliinae (Col., Staphylinidae) (7) The genus Xantho- nomus Bernhauer. Entomol. Mon. Mag. 91: 275-278. ASSOCIATIONS BETWEEN FLIES AND SPIDERS: BIBIOCOMMENSALISM AND DIPSOPARASITISM?* By Michael H. Robinson and Barbara Robinson Smithsonian Tropical Research Institute P.O. Box 2072, Balboa, Canal Zone, Panama There are numerous records in the arachnological and entomo- logical literature of relationships between spiders and flies other than the simple case of predator and prey. Bristowe ( 1 94 1 :362— 370) reviewed a number of cases of parasitism and commensalism. Flies of the superfamily Drosophiloidea are involved in a number of more or less complex relationships with spiders. Chloropids parasitize spiders’ egg cocoons and may actually perch on adult spiders (Bris- towe, 1941:367) while milichiids share food with spiders (Richards, 1953). McMillan (1975) has recorded an association between mili- chiid flies of the genus Desmometopa and two species of large Aus- tralian orb-weaving spiders. The flies moved about the host web and fed on prey items as they were being consumed by the spiders. In addition, the milichiid moved onto the host and apparently cleaned the mouthparts and anal region of the spider. McMillan does not state whether the flies remained on the spiders when they were not actively cleaning them nor does he state how many flies were present on the spider at any one time. We here report on several different associations between flies and spiders, all of which are commensal (in the broadest sense). We found milichiids asso- ciated with the golden-web spider Nephila clavipes, unidentified flies were found as commensals of Argiope savignyi, and chloropid flies were found in a similar relationship with Argiope argentata. All these relationships were discovered in Panama. We describe a case of milichiid commensalism with a predatory hemipteran and suggest that the complex relationship between Nephila and the milichiids may have evolved from such a relatively simple stage. We think that the term commensalism is not sufficiently specific to describe some of the relationships reviewed here and suggest two possible additions to the terminology of symbioses. * Manuscript received by the editor October 31 , 1977. 150 1977] Robinson & Robinson — Flies and Spiders 151 Figure 1. Dorsal surface of cephalothorax and part of abdomen of adult female Nephila clavipes. Eight milichiid flies can be counted on the spider. 152 Psyche [June Nephila clavipes and flies of the genus Phyllomyza We first found flies associated with Nephila clavipes in January 1976, on Barro Colorado Island, Canal Zone, Panama. We sub- sequently found similar flies associated with this spider at a number of localities in the Canal Zone and elsewhere in Panama. Flies from four adult female N. clavipes at four different sites were col- lected. They were identified (see acknowledgments) as belonging to the genus Phyllomyza and all belonging to the same (unde- termined) species. All eleven insects were females. The flies rest on the dorsal surface of the spider and usually aggregate on the cephalothorax. Figure 1 shows eight flies resting on this area. The flies remain on the spider for long periods of time and are virtually inactive. When we set out to determine what the flies were doing sitting on the body of the spider we ran into a major practical problem. The Drosophila-sized flies were really too small to observe with the unaided eye. This problem was solved by adapting a stereo-binocular microscope for horizontal use, mounted on a camera tripod (Robinson & Smythe, 1976). With this device, under field conditions we could watch the insects under 10X or 20X magnification. It became apparent that on the cepha- lothorax of the spider the flies were not doing anything other than grooming themselves, sporadically shifting position and occasion- ally defecating. The bodies looked entirely normal and there was no evidence of oviposition or of penetrative feeding on the spider itself. (The mouthparts of milichiids could clearly not be used for piercing the spider’s cuticle and sucking its internal fluids, but at this stage we did not know what the flies were.) Eventually we decided to feed the spider. This went through all phases of its predatory behavior without disturbance to the flies. Prey capture involved rushing out to attack the prey, biting it, wrapping it in silk, removing it from the web, transporting it back to the hub and there wrapping it once again before hanging it and feeding (details in Robinson & Robinson, 1973, for Nephila maculata ap- ply broadly to N. clavipes). This predatory sequence involves a great deal of violent movement, in space and of the spider’s legs, throughout which the flies simply sat tight. At the hub the spider passed secretions into the prey and after about eight minutes the whole surface of the insect was covered in a film of liquid. At this stage the flies left the body of the spider and clustered on the surface of its prey. There they could be seen 1977] Robinson & Robinson — Flies and Spiders 153 dabbing at the liquid with their extended mouthparts. They quickly became swollen with food and their abdomens in particular were distended and almost spherical. The intersegmental membranes became very clearly visible and extended. After feeding the flies returned to the spider’s cephalothorax. This timing of movement onto the prey item to coincide with its liquifaction seems to be fairly precise. It occurred in five out of five instances in which we pro- vided the spider with prey and watched the whole process from its inception. The spider on which we made these observations dis- appeared after four days and we replaced it (on the still-intact web) with an adult female Argiope argentata. Two flies settled on this spider and fed once on her prey before disappearing. (It is note- worthy that the Argiope was able to locate and successfully attack prey on the structurally very different Nephila web. This has pro- vided us with a useful tool for further studies of araneid predatory behavior.) We saw very few cases where the spider reacted to the presence of the flies. Araneids seem to make very few responses to the ac- tivities of their larger kleptoparasitic associates, the theridiids of the genus Argyroides (see Robinson & Olazarri, 1971:34-5; Robin- son & Robinson, 1973:32). Argiope savignyi and unidentified flies While carrying out observations on Argiope savignyi in an in- sectary at Curundu, Canal Zone, Panama, the prey of two separate adult female spiders was visited by flies that did not alight on the spider at any stage. The flies “appeared from nowhere” and fed on liquifying prey items from which the spider was simultaneously feeding. On one occasion the spider’s prey was a pentatomid and the entire insectary in which we were working was flooded with the penetrating odor of the hemipteran’s defensive secretion. In the second case (Figure 2), the prey was a moth. In both instances the flies alighted on the spider’s prey and never moved onto the spider at any stage. After feeding they simply flew off. We were unable to catch the three flies involved. Argiope argentata and Conioscinella sp. While censusing Argiope argentata along the Old Gamboa Road, Summit, Canal Zone, we found an adult female of this species con- suming a half-digested acridiid. On this prey were two small flies 154 Psyche [June Figure 2. Argiope savignvi feeding on a moth, two flies (marked) are visible on the prey. that we succeeded in collecting. They were identified as chloropids of the genus Conioscinella. Several species of chloropids are known to parasitize the egg cocoons of spiders (see, for instance, Bristowe, 1941:366 7). Milichiids and a reduviid At the La Fortuna dam site, Chiriqui, Panama, one of us (MHR) observed a reduviid Zelus trimaculatus Distant, feeding on a sting- 1977] Robinson & Robinson — Flies and Spiders 155 less bee ( Trigona cupira Sm.) which was impaled on the bug’s pro- boscis. Around the prey item was a large number of small flies all apparently feeding. Eight of these flies were captured, many es- caped. The flies were identified as Neophyllomyza sp. Discussion The relationship between Phvllomyza sp. and its host Nephila clavipes is one that involves prolonged contact between the two species. We suspect that the same flies may remain on the spider for days at a time, leaving only to make very short feeding forays. Our attempts to paint-mark the tiny flies failed utterly so we cannot be certain on this point. In any case, the association seems to us to be distinctly more specialized than that described by McMillan (1975) for Desmometopa sp. Of course, this is a matter of inter- pretation. However, it is possible to suggest an evolutionary path- way from commensalism without contact (the reduviid and Argiope savignyi associates) through commensalism plus feeding excursions onto the host, to commensalism with sustained non-trophic con- tact. If the flies were cued into food sources by olfactory stimuli, as seems possible, then the pathway would involve a reduction of the detection distance. The strategy of waiting on the host must involve some interesting mechanism that allows the fly to “evalu- ate” the odds on food being available within its own feeding time- scale. At some stage the fly may be faced with “deciding” whether to remain with a spider on the off-chance that it will catch food or using its remaining food reserves to fly off in search of another (more successful?) host. In this respect, the fly may be at an ad- vantage over the kleptoparasitic theridiid spiders that also asso- ciate with N. clavipes. It can probably range over greater distances, more quickly, in search of a new host than can the spiders. At least three species of theridiids associate with the golden-web spi- der; at least one of these regularly shares the host’s meal, at the hub of the web (Vollrath, in press). Such kleptoparasitic spiders are small, but differ from the milichiids in having mouthparts ca- pable of penetrating insect cuticle. Spiders of the genus Nephila may be particularly suitable as hosts for this kind of associate. They are large, build very efficient webs that are operated 24 hours per day, and show a considerable degree of web-site tenacity. The other large, diurnal, orb-web spi- 156 Psyche [June ders in Panama build much more ephemeral webs that are more susceptible to damage (they are nothing like as strong). They prob- ably spend much less time on their webs at any one site, a situation which may be less favorable to the development of a protracted association. The fact that the only time we have seen milichiids resting on an Argiope was when this spider had been placed on a Nephila web could indicate that the flies respond to some char- acteristic of the web in finding their hosts. On a recent trip to Papua New Guinea (May 1977), one of us (MHR) looked at over 500 adult Nephila maculata, hoping to find flies resting on the spi- der. None were found. Bristowe (1941:369) reports that R. N. Champion Jones saw a small fly crawl over the palps of Nephila maculata in India. Conceivably this is a case where a less sustained association has evolved. As far as the milichiids are concerned it is at least possible that their relationship with Nephila may be more than a trophic one. The fact that all the Phyllomyza sp. that we collected were females is disturbing. The fact that we found Conioscinella feeding on the prey of Argiope argentata makes it possible that the egg parasites of araneids could also be commensals. The reverse could be true. There is some problem about finding terms that accurately de- scribe the relationship of the milichiids to their host(s). They are clearly commensals (in the broad sense) since they “share a table” with their hosts. However, they rely on the host liquifying the prey, they drink alongside the host and could perhaps be called bibiocommensals. The presence of fairly large numbers of flies feeding on a prey item could reduce the amount of food available to the host in a significant way, in which case the term parasite would be justifiable. A drinking parasite would be a dipsoparasite and this term is (to us) more euphonious than bibiocommensal. Acknowledgments We thank the Insect Identification and Beneficial Insect Intro- duction Institute, USDA, Beltsville, Maryland (Chairman, Dr. L. Knutson), for identifying all the insects involved in this study. The experts involved were: C. W. Sabrosky, J. L. Herring, and S. W. Batra. Dr. Sabrosky drew our attention to a number of published works on the relationship between milichiids and spiders; this ad- ditional help is greatly appreciated. 1977] Robinson & Robinson — Flies and Spiders 157 References Bristowe, W. S.: The comity of spiders. Volume 2. Ray Society, London 1941 McMillan, R. P.: Observations on flies of the family Milichiidae cleaning Araneus and Nephila spiders. West Australian Naturalist, 13: 96 (1965) Richards, O'. W.: On commensalism of Desmometopa with predacious insects and spiders. Proc. roy. ent. Soc. (Lond.) Ser. C. 18: 55-56 (1953) Robinson, M. H. and Olazarri, J.: Units of behavior and complex sequences in the predatory behavior of Argiope argentata (Fabricius): (Araneae: Araneidae). Smith. Contr. Zool. 65: 1-36 (1971) Robinson, M. H. and Robinson, B.: The ecology and behavior of the giant wood spider Nephila mareulata (Fabricius) in New Guinea. Smith. Contr. Zool. 149: 1-76 (1973) Robinson, M. H. and Smythe, N.: A technique for observing the behavior of small animals under field conditions. Psyche 83: 210-212 (1976) THE LARVA OF PLATYSTETHUS SPICULUS ERICHSON (COLEOPTERA : STAPH YLINI DAE) AND ITS OCCURRENCE IN BOVINE FECES IN IRRIGATED PASTURES* By E. F. Legner and Ian Moore Department of Entomology, University of California, Riverside, California 92521 Platystethus is a relatively small genus (35 described species) with only two species being known to be indigenous to the United States. The Nearctic species were reviewed by Moore and Legner (1971). The larva of one of these, P. americanus Erichson, was described and illustrated by Paulian (1941). The larvae of the other species, P. spiculus Erichson, is described here. It goes to Platy- stethus in Paulian’s (1941) key to the genera of the larvae of the Staphylinoidea. Several larvae were taken by Berlese extraction as previously described (Legner et al., 1975) in company with many adults of P. spiculus in bovine manure that, at the time, contained no other staphylinids. Ten to 25 1 L samples were taken at random from bovine feces deposited in green and dry irrigated pastures at 3 sites near Calexico, California between 9 AM and 11 AM over several sample dates from October 22, 1974 to April 22, 1975. Two age groups of manure were distinguished: 12 hr and 24-48 hr old. KEY TO THE LARVAE OF THE SPECIES OF PLATYSTETHUS INDIGENOUS TO THE UNITED STATES 1 . Acorn-like seta at apex of second antennal segment equal in size to third antennal segment spiculus Erichson Acorn-like seta at apex of second antennal segment only half as long as third antennal segment americanus Erichson Larva of Platystethus spiculus Erichson Length 3.8 mm. Body elongate, pale, integuments mostly trans- parent with the head and mouthparts tinged with brown, the man- * Manuscript received by the editor June 10, 1977. 158 1977] Legner & Moore — Platystethus spiculus 159 3 Figures 1-5. Larva of Platystethus spiculus. Fig. 1, antenna; Fig. 2, maxilla; Fig. 3, labial palpus; Fig. 4, right mandible, dorsal view; Fig. 5, pseudopod and urogomphus. NO / L (semi -log scale ) 160 Psyche [June 23 22 1975 SAMPLE OATE Figure 6. Average density of Platvstethus spiculus and Haematobia irritans larva, in 12-hr-old bovine feces from green, irrigated pastures; Calexico, California, 1975 (+' 05 Conf. limit shown above each mean). t 1 r 22 N '9 OC» Nov • 974 MEAN NO- / L. ( s«mi -log 1977] Legner & Moore — Platystethus spiculus 161 — T“ 22 Oct l»T4 II Nov T 19 ~r 23 SAMPLE DATE Figure 7. Average density of Platystethus spiculus and Haematobia irritans larvae, in 24-48-hr-old bovine feces from green, irrigated pastures; Calexico, Cali- fornia, 1975 (+‘ 05 Conf. limit shown above each mean). 162 Psyche [June dibles darkest. Head oval, about one-fourth wider than long, with a single ocellus very near outer apical angle of the head at the base of mandible. Labrum simple, a little wider than long. Antennal fossa located near the outer, apical angle of the head at the base of the mandible. Antenna three-segmented; first segment about as wide as long; second segment about one-fourth wider than first and twice as long, widest near apical fourth, with a large acorn-like seta at obliquely truncate component of apex, about as long as third segment; third segment less than one-third as long as second segment and a little more than one-third as wide. Mandibles stout at base, arcuate, with two teeth at apex in dorsal view. Maxillary palpus three-segmented; first segment longest, about three times as long as wide; second segment about as wide as first and almost half as long; third segment a little longer than second but only about half as wide, apex pointed. Lacinia gradually, irregularly narrowed to apex, with a series of stout setae internally in apical half and a single large seta in the middle on the outer edge. Labial palpus two-segmented, segments of about equal length, first segment wider than second. Pronotum about as wide as head but a little shorter, with two setae at anterior margin and two in each lateral series. Mesonotum and metanotum each about as wide as pronotum but a little shorter, each with four transversely arranged discal setae and with one large seta and two small setae in the lateral series. Abdomen with nine segments, first segment about as wide as meta- notum but not quite as long, segments progressively slightly in- creasing in width and length through eighth which is slightly wider and about as long as metanotum; each with four setae along the posterior margin and three in the lateral series. Pseudopod short and stout. Urogomphus one-segmented, cylindrical, about four times as long as wide, tapered to apex. Five specimens taken from field-manure in company with adults at Calexico, Imperial county, California, 21 October 1974, E. F. Legner collector. This species is similar to Paulian’s (1941) description and illus- tration of P. americanus, except that the acorn-like seta at the apex of the second antennal segment is as large as the third antennal segment whereas in P. americanus it is only half as long. 1977] Legner & Moore — Platystethus spiculus 163 12 hr old manure Dry Pasture 24-48hr old manure SAMPLE DATE Figure 8. Average density of Platystethus spiculus and Haematobia irritans larvae, in 12- and 24-48-hr-old bovine feces from dry pastures; Calexico, California | (+l 05 Conf. limit shown above each mean). 