1999. The Journal of Arachnology 27:135±141

SEXUAL SELECTION IN PHOLCID (ARANEAE, ): ARTFUL CHELICERAE AND FORCEFUL GENITALIA

Bernhard A. Huber: Department of Entomology, American Museum of Natural History, Central Park West at 79th Street, New York, New York 10024 USA

ABSTRACT. Two aspects of pholcid reproductive biology are reviewed and appear best explained by sexual selection by female choice: the rapid and divergent evolution of male chelicerae (and clypei in some groups) which contact the female epigynum during copulation and probably act as copulatory court- ship devices; and the often exceptionally strong pedipalps in males, which possibly function in correlation with the `valve' in the internal female genitalia.

The last decades have seen a promising in- particular aspects: on non-genitalic contact crease of studies examining reproduc- structures which appear to evolve under se- tion from an evolutionary perspective (review: lection similar to that acting on genitalia, and Elgar 1998). In most cases, the mechanisms on the unusual phenomenon of copulatory of sexual selection in spiders are much the courtship associated with vigor. same as those documented in insects and other Voucher specimens of all unnamed species major groups (see e.g., Eberhard 1996, where are deposited at the American Museum of almost every spider example used to docu- Natural History, New York, and labeled with ment a speci®c mechanism of cryptic female an I.D. number (``B.A.H. 1999 I.D.# 1±6''). choice is accompanied by at least one insect or mammal example). Some details, however, ARTFUL CHELICERAE make spiders either especially useful (e.g., the Pholcids are not unique in having species- pairedness of genitalia for studies of ¯uctu- speci®c copulatory contact structures (Eber- ating asymmetry - Huber 1996b), or especial- hard 1985). However, pholcids are unique, at ly interesting (e.g., the apparent lack of both least among spiders, with respect to the wide muscles and nerves in the male intromittent range of non-intromittent male structures that genitalia - Eberhard & Huber 1998). (For fur- are sexually modi®ed (practically the entire ther, though less unique, spider characteristics, palp is sexually dimorphic in most pholcids, see Elgar 1998.) including coxa and trochanter). Two male In the present paper I will brie¯y review structures deserve special attention: the che- some recent advances in one particular spider licerae and the clypeus. At least one of them family, the pholcids. Pholcids are the only contacts the female during copulation in all non-entelegyne spiders whose reproductive species studied (Huber 1994, 1995, 1997b, biology has been carefully studied in several 1998b; Huber & Eberhard 1997; Uhl et al. species (Eberhard 1992; Eberhard & BricenÄo 1995), and the chelicerae in particular are of- 1983, 1985; Huber 1994, 1995, 1996a, b, ten the most species-speci®c and taxonomi- 1997a, b, 1998a, b, c; Huber & Eberhard cally useful structures. Modi®cations range 1997; Kaster & Jakob 1997; Uhl 1993, 1994; from hairs of different shapes to cones, round- Uhl et al. 1995; Yoward 1998). Further ad- ed, pointed, hooked and blade-shaped apoph- vantages for the study of sexual selection are yses, and even to sexually dimorphic fangs the number of synanthropic species that are (Figs. 1±4, 10, 11, 13, 14). Several hypotheses available worldwide and readily maintained in might explain this phenomenon: (1) reproduc- the laboratory for in depth single-species stud- tive isolation hypotheses (lock-and-key and ies, and a rich and diverse (mainly tropical) genitalic recognition - reviewed in Eberhard fauna for comparative studies. 1985); (2) the ``con¯ict of interest hypothe- For reasons of space, I will focus on two sis'' (Alexander et al. 1997); (3) sexual selec-