164 Psyche [June Field Occurrence Larvae of the horn fly, Haematobia irritans Linnaeus, occurred in bovine feces in pastures in association with adults and larvae of Platystethus spiculus. There was no continuous coincidence of P . spiculus larvae and adults with H. irritans larvae (Figs. 6-8). Therefore, alternative food would have been necessary to sustain P. spiculus during period of H. irritans scarcity or absence. The occurrence of P. spiculus larvae was exceptionally low in the manure habitat, certainly not sufficiently dense to account for the comparative high adult abundance (Figs. 6 & 7), and no larvae were detected in dry pastures (Fig. 8). It appears that P. spiculus adults are initially attracted to fresh bovine feces in green pastures where some eggs are laid (Fig. 6). Attraction to dry pastures is not as great, and apparently no eggs are laid judging from the absence of larvae (Fig. 8). The adult density is maintained for at least 24 hrs (Fig. 7). The majority of the larvae of this species may occur subterraneously below and sur- rounding the bovine fecal droppings. Hinton (1944) has shown that in the European P. arenarius (Fourcroy) larvae and adults feed on manure and are facultative feeders on carrion; that eggs are laid in a brood chamber in or partly beneath manure and protected by the female until hatched. Whether similar sub-social behavior exists with the American spe- cies has not been determined. Literature Cited Hinton. H. E. 1944. Some remarks on sub-social beetles, with notes on the Biology of the staphylinid, Platystethus arenarius (Fourcroy). Proc. Royal Ent. Soc. London. (A): 19: 115-118, 15 Figs. Legner, E. F., G. S. Olton, R. E. Eastwood and E. J. Dietrick. 1975. Seasonal density, distribution and interactions of predatory and scaven- ger arthropods in accumulating poultry wastes in coastal and interior southern California. Entomophaga. 20 (3): 269-283. Moore. I. and E. F. Legner 1971. A review of the Nearctic species of Platystethus (Coleoptera:Staphylini- dae). Pan-Pac. Entomol. 47: 260-264. 1 Fig. Paulian, R. 1941. Les premier etats des Staphylinoidea. Etude de Morphologie Comperee. Mem. Mus. Nat. Paris, n. ser., 15: 1-361; 1365 Fig., 3 L. DRAGLINE-FOLLOWING BY MALE LYCOSID SPIDERS' By William J. Tietjen Department of Zoology, Ohio University, Athens, Ohio 45701 Wolf spiders stalk or ambush their prey rather than build prey- capture webs. Production of silk by lycosids is important during aerial dispersal via ballooning (Richter, 1970b) and, in some, dur- ing ’post-immobilization wrapping of prey (Rovner and Knost, 1974), and construction of egg sacs or sperm webs (Richter, 1970a). During locomotion both sexes lay down silk draglines that are generally assumed to have a stabilizing or security function (Rich- ter and Van der Krann, 1970). In addition, female draglines induce courtship behavior in male conspecifics and, in some species, males are reported to have followed a female trail, but no systematic analyses of the dragline-following behavior have been attempted (Bristowe and Locket, 1926; Engelhardt, 1964; Kaston, 1936). The present study is concerned with the trail-following behavior of male Lycosa rabida and L. punctulata. Specific areas of inves- tigation include determination of the use of various appendages during following, analysis of cues involved in initiating and main- taining trail-following, and an interspecific comparison of trail- following behavior. Methods Penultimate and young adult instars of L . rabida and L. punctu- lata were collected near Amesville, Athens County, Ohio, USA, during 1974 and 1976. L. rabida were collected late June through July while L. punctulata were collected mid- to late September. Animals were housed separately in 13 X 7 X 6.5 cm plastic cages painted on two sides to afford visual isolation between animals in adjacent cages. A paper substratum in each cage collected silk used in some experiments and facilitated periodic cleaning of the cages. Water was provided ad libitum via a cotton-stoppered vial, and two Tenebrio sp. larvae were provided per week as food. Cages 'This study was supported by National Science Foundation Grant BNS 76-15009 to J. S. Rovner. Manuscript received by the editor September 9, 1977. 165 166 Psyche [June containing males were located on shelves along one wall of the laboratory that received indirect sunlight. Laboratory humidity was not controlled (range 36-56% RH) and temperature ranged from 23-28° C. Cages containing females were placed in a con- trolled environmental chamber (Frease model 818) at 15°C night and 20° C day and 53-56% RH, with a photoperiod approximating natural conditions. The lower temperatures were necessary to re- tard female development and to delay onset of eggsac production, which even occurs eventually in unmated females, making them useless for further experimentation. Females were removed from the environmental chamber and allowed sufficient time to warm to room temperature before experimental use. In order to lay draglines along predetermined trails spiders were anesthetized with CCF. A thread leash, secured to a hemostat, was tied around the cephalothorax of each spider between the second and third pairs of legs. Upon recovery, the spider was led along a path such that the dragline was laid either directly on the sub- stratum to form “ground lines” or suspended above the substratum by being laid across horizontally placed glass rods to form “aerial lines” (Fig. 1). Lines connecting parts of an artificial plant made of wood dowels were also called “aerial lines.” Male draglines or imitation draglines (nylon thread or human hair) were laid in either a continuous or discontinuous manner with the female trail during some experiments. The test trail was then surrounded by glass walls to provide an arena (54 X 64 cm), the position of which was rotated in a random manner between experimental runs. The position, length, and direc- tion of the dragline, including attachment disks (produced by the spider to fix the dragline to a substratum) were recorded on a dia- gram of the completed path. Test males were of two physiological conditions: (1) unprimed males, which had not encountered female silk for the previous 72 hr and (2) primed males, which were induced to court in response to substratum silk from a female cage, and then tested 1 hr later for dragline-following behavior. Experimental trials began when a male released into the arena contacted and explored a female line with his palps. Trials lasted for 5 min, during which time data were recorded in shorthand notation from a distance of 0.5- 1.0 m from the experimental males. To test the response of males to various types of silk, males were released into cages containing either lepi- 1977] Tietjen — Dragline- Following by Spiders 167 Figure 1. Diagrammatic summary of a typical trail including attachment disks (D) and female aerial lines (A). Female ground lines (G) and imitation draglines are laid contiguously (C) and discontiguously (DC) with the female dragline. Small strips of masking tape (T) attach imitation dragline to the wall of the arena and to 5-mm diameter glass rods (R). Glass rods were usually parallel and 9-11 cm apart. When artificial “plants” (P) were incorporated into the experiment, female trails up to 236 cm in length were constructed. dopteran silk or conspecific male or female silk. Male L. punctu- lata were also exposed to female L. rabida silk. Clean cage cards were used as controls. A positive response by males to the silk consisted of exploring the silk and initiating courtship behavior within 5 min. Statistical analyses of data were performed according to the methods of Conover (1971) and Sokal and Rohlf (1969). All means are accompanied by their standard errors. Results COMPARISONS OF YEARLY SAMPLES No difference was found in trail-following behavior between 1974 and 1976. Unprimed male L. rabida showed no difference in fre- quency of following-behavior (\2 = 0.0, df = 1, P > .50) and in 168 Psyche [June total distance followed during the 5-min test period (Mann- Whitney test, T= 173 ,P> .80). Primed male L. rahida showed no difference in frequency of following (x2 = .504, df = 1, P > .70) or in total distance followed (Mann-Whitney test, T — 229.5, P > .56). Male L. punctulata showed no difference in the frequency of following (x2 = 2.5, df — 1, P >.90) or in total distance followed (Mann- Whitney test, T = 235.5, P > .84). Therefore, data for the two years were pooled for analysis. DESCRIPTION OF DRAGLINE-FOLLOWING Males of both species followed conspecific female lines. Al- though high speed (36-180 fps) cinematographic analysis indicates that there are slight differences in the use of appendages during dragline-following (Tietjen, unpublished data), the two species ex- hibit many similarities. Upon contacting the silk line for the first time, males tend to examine it with alternating movements of the palps such that the dorsal surface of the palp contacts the line and moves anteriorly along it. Such behavior is soon terminated if the line is a male’s or is an imitation. Dragline-following results if the line is that of a female conspecific. Males usually wander to the edge of the arena and walk around the periphery if chemo- and mechano- exploratory behavior is terminated. In both species, dragline-following is characterized by the male straddling the silk so that it passes medially beneath him. Although the first pair of legs may occasionally pluck at the line, the palps are the most important appendages used. The palps are alternately moved forward and back, making contact with the silk on their medial surface rather than on the dorsal surface, which is used dur- ing the initial contact with the line. Experiments conducted under red light (Kodak safelight Filter No. 1) and with blinded animals indicated that vision is not necessary for dragline-following. Males may accompany their dragline-following behavior with elements of courtship or, more typically, interrupt dragline-following with sep- arate bouts of courtship behavior. DRAGLINE-FOLLOWING BY LYCOS A RABID A Both primed and unprimed male L. rabida followed female aerial draglines. Unprimed males exhibited a lower incidence of dragline- following, with 84.6% of primed and 65.2% of unprimed following (x2 = 5.83, df — 1,P < .025). Comparing only those males that 1977] Tietjen — Dragline- Following by Spiders 169 Table I: Performance of Male Lycosa rabida and Male L. Dragline-Following of Conspecific Male, Female, L. rabida punctulata During and Imitation Silk. L. punctulata Primed Unprimed Pooled Number of males tested 52 69 57 number following 44 45 40 number not following 8 24 17 Distance (cm) Y total distance 23.1 ±2.3 26.7 ±.2 56.9 ±8.7 Y polarity + 7.9 ± 3.3 -6.5 ±4.2 + 5.3 ±2.4 Courtship Frequency 27.2% 26.6% 12.5% Y latency (min) 1.4 ±.4 1.1 ± .3 1.4 ±.8 Imitation and male draglines number followed 7 4 0 number not followed 6 12 16 Ground lines number followed 0 0 13 number not followed 18 9 4 followed draglines, no significant difference was found in the total distance followed between primed and unprimed animals (Mann- Whitney test, T = 1069.5, P > .72). No males of either group fol- lowed ground lines, although they would usually examine such lines with their palps (Table I). None of the male L. rabida followed isolated imitation draglines; and chemoexploratory behavior to such lines was exhibited only five times. Males in the process of following a female line would occasionally follow imitation and male draglines laid contiguously with the female trail (Table I). Once, a male following a female line continued on and followed the entire length of a human hair that constituted an 8.5 cm interruption in the female trail, losing no momentum in his following behavior. Male draglines were fol- lowed four times, human hairs were followed six times and nylon thread was followed once. The polarity of each male’s trail was calculated as the difference between total distance traveled in the same direction as the female 170 Psyche [June trail was laid and total distance traveled in the opposite direction. Primed males exhibited a positive polarity by following the correct direction more often than the incorrect direction (x2 = 7.36, df— 1, P < .01), and by traveling further in the correct direction along the female trail (Wilcoxon test, T— 252.5, P< .01). Unprimed males did not show a directional preference (x2 = 0.0, df = 1, P > .50; Wilcoxon test, T= 525, P > .64; Table I). Primed and unprimed male L. rabida showed no difference in the frequency of courtship during dragline-following (x2 = .613, df — 1, P > .50) or in the elapsed time during a test before court- ship was observed (= courtship latency) (Mann-Whitney test, T = 70.0, P > .54; Table I). Primed males that courted showed no correlation between total distance of dragline followed and court- ship latency (Spearman’s rho, p = + .026, P > .53). Unprimed males courting early in the test showed decreased total dragline- following (Spearman’s rho, p = + .692, P < .01). Comparing primed and unprimed males that both courted and followed drag- lines, no difference was found in the total amount of dragline- following (Mann-Whitney test, T — 747.0, P > .65). DRAGLINE-FOLLOWING IN LYCOS A PUNCTULATA L. punctulata was a more difficult species to study experimentally than L. rabida. Females often would not lay the required lines and both sexes tended to break the lines by catching a claw on them, and the males by running through the lines. In 1974, 9 of 10 males in each group exhibited dragline-following. The total distance followed for primed (Y = 63.1 ± 20.7 cm) and unprimed (Y = 63.1 ± 17.5 cm) males did not differ significantly between the two groups (Mann-Whitney test, T — 47.5, P> .57). Incidence of courtship did not differ between the two groups with primed males courting during one and unprimed males courting during three of the trials (x2 = 1.25, df = 1, P >.70). Data for the two groups were pooled for further analysis. In 1976 only unprimed males were used. During 34 trials 23 males followed a mean total distance of 52.4 ±11.3 cm. No dif- ferences were found between pooled 1974 data and the 1976 (all unprimed) data for incidence of following (x2 = 3.41, df — 1, P> .05), or for distance followed (Mann-Whitney test. T— 235.5, P > .77). Unprimed males in 1976 exhibited courtship three times during dragline-following; no difference in courtship frequency was 1977] Tietjen — Dragline- Following by Spiders 171 found between the pooled 1974 data and those for 1976 males (x2 = 0.34, df = 1, P > .70). In order to have a larger data base the data were pooled for both years. Unlike male L. rabida, male L. punctulata initiated dragline- following at ground lines and appeared to use their palps in a manner similar to chemoexploratory behavior. Male L. punctulata rarely examined isolated male lines or imitation lines; of 16 that did palpate these lines, none followed them (Table I). Males ex- hibited a positive polarity (\2 = 4.33, df — 1, P< .05), and traveled a greater distance in the correct direction along the female trail (Wilcoxon test, T= 255.5, P < .03). Males that courted during dragline-following showed no corre- lation between courtship latency and distance followed (Pearson’s r, r = + .700, P > .90). No difference in distance followed was found between males that courted during dragline-following and those that did not (Mann-Whitney test, T— 93.0, P> .58). Court- ship latency of males placed on a female substratum was affected after dragline-following; those which had recently followed a drag- line exhibited a mean courtship latency of 0.66 ± .29 min (N = 23) compared with 2.30 ± 0.25 min (N = 20) for those which had not (Mann-Whitney test, T— 52.0, P< .0001). COURTSHIP RESPONSES TO VARIOUS TYPES OF SILK Male L. rabida courted in response to silk in a female cage in 50% of the trials. Courtship latency in response to female silk was 2.30 ± .25 min. The incidence of courtship in response to female silk differed from the courtship frequency in response to a clean cage, lepidopteran silk and male L. rabida silk (x“ = 33.99, df = 3, P < .001). The incidence of courtship did not differ among the latter three conditions (x2 — 2.05, df = 2, P> .70). (Table II). Male L. punctulata courted in response to female conspecific silk in 61.1% of the trials and exhibited a mean courtship latency of 2.55 min ± .44 min. The incidence of courtship in response to female conspecific silk, female L. rabida silk, lepidopteran silk, a clean cage and male L. punctulata silk differed significantly (x2 — 35.93, df — 4, P< .001), while courtship frequency did not differ among the latter four test conditions (x2 — 2.10, df =3, P> .50). To simulate the effects of dew under natural conditions, silk trails of females of both species were sprayed with a fine mist of distilled water and were air dried (= washed lines). Draglines so 172 Psyche [June Table II. Courtship Responses to Various Types of Silk by Male Lycosa rabida and Male L. punctulata. Experimental Silks Clean Female con- Male con- Female hetero- lepidop- cage specific specific specific teran L. rabida number tested 20 40 20 0 20 number courting 1 20 0 — 0 number not courting 19 20 20 — 20 Y courtship latency (min) 1.27 2.30 ±.2 — — — . L. punctulata number tested 20 18 20 20 20 number courting 2 11 1 0 1 number not courting 18 7 19 20 19 Y courtship latency (min) 4.06 ±.7 2.55 ±.7 4.93 4.57 treated lost elasticity and drooped between the supporting glass rods. Both unprimed male L. rabida (N = 10) and unprimed male L. punctulata (N — 10) exhibited short bouts of chemoexploratory behavior at washed lines, but neither species exhibited courtship or trail-following behavior during the 5-min trials. Discussion Trail-following has been observed in a variety of arthropods. Terrestrial chemical trails are often employed by eusocial insects, but such trails are detected by the receiver as a rapidly decaying vapor cloud above the substratum and provide no directional in- formation (Wilson, 1971). Lepidopteran larvae ( Malacosoma , Hy- po no m cut a and Thaumetopoed) have been observed to follow silk trails from their nest to a food source; but tactile information provided by the silk is unimportant, and the pheromone decays within a few minutes (Fitzgerald, 1976; Wigglesworth, 1966). Wolf spider trails differ from the above examples in function (mating vs. food), in the use of a contact rather than an olfactory phero- mone, and in the presence of an inherent polarity. Silk lines in the 1977] Tietjen — Dragline- Following by Spiders 173 laboratory up to 4 days old will elicit following-behavior in male L. rabida and male L. punctulata; and a cage substratum with female silk elicits courtship behavior for over a month (Tietjen, unpub. data), indicating that the spider pheromone does not decay over a relatively long period of time. Female lycosid spiders exhibit low motility during the breeding season and adopt a “sit and wait” reproductive strategy; the highly active males seek out the females (Hallander, 1967a, 1967b; Hol- lander, 1972; Muma, 1973; Richter, et ah, 1971). Mathematical modeling of females laying no silk trails and silk trails of various lengths indicates that the presence of silk effectively increases the size of the female target as perceived by the searching male (Tietjen, unpub. data). Thus, the ability of male L. rabida and L . punctu- lata to follow conspecific female draglines to the source increases the reproductive success of these males over spiders that depend on chance encounters in a heterogeneous environment. Female silk also alerts the male to the presence of a female conspecific and, as shown in L. punctulata, decreases the courtship latency. The latter two functions effectively reduce the likelihood of intraspecific can- nibalism (Platnick, 1971). Contact sex pheromones associated with the female dragline are known to induce courtship in conspecific male spiders (Dijkstra, 1970; Dondale and Hegdekar, 1973; Hegdekar and Dondale, 1969). Results from the present study support the hypothesis that a phe- romone associated with the female dragline elicits courtship in re- ceptive males, i.e., these behaviors are not dependent exclusively upon tactile information provided by the silk. Both L. rabida and L. punctulata are found with a variety of other spider species. Dependence on a pheromone to elicit sexual behavior in males would reduce the chances of males following draglines of other species and thereby adversely affecting the time budget allocated for reproductive purposes. The propensity of male L. rabida not to follow ground lines which cannot be plucked by the palps (thereby providing little tac- tile input) and their occasional following of imitation draglines, which provide no chemical cues, indicate that the pheromone is responsible only for initiating the following behavior in males. Once a male is following a dragline, tactile cues appear to be ade- quate. Males are expected, however, to occasionally sample the trail for female pheromone, as only 38% of male and imitation 174 Psyche [June dragline contacts during female dragline-following resulted in fol- lowing by males, and most imitation draglines were not followed the entire length. Male L. punctulata followed ground lines, but did not follow male and imitation draglines. This species appears to depend upon the female pheromone for initiating and for maintaining dragline- following. Apparently, male L. punctulata are either more sensi- tive to tactile input provided by ground lines, or more sensitive to pheromonal cues than male L. rabida, or both. The differential sensitivity to tactile and chemical cues exhibited by the two species may be related to microhabitat preferences. L. punctulata is found in the lower levels of the herbaceous stratum (Eason and Whitcomb, 1965) in which one expects a greater di- versity and density of spiders (Whitcomb, Exline and Hite, 1963) and consequently more draglines. Many lycosids of similar or larger size are found within this microhabitat of which male L. punctulata could be potential prey, including L. helluo and L. caro- linensis. Continuous sampling for female pheromone during drag- line following reduces the likelihood that males will follow another species’ trail in a high silk density microhabitat. In addition, due to the greater foliage density at this level, many draglines could be expected to be laid directly on a substratum. A greater sensitivity to tactile and/or chemical cues could allow male L. punctulata to follow such ground lines. L. rabida , on the other hand, is found high in the herbaceous stratum (Eason and Whitcomb, 1965; Kuenzler, 1958) and is asso- ciated with lower interspecific silk densities derived mainly from smaller species such as salticids and oxyopids. Foliage density is less, and a greater volume of open space is found at this level. Fewer female silk trails would be expected to be found directly on this substratum. Under these conditions the sensitivity to tactile and/or chemical cues exhibited by L. punctulata would not be required for efficient dragline-following by male L. rabida. Silk lines are destroyed mechanically in the field by wind and the activities of animals. Moisture resulting from rain and early morning dew inactivates the female pheromone found on unbroken lines (Dondale and Hegdeker, 1973; Hegdeker and Dondale, 1969). All of these effects reduce the likelihood that males will follow draglines laid the previous day. 1977] Tietjen — Dragline- Following by Spiders 175 Male L. punctulata and primed male L. rabida may be able to extract directional information from the female dragline, though it is unlikely that a pheromone concentration gradient could pro- vide the directional information since the pheromone remains active over long periods of time. Examination of female L. rabida silk under a light microscope (450X) as well as scanning electron micro- scope inspection of female L. punctulata silk (1000X) indicate no evidence of structural features on the dragline which could provide directional information to males. Spiders are able to detect slight changes in web tension through vibration receptors during prey capture, web building, and court- ship (Robinson, 1969; Walcott, 1969; Witt, 1975). Such variations in tension, related to the direction in which the dragline was laid, could provide directional information to dragline-following males. If present, the tension differential may be related to the structure of the attachment disk by which the silk is fixed to a substratum. As the dragline enters the attachment disk it is tightly wound, whereas the silk leaving the attachment disk is composed of single threads which later join and wind to form the dragline. Single silk threads leaving an attachment disk would be expected to damp transients produced along the line more readily than a complete dragline entering an attachment disk. If this were the case, drag- line-following toward the pole with the higher resonant frequency would constitute movement in the direction the silk was laid (Tiet- jen, unpub. data). Such a system would require that males im- mobilize a portion of the dragline, perhaps with a palp, while plucking on either side of the fixed portion. Male spiders en- countering a female line often wipe the dorsal surface of each palp alternately over the silk and often pluck at the silk with the first legs as they swivel their bodies along a short length of the trail. This behavior may represent both a sampling for female phero- mone and a testing for differential mechanical properties of the line related to the direction in which it was laid. Unprimed male L. rabida did not exhibit a polarity in dragline- following, but a positive correlation between courtship latency and total distance followed was observed. The above correlation indi- cates that unprimed males probably have a lower threshold for chemical cues than do primed males. If unprimed males are at- tending more to chemical cues, they would be less likely to follow 176 Psyche [June a dragline in a preferential direction if directional cues were de- pendent on the mechanical properties of the line. Summary The dragline-following behavior of male wolf spiders was ob- served in response to male, female and imitation draglines. Male L. rabida followed female draglines suspended above the sub- stratum but not those laid directly on the substratum. These males followed male silk and imitation lines if laid contiguously with female draglines, suggesting that they depend on tactile cues dur- ing dragline-following. On the other hand, male L. punctulata followed both aerial lines and ground lines laid by females but did not follow imitation or male lines. This suggests that male L . punctulata are more sensitive to tactile and/or chemical cues than are male L. rabida. Microhabitat preferences of the two species may explain the above differences. Data also suggest that males of both species are able to extract directional information from the dragline, perhaps by tensional cues. Acknowledgments The author wishes to express his gratitude to Dr. J. S. Rovner for his support, valuable discussions and assistance in the prepara- tion of the manuscript. Thanks are also due Dr. W. D. Hummon, Dr. C. P. Spirito and Dr. G. E. Svendsen for their helpful criticism of this work. Anne Tietjen assisted in the preparation of the man- uscript. Literature Cited Bristowe, W. S. and G. H. Locket 1926. The courtship of British lycosid spiders and its probable significance. Proc. Zool. Soc. 22: 317 347. Conover, W. J. 1971. Practical Nonparametric Statistics. New York: John Wiley and Sons, Inc. 492 pp. Dijkstra, H. 1970. Comparative research of the courtship behaviour on the genus Pardosa (Arach. Araneae) III. Agnostic behaviour in Pardosa amentata. Bull. Mus. Nat. Hist. Natur. 41: 91-97. Dondale, C. D. and B. M. Hegdekar 1973. The contact sex pheromone of Pardosa lapidicina. Emerton (Araneida; Lycosidae). Canad. J. Zool. 51: 400-401. 1977] Tietjen — Dragline- Following by Spiders 111 Eason, R. and W. H. Whitcomb 1965. Life history of the dotted wolf spider. Lycosa punctulata Hentz (Ara- neida; Lycosidae). Arkansas Acad. Sci. Proc. 19: 1 1 20. Engelhardt, W. 1964. Die mitteleuropischen Arten der Gattung Trochosa C. L. Koch, 1848. Z. Morph. Tiere. 54: 219-393. Fitzgerald, T. D. 1976. Trail marking by larvae of the eastern tent caterpillar. Science 194: 961-963. Hallander, H. 1967a. Courtship display and habitat selection in Pardosa chelata (O. F. Mul- ler). Oikos 18: 145-150. 1967b. Range and movements of the wolf spiders Pardosa chelata (O. F. Mul- ler) and P. pullata (Clerck). Oikos 18: 360-364. Hegdekar, B. M. and C. D. Dondale 1969. A contact sex pheromone and some response parameters in lycosid spiders. Canad. J. Zool. 47: 1-4. Hollander, J. Den. 1972. Differential use of the habitat by Pardosa pullata (Clerck) and Pardosa prativaga (L. Koch) in a mixed population (Araneae, Lycosidae). Tijd. Joor. Entomol. 115: 205-215. Kaston, B. J. 1936. The senses involved in the courtship of some vagabond spiders. Entomol. Amer. 16: 97-169. Kuenzler, E. J. 1958. Niche relations of three species of lycosid spiders. Ecology 39: 494-500. Muma, M. H. 1973. Comparison of ground surface spiders in four central Florida ecosys- tems. Florida Entomol. 56: 173-196. Platnick, N. 1971. The evolution of courtship behaviour in spiders. Bull. Brit. Arach. Soc. 2: 40-47. Richter, C. J. J. 1970a. Morphology and function of the spinning apparatus of the wolf spider Pardosa amentata (Cl.) (Araneae, Lycosidae). Z. Morph. Tiere. 68: 37-68. 1970b. Aerial dispersal in relation to habitat in eight wolf spider species {Par- dosa, Araneae, Lycosidae). Oecologia 5: 200-214. Richter, C. J. J. and J. Den Hollander 1971. Differences in breeding and motility between Pardosa pullata (Clerck) and Pardosa prativaga (L. Koch) (Lycosidae, Araneae) in relation to habitat. Oecologia 6: 318-327. Robinson, M. H. 1969. Predatory behaviour of Argiope argentata (Fabricus). Amer. Zool. 9: 161-174. Rovner, J. S. and S. J. Knost 1974. Post-immobilization wrapping of prey by lycosid spiders of the herba- ceous stratum. Psyche 8: 398 415. 178 Psyche [June SOKAL, R. R. AND F. J. ROHLF 1969. Biometry. San Francisco: W. H. Freeman. 766 pp. Walcott, C. 1969. A spider’s vibration receptor: its anatomy and physiology. Amer. Zool. 9: 133-144. Whitcomb, W. H., H. Exline and M. Hite 1963. Comparison of spider populations of ground stratum in Arkansas pas- ture and adjacent cultivated field. Arkansas Acad. Sci. Proc. 17: 1-6. WlGGLESWORTH, V. B. 1966. The Life of Insects. London: William Clowes and Sons, Ltd. 360 pp. Wilson, E. O. 1971. The Insect Societies. Cambridge: Belknap. 548 pp. Witt, P. N. 1975. The web as a means of communication. Biosci. Commun. 1: 7-23. THE BIOLOGY OF PHANETA IMBRIDANA (LEPIDOPTERA : TORTRICIDAE), A SEED PREDATOR OF XANTHIUM STRUMA RIUM (COMPOSITAE) By J. Daniel Hare Department of Ecology and Evolution* State University of New York Stony Brook, New York 11794 Of the more than sixty North American species of Phaneta, host plants are known for less than one third. All of the known hosts are in the family Compositae, and most species feed only on the flowers or seeds of their host plant (Heinrich, 1923, Mackay, 1959). (Host plants are listed by these authors for the species of the genus, Thiodia, the North American members of which have been trans- ferred to Phaneta (Obraztsov, 1952)). Although Phaneta imbridana (Fernald) has been known to taxonomists for years (Fernald, 1905, Miller, 1970), nothing is known of its biology or life history. I therefore report certain aspects of the ecology of P. imbridana and its relationship with a local host plant, Xanthium strumarium, unique among the Compositae by having relatively large fruits and seeds. This information was obtained as part of a larger study of the variation in susceptibility of populations of X. strumarium to seed predation by more than one species of seed predator along Long Island beaches. A. Life Cycle Adults emerge in late August and can be found until late Sep- tember, with oviposition occurring throughout the adult period. Females oviposit directly on the surface of the full-sized but im- mature burrs of X. strumarium. Eggs soon hatch and the larvae bore through the burr wall and begin to feed on one of two seeds of the burr. If one seed is insufficient for complete larval develop- ment, larvae will attack the other seed within the same burr, or rarely, seeds of another burr on the same plant. Full larval devel- opment is completed by late September or early October, at which * Present address: Department of Entomology, Connecticut Agricultural Experi- ment Station, Box 1106, New Haven, CT 06504. Manuscript received by the editor October 1, 1977. 179 180 Psyche [June Table 1 Distribution of Phaneta imbridana among Populations of Xanthium strumarium Mean Proportion Seeds Attacked (1973-1975) 1 2 3 Population Number 4 5 6 7 8 9 10 Upper Seed .03 .12 .05 .09 .01 .05 .03 .07 .07 0.0 Lower Seed .07 .20 .24 .20 .07 .15 .10 .22 .12 .05 time larvae leave the burr through a hole bored near its basal end. Since burrs reach full maturity and are easily dislodged and dis- persed before larvae leave the burr, passive long-range dispersal of P. imbridana may occur in the larval stage. Local populations of P. imbridana overwinter as last-instar larvae in the dry pithy stems of X. strumarium. It is unlikely that P. im- bridana is limited to X. strumarium for overwintering, however, the other common herbacepus species associated with X. strumar- ium do not contain overwintering larvae. Pupation occurs in the stem fragments in the following summer. Mating behavior was not observed. B. Role as a Seed Predator Levels of seed predation were measured for ten populations of X. strumarium over a three-year period. Consistent, significant differences in the abundance of P. imbridana were observed among plant populations (Table 1), however, mean seed loss was less than 10%. The two seeds within a burr of X. strumarium differ in size and germination requirements (e.g. Wareing and Foda, 1957). The lower seed is larger and germinates the spring following produc- tion, while the smaller, upper seed remains dormant for one year or more if its seed coat remains intact. Phaneta imbridana is more commonly found in the lower, non-dormant seed within a burr (p less than .001). Although one cannot exclude the possibility that larvae or ovipositing females may be choosing seeds on the basis of their dormancy properties, differential seed predation within burrs is best explained by burr asymmetry. Since the larger seed occupies more than half of the burr cavity, it is covered by more 1977] Hare — Biology of Phanta imbridana 181 than half of the burr surface, and oviposition is more likely to occur on burr surface adjacent to lower than to upper seeds. C. Interactions with Other Seed Predators The tephritid fly, Euaresta aequalis Loew, is another common seed predator of X. strumarium. The abundance of E. aequalis also varies significantly among populations, and larvae are more frequent in lower than upper seeds. The oviposition periods of both insect species coincide. Most local populations of X. stru- marium are not attacked by both insect species, however, in those plant populations which experience at least 5% seed predation by both species, the abundance of the two species on individual plants is significantly negatively correlated ( r = -.42, p less than .01). An oviposition experiment was performed using plants from several populations grown under uniform conditions and then simulta- neously exposed to both insect species. The number of burrs at- tacked by both species was much less than expected assuming that their oviposition behaviors were independent (Table 2), and the number of burrs containing one larva of P. imbridana and one undamaged seed was much greater than expected. These results indicate that within populations, some plants may produce burrs more susceptible to one insect species than the other. Table 2 Frequency of Attack of Seeds Within Burrs Disposition PP UP PE EE UE UU Observed 16 70 43 110 46 87 Expected 14.1 56.5 60.2 64.2 130.5 56.5 Difference 1.9 13.5 -17.2 45.8 -74.5 30.5 G-test Statistic = 1 10.024, p less than .005 PP= Both seeds containing P. imbridana. UP = One seed containing P. imbridana and the other undamaged. PE = One seed containing P. imbridana and the other containing E. aequalis. EE = Both seeds containing E. aequalis. UE = One seed containing E. aequalis and the other undamaged. UU = Both seeds undamaged. 182 Psyche [June and also that within plants, P. imbridana may avoid ovipositing on burrs previously attacked by others of its own or different species. Further investigations are in progress to determine which particu- lar aspects of burr morphology and chemistry most strongly influ- ence susceptibility of burrs to each insect species. Acknowledgements I thank William E. Miller and a reviewer for their information concerning the taxonomy and ecology of the genus Phaneta. Spec- imens were kindly identified by D. R. Davis of the U. S. National Museum. Contribution #226 from the program of Ecology and Evolution at the State University of New York at Stony Brook. References Fernald, C. H. 1905. North American Tortricidae. Can. Ent. 37: 399 400. Heinrich, C. 1923. Revision of the North American Moths of the Subfamily Eucosminae of the Family Olethreutidae. Bull. U. S. Nat. Mus. #123. 298 pp. MacKay, M. R. 1959. Larvae of the North American Olethreutidae (Lepidoptera). Can. Ent. Suppl. #10. 338 pp. Miller. W. E. 1970. Fernald Types of North American Olethreutinae (Lepidoptera : Tortri- cidae). Proc. Ent. Soc. Wash. 72: 288-294. Obraztsov, N. 1952. Thiodia Hb. as not a North American Genus (Lepidoptera, Tortrici- dae). Ent. News 63: 145-149. Wareing, P. F., and H. A. Foda 1957. Growth Inhibitors and Dormancy in Xanthium seed. Phys. Plant. 10: 266-280. EVIDENCE FOR OBLIGATE MONOPHENISM IN RELIQUIA SANTAMARTA , A NEOTROPICAL-ALPINE PIERINE BUTTERFLY (LEPIDOPTERA:PIERIDAE) By Arthur M. Shapiro1 Department of Zoology University of California Davis, California 95616, U.S.A. Introduction The phenomenon of seasonal polyphenism under photoperiodic control is now well established in a variety of butterflies, especially members of the family Pieridae (Shapiro, 1976a). Although it is expressed in nature primarily by multivoltine populations at low to middle altitudes in strongly seasonal mid-latitude climates, po- lyphenism has been found in a latent form in univoltine species which are ordinarily monophenic. These include species coevolved with vernal-ephemeral host plants ( Pieris virginiensis Edwards, Shapiro, 1971; P. napi microstriata Comstock, Shapiro, 1975a) or facing short growing seasons due to altitude ( Pieris occidentalis “calyce” Edwards, Shapiro, 1974) or latitude ( P.o . nelsoni Edwards, Shapiro, 1975b). To date no population of either the Pieris calli- dice Hiibner or P. napi Linnaeus complexes has been found to be obligately monophenic, although some P. napi (Yukon Territory and central New Mexico, Shapiro, 1976a) produce only more-or- less heavily dark-veined phenotypes. The ability of univoltine Pierines to produce phenotypes analogous to normal seasonal ones produced by their multivoltine relatives, when reared in photo- period-temperature regimes which do not occur in their natural habitat, has been interpreted (Shapiro, 1976a) as evidence for the derivation of univoltinism/monophenism from multivoltinism/ po- lyphenism in the course of adaptation to new or changing climates. 