135 136 THE JOURNAL OF ARACHNOLOGY

Figures 1±7.ÐSexually dimorphic structures in male pholcids, SEM. 1. Cheliceral apophyses in Uthina sp. (I.D. #1); 2. Modi®ed hairs on the chelicerae of Modisimus dominical Huber; 3. Modi®ed hairs on the chelicerae of Spermophora senoculata (DugeÁs); 4. Sclerotized cones on the chelicerae of Physocyclus guanacaste Huber; 5±6. Eye turret of Modisimus culicinus (Simon), in lateral and frontal view, showing frontal lobe; 7. Femur of Modisimus tortuguero Huber, showing a spine, a ``normal'' tactile sensillum, and several almost perpendicular hairs that cover the femora of only male walking legs. Scale bars: 0.01 mm (1±3); 0.05 mm (4, 7); 0.1 mm (5, 6). HUBERÐSEXUAL SELECTION IN PHOLCID SPIDERS 137 tion by male-male competition (Eberhard & epigynum also carries a large apophysis on BricenÄo 1985); (4) the ``sperm holder hypoth- each side of the hood (Figs. 8, 9), so that only esis'' (Brignoli 1973); (5) sexual selection by males with exceedingly long cheliceral apoph- female choice (Eberhard 1985, 1996). yses can reach the hood (Figs. 10, 11). Thus, Reproductive isolation hypotheses assume ``genitalic arms races'' of this sort may re¯ect that species-speci®c differences in pholcid selective female cooperation (the female pro- chelicerae evolved because they prevent hy- vides a hood for those males able to overcome bridization. The lock-and-key version does her obstructive apophyses) rather than female this on a mechanical level, the genitalic rec- resistance to coercive males. Revealingly, the ognition hypothesis on a sensory level. Both females of several putative close relatives seem unlikely to account for the phenomenon have the cooperative structure (the hood) but in a general way. Often there is no female not the barrier (the apophyses) (Fig. 12). Ac- ``lock'', for instance in most Modisimus spe- cordingly, their males' cheliceral apophyses cies where the epigynum against which the are rather inconspicuous (Fig. 13), but nev- male chelicerae are pressed during copulation ertheless species-speci®c in form. is just a ¯at plate, but the chelicerae are nev- Male-male competition is equally unlikely ertheless species-speci®c (Huber 1998a). Even to provide a general explanation (Huber in cases with a lock-and-key like ®t, the hy- 1994). In most species the modi®cations seem pothesis that such ®t evolved to avoid cross- highly inef®cient for combat, and in some cas- speci®c pairings is dubious because natural es where their shape might suggest such a selection should favor early species recogni- function (as in the Ecuadorian pholcid illus- tion (Eberhard 1985, and references therein). trated in Fig. 14; I.D. #3) the female mor- However, transitory selection on cheliceral phology strongly indicates their use during morphology in a species-isolation context can- copulation (the epigynum is unusually broad not be ruled out, and may have been important and at the lateral extremes provides two blind- in the past (Shapiro & Porter 1989). ended cuticular tubes, whose location and The ``con¯ict of interest hypothesis'' (Al- spacing indicates that they are used to accom- exander et al. 1997), applied to pholcid che- modate the male fang-apophyses: Figs. 15, licerae, would explain their often complex and 16). However, inter- and intrasexual functions species-speci®c design by physical coercive are not mutually exclusive, and ®ghting with mating during which the male has to over- chelicerae must be considered a possibility in come female resistance and for this purpose single cases. uses his cheliceral modi®cations. One predic- Brignoli's (1973) speculation that pholcid tion is that one sex (usually the male) changes chelicerae may function to hold the sperm to increase the match and the other sex during sperm uptake is probably based on the evolves to either decrease it or it does not observations of Gerhardt (1921±1933, refer- evolve at all in that context (Alexander et al. ences in Huber 1998d) that pholcid males 1997, p. 5). The data available for pholcids do make no sperm webs but transfer the drop of not support this scenario. To the contrary, fe- sperm to the chelicerae and take it up from male epigyneal structures usually appear ei- there. This hypothesis obviously fails to ex- ther neutral (¯at plates) or even cooperative plain why males should evolve such a variety (hoods, grooves, pits, scapes) in that they help of modi®cations to perform the same simple the male to lodge his chelicerae and thus to task of holding a drop of sperm. position his body correctly. It should be em- Thus, by elimination, the hypothesis that phasized that con¯ict per se is of course not seems to ®t the available data best is cryptic a distinguishing characteristic between the female choice (as also for many other non- ``con¯ict of interest'' and the ``female choice'' genitalic male contact structures - Eberhard hypotheses. Con¯ict is a necessary result of 1985). Much like genitalia, chelicerae may female choosiness, and some morphological function as copulatory courtship devices, details re¯ect this con¯ict particularly clearly: whose elaborate morphology is used to stim- in a Peruvian pholcid (I.D. #2) the male pre- ulate or ®t the female in a way that increases sumably (judging by male and female mor- the male's chances of fathering her offspring. phology) lodges his cheliceral apophyses into In this hypothesis two main factors account a hood in the female epigynum. However, the for the diversity and relatively rapid evolution 138 THE JOURNAL OF ARACHNOLOGY

Figures 8±19.ÐCharacters discussed in the present paper. 8±9. Epigynum of a Peruvian pholcid (I.D. #2) in ventral and lateral (anterior side on left) view, showing the median hood and the lateral apophyses; 10±11. Male chelicerae of the same species, lateral and frontal view; 12±13. Epigynum, ventral view, and male chelicerae, frontal view, of `Blechroscelis' cyaneotaeniata (Keys.); 14. Portrait of an Ecuadorian species (I.D. #3), with modi®ed fangs; 15±16. Epigynum of the same species, in ventral and lateral (anterior side on left) view, showing the blind-ended tubes into which the male apophyses are presumably inserted during copulation; 17±18. Clypeal modi®cations in two Metagonia species from Peru (17; I.D. #4) and Brazil (18; I.D. #5); 19. Right pedipalp of a Bolivian species (I.D. #6), in which the patella is reduced. Drawn to different scales. HUBERÐSEXUAL SELECTION IN PHOLCID SPIDERS 139 of male chelicerae: the unpredictability of fe- gression of FA on size is often interpreted as male criteria and the never ending competition evidence for handicap models of sexual selec- among males for access to female eggs. How tion, in which only genetically ``good'' males females evaluate the minimal differences can produce display structures that are both among conspeci®c males' chelicerae, i.e., the large and symmetric (Mùller & Pomiankowski sensory and neuroanatomical basis for doing 1993; Watson & Thornhill 1994). so, remains an open question. From a mechanical point of view, the phol- Sexual modi®cations of the male clypeus cid male pedipalp works like a clamp, with are less common in pholcids, but have appar- the most distal segment (cymbium with pro- ently evolved several times convergently (e.g., cursus) acting against the femur. The economy in Metagonia: Figs. 17, 18, Holocneminus - of such a clamp is decreased by the two seg- Huber 1997b, Deeleman-Reinhold 1994). ments in-between (patella, tibia) and could be Like the chelicerae, the clypeal modi®cations improved by elimination of one or both seg- are highly species-speci®c and in one species ments. In fact, in many pholcids (e.g., in P. (Metagonia rica Gertsch 1986) it has been globosus), the patella is functionally reduced shown that they also contact the female gen- in that part of the muscles of the femur insert ital area during copulation (Huber 1997b). in the tibia (Huber & Eberhard 1997) and not A special case of non-genitalic contact as usually in the patella (Ruhl & Rathmayer structure is the frontal lobe in male Modisimus 1978). And in at least one species (sp. n. from culicinus (Simon 1893) (Figs. 5, 6). Clypeal Bolivia; I.D. #6) the reduction is complete, glands open at the lobe, and during copulation with the femur directly articulating with the the female mouth is in contact with the lobe, tibia and no external trace of the patella left suggesting gustatorial courtship (Huber (Fig. 19). Yet another characteristic apparently 1997a). However, the nature and function of functioning to increase the force applied to the gland products are unknown (trigger fe- one critical point is realized in P. globosus male responses that are favorable to male? - (and probably in all Physocyclus species and signal the female that copulation has oc- in Artema atlanta Walckenaer 1837): the pro- curred? - nourish the female?) meaning that a cursi are locked to each other, but the tip of decision between natural and sexual selection only one is inserted into the female, moved by is not yet possible (see Eberhard 1996 for ar- the muscular power of both pedipalps (Huber guments linking sexual selection and male & Eberhard 1997). seminal products). Thus, there seems to be an ultimate advan- tage for males with strong palps, but the prox- FORCEFUL GENITALIA imate function of this vigor is poorly under- It has been noted that ``details of copulatory stood. A possible solution may be in a courtship often seem to have little relationship structure of the female genitalia that is appar- to male size or vigor'' (Eberhard 1997: 35). If ently unique to pholcids: the so-called this is the rule, then many pholcids might be ``valve'', an often complex ``three dimension- exceptional: their genitalia are obviously their al zipper'' between copulatory pouch (uterus strongest organs (provided with the largest externus) and oviduct (uterus internus). An muscles), and in Physocyclus globosus (Tacz. apparent correlation has been documented be- 1873) this force is apparently used to rhyth- tween the strength and complexity of the mically squeeze parts of the female genitalia ``valve'' and the strength of the male pedipalp during copulation (Huber & Eberhard 1997). (Huber 1998c). The correlation may be phy- Moreover, a morphometric study of genitalic logenetically biased, however, so it is dif®cult and non-genitalic structures in the same spe- to interpret. cies also apparently supported the notion that It is not surprising that the recently inten- there is sexual selection on male vigor: ¯uc- si®ed research on pholcids has raised more tuating asymmetry (FA: deviations from per- questions than it has answered. Thus, I would fect bilateral symmetry that are thought to re- like to close this short review with yet another ¯ect the degree of developmental stability) in riddle. The males (but not females) of most large (strong) genitalia tended to be lower species of several mainly Central American than in small genitalia (Huber 1996b). In the genera have the femora of their walking legs recent literature on FA such a negative re- covered with short, almost perpendicular hairs 140 THE JOURNAL OF ARACHNOLOGY

(Huber 1998a), resembling taste hairs (Foelix Elgar, M.A. 1998. Sperm competition and sexual & Chu-Wang 1973) (Fig. 7). Nothing is selection in spiders and other . Pp. 307± known of these hairs, apart from the approx- 339. In Sperm Competition and Sexual Selection imate systematic and geographical distribution (T.R. Birkhead & A.P. Mùller, eds.). Academic of the character, the improbability of taste Press. Foelix, R.F. & Chu-Wang I-Wu. 1973. The mor- hairs being concentrated on femora, and the phology of spider sensilla. II. Chemoreceptors. apparent lack of terminal pores necessary for Tissue & Cell, 5(3):461±478. chemosensory function. Huber, B.A. 1994. Genital morphology, copulatory ACKNOWLEDGMENTS mechanism and reproductive behaviour in Psil- ochorus simoni (Berland, 1911) (Pholcidae; Ar- I thank James Berry, William Eberhard, aneae). Netherlands J. Zool., 44(1±2):85±99. Gustavo Hormiga and Brent Opell for valu- Huber, B.A. 1995. Copulatory mechanism in Hol- able comments on a previous draft. This study ocnemus pluchei and Pholcus opilionoides, with was supported by a Theodore Roosevelt Fel- notes on male cheliceral apophyses and stridu- lowship from the American Museum of Nat- latory organs in Pholcidae (Araneae). Acta Zool., ural History, New York. Stockholm, 76:291±300. Huber, B.A. 1996a. On the distinction between LITERATURE CITED Modisimus and Hedypsilus (Araneae, Pholcidae), Alexander, R.D., D.C. Marshall & J.R. Cooley. with notes on behavior and natural history. Zool. 1997. Evolutionary perspectives on insect mat- Scripta, 25:233±240. ing. Pp. 4±31. In The Evolution of Mating Sys- Huber, B.A. 1996b. Genitalia, ¯uctuating asym- tems in Insects and Arachnids (J.C. Choe & B.J. metry, and patterns of sexual selection in Phy- Crespi, eds.). Cambridge Univ. Press, U.K. socyclus globosus (Araneae: Pholcidae). Rev. Brignoli, P.M. 1973. Notes on spiders, mainly Suisse Zool., suppl. 1996:289±294. cave-dwelling of Southern Mexico and Guate- Huber, B.A. 1997a. Evidence for gustatorial court- mala (Araneae). Accad. Naz. Lincei, 171:195± ship in a haplogyne spider (Hedypsilus culicinus: 238. Pholcidae: Araneae). Netherlands J. Zool., 47: Deeleman-Reinhold, C.L. 1994. Redescription of 95±98. Holocneminus multiguttatus Simon and descrip- Huber, B.A. 1997b. On American ``Micromerys'' tion of two new species of pholcid spiders from and Metagonia (Araneae, Pholcidae), with notes Australia (Arachnida: Araneae: Pholcidae). Beitr. on natural history and genital mechanics. Zool. Araneol., 4:31±42. Scripta, 25:341±363. Eberhard, W.G. 1985. Sexual Selection and Huber, B.A. 1998a. Notes on the neotropical spider Genitalia. Harvard Univ. Press, Cambridge, Mas- Modisimus (Pholcidae, Araneae), with de- sachusetts. scriptions of thirteen new species from Costa Eberhard, W.G. 1992. Notes on the ecology and Rica and neighboring countries. J. Arachnol. behaviour of Physocyclus globosus (Araneae, 26(1):19±60. Pholcidae). Bull. British Arachnol. Soc., 9:38± Huber, B.A. 1998b. Genital mechanics in some 42. neotropical pholcid spiders (Araneae; Pholcidae), Eberhard, W.G. 1996. Female Control: Sexual Se- with implications for systematics. J. Zool., Lon- lection by Cryptic Female Choice. Princeton don, 244:587±599. Univ. Press, New Jersey. Huber, B.A. 1998c. On the ``valve'' in the genitalia Eberhard, W.G. 1997. Sexual selection by cryptic of female pholcids (Pholcidae, Araneae). Bull. female choice in insects and arachnids. Pp. 32± British Arachnol. Soc., 11(2):41±48. 57. In The Evolution of Mating Systems in In- Huber, B.A. 1998