'This work was made possible by grants from the National Geographic Society (USA) and the National Science Foundation (USA) and with the help of Colombian friends too numerous to mention. Special thanks go to Ms Adrienne R. Shapiro and Dr. Arthur S. Weston for their companionship and constant help afield in the Sierra Nevada. Manuscript received by the editor December 15, 1977. 183 184 Psyche [June Reliquia santamarta Ackery is an unusual Pierine of uncertain affinities which is known only from above 3500 m in the Sierra Nevada de Santa Marta of northeastern Colombia. It was dis- covered in 1971 and described four years later (Ackery, 1975). Phenotypically, the adult closely resembles high-altitude and -lati- tude members of the Holarctic Pieris callidice complex (figs. 1, 2) and is unlike the distinctive Andean montane and alpine Pierines ( Tatochila , Phulia, Piercolias). In those genera the submarginal black chevrons on the hindwing above and below point outward in the interspaces. In Holarctic Pieris and in R. santamarta they point inward. Morphologically R. santamarta is also close to Pieris, and indeed would be included therein under the traditional broad concept of the genus, which is beginning to break down (Kudrna, 1974). These facts suggest that R. santamarta might represent a relict of a Holarctic stock of the P. callidice complex which in- vaded northern South America during a cold period, presumably in the Pleistocene. The Sierra Nevada de Santa Marta, although only 62 km from the Serrania de Valledupar which connects to the northern Andes, shows a very high degree of faunal and floral endemism. The entomologist who knows it best, Michael Adams, is convinced (1973, 1975, and personal communications) that its butterfly fauna had not begun differentiating before the eastern Andean orogeny — thus precisely contradicting the hypothesis of Todd and Carriker (1922) of an eastern Andean origin for the Sierran alpine avifauna — and that certain groups speciated and underwent character- and altitudinal displacement in the Pleisto- cene. A.S. Weston (personal communication) has noted a floristic connection between the Sierran paramos and those of Costa Rica. At least one butterfly, Nathalis iole Boisduval (Pieridae, Coliadinae) is perhaps a Nearctic relict in the Sierra Nevada de Santa Marta. R. santamarta has been recorded from both dry seasons, the “verano” (“summer”; actually trade-wind season) in December- March and the shorter and less reliable “veranillo” in July — these being the only times of the year when weather conditions in the high Sierra would make butterfly collecting feasible. No pheno- typic differences are apparent among the putative broods. This is perhaps not surprising. Pierines in middle latitudes respond phe- notypically to daylength, but the Sierra lies at a latitude of 10°44' N, and the longest and shortest days of the year there differ in length by only about 70 minutes. The literature is devoid of photo- 1977] Shapiro — Reliquia phenotypes 185 periodic studies of circumequatorial insects, except for one paper by McLeod (1968) who excluded photoperiod as a factor in the seasonal polyphenism of an African Nymphalid. If, however, Re- liquia santamarta were a Pleistocene derivative of the Pieris calli- dice complex, it might be expected to show a latent polyphenism when reared under a laboratory regime that induces light pheno- types in that group. Long days, particularly continuous light, coupled with high temperatures are very effective in this regard (Shapiro, 1976a). Materials and Methods Eight females of R. santamarta were collected January 18-22, 1977 at and near the type locality (headwaters of the Rio Cam- birumeina and south slope of Cerro Icachui, 3950-4400 m). They were induced to oviposit in camp by confining them in sunlight in cylindrical tins 9.5 X 10.5 cm, covered outside with white glazed paper and topped with gauze, containing fresh sprigs of local Cru- cifers as oviposition substrates and Composite flowers as nectar sources. One female was a virgin, but all the others laid at least a few eggs. Eggs were placed on the plants, the gauze, and the tins. They were transported by ground to Valledupar, Department of Cesar, on January 26 and thence by air to Cali, Department of Valle del Cauca, the next day, where the first hatch occurred in the afternoon. Rearing was carried out in Cali on continuous light from a 60w bulb in plastic Petri dishes 18 cm X 3 cm on fresh sprigs of the Crucifer Lepidium virginieum L. collected from a vacant lot; this common weed has often been used in experiments with Pierines and is a frequent wild host of North American mem- bers of the callidice group. The rearing temperature was 26.5° ± 2° C. These conditions would induce light phenotypes in any Nearctic member of the callidice group which has been tested. Due to electricity rationing in Colombia, it was necessary to substitute a powerful candle for the lamp from 1800 to 1900 hours daily throughout the rearing period. My impression, based on pre- vious work with Pierines, is that this was read as “day” by the animals. Even if it was read as “night,” a 23-hour photophase has always been read as a “long day” by Nearctic species. R. santamarta proved difficult to rear under the experimental conditions in Cali. The culture started well, but about half the larvae died in the penultimate and ultimate instars of apparent 186 Psyche [June Figure 1. Dorsal surfaces of three Pierines from extreme climates, males at left. Top row: Reliquia santamarta, 4200 m. Sierra Nevada de Santa Marta, Colombia, 10°44' N. Centre: Pieris occidental is “calyce,” 3640 m, Colorado front range, USA, 40o01' N. Bottom: Pieris occidentalis nelsoni, Fairbanks, Alaska, 64°51' N. 1977] Shapiro — Reliquia phenotypes 187 Figure 2. Same as figure 1, ventral surfaces. 188 Psyche [June bacterial septicemia, despite stringent sanitation and daily replace- ment of the host plant. Ultimately 16 pupae were obtained, but 9 of these died without showing development. The remaining 7 pu- pae proceeded rapidly to the pharate adult, but failed to eclose. On February 15 they were all dissected to determine the pheno- types of the pharate adults — an easy and reliable procedure. Ex- amples of the early stages were preserved and the life-history will be described elsewhere. Results and Conclusions All of the pharate adults (5 males, 2 females) were in good enough condition for the ventral hindwing phenotype to be deter- mined (one female had to be degreased). All were completely normal, with the dark ventral pattern precisely as it occurs on wild specimens from the mountains. Despite the small sample size, the use of the candle, and the overall logistical difficulty of the experiment, this is a definitive result since the phenotypes were so consistent and because no known Nearctic Pierine reared under the same conditions would have given the same result. Obviously we cannot exclude the pos- sibility of a latent polyphenism, but it is made much less probable by the demonstration that it cannot be exposed under the most effective rearing regime known for that purpose. Since no Nearctic species yet tested has so canalized its phenotype, the likelihood that R. santamarta is a close relative of the species it most resembles is diminished. Why lose the potential for polyphenism? There is no obvious selective advantage in doing so. In Holarctic populations it is merely submerged when selection for an appropriate phenology alters photoperiodic thresholds. This is especially easy in taxa in which phenotype is somehow coupled to pupal diapause, since all diapaused pupae will yield dark adults. We do not know if R. santamarta is capable of diapause; certainly it would make sense during the wettest months (October and November), but we do not know what the environmental cues might be. At any rate, the animals collected in January 1977 were in very mixed condition, suggesting overlapping broods during the “verano.” As noted be- fore no sign of seasonal phenotypes has been detected even though 1977] Shapiro — Reliquia phenotypes 189 some of these adults would have been from diapause pupae if there are any such. We know from the experiment what non-diapause animals look like. The dark-veined phenotype is undoubtedly optimal at all sea- sons at Cambirumeina. In the two dry seasons the normal weather is clear in the morning and foggy in the afternoon. Night tempera- tures drop below freezing; before the fog asserts itself the tempera- ture may climb to 18° C, but then it drops to 5-7° C and remains there until after nightfall, when clearing occurs. During my stay fog set in from 1000 to 1400 hours on different days, i.e. from 2 to 6 hours after the initiation of flight activity. This was at the sunniest time of the year; during the rainy seasons the temperature probably hovers between 2° and 6° C most of the time, and sun- shine occurs only fleetingly. The R. santamarta phenotype is of a sort known to be thermoregulatorily used by Nearctic Pierines (Shapiro, 1975c, 1976a) and the behavior of R. santamarta afield precisely matches theirs (Shapiro, 1977). Pieris occidentalis “ca- lyce” Edwards in the Colorado front range has the same kind of weather during its flight season in August and matches R. santa- marta in pattern, almost scale for scale. However, it retains a latent polyphenism (Shapiro, 1974, 1976b). It is, of course, possible that the phenotype of R. santamarta is merely convergent to Pieris, and that no polyphenism has ever existed in its ancestry. If this is the case, the phytogeny of the Pierini is more confused than ever, and R. santamarta has no known close relatives. Summary Reliquia santamarta is a multivoltine Pierine butterfly from above 3500 m in the Sierra Nevada de Santa Marta, Colombia, latitude 10°44' N. Its single natural phenotype is extremely similar to the high-altitude and -latitude members of the Holarctic Pieris callidiee complex. When reared on continuous light at 26.5° ± 2°C, R. santamarta produced only its usual dark-veined, presumably thermoregu- latory phenotype. In this regard it differs from all previously tested, Holarctic Pierines, which display latent polyphenism attributable to their evolution from multivoltine, phenotypically plastic ancestors. The seeming lack of a latent po- lyphenism in R. santamarta casts doubt on the close affinity of that animal to the P. callidiee complex. Its relationships remain important for understanding the biogeography of the unusual endemic group of Andean Pierini. 190 Psyche [June References Ackery, P. R. 1975. A new Pierine genus and species with notes on the genus Tatochila (Lepidoptera: Pieridae). Bull. Allyn Mus. 30: 1-9. Adams, M. 1973. Ecological zonation and the butterflies of the Sierra Nevada de Santa Marta, Colombia. J. Nat. Hist. 7: 699-718. 1975. Full report of the third “North Colombia Butterflies Expedition 1974/ 75”. Mimeograph, 42 pp. Kudrna, O. 1974. Artogeia Verity 1947 gen. rev. for Papilio napi Linnaeus (Lep., Pieri- dae). Entomol. Gaz. 25: 9-12. McLeod, L. 1968. Controlled environment experiments with Precis octavia Cramer (Nym- phalidae). J. Res. Lepid. 7: 1 18. Shapiro, A. M. 1971. Occurrence of a latent polyphenism in Pieris virginiensis (Lepidoptera: Pieridae). Entomol. News 82: 13-16. 1974. Ecotypic variation in montane butterflies. Wasmann J. Biol. 32: 267- 280. 1975a. Developmental and phenotypic responses to photoperiod in uni- and bivoltine Pieris napi in California. Trans. R. ent. Soc. Lond. 127: 65-71. 1975b. Photoperiodic control of development and phenotype in a subarctic population of Pieris occidentalis (Lepidoptera: Pieridae). Can. Ento- mol. 107: 775-779. 1975c. Ecological and behavioral aspects of coexistence in six Crucifer-feeding Pierid butterflies in the central Sierra Nevada. Amer. Midi. Nat. 93: 424-433. 1976a. Seasonal polyphenism. in M. K. Hecht, W. C. Steere, and B. Wallace, eds.. Evolutionary Biology, vol. 9, pp. 259-333. Plenum Press, New York and London. 1976b. Photoperiodic responses of phenologically aberrant populations of Pierid butterflies (Lepidoptera). Great Basin Nat. 35: 310-316. 1977. Notes on the behavior and ecology of Reliquia santamarta, an alpine butterfly (Lepidoptera: Pieridae) from the Sierra Nevada de Santa Marta, Colombia, with comparisons to Nearctic alpine Pierini. Studies in the Neotrop. Fauna Envt., in press. Todd, W. E. C. and M. A. Carriker, Jr. 1922. The birds of the Santa Marta region of Colombia: a study in altitudinal distribution. Ann. Carneg. Mus. 14: 3-611. SEXUAL BEHAVIOR OF MURGANTIA HISTRIONICA (HEMIPTERA:PENTATOMIDAE) By Patrick J. Lanigan1 and Edward M. Barrows2 Introduction Murgantia histrionica (Hahn) is an economically important pest of brassiaceous crops in Central and Southern United States. As- pects of its biology have been studied by Paddock (1918), Chitten- den (1920), White and Brannon (1933), and Canerday (1965); how- ever, its sexual behavior has not been studied in detail. Therefore, we studied courtship, copulation, polygyny, and polyandry in this bug. Fish and Alcock (1973) and Gamboa and Alcock (1973) have reviewed the literature on pentatomid courtship and copulation. Materials and Methods In July and August, 1976, nymphs and adults of M. histrionica were collected in Washington, D. C. from cabbage, Brassica olera- cea capitata L.; broccoli, B. o. cymosa L.; and radish, Raphanus sativa L. Groups of four to seven adult or nymphal bugs were maintained in petri dishes in the laboratory. The laboratory was illuminated with flourescent light and indirect sunlight, and tem- peratures varied from 24 to 38° C. Bugs were fed pieces of leaves of B. o. capitata or R. sativa that were changed about daily. When conspicuous mold appeared on bug feces in petri dishes, bugs were transferred to clean dishes to reduce mortality. Only virgin adults were used in the study of courtship and copu- lation to reduce possible behavioral variability due to learned sex- ual behavior. In each trial of the experiment a male was removed from his dish with forceps and placed in a dish with a female. The female’s dish was placed on white paper to provide a good back- ground for observation. Food was removed from a female’s cham- 1 Department of Biology, Georgetown University, Washington, D.C., 20057. department of Biology, Georgetown University, Washington, D.C., 20057, and The University of Michigan Biological Station, Pellston, Michigan, 49769. Send reprint requests to E. M. Barrows at Georgetown University. Manuscript received by the editor November 10, 1977. 191 192 Psyche [June ber prior to introduction of a male. Males were marked on their scutella with dots of fast-drying, enamel paint so that they could be readily distinguished from females. In November, 1976, M. histrionica were collected from the field for a study of polygamy. Twenty pairs were used, and the bugs may or may not have been virgins. One male and one female were placed with food in each of 20 petri dishes. After a pair copu- lated, the female was transferred to a dish with a different male. A sample mean is denoted by X; median, M; and size, N. Results Fourteen of 20 courtships of M. histrionica led to copulation (Fig. 1 and 2). Durations from introduction of a male to copula- tion initiation ranged from 8.5 to 47.3 min (X — 26.8, M = 24.5). Main steps in this bug’s sexual behavior are as follows: (1) The male approaches the female with his antennae waving while she is mobile or immobile. She may become motionless and “crouch down” on the substrate if she is approached when moving, or she may escape and crawl away from the male. (2) The male antennates the female while she is motionless. If his approach is from her front, he antennates her antennae and then her scutellum. If his approach is from behind, he antennates her posterior abdominal segments and posterior portions of her folded wings. Eighteen antennations lasted from 3 to 15 sec (X = 7.0, M = 8.5). (3) After vigorous male antennation, the female rapidly and vio- lently jerks her body sideways for approximately 3 sec. If the male approaches her from behind, he moves in front of her and anten- nates her antennae, and then the female shows jerking movements. Females may escape from males during this step in sexual behavior. (4) The male moves to the female’s side, stops, and antennates her side in different places. Antennating males frequently orient approximately 60, 90, 120° with respect to the female longitudinal axis. (Zero degrees is considered to be when the bugs are head to head.) Eleven males used all three orientations and stopped to antennate for from 3 to 5 sec before moving to female posteriors. Females respond to antennation by violent sideways jerking. (5) When the male reaches the female posterior, he antennates and she shows sideways jerking for from 2 to 5 sec. He strokes the 1977] Lanigan & Barrows — Murgantia histrionica 193 male approaches front of 12 female male approaches rear of g female 12 18 male ante-mates female antennae 1 7 male moves to side ' of female and antennates the side of her abdomen 16 male moves to rear of female and antennates posterior part of female abdomen, female raises her abdomen , , 15 male extends his aedeagus and turns away from the female 15 male antennates posterior part of g female abdomen female crawls away from the male 6 14 pair copulates 14 Figure 1. Behavioral sequence in courtship involving 20 pairs of Murgantia histrionica. See the text for further explanation. 194 Psyche [June female venter alternately with his left and right antennae, and then the female raises her abdomen to approximately 30° above the substrate. Females may escape during this step. (6) With his aedeagus extended, the male makes a 180° turn which maneuvers him into a position from which he initiates linear copulation. Unreceptive females may also escape at this time. (7) The male elevates his abdomen and backs directly against the female, and he initiates copulation. (8) Once the pair starts copulating, they use their middle and hindlegs to jerk their bodies up and down. In 14 observations, jerking lasted from 10 to 185 sec (X = 52, M = 48). This behavior also occurred from 5 min to 6 hr after copulation initiation. (9) The male rapidly strokes up and down the sides of the fe- male abdomen with both his hindlegs. This occurs during and after body jerking. Male leg stroking of the female lasted from 4 to 27 sec (X = 13, M = 11, N 1 14). (10) The pair crawls in the petri dish with the female, which is usually larger, pulling the male in most cases. In 71 observation periods, copulating pairs remained motionless from 5 to 1260 sec (X = 453, M = 315). When food was placed in the chamber after several hours of its absence, bugs often fed while in copula. Copu- lation duration ranged from 5.4 to at least 8.3 hr. The termination of copulation was not observed in most pairs. In petri dishes, a common female escape behavior was quick- crawling to dish sides with the male in pursuit. Females on dish sides continued at steady paces and males followed them with quick-crawling bouts interspersed with rests. Males waved anten- nae as they approached females. If males reached females, or if females stopped, courtship was hindered because females often escaped when males tried to move in front of them. Females also escaped while males antennated their sides, venters, and when males made 180° turns before aedeagal insertions. Males that were on dish tops or bottoms courted females which were on dish sides, but males did not crawl on dish sides to attain typical positions that occur during copulation initiation. In other situations, two females raised their wings as males approached, and they crawled away as males attempted to court them. Murgantia histrionica was polygynous and polyandrous in the laboratory. In the polygamy experiment, two males copulated with four different females; four males, with three females; and 1977] Lanigan & Barrows — Murgantia histrionica 195 Figure 2. Murgantia histrionica in copula on radish foliage. The male faces to the right and lacks part of his right antenna. two males, with two females. One male copulated once, and 1 1 other males did not copulate. Two females copulated with three different males; six females, with two males; seven females, with one male; and five females did not copulate. Discussion Gamboa and Alcock (1973) reported three major methods by which pentatomids initiate copulation, one of which involves males and females facing in opposite directions, which we found in M. histrionica. Fish and Alcock (1973) noted that other species which employ this method have similar courtship behaviors. They listed four main behaviors shared by species of this group: (i) male an- tennation of the female, (ii) male attempt to lift the female abdo- men with his head, (iii) abdominal elevation by receptive females, and (iv) male antennal and aedeagal stimulation of female abdo- mens. Fish and Alcock conclude that males attempt to induce fe- males to assume positions that facilitate aedeagal insertion. Mem- bers of at least six other genera of pentatomids, besides Murgantia, 196 Psyche [June are known to initiate copulation when males and females face op- posite directions (Olsen, 1910; Esselbaugh, 1948; Mitchell and Mau, 1969; Alcock, 1971; Fish and Alcock, 1973; and Gamboa and Alcock, 1973). In M. histrionica, male antennation of the female abdomen is followed by female abdominal-raising. Male M. histrionica did not attempt to raise female abdomens with their heads as do other males of genera in which bugs initiate copulation while facing op- posite directions (Gamboa and Alcock, 1973). In this species, male antennation might be a more derived behavior than male use of heads to raise female abdomens as displayed in other species. John Alcock (pers. comm.) observed about 25 courting pairs of M. histrionica on C/eome jonesii (MacBride) Tibestrom (Cappa- ridaceae) on which they also fed, in Patagonia, Arizona, in July 1976. Bug courtship in Arizona was similar to that which we ob- served, but in addition, males used their abdominal tips to push at and lift female abdomens. Alcock saw many pairs in copula; however, none of the observed courtships led to copulation. As noted above, Murgantica histrionica was polyandrous and polygynous in our laboratory study. These behaviors also occur in Podisus modestus (Dallas) (Tostowaryk, 1971) and in Nezara viridu/a (L.) (Mitchell and Mau, 1969). Acknowledgements Jon Herring (USDA, Washington, D. C.) made available his pentatomid literature file for this study. John Alcock (Arizona State University) and Lawrence S. Oliver (Georgetown University) offered important suggestions regarding a draft of this note. Literature Cited Alcock, J. 1971. The behavior of a stink bug, Euschistus conspersus Uhler (Hemiptera: Pentatomidae). Psyche 78: 218-228. Canerday, T. 1965. On the biology of the harlequin bug, Murgantia histrionica (Hemiptera: Pentatomidae). Ann. Entomol. Soc. Amer. 58: 931-932. Chittenden, F. H. 1920. Harlequin cabbage bug and its control. U. S. Dept. Agr. Farmers Bull. 1061. 1977] Lanigan & Barrows — Murgantia histrionica 197 Esselbaugh, C. O. 1948. Notes on the bionomics of some midwestern Pentatomidae. Entomol. Americana 28: 1-73. Fish, J. and J. Alcock. 1973. The behavior of Chlorochroa ligata (Say) and Cosmopepla bimaculata (Thomas) (Hemiptera: Pentatomidae). Entomol. News 84: 260-268. Gamboa, G. and J. Alcock. 1973. The mating behavior of Brochymena quadrapustulata (Fabricius). Psyche 80: 265-270. Mitchell, W. C. and R. F. L. Mau. 1969. Sexual activity and longevity of the southern green stinkbug, Nezara viridula. Ann. Entomol. Soc. Amer. 62: 1246-1247. Olsen, C. E. 1910. Notes on breeding Hemiptera. J. N. Y. Entomol. Soc. 18: 39-42. Paddock, M. S. 1918. Studies on the harlequin bug. Texas Agr. Exp. Sta. Bull. 227. Tostowaryk, W. 1971. Life history and behavior of Podisus modestus (Hemiptera: Pentatomi- dae) in a boreal forest in Quebec. Can. Entomol. 103: 662-674. White, W. H. and L. W. Brannon. 1933. The harlequin bug and its control. U. S. Dept. Agr. Farmers Bull. 1712. CAMBRIDGE ENTOMOLOGICAL CLUB A regular meeting of the Club is held on the second Tuesday of each month October through May at 7:30 p.m. in Room 154, Biological Laboratories, Divinity Ave., Cambridge. Entomolo- gists visiting the vicinity are cordially invited to attend. BACK VOLUMES OF PSYCHE Requests for information about back volumes of Psyche should be sent directly to the editor. F. M. Carpenter Editorial Office, Psyche 16 Divinity Avenue Cambridge, Mass. 02138 FOR SALE Reprints of articles by W. M. Wheeler The Cambridge Entomological Club has for sale numerous reprints of Dr. Wheeler’s articles that were Filed in his office at Harvard University at the time of his death in 1937. Included are about 12,700 individual reprints of 250 publications. The cost of the re- prints has been set at 5c a page, including postage; for orders under $5 there will be an additional handling charge of 50c. A list of the reprints is available for $1.00 from the W. M. Wheeler Reprint Committee, Cambridge Entomological Club, 16 Divinity Avenue, Cambridge, Mass. 02138. Checks should be made payable to the Cambridge Entomological Club. PSYCHE A JOURNAL OF ENTOMOLOGY founded in 1874 by the Cambridge Entomological Club Vol. 84 September-December, 1977 No. 3-4 CONTENTS Dispersal Behavior of Honey Bee Swarms. Thomas D. Seeley and Roger A. Morse 199 Seasonality and the Flight of Paussids (Coleoptera) in West Africa. Dennis Lest on 210 An Aberrant New Genus of Myrmicine Ant from Madagascar. William L. Brown, Jr 218 Symbioses Between Insects and Spiders: An Association Between Lepidopteran Larvae and the Social Spider, Anelosimus eximus (Araneae: Theridiidae). Michael H. Robinson 225 Population Structure and Polymorphism in the Slave-Making Ant, Harpagox- enus americanus (Emery) (Hymenoptera: Formicidae). Alfred Buschinger and Thomas M. Alloway 233 New Records and Species of Leiodinae and Catopinae (Coleoptera: Leiodidae) from Jamaica and Puerto Rico, with a Discussion of Wing Dimorphism. Stewart B. Peck 243 Observations on the Nests and Prey of Eumenid Wasps (Hymenoptera, Eu- menidae). Howard E. Evans 255 New Name for a Triassic Mayfly from South Africa. Michael D. Hubbard and E. F. Riek 260 The Larva of Rothium sonorensis Moore & Legner, with a Key to the Known Larvae of the Genera of the Marine Bolitocharini (Coleoptera: Staphylini- dae). Ian Moore 262 Comparative Studies of Dictyna and Mallos (Araneae, Dictynidae). III. Prey and Predatory Behavior. Robert R. Jackson 267 A Supplement to the World Revision of Odontomachus (Hymenoptera: For- micidae). William L. Brown, Jr 281 New Observations of Maternal Care Exhibited by the Green Lynx Spider, Peucetia viridans Hentz (Araneida:Oxyopidae). John B. Randall 286 Fighting Behavior of Male Golofa porteri Beetles (Scarabeidae:Dynastinae). William G. Eberhard 292 A Review of the Distribution and Biology of the Small Carrion Beetle, Pri- onochaeta opaca of North America. (Coleoptera; Leiodidae; Catopinae). Stewart B. Peck 299 The Genera of Eastern North American Chloroperlidae (Plecoptera): Key to Larval Stages. Sandy B. Fiance 308 Author and Subject Index to Volume 84 319 CAMBRIDGE ENTOMOLOGICAL CLUB Officers for 1977-1978 Gary D. Alpert John A. Shetterly Robert Robbins Frank M. Carpenter Martha Fisher Katherine Horton EDITORIAL BOARD OF PSYCHE F. M. CARPENTER (Editor), Fisher Professor of Natural History, Emeritus, Harvard University ALFRED F. Newton, JR., Curatorial Associate in Entomology, Har- vard University W. L. BROWN, Jr., Professor of Entomology, Cornell University, and Associate in Entomology, Museum of Comparative Zoology P. J. DARLINGTON, Jr., Professor of Zoology, Emeritus, Harvard University B. K. HOLLDOBLER, Professor of Biology Harvard University H. W. LEVI, Alexander Agassiz Professor of Zoology, Harvard University R. E. SlLBERGLIED, Assistant Professor of Biology, Harvard University E. O. WILSON, Baird Professor of Science, Harvard University President Vice-President Secretary Treasurer Executive Committee PSYCHE is published quaterly by the Cambridge Entomological Club, the issues appearing in March, June, September and December. Subscription price, per year, payable in advance: $8.00 for United States and Canada, $9.50 for other countries. Single copies, $2.50. Checks and remittances should be addressed to Treasurer, Cambridge Entomo- logical Club, 16 Divinity Avenue, Cambridge, Mass. 02138. Orders for missing numbers, notices of change of address, etc., should be sent to the Editorial Office of Psyche, 16 Divinity Avenue, Cambridge, Mass. 02138. For previous volumes, see notice on inside back cover. IMPORTANT NOTICE TO CONTRIBUTORS Manuscripts intended for publication should be addressed to Professor F. M. Car- penter, Biological Laboratories, Harvard University, Cambridge, Mass. 02138. Authors are expected to bear part of the printing costs, at the rate of $23.50 per printed page. The actual cost of preparing cuts for all illustrations must be borne by contributors: the cost for full page plates from line drawings is ordinarily $18.00 each, and for full page half-tones, $30.00 each; smaller sizes in proportion. The June, 1977, Psyche (Vol. 84, No. 2) was mailed March 28, 1978 The Lexington Press, Inc., Lexington, Massachusetts PSYCHE Vol. 84 September-December, 1977 No. 3-4 DISPERSAL BEHAVIOR OF HONEY BEE SWARMS* By Thomas D. Seeley and Roger A. Morse Museum of Comparative Zoology Harvard University, Cambridge, Massachusetts 02138 and Department of Entomology Cornell University, Ithaca, New York 14853 Introduction Food shortages were probably a major ecological force upon the European races of honey bees ( Apis mellifera) in their natural habitat of temperate deciduous forest. And many features of honey bee biology are reasonably interpreted as techniques of competition for food. For example, the demographic properties of descendant honey bee colonies in North America, such as low reproductive rate and infrequent but expensive offspring, probably reflect selec- tion for competitive ability rather than productivity (Seeley 1978). Also the honey bee’s sophisticated recruitment system involving dance language and olfactory recruitment (von Frisch 1967, Gould 1975) seems ideal for a “scramble” type competitive device involv- ing rapid discovery and exploitation of food sources. Furthermore bees from different colonies will fight at feeding dishes when the food is in short supply (Kalmus 1941) and will reduce each others’ foraging range (Levin 1961, Levin and Glowska-Konopacka 1963, Gary et al. 1972, 1973, 1975). Thus honey bee colonies can appar- ently also compete for food sources using techniques of “contest” competition. * Manuscript received by the editor March 15, 1978. 199 200 Psyche [September-December Behaviors promoting colony spacing are another line of adapta- tions to limited food supplies and are widespread among the social insects (Brian 1965, Wilson 1971). For honey bees these behaviors fall logically into two classes: ( 1 ) attack by established colonies upon adjacent colonies, and (2) avoidance of established colonies by swarms when selecting nest sites. Behavior of the first category is apparently of minor importance with honey bees since bee colonies can be crowded into peaceful apiaries. However strong colonies occasionally plunder nearby weak colonies. Regarding the second category, Lindauer (1955) provides evidence suggesting that honey bee swarms avoid their parent colonies by selecting new nest sites at least a few hundred meters from the original nest. Given the importance of understanding colony spacing to a clear understand- ing of honey bee ecology, especially intraspecific foraging compe- tition, we decided to investigate the dispersal behavior of honey bee swarms. Materials and Methods The honey bees used in this study came from the Dyce Honey Bee Laboratory, Cornell University, and were hybrids of the Euro- pean races of honey bees imported for American apiculture. These races include primarily Apis mellifera ligustica Spinola, A. m. cau- casica Gorbatschew, A. m. carnica Pollmann and A. m. mellifera L. (Ruttner 1975). The study of swarm dispersal distance was con- ducted during the summers of 1976 and 1977 on Mount Pleasant, a large area of mature forest near Ithaca, New York. The test of swarms’ preferred dispersal distance was performed during Decem- ber, 1977 and January, 1978 at the Archbold Biological Station, Lake Placid, Florida. There the study area consisted of a sandy plain which extends for many kilometers and which is primarily covered by scrubby vegetation. The widely dispersed pine trees in this area offer very few, if any, natural nest sites for honey bees. Additional methodological details will be given with the descrip- tions of the individual experiments. Experiments and Results Distribution of Swarm Dispersal Distances Upon departing its parent colony, a honey bee swarm flies only a few tens of meters before assembling to form a hanging swarm 1977] Seeley & Morse — Honey Bee Swarms 201 cluster (Ambrose 1974). Scout bees fly from this cluster in search of a nest site and later recruit other scouts to newly discovered nest sites using the dance language (Lindauer 1955). These communica- tion dances are conspicuously performed on the surface of the swarm. Thus one can measure approximately how far swarms move between parent and new colony sites by reading the recruit- ment dances of the scout bees on swarms. We used artificial swarms of honey bees which were prepared as follows. First, worker bees were shaken off frames of a beehive into a swarm cage (15 X 25 X 35 cm) of wood and wire screen sides using a large funnel. Then the swarm’s queen was removed from the beehive and confined in a standard queen mailing cage (3.2 X 10 X 1.6 cm) which was suspended amidst the worker bees in the larger swarm cage. The bees were kept confined and liberally fed with a 50% sucrose solution for at least 24 hours. Bees treated in this way behave like a natural swarm. If placed near their parent hive, they do not return to it but instead search for a new nest site. We controlled swarm size by weighing the workers shaken into the swarm cage. The swarms all weighed approximately 2 kg (about 15,000 bees), a typical size for natural swarms (Fell et al. 1977). Each swarm was placed on a wood cross (120 cm high with a 46 cm long cross member) in the study area by tying the caged queen to the cross. The worker bees, upon being shaken from the swarm cage, would cluster about the caged queen. A 1 -liter, gravity feeder jar provided sugar syrup continuously for each swarm. We posi- tioned the swarms in a small clearing surrounded by forest for at least 1 km and generally 2 or more km. Thus the swarms were surrounded at both small and large distances by a presumably ran- dom distribution of natural nest sites. The swarms were run one at a time, except for three swarms which we observed simultaneously. The three concurrently run swarms were positioned at least 30 m from each other. We followed each swarm’s selection of a nest site from start to finish by reading the scout bees’ dances to determine the distances and directions to the nesting sites they had discovered. The cali- bration curves of von Frisch (1967) and Lindauer (1971) for Apis mellifera ligustica were used to translate dance tempos into dis- tances to the advertised nest sites. The dances representing the site finally selected by each swarm were easily recognized by the frenzy with which they were performed and by their heavy preponderance 202 Psyche [September-December Fig. 1 . Distributions of distances from swarm cluster sites to new nest sites, as calculated from the recruitment dances of scout bees on swarms. The curves are least-squares fits to points centered in the top of the histogram bars. Upper figure is original data; lower figure is after Table 1 in Lindauer (1955). over all other dances just before each swarm lifted off to fly to its chosen site. The results from observing 13 swarms are shown in Fig. 1. This figure also includes data gathered in similar fashion by Lindauer (1955) who observed swarms inside Munich and in various rural locations in West Germany. The curves in Fig. 1 were fitted to the histograms as described in Seeley and Morse (1978). The striking features of the two distributions are (1) their similarity despite widely separated study areas, and (2) the low frequency of swarms 1977] Seeley & Morse — Honey Bee Swarms 203 travelling less than 300 m to a new nest site. This pattern may simply reflect the smaller number of nest sites within a small radius area relative to a large radius area. But it could also represent a preference by swarms for nest sites beyond 300 meters from their cluster sites. Since swarms generally travel only a few tens of me- ters from the parent colony before settling at an interim cluster site, a preference for nest sites far beyond the cluster site would promote the dispersion of parent and daughter colonies. The fol- lowing section reports a test for this preference. Test of Preference for Distant Nest Sites To test whether swarms prefer distant nest sites, we offered swarms a choice between two nestboxes which were constructed and positioned as identically as possible, but with one 20 and the other 400 m from each swarm’s cluster site. These distances cor- respond to the low end tails and the modes of the distributions in Fig. 1. Lindauer (1955) performed a similar experiment with nest- boxes 30 and 250 m from a swarm. His swarm chose the 250 m site. However, the lack of repetitions and of controls for differ- ences in nest site exposure, to which bees are highly sensitive, make this experiment’s result suggestive rather than conclusive. The nestboxes for our experiment were nailed onto two very similar sand pines 380 m apart. Each nestbox was cube-shaped, 40 liters in volume, and had a 3 cm diameter entrance hole posi- tioned midway across the front, 8 cm up from the nestbox floor. A nail driven horizontally across the entrance prevented occupa- tion by birds. The nestboxes were constructed of 1.5 cm thick plywood and were painted dark green on the outside. Nestbox floors were removable to permit interior inspections. The seam between the floor and the walls of each nestbox was sealed with opaque photographic tape. The entrance of both nestboxes faced south and was 3.75 m above the ground. The wind, sun and rain exposures of both nestboxes were carefully matched by trimming off branches about the nestboxes and by nailing a shade board (56 X 100 cm) atop each nestbox. Each trial of the test was started by introducing a colony of bees into the study area, positioning it 30 m from one of the nestboxes as shown in Fig. 2. We left each colony undisturbed for at least two days of good weather to provide time for the colony’s orienta- tion to its new home range. On the third day or later an artificial 204 Psyche [September-December li Nbi i Nb2 H Sw 20 380 m Fig. 2. Experimental array for testing preference in swarm dispersal distance. H, hive; Sw, swarm on wooden cross; Nb, nestbox. In the five trials of the test, the hive and swarm were alternately positioned near nestbox 1 or nestbox 2, always maintaining the distance relationships shown. All distances are in meters. The ob- jects in this figure and the spacings between objects are not drawn to the same scale. swarm was prepared from the hive as described above, but with the difference that each swarm was confined and fed in the swarm cage for only one hour. Following this brief confinement each swarm was placed on a wooden cross located between the parent hive and the nearby nestbox, 10 m from the former and 20 m from the latter. Thus a swarm cluster was established a natural distance from its parent colony and 20 and 400 m from two otherwise closely matched nest sites, as shown in Fig. 2. The parent hive and swarm were placed at opposite ends of the nestbox array on alternate trials. This provided control for possible differences be- tween the nestboxes besides distance from the swarm. We monitored each swarm’s selection of a nest site by reading the recruitment dances as described above. But besides following the dances on the swarm, we periodically (at least hourly) meas- ured the number of scout bees visible at each nestbox by making 10 counts, each count 15 seconds apart, while standing directly in front of a nestbox. Both near and far nestboxes were always rap- idly discovered by the scout bees. Each morning, before the bees started flying, we inspected the interior of both nestboxes. Often 2 to 10 ants were found in the nestboxes and were promptly re- moved and killed. One morning during the second trial a complete ant colony (queen, workers and brood) was discovered in the far nestbox. This nestbox was quickly sealed with the ants inside, re- moved and replaced with a new nestbox. This was the only nest- box change performed during the experiment. 1977] Seeley & Morse — Honey Bee Swarms 205 Table 1. Outcomes of five choices by honey bee swarms between near and far nestboxes. Trial Nestbox Near Swarm Nestbox Chosen Distance to Selected Nestbox (m) 1 1 1 20 2 2 2 20 3 1 1 20 4 2 2 20 5 1 2 400 We prevented swarms from occupying the nestboxes by keeping each swarm’s queen caged on the wooden cross in a standard queen mailing cage. Every swarm’s attempt to move to a nestbox ended with its return to the caged queen. But even though swarms never occupied the nestboxes, their nestbox preferences were al- ways clearly indicated by large differences in the number of dances for and scouts at the two nestboxes. The outcomes of five swarms’ selections of a nest site are shown in Table 1. Apparently the nestboxes 1 and 2 were well matched, because with 2 and 3 selections respectively, no significant prefer- ence for either nestbox was shown. More important was the pat- tern of choice between near and far nestboxes: 4 to 1, respectively. The estimated probability of a swarm choosing the near nestbox is 0.80 and the 95% confidence limits on this probability are 0.48 and 0.95. Thus these results do not support the hypothesis that swarms prefer distant nest sites. Instead, they suggest that swarms prefer nearby nest sites. Unfortunately, because of lack of time, we were unable to perform further trials of this experiment. The one selection of the distant nest site may not even be a valid test result. For in the fifth trial the swarm’s choice between the nestboxes proceeded differently than with the previous four swarms. As is shown in Fig. 3, this swarm’s preference developed initially in favor of the near nestbox and reached a point at which we expected the swarm to lift off and attempt moving to the near nestbox. Then suddenly the situation reversed. The scouts decreased at the near nestbox and increased at the far nestbox. Finally the swarm lifted off and flew past the near nestbox en route to the far nestbox. In the previous four trials, each swarm’s preference between nestboxes 206 Psyche [September-December Fig. 3. Record from the fifth trial of a swarm’s selection between near and far nestboxes, as monitored by counting the scouts visible at each nestbox. Vertical bars denote plus and minus one standard deviation for 10 counts at 15 second intervals. developed smoothly and steadily in favor of the nestbox which was ultimately chosen. Moreover, when we inspected both nestboxes shortly after the lift-off in the fifth trial, we found the far nestbox empty inside except for a few scout bees, but we discovered four ants in the near nestbox. Similar nestbox inspections in the previ- ous four trials had not disclosed any ants in either nestbox. These observations suggest that ants interfered in the fifth trial by enter- ing the near nestbox. If so, then this may have created a difference in the nestboxes’ qualities which outweighed any quality difference based upon the nestboxes’ different distances from the swarm. Discussion The two experiments reported here appear to give conflicting results concerning the dispersal behavior of honey bee swarms. In the first experiment we observed swarms generally travelling a large distance, at least 300 meters, to new home sites. But in the second experiment, wherein we provided nest sites at 20 and 400 meters, the swarms showed no preference for the more distant nest site. This difference in dispersal behavior probably does not reflect differences between the bees used in the experiments. In both ex- 1977] Seeley & Morse — Honey Bee Swarms 207 periments the bees came from the same source, the apiaries of Dyce Laboratory, and in both experiments the bees were prepared as artificial swarms using nearly identical techniques. We suspect the disparity in experimental outcomes simply reflects a lack of nearby nest sites in the first experiment which forced the swarms to choose distant nest sites. If so, then our findings suggest that in nature the spacing out of feral honey bee colonies is based more upon the dispersion of suitable nest sites than upon programmed dispersal behavior in honey bee swarms. Our findings also suggest that swarms prefer moving only a short distance to a new home site. Minimizing dispersal distance may be advantageous to swarms in reducing the hazard of losing poor fly- ing queens. It might also help keep the daughter colony near the closely related, and thus perhaps minimally aggressive, mother col- ony. Robbing and foraging range restriction are probably the most common forms of aggression between bee colonies. Furthermore, because the honey bee’s flying ability enables it to forage over very large areas, colony dispersal may not significantly reduce the com- petition, if any, between colonies for food. We close this report by stating a possible weakness of this study: use of artificial swarms. Because this study’s experiments required many repetitions, they would have proceeded exceedingly slowly had we used only swarms emerging naturally from colonies placed at the study sites. Thus we used the readily available artificial swarms. And these swarms appear to behave normally while se- lecting a nest site. They form a quiet cluster, dispatch scouts which discover and select the new home site, and finally fly to the chosen site. However, if a swarm’s dispersal behavior is stimulated by its scouts’ close familiarity with the surrounding region or is depen- dent upon the natural process of swarm formation, then our arti- ficial swarms would have shown abnormal dispersal behavior. Acknowledgments We thank Richard Nowogrodzki for field assistance and the Archbold Biological Station for the use of their research facilities. Bert Holldobler critically reviewed the manuscript. Supported by the National Science Foundation (Grant No. BMS 76-15008), and the Anderson Fund and the Parker Fellowship, both of Harvard University. 208 Psyche [September-December Summary Insofar as normal honey bee behavior was observed in these studies with artificial swarms, our results indicate that swarms fre- quently move at least 300 meters from their parent colony to a new nest site, but that they do not prefer nest sites far from their parent colonies. Instead, swarms may prefer a nest site which is near the parent colony. Therefore the spacing of suitable nest sites appears to be a major determinant of the spacing of feral honey bee colo- nies, and behaviors promoting colony spacing to reduce foraging competition may not exist in the European races of honey bees. References Ambrose, J. T. 1975. A study of honey bee ( Apis mellifera L.) swarms while in transit to a new homesite. Ph.D. thesis, Cornell Univ., Ithaca, N. Y. Brian, M. V. 1965. Social insect populations. Academic Press, London and New York, 135 pp. Fell, R. D., J. T. Ambrose, D. M. Burgett, D. DeJong, R. A. Morse and T. D. Seeley 1977. The seasonal cycle of swarming in honey bees. J. apic. Res. 16, 1 70— 173. Frisch, K. von 1967. The dance language and orientation of bees. Belknap Press of Harvard University Press, Cambridge, Mass., 566 pp. Gary, N. E., P. C. Witherell and J. Marston 1972. Foraging range and distribution of honey bees used for carrot and onion population. Environ. Entomol. 1, 71-78. Gary, N. E., P. C. Witherell and J. M. Marston 1973. Distribution of foraging bees used to pollinate alfalfa. Environ. Ento- mol. 2, 573-578. Gary, N. E., P. C. Witherell and J. M. Marston 1975. The distribution of foraging honey bees from colonies used for honey- dew melon pollination. Environ. Entomol. 4, 277-281. Gould, J. L. 1975. Honey bee recruitment: the dance-language controversy. Science 189, 685-693. Kalmus, H. 1941. Defence of source of food by bees. Nature 148, 228. Levin, M. D. 1961. Interactions among foraging honeybees from different apiaries in the same field. Insectes sociaux 8, 195-201. Levin, M. D. and S. Glowska-Konopacka 1963. Responses of foraging honeybees in alfalfa to increasing competition from other colonies. J. apic. Res. 2, 33-42. 1977] Seeley & Morse — Honey Bee Swarms 209 Lindauer, M. 1955. Schwarmbienen auf Wohnungssuche. Z. vergl. Physiol. 37, 263-324. Lindauer, M. 1971. Communication among social bees. Harvard University Press, Cam- bridge, Mass., 161 pp. Ruttner, F. 1975. Races of bees, pp. 19-38 In Dadant and Sons (Eds.), The hive and the honey bee. Dadant, Hamilton, 111., 740 pp. Seeley, T. D. 1978. Life history strategy of the honey bee. Apis mellifera. Oecologia. 32, 109-118. Seeley, T. D. and R. A. Morse 1978. Nest site selection by the honey bee, Apis mellifera. Insectes sociaux. In press. Wilson, E. O. 1971. The insect societies. Belknap Press of Harvard University Press, Cam- bridge, Mass., 548 pp. SEASONALITY AND THE FLIGHT OF PAUSSIDS (COLEOPTERA) IN WEST AFRICA By Dennis Leston* Biological Sciences Group University of Connecticut Storrs, CT 06268 Introduction The paussids of this paper are the brown to black beetles, less than 8mm in length, with aberrant antennae, variously regarded as a superfamily, family or subfamily within Carabidae, or of even lower taxonomic rank (Darlington, 1950): all are myrmecophiles. That these insects come to light is well known; that the emission of adults is seasonal rests, as far as I am aware, unreported. In West Africa almost all insects are markedly seasonal in breed- ing, dispersal, population growth or other life history phenomena. The insect periodicities can be tied to underlying events at the primary producer and/ or decomposer levels and it has been dem- onstrated that these are associated with periodicities delimited by a combination of rainfall and sunshine (not daylength) factors (Gibbs & Leston, 1970; Leston, 1972, 1978; Leston & Gibbs, 1971). This paper presents the evidence for seasonality in Ghana, at- tempting to place the ultimate factors within the framework of West African phenology. Material and Method The material was named using the collection of the British Mu- seum (Nat. Hist.), London, where voucher specimens have been deposited. Taxonomists have clearly oversplit and the polytypic concept is not used by them. Paussus cilipes Westwood — represented in West Africa by a subspecies other than the nominate. ♦Present address: Department of Biology, University of Miami, Coral Gables, Florida 33124. Manuscript received by the editor February 28, 1978. 210 1977] Lest on — Paussids in West Africa 211 Paussus klugi Westwood — I suspect this and P. latreillei West- wood are conspecific, with the West African subspecies distinct from the nominate. Paussus setosus Westwood. Paussus sphaerocerus Afzel — outside of the sampling area this is abundant in the forest zone of Ghana. Paussus spinicoxis Westwood — again, the West African form differs at the subspecies level from the nominate one. Paussus sp 4 — this is not in the British Museum collection. Platyrhopalopsis laevifrons (Westwood) — not found in my sam- ples and known to me only from a specimen from Tumu, Upper Region (P. M. Room). The seven listed comprise all the Paussina (Darlington’s subtribe) found in Ghana. A 125 watt Robinson ultraviolet light-trap was run for 400 days on the campus of the University of Ghana, Legon, Accra District, Ghana. The catches were grouped into 20-day classes, a method found of value in previous investigations (Gibbs & Leston, 1970; Leston, 1973a). Sudden rain caused breakdowns on 14 nights: the figures were corrected by dividing the total for each species, for the relevant 20-day period, by the number of days actually sampled and adding this result for each missing sample — however, the overall results would have been the same even uncorrected. The original data sheets are deposited in the library of the Royal En- tomological Society of London. Legon, 5°40'N, was once forested but is now an area of derived savanna at the edge of the dry Dahomey Gap. Food-farms, gar- dens and buildings cover the region but once shade has been arti- ficially reestablished forest zone crops such as cocoa and robusta coffee can be grown successfully. Numbers caught were Results 1. Paussus spinicoxis 758 2. P. sphaerocerus 165 3. P. cilipes 67 4. P. setosus 54 5. P. klugi 13 6. P. sp 4 6 Total 1063 212 Psyche [September-December Fig. 1 . Log frequency distribution and regression for the six paussid species trapped. The log frequency was plotted, together with the calculated regres- sion (Fig. 1). That the six species have a logarithmic distribution pattern can be seen by inspection. The total catch (all species) for each sampling period is shown, together with the catches of the three most frequent species (Fig. 2). It would appear spinicoxis, sphaerocerus and cilipes frequencies are varying in parallel. The figures for these three species and the next most abundant, setosus, were tested for the significance of this ap- parent correlation by Kendall’s coefficient of concordance (Siegel, 1956). The results: n (number of sampling classes) = 20 K (number of species) = 4 W = 0.66 X = 50.22 df = 19 /?< 0.001 There is a highly significant correlation in the periodic trends in the four species, these accounting for 98.2 percent of the material caught (W was calculated uncorrected for ties; with a correction 1977] Lest on — Paussids in West Africa 213 Table 1. Days of sampling and numbers sampled to give an additional species. Sampling day Cumulative total sampled Species added Total species 1 0 0 0 4 1 1 1 7 4 2 3 44 52 1 4 47 56 1 5 116 220 1 6 400 1063 0 6 the value of would have been even larger). We can therefore analyse the total paussids with some confidence as reflecting the trends in each. It is usual in tropical phenology to attempt to tie phenomena to the rainfall pattern (Karr, 1976). The major peak (Fig. 3) occurred in April-May, when the rains were building up to their maximum, but the minor peak of January-February occurred in the dry period when the rains, although increasing, did not reach the effective level, lOcms/ month. The August trough coincided with the so- called “little dry season” but no simple correlation of paussid fre- quency with rainfall amounts was detected. Table 1 gives the arrival day and cumulative total for the capture of the ith species. Discussion It is likely the six species trapped represented all of the group present in the locality; extrapolation from the figures of Table 1 suggests an additional species should have been found, if present, before the 1000th individual was trapped. The lognormal species distribution parallels that found in sam- ples from Ghana of birds, ants, snakes and several other taxa (Leston, 1972): no explanation is offered here (but see Williams, 1964). Paussids are emerging throughout the year but with marked fluctuations in frequency. The four most frequent species show parallel trends in their respective frequencies, indicative of the same set of factors influencing all. The peak corresponds closely 214 Psyche [September-December 1969 1970 Fig. 2. Frequencies of all species together and the three most abundant species, in 20-day classes. Logarithmic. to that found at Legon in the flight of male doryline ants (Leston, 1978), falling within the first wet sunny season (Gibbs & Leston, 1970). It coincides too with that of alate sewing-ants, Oecophylla longinoda (Latreille), in the forest zone: again a wet sunny season event (Leston & Gibbs, 1971). But the apparent trimodality of the paussids does not fit just into the two wet sunny seasons of the forest zone although the October peak falls within the second of these. Were it not that a similar trimodality has been detected elsewhere it might have been dis- missed as a peculiarity of the particular period sampled but a range of mantids sampled at Tafo, 35 miles north of Legon and in the forest zone, had a similar pattern (Leston, 1972). 1977] Lesion — Paussids in West Africa 215 3 cr CD 0 N D J F MAMJ JASO 1969 1970 Fig. 3. Total paussids trapped (histogram) and rainfall, in monthly classes. Paussids are entirely dependent for food upon ants (LeMasne, 1961). Ants in the West African forest zone are essentially preda- tory. Prey abundance there is more or less trimodal. There is a major peak in the first wet sunny season, around April, tied to the new flush and essentially comprising foliage feeders (augmented by decomposers). A second, less obvious, peak occurs in October- November, the second wet sunny season, and is again made up of foliage feeders, augmented this time by the fungivores. A third peak occurs around January, its date less reliable because of the variability in the intensity of the dry period from year to year: it sees a peak in insects associated with fruit or seeds (Gibbs & Leston, 1970; Leston, 1972). It is suggested that the ultimate factors concerned in the pro- duction of paussids are directly tied to the seasonal availability of prey to their host ants, just as in Mantodea the ultimate factors are available prey. Elsewhere (Leston, 1978) it is indicated the seasonal production of alate male doryline ants, as with paussids reaching a peak in the first wet sunny season, is geared to the op- timal availability of prey to the workers of the species involved. Once again it has been found that the seasonal pattern in a tropical 216 Psyche [September-December insect taxon cannot be explained by a simple wet season/dry season cycle: the biological periodicities — the true seasons — are those which follow from periodicities in rainfall and sunshine taken to- gether (Gibbs & Leston, 1970). It is likely that the majority of the paussid species trapped at Legon came from the forest zone but nothing is known as to their host ants in West Africa. In view of their absence from the wide range of colonies of dominant, arboreal, ants sampled in Ghana (Leston, 1973b) it is probable their hosts are for the most part ground-nesting species. Acknowledgements My research has been supported by the Ghana Cocoa Growers’ Research Association, the Cocoa, Chocolate and Confectionery Alliance (U.K.), the University of Ghana and the Research Foun- dation of the University of Connecticut. Conclusions 1. Six paussid species were taken in an ultraviolet light-trap run for 400 days at Legon, Ghana, the catch totalling 1063. 2. The frequencies of the trapped species had a lognormal dis- tribution. 3. The frequencies showed parallel periodic changes between the species. 4. Peak emissions occur in the 1st wet sunny period, lesser max- ima in the 2nd wet sunny and the dry sunny periods of the Gibbs- Leston model. 5. The ultimate factor, it is suggested, is prey availability to the host ants. References Darlington, P. J. (1950) Paussid beetles. Transactions of the American Ento- mological Society 76: 47-142. Gibbs, D. G. & Leston, D. (1970) Insect phenology in a forest cocoa-farm lo- cality in West Africa. Journal of applied Eclogy 7: 519-548. Karr, J. R. (1976) Seasonality, resource availability, and community diversity in tropical bird communities. The American Naturalist 110: 973-994. LeMasne, G. (1961) Recherches, sur la biologie des animaux myrmecophiles: observations sur le regime alimentaire de Paussus favieri Fairm., hote de la fourmi Pheidole pallidula Nyl. C. R. Acad. Sci. Paris 253: 1356-1357. 1977] Lest on — Paussids in West Africa 217 Leston, D. (1972) Insect interrelations in cocoa: a contribution to tropical ecol- ogy. Ph.D. thesis, University of Ghana. Leston, D. (1973a) The flight behaviour of cocoa capsids (Hem., Miridae). En- tomologia experimentalis et applicata 16: 91-100. Leston, D. (1973b) The ant mosaic, tropical tree crops and the limiting of pests and diseases. PANS, London 19: 311-341. Leston, D. (1978) Dispersal by male doryline ants in West Africa. Ecology (in press). Leston, D. & Gibbs, D. G. (1971) Phenology of cocoa and some associated in- sects in Ghana. Proceedings of the 3rd international Cocoa Research Confer- ence, Accra, 1969: 197-204. Siegel, S. (1956) Nonparametric statistics for the behavioral sciences. McGraw- Hill; New York. Williams, C. B. (1964) Patterns in the balance of nature. Academic Press; London. AN ABERRANT NEW GENUS OF MYRMICINE ANT FROM MADAGASCAR1 By William L. Brown, Jr. Department of Entomology Cornell University Ithaca, New York 14853 During a research trip to the Old World tropics during January to April 1977, I was allowed by M. A. Peyrieras, of the Institut de Recherche Scientifique of the Malagasy Republic in Tananarive, to sort through some berlesates of humus and leaf litter collected by him in various parts of Madagascar. Among these samples, I dis- covered a single worker example of an extraordinary new genus and species of Myrmicinae. M. Peyrieras has my thanks for these and many other interesting samples, including several undescribed species of ants. Among these are the first recorded representatives of Discothyrea and Amblyopone found on Madagascar, to be de- scribed in a separate publication. The new genus is described next below. Pilotrochus new genus Worker: Subfamily Myrmicinae, tribe unknown. Integument thick and rigid. Head subpyriform, slightly depressed but convex, broadest behind, without posteromedian excision; frontal carinae far apart, at sides of head, produced laterad angularly above an- tennal insertions, continuous posteriad with the sharp upper mar- gins of deep and broad antennal scrobes occupying about 2/3 of length of sides of head; eyes small, situated on ventral borders of scrobes near their posterior ends, slightly posterior to midlength of head. Antennae 8-merous, scapes short and thick, much nar- rowed basad; club distinctly 2-merous, slightly longer than re- mainder of funiculus, apical segment about twice as long as •Hymenoptera: Formicidae. A report of research from the Cornell University Agricultural Experiment Station. Research supported by National Science Founda- tion Grant DEB 75-22427. Manuscript received by the editor February 3, 1978. 218 1977] Brown — New Genus of Myrmicine Ant 219 penultimate; funicular (ring) segments II-V small, subequal in length, broader than long; funicular segment I (pedicel) about equal in length to the next 2 to 3 (ring) segments. Clypeus broad, shield-shaped, narrowed sharply laterad on each side, its surface gently convex; anterior margin broadly arcuate, with a complex median notch. Mandibles triangular, convex, with opposable, serially-toothed mesial margins (teeth coarse, sharp, 7 in number on each mandi- ble); these margins arcuate in side view and finally directed ventrad at apex. In the vee between each of the larger teeth is a minute piligerous denticle. Labrum thick, linguiform and narrowly rounded at apex, but appearing truncate in side view; its dorsal surface bear- ing a narrow pencil of fine white setae that arches forward to ex- tend beyond the mandibular apices, probably serving as a “range- finder” trigger hair. Palpi not visible and undoubtedly short, but segmentation not determined. Trunk compact, its dorsal outline forming one continuous arc from base of pronotal cervix to petiolar insertion, dominated by pronotum. which makes up about half its length and is wider than the rest (a little more than half as wide as head). Promesonotal suture marked by a faint curved transverse line paralleled by a costa on the dorsum, but completely fused here, though complete on the pleura. Pronotum with blunt, barely suggested humeral angles as seen from above, but not distinctly marginate in front or on the sides; upper sides bulging and overhanging lower sides; ventral margins each forming a curved, cultrate, projecting flange or lamella. Metanotal groove obsolete, its position perhaps marked by a transverse carina at the top of what appears to be the pro- podeal declivity, but the true declivity probably is confined to the lower part of this slope, beneath a lower and weaker transverse carina or costula. Mesopleura narrow and impressed, but their lower central parts (mesepisterna) are occupied on each side by a peculiar organ con- sisting of a large subcircular pit filled with a silvery-white, convex pad of fine, radially-arranged hairs. The anterior edge of this struc- ture forms the posterior side of the ventral invagination between pro- and mesonotum, but a broad piece of mesokatepisternal cuticle separates the organ from the mesocoxa. This organ appears to be the external part of an exocrine gland or glands, though the gland openings, if any, are hidden by the pad of fine hairs. 220 Psyche [September-December The (probably) true propodeal declivity small, flat, steep, un- armed, bounded by an inverted U-shaped carina, flanked on each side by the large, circular openings of the propodeal spiracles, which are directed caudad. Inferior propodeal plates low and rounded. Petiole curved-clavate, with a long, subcylindrical anterior pe- duncle, a long, low, rounded node, and poorly-defined posterior peduncle, or collar. Postpetiole rounded above and on the sides, broader behind than in front, and about as broad as long; sternum boxlike, with inwardly sloping sides and a flat, subrectangular ventral face limited on all 4 sides by sharp carinae; broadly at- tached to gaster behind, but with a moderate constriction between the 2 tagmata. Gaster rounded at base to the basal constriction, which has a few short longitudinal costulae hidden within it. Basal segment extending over about 2/3 length of gaster, the remaining 3 visible segments curving to a pointed apex; terminal external (seventh) sternite (hypopygium) acutely pointed; sting with acute tip ex- serted, some of the rest of the shaft visible through transparent cuticle. Legs moderately long. Femur robust, its flexor surface sulcate apicad to receive the folded tibia; tibia claviform, thickest near apical third, lacking spurs on middle and hind legs; tarsus slender, cylindrical; claws small, slender, simple. Sculpture mostly shining; head both above and below (except clypeus and antennal scrobes) covered with a coarse reticulum of costulae forming large, shallow-polygonal fossae, each of which bears a long, fine, curved hair arising from an inconspicuous punc- ture, usually near the margin of the fossa. Similar sculpture on truncal dorsum, though here 10-11 longitudinal costae predomi- nate. A broad median strip is nearly smooth, forming a very shal- low median sulcus, which has a modest costa or carina in the middle. (The median posterior vertex is feebly subsulcate in much the same way.) Propodeum (sloping dorsum and declivity) nearly or quite smooth. Sides of trunk smooth and shining, but the pos- terior half with 5 strong costae radiating outward from the meso- pleural organ. Petiolar and postpetiolar nodes loosely longitudi- nally rugose, with broad, shining interspaces; peduncles and sides of these segments becoming finely and densely punctate-reticulate and more opaque; ventral rectangle of postpetiole smooth and shining. Gaster smooth and shining, except posterolateral margins 1977] Brown — New Genus of Myrmicine Ant 221 of segments, which are delicately reticulate-striolate, but still shin- ing. Mandibles densely and finely punctulate, opaque (teeth in- fuscate and shining), lateral surfaces finely longitudinally costulate basad. Antennal scapes and legs shining, finely longitudinally rug- ulose to smooth; antennal funiculi densely punctulate, pubescent and opaque. Antennal scrobes basically smooth and shining, but each is crossed by 7-9 vermiculate transverse rugulae. Pilosity consisting of long, fine, flexuous erect to decumbent hairs distributed widely over dorsal surfaces of body, and on scapes and legs, sparser on underside of head and gaster. Mandibles, an- tennal funiculi, and flexor surfaces of tibiae and tarsi with fine, appressed to decumbent pubescence. Color light ferruginous red; antennae and legs more yellowish. Queen, male and larva still unknown. Type species: Pilotrochus besmerus, new species, described below. Distribution as far as known limited to Madagascar. Pilotrochus besmerus new species Figs. 1, 2. Holotype worker: Total length (TL) 2.9, head length (HL) 0.68, head width (HW) 0.60, mandibles extend beyond median clypeal free margin (ML) 0.11 mm; when head is tilted back a little from full-face view, mandibles may extend beyond clypeus as much as 0.18 mm; trunk length (WL) 0.64, scape L (excluding radicle) 0.32, greatest length of eye 0.07 mm. Petiole length (chord of arc) 0.45, hind femur length 0.43, hind tibia length 0.34, hind tarsus length 0.60 mm, of which metatarsus is half. Details of form and sculpture are well shown in the figures. As seen in dorsal view, eyes hidden in full-face view, barely visible when head is tilted back slightly, as in fig. 1. Pronotum slightly broader than long (width 0.38 mm), rounded in front, excised be- hind; mesonotum subquadratic, about 0.20 mm wide, with feebly convex borders on all four sides. Petiolar node 0.17 mm wide, postpetiolar node with rounded sides, slightly wider behind (width 0.21 mm, length 0.20 mm). Pronotum smooth and shining, with about 10 costulae running longitudinally at different lengths; mesonotum with 5 longitudinal costulae on rugulae. About 5 longitudinal rugules each on petiolar and postpetiolar nodes, but weaker on the postpetiole, so that its disc is primarily smooth and shining like the gaster. 222 Psyche [September-December Fig. 1, Pilotrochus besmerus, new genus and species, holotype worker, head in dorsal view, tilted back slightly from the full-face position so as to show the man- dibular dentition better, X67. Drawing by Susan Poulakis. Hairs mostly a little less than 0.1 mm long on anterior head and scapes, a little more than 0.1 mm on posterior vertex; 0.15-0.25 on trunk, petiole, postpetiole and gaster, becoming shorter again at gastric apex; about 0.1 to about 0.3 mm on legs; flagelliform, many with tips reflexed or even looped back. Holotype (Museum of Comparative Zoology at Harvard Univer- sity) a unique worker taken in a Berlese sample of forest humus and litter from along the road to Anosibe, 33 km south of Mora- manga, in east central Madagascar, 4-12 April 1975 (A. Peyrieras). The relationships of Pilotrochus are obscure. The shape of the head, with its broad and deep scrobes and small ventrolateral eyes, recalls that of Dacetinops, or the Codiomyrmex group of Dacetini, or even Tatuidris, but Pilotrochus differs strongly from all of these in its 8-segmsnted antennae, in the form of its mandibles and their teeth, and in the form of its trunk and petiole. The lateral “hair- k 1977] Brown — New Genus of Myrmicine Ant 223 4 Fig. 2, Pilotrochus besmerus, new genus and species, holotype worker in side view, X67. Drawing by Susan Poulakis. 224 Psyche [September-December wheel” organs are also striking, and so far as I am aware are un- matched among the ants, although the region of the ventral furrow between pro- and mesothorax is sometimes modified and apparently glandular in some dacetines and a few other myrmicines. The 2-segmented antennal club, together with some points of habitus (especially coarse sculpture and long, flexuous pilosity, plus the antennal scrobes), recalls the neotropical genus Lachno- myrmex , but the shape of the trunk in Pilotrochus is completely different, the propodeal teeth so prominent in Lachnomyrmex are completely absent, and the mandibles in the two genera are very different. At the moment, all one can say without seeing the winged forms and larvae of Pilotrochus, and without knowing something of its lifeway, is that it is a member of subfamily Myrmicinae, but one not belonging to any of the well-circumscribed “higher” tribes (Crematogastrini, Dacetini, Basicerotini, Cataulacini, Attini, etc.). Thus, it falls among the mass of generic complexes related to Myrmica, Pheidole, Myrmecina, Rogeria, etc., among which tribal boundaries are impossible to define for the present, or at least are in dispute. Considering the revisionary work that must be done before we have a rational tribal classification of Myrmicinae, it seems to me that nothing would be gained by erecting a new tribe for Pilotro- chus, even though for the time being I am able to fit it comfortably into any existing myrmicine tribe. The generic name is derived from the Greek ‘pilos’ (hair) + ‘tro- chos’ (wheel), while the specific name besmerus combines the Latin ‘bes’ (eight of twelve) with the Greek ‘meros’ (part), in reference to the 8-segmented antennae as compared to the primitive myrmicine (and formicid) 12-merous condition. SYMBIOSES BETWEEN INSECTS AND SPIDERS: AN ASSOCIATION BETWEEN LEPIDOPTERAN LARVAE AND THE SOCIAL SPIDER ANELOSIMUS EXIMIUS (ARANEAE: THERIDIIDAE)* By Michael H. Robinson Smithsonian Tropical Research Institute P.O. Box 2072, Balboa, Canal Zone, Panama Introduction There are many instances of relationships between insects and spiders that are not simply relationships between predators and prey. Bristowe (1941) cites numerous examples either from his own extensive experience or from a broad review of the diverse litera- ture. Moths have been reported to associate with spiders’ webs both as adults and larvae. Thus Pocock (1903) reported a case of commensalism between the gregarious spider Stegodyphus sp. (Eri- sidae) and the moth Batrachedra stegodyphobius Walsingham. The unnamed species of Stegodyphus from South Africa had small lepi- dopteran larvae crawling about within the communal web. These fed upon “the carcases of the flies or other insects which, with in- finite labour and patience, the spiders hauled up as near their nest as possible. . . .” Pocock states that pupation occurred within the nest (= web) and that, after emergence, adult moths moved about the web walking, leaping and fluttering. Reportedly the moths did not get caught in the sticky (cribellate) silk “being gifted apparently, like the spiders themselves, with some safeguard against the sticki- ness of the threads, which proved so fatal to other insects” (1903: 169). Brach (1977) reports that the webs of Anelosimus studiosus, in Florida are shared by a host of other arthropods including py- ralid “webworms.” He comments that the relationship between these other arthropods and the Anelosimus is not clear, but that the majority “are found in the periphery of senescent webs and may be physically isolated from contact with colony members by their own silken retreats” (1977:155). Robinson and Robinson (1976:12-16) report on a pyralid moth that, as an adult, rests on * Manuscript received by the editor April 27, 1978 225 226 Psyche [September-December the silk lines of araneid webs. They conclude that the moth gains protection by the association. In this paper I describe the associa- tion between larvae of the noctuid moth Neopalthis madates Druce and the colonial spider Anelosimus eximius Simon. In addition, three other instances of associations between lepidopteran larvae and web-building spiders are briefly reported. These involve the araneid Cyrtophora nympha Simon, an undetermined diplurid and an undetermined social theridiid in Papua New Guinea. The rela- tionship between Neopalthis larvae and A. eximius is a symbiosis that primarily involves scavenging but may occasionally involve the loss to the spider of usable food resources. Thus the symbiosis is probably commensal for the most part but sometimes (or poten- tially) deleterious. The terminology of symbioses for such “border- line” cases is in a currently unsatisfactory state. The Neopalthis/ Anelosimus Symbiosis The nature of the symbiosis can only be understood if some de- tails of the biology of Anelosimus eximius are given. There have been a number of notes on aspects of the biology of this species (e.g., Levi, 1955; Brach, 1975) and it is currently under study at the Smithsonian Tropical Research Institute (by Dr. F. Vollrath). The following notes are based on my own studies and those of Vollrath (pers. comm.). Anelosimus eximius webs are built and occupied by a variable number of spiders, from less than a hundred to at least several thousand. They may persist in one place for many years. One on Barro Colorado Island, Canal Zone, was in the same tree from 1965 to 1971. Colonies may reach striking proportions, occupying many cubic meters of space. Essentially all webs have a simple basic structure. The lower web consists of a continuous sheet of silk that is concave and often basin-like, being raised at its periphery. Above this and partly attached to it is an aerial snare of threads that are preponderantly oriented more or less perpen- dicular to the sheet. This is the part that is the effective prey-capture structure. The spiders attack prey within the aerial snare and on the basal sheet. Since the basic web is persistent for long periods, it acquires a litter of plant debris and prey remains. (In captivity the spiders seem to indulge in occasional web-cleaning bouts and carry prey remains to the edge of the web where they are tipped out.) 1977] Robinson — Insects and Spiders 227 I found the first caterpillars in an Anelosimus colony in January 1976. A small colony of less than one hundred individuals was collected from Cerro Galera, Canal Zone, Panama, and taken into the laboratory for behavioral studies. The collection was made by bagging the entire colony on the branch of a tree and then cutting off the branch. This was then set up in a cage. The colony was thus collected intact and complete with all spiders, debris, and inquilines. After some period of observations on the spiders, the caterpillars were seen. There appeared to be several of these; at least two were seen at one time, and eventually four pupae were recovered from the web. When not actively feeding, the caterpillars rested beneath the plant debris in the web or stretched out in stick- like postures. Feeding on insect remains was seen both by day and by night. Movement within the web seemed to be purposively towards prey remains rather than exploratory, but this proved difficult to quantify since the caterpillars had to weave their bodies in and out of the maze-like strands of the snare. In an attempt to investigate the possible cues used by the caterpillars in finding their food, I excised a section of the web away from a spider aggregation and put two caterpillars in this. (Cutting sections of spider web is easy, a hot soldering iron or glowing tip of a lighted cigarette cuts the silk with little pressure.) In this spider-free web I placed a freshly killed tettigoniid (Orthoptera) of about 100 mg weight. The caterpillars made no immediate move towards the katydid, but some hours later both were feeding on it and one was almost inside the body of the dead insect. Figure 1 shows a caterpillar in action. Clearly the caterpillars could feed on an entire insect as well as on exoskeletal fragments. They may well do this in the natural situa- tion. Both caterpillars pupated the morning following this massive meal and this ended the investigation. Caterpillars have been found in other Anelosimus eximius colonies by me and by F. Vollrath (pers. comm.). Vollrath raised Neopalthis madates from his cater- pillars and also another moth species that was clearly not a noctuid. This species has not been determined. It is entirely possible that still other species of lepidoptera could be involved in this kind of symbiosis with Anelosimus eximius; this matter is discussed later. Further investigation of the relationship awaits a situation where infested colonies are abundant and can be subjected to traumatic or destructive experimentation. 228 Psyche [September-December Figure 1. Two larvae of Neopalthis madates feeding on a katydid (see text). One larva is in profile; the other, arrowed, is facing the camera and only the head is visible. 1977] Robinson — Insects and Spiders 229 Figure 2. Larva (unidentified) feeding on a moth within the web of Cyrtophora nympha. The characteristic fine mesh of the Cyrtophora web is visible in the back- ground. Other Moth Caterpillar/ Spider Symbioses The following observations are fragmentary but are worth re- cording to alert workers in this field to the possibility of widespread symbioses between web-building spiders and lepidopterans. The most excitingly suggestive observation is the discovery, during my brief visit to Wau Ecology Institute, Papua New Guinea, in May 1977, of caterpillars living in colonies of a social theridiid there. The theridiid has not yet been determined and its relationship to the Neotropical Anelosimus eximius is thus unknown. Despite this, it is reasonable to assume that the symbiotic caterpillars represent a case of convergent evolution. I would guess that they were not noctuid caterpillars, but none were collected. 230 Psyche [September-December In Panama a web of the araneid spider Cyrtophora nympha Simon, collected in July 1976 on the Navy Pipeline Road, Gamboa, Canal Zone, contained a single caterpillar moving about in the web in the same way as the larvae in the Anelosimus colonies. This was collected together with the web and the host. The caterpillar is shown in Figure 2. It disappeared without trace. Finally, F. Vollrath reports (pers. comm.) that he found a cater- pillar living in the web of a diplurid and feeding on prey remains; it also apparently fed on the web silk, where it made “dime-sized holes.” Neither the diplurid nor the moth have been determined. Discussion So far all the associations between lepidopteran larvae and spiders occur where the hosts build webs that persist for long periods at the same site. This is certainly true of Stegodyphus and Anelosimus colonies and also true of Cyrtophora nympha. The webs of the latter are presumably high investment structures like those of C. moluccensis (see Lubin, 1973, 1974); they are soli- tary and become littered with leaves and debris and often look defunct. An indication of the persistence of C. nympha webs is given by the fact that colonies of Uloborus republicanus often es- tablish themselves in the upper snare (personal observations). In Panama, ground-living diplurids build their sheet webs at one site for long periods and have a wide range of arthropod commensals, kleptoparasites and other symbionts (Vollrath, Kirkendall, pers. comm.). The correlation between persistent webs, or persistent utilization of web sites, and the occurrence of lepidopteran sym- bionts suggests that the webs of other perennial (or long-term) web-site occupiers should be examined for caterpillar cohabitants. The various gregarious Cyrtophora species are clear candidates for such studies. Another factor is necessary to provide a niche (within a web) for scavenging cohabitants: clearly there must be an accumulation of prey-remains. If prey-remains were rapidly ejected from the web, there would be no resource for a scavenger to exploit; kleptopara- sitism or commensalism would be the only feeding niches available to symbionts. 1977] Robinson — Insects and Spiders 231 The evolution of finely adjusted interspecific relationships must have involved innumerable adaptive steps that are almost incon- ceivable in their probable complexity. Lepidopteran larvae (and other spiders) may be, as producers and manipulators of silk, some- what preadapted to evolve symbioses with web-building spiders. Adult butterflies and moths are probably less endangered by sticky spider silk than are most other insects (Eisner, Ettershank and Alsop, 1964) and this could reduce some of the dangers involved in evolving symbioses with spiders. Nonetheless, the caterpillars had to solve two major problems. They had to in some way sup- press the predatory responses of the spiders to objects moving on the web and also develop a system of detecting their own food within the web. From my own observations I would conclude that both Stegodyphus sp. and Anelosimus eximius are much less re- sponsive to gently moving objects in their webs than are many orb-weavers. It may have thus been slightly less dangerous for moving caterpillars to invade these webs in the first place, but I strongly suspect that the present immunity to spider attacks de- pends on something more than the caterpillars “walking softly.” Acknowledgments I am extremely grateful to Dr. E. L. Todd, Systematic Ento- mology Laboratory, IIBII Institute, U.S. Department of Agricul- ture, for identifying the moths. My thanks to Dr. F. Vollrath of STRI for access to unpublished observations and for helpful com- ments. References Brach, V. 1975. The biology of the social spider Anelosimus eximius (Araneae: Theri- diidae). Bull. So. Calif. Acad. Sci. 74: 37-41. 1977. Anelosimus studiosus (Araneae: Theridiidae) and the evolution of quasi- sociality in theridiid spiders. Evolution 31: 154-161. Bristowe, W. S. 1941. The Comity of Spiders. Vol. II. Roy. Soc. London. Eisner, T., R. Alsop and Ettershank, G. 1964. Adhesiveness of spider silk. Science 146: 1058-1061. Levi, H. W. 1955. The spider genera Neottiura and Anelosimus in America (Araneae: Theridiidae). Trans. Amer. Microscop. Soc. 75: 407-422. 232 Psyche [September-December Lubin, Y. D. 1973. Web structure and function: the non-adhesive orb-web of Cyrtophora moluccensis (Doleschall) (Araneae: Araneidae). Forma et Functio 6: 337-338. 1974. Adaptive advantages and the evolution of colony formation in Cyrto- phora (Araneae: Araneidae). Zool. J. Linn. Soc. 54: 321-339. Pocock, R. I. 1903. Notes on the commensalism subsisting between a gregarious spider Stegodyphus sp. and the moth Batrachedra stegodyphobius Wlsm. Ent. Mon. Mag. 39: 167-170. Robinson, M. H. and B. Robinson 1976. The ecology and behavior of Nephila maculata: a supplement. Smith- son. Contrib. Zool. 218: 1-22. POPULATION STRUCTURE AND POLYMORPHISM IN THE SLAVE-MAKING ANT HARPAGOXENUS AMERICANUS (EMERY) (HYMENOPTERA: FORMICIDAE) By Alfred Buschinger1 and Thomas M. Alloway2 Fachbereich Biologie, Institut fur Zoologie, der Technischen Hochschule, D6100 Darmstadt, Schnittspahnstr. 3; and Erindale College, University of Toronto, Mississauga, Ontario, L5L 1C6, Canada Introduction The biology of the slave-making ant, Harpagoxenus americanus, has been studied by several previous investigators. Creighton (1927, 1929) described slave raids and performed some experiments on colony foundation. Wesson (1939) provided a classic account of numerous aspects of the natural history of H. americanus including its distribution, colony demography, hibernation, brood develop- ment, and its colony-foundation, mating, and slave-raiding behav- iours. Alloway (in press) provides a detailed account of its raiding behaviour in comparison to that of the closely related, but more primitive, species Leptothorax duloticus. Wesson (1939) found that, in addition to a rather small propor- tion of “normal” queenright colonies, there are many queenless “branch colonies” which sometimes contained apparently fertile “gynecoid” workers. Moreover, he observed the formation of such “branch colonies” as a consequence of slave raids which occurred late in the raiding season. Previously, Sturtevant (1927) had dis- sected americanus workers and found that they typically have 6 ovarioles and “a sac arising from the anterior end of the common oviduct” which he interpreted as being a seminal receptacle. Sturte- vant also found some workerlike individuals with rudimentary ocelli and presumed that H. americanus, like the European H. sublaevis, has ergatoid queens. However, Wesson (1939) wrote that he had Supported by a grant from the Deutsche Forschungsgemeinschaft. Supported by Grant A0301 from the National Research Council of Canada. Manuscript received by the editor April 27, 1978. 233 234 Psyche [September-December failed to note any “ergatoids such as occur with European H. sub- laevis.” Nevertheless, Creighton (1950, p. 281), comparing H. amer- icanus with H. canadensis and H. sublaevis, stated that in ameri- canus “the ergatoid female is rarely produced.” In a recent study (Buschinger & Alloway, in press), we demon- strated that in all probability there are no ergatoid queens at all in H. canadensis. In contrast, although “normal” full queens occur occasionally in H. sublaevis, the usual reproductive female in this species is an ergatoid form which closely resembles the worker caste in its external morphology. Moreover, polymorphism among re- productive females is genetically mediated in H. sublaevis (Busch- inger, 1975, 1978, in press). Thus, the main objective of the present study was to determine whether ergatoid queens analogous to those found in H. sublaevis occur in H. americanus and, if so, what role they play. We especially wondered whether they might be the usual reproductives in the numerous “branch colonies” that lack a dealate full queen. Material and Methods Colonies of H. americanus were collected from various localities in the regional municipalities of Halton and Peel in southern On- tario, Canada. A few additional colonies were found near Cleve- land and Ashtabula, Ohio, in the United States. Most of the col- onies were nesting in old hollow acorns and hickory nuts lying on the ground. The colonies were removed from their nests using an aspirator either immediately in the field or later in the laboratory. Colonies whose broods consisted only of larvae and prepupae at the time of collection were kept alive in the laboratory for a few weeks to determine the sex and caste of the Harpagoxenus pupae that were produced. Some of the Harpagoxenus queens, all inter- morphs, and most of the Harpagoxenus workers were dissected using a method which we have described fully elsewhere (Buschinger & Alloway, in press). Results 1. Population data. No effort was made to determine the ratio of Harpagoxenus to host-species colonies or the number of colonies per unit of area. Both variables fluctuate widely and depend upon a large number 1977] Buschinger & Alloway — Harpagoxenus americanus 235 Table I. Occurrence of Lepothorax longispinosus and L. ambiguus as host species in colonies of H. americanus. H. americanus with slave species colonies L. longispinosus L. ambiguus L. Long. + L. amb. incipient 2 queenright 12 queenless 13 of factors such as the type of vegetation, surface drainage, the dis- tribution of habitable nesting sites, etc. Sturtevant (1927) recorded an infestation rate of 13% for H. americanus (i.e. 17 americanus colonies to 132 colonies of one of its three host species, Leptothorax curvispinosus Mayr). In the areas where we collected our material, the host species were Leptothorax ambiguus Emery and L. longi- spinosus Roger. Although ambiguus was generally the more abun- dant of the two host species in acorn and hickory-nut nests, longi- spinosus was more frequently enslaved (Table I). These data sug- gest that colony-founding Harpagoxenus queens preferentially seek out longispinosus colonies but that established Harpagoxenus col- onies, especially the so-called “branch colonies,” raid the nests of both hosts. Our material consisted of a total of 41 Harpagoxenus colonies or “branch colonies” in the sense of Wesson (1939). Thirteen “pri- mary colonies” contained a dealate full Harpagoxenus queen and from 1 to 8 Harpagoxenus workers. The average number of Har- pagoxenus workers in these colonies was 3.0. Three incipient col- onies contained only a Harpagoxenus queen, several host-speices workers, and a brood which consisted of Harpagoxenus worker pupae or larvae which developed into Harpagoxenus worker pupae. The 25 “branch colonies” lacked a dealate full queen but contained from 1 to 9 Harpagoxenus workers (with an average of 2.8), a variable number of host-species workers, and a brood. Six of these ’’branch colonies” produced only male pupae of the parasite species; but the rest yielded female, worker, and male pupae of the parasite species. In one acorn which contained no adult Harpagoxenus when it was collected, there were nevertheless Harpagoxenus male, female, and worker pupae which were being attended by several longispinosus workers. Apparently, the Harpagoxenus adults had 236 Psyche [September-December either died or were away for some reason (perhaps on a raid) when their nest was collected. Finally, we made one quite peculiar ob- servation. We found a dealate Harpagoxenus queen crawling around on the outside of an acorn which contained 3 Harpagox- enus workers, several longispinosus slaves, and a brood. After col- lecting the queen and the “colony” together in a vial, the Harpa- goxenus queen was heavily mutilated by the Harpagoxenus work- ers, thus indicating that the queen and the workers did not belong together. Dissection revealed that the queen was an inseminated, egg-laying individual and that all 3 of the workers in the nest were completely sterile. In all probability this collection represents a newly founded Harpagoxenus colony that had been raided by workers of another Harpagoxenus colony. 2. Dissections Dealate queens: The only dealate females from 8 colonies and both dealate females from a single colony which contained 2 such individuals were dissected. All 10 females had 6 ovarioles and a spermatheca. One of the females from the two-queen colony had short ovarioles that contained no developing eggs, and an empty spermatheca. However, the other 9 individuals were functional colony queens, individuals with receptacula filled with sperm and ovarioles which were about their total body length and which con- tained eggs in different stages of growth and corpora lutea. Intermorphs: In 6 different colonies (2 queenright colonies and 4 “branch colonies”), we found a total of 6 individuals that were morphologically intermediate between worker and full queens and which could thus be regarded as intermorphs. These individuals had more or less well developed ocelli on their heads and several distinct sclerites on the alitrunk. Workers lack these structures. All of these intermorphs except one had 6 short, empty ovarioles like those of a virgin queen. None of them had a spermatheca. One individual had ovarioles which were about as long as the gaster and which contained developing oocytes and corpora lutea. These data indicate that H. americanus intermorphs have a reproductive func- tion which cannot be distinguished from that of ordinary workers. Although they cannot be inseminated, they do sometimes become egg-layers. Workers: We dissected a total of 91 workers, 28 from 8 queen- right colonies and 63 from 22 “branch colonies” which contained 1977] Buschinger & Alloway — Harpagoxenus americanus 111 Table II: Numbers of fertile and sterile workers in queenright and queenless colonies. Colony no. queen workers and intermorphs (I) sterile with growing with grow.ooc. oocytes and corpora lutea Harpagoxenus d5S produced +) date of collection 1 1 _ 3 1 3 5 S VI-12-77 2 1 1 2 - 3 * s VI-12-77 3 1 3 - - 3 5 S VI-12-77 4 1 8 - - s VI-13-77 5 1 1 - - s VI-24-77 6 1 2 - - 25 4s VI-16-77 7 1 4 - - s VII-1 7-77 8 - 1 - - 4tJ 4$ 2S VI-12-77 9 - 1(1) 1 - 23 25 2S VI-12-77 10 - 1 - - 17S Vl-24-77 11 - 2 1 - 13 1 ¥ 7S VI-16-77 12 - 2 1 - 3 5 S VI-22-77 13 - - - 1 3 S VII-1 -77 14 - - 4 4+1(1) 113 1 5 8S VI-12-77 15 - - - 1 5d 4s VI-14-77 16 - 2 - 3 73 VI-14-77 17 - - 3 1 1 03 IS VI-22-77 18 - - - 1 4s VI-24-77 19 - - 1 1 34 3 1 5 VI-27-77 20 - 3 - 1 10 z r- z r; ° 'f^\ w 2 J&t, a ^ CD 73 > 73 rn \W' ^ N^n^sv^ix m CO — GO VARIES SMITHSONIAN INSTITUTION NOIlfUllSNI NVINOSH1IWS S3ld\j z \ to z to X to IT I S? > s (/}'** zz nillSNI_NVIN0SHillMSwS3 i y va a n_Li B raries smithsonian_instit M ^ W xf55*3x “ M O " Q ^ -J z _ _ RARIES SMITHSONIAN INSTITUTION NOliniUSNI NVINOSHJLIWS S3l*n 2 2 ^ ^ z rr Z 2 .< m 2 ™ O to CD 73 > 73 *•*” m LfULLSNI NVIN0SH1IWS S3 I H Vd 8 n~LI B R AR I ES^SMITHSONIAN INSTIi Z to Z - * . to z ~ S ,. '«** 5 v > 05 Z 05 * Z 05 RARIES SMITHSONIAN INSTITUTION NOlifUliSNI NVINOSHJ.IWS S3 IS to 5 to ~ . to SI°£3X “ , .. .«(/> w ,#$\x “ ,# * =! K < ex CD .nuiSNi NViNOSHims ssiavaan libraries Smithsonian instc ' r* v Z r- Z RARIES SMITHSONIAN INSTITUTION NOUDillSNI NVINOSHIMS S3 Id' to z \ to z to x to o 2 2 v \^v > ' 2 >‘ -fUUSNI N.VIN0SHilHISWS3 I U VU 8 H^LI B R AR I ES^SMITHSONIAN^INSTIl ^ \ to ~ CO ~ id w, NOiini iSNi NViNOSHims bdidvdan u b kak i tb bMimbumaiN m - V Z r~ 2: . r. ) / ±1 •r m ^ m (LIBRARIES SMITHS0NIAN~ IN STITUT I ON^ NOIlfUllSNl ~"NVIN0SH11WS S3 0> z \ 03 z 5 m ji) ° 3Wk - ° > V««v^/ 2 > • s NouniiisN^NviNosHiiws^sa 1 ava a ii^li b rar i es”smithsonian in ■7 \ cn “ v.ssssss/ _ x'firrr? v-V /n H -r (ft — 03 = C/> NOIlfUllSNl NVIN0SH1IWS S3IBVBail LIBRARIES SMITHSONIAN IN w 2 03 Z . 03 r>- 1 . -A Isa J 2 Xi2S°£Z 5 ■'•IP®’' 2 > > z co z CO 1 z — CO LIBRARIES SMITHSONIAN INSTITUTION NOIlfUllSNl NVIN0SH1HMS S 1 03 zz 03 ~ 03 NOliniUSN! NVINOSHlli/MS S3IBVUan LIBRARIES SMITHSONIAN IN 2 ___ r- > 2: 0 33; V o x; m co 33 * W Ej _ JLI B RAR I ESWSMITHSONIAN~INSTITUTIONC,>NOIinillSNI-NVINOSHllWSC/,S; 2 \ 03 Z v 03 Z 03 < V 2 V S < 2 * 0 z w V 03 ■H X 03 pOIinillSNI^ N VI N0SH11 WS^ S 3 I H VH 0 Ilf LI B RAR ! ES^SMITHSONIAN JN ? 3 V* 1 \ 5 4t‘ « m C ..u/ . x- w *»»»■<-, /V ' A/ rr \4MPK/ m»y O X^os^X _ o Z j z SMITHSONIAN INSTITUTION NOIlfUllSNl NVIN0SH1IWS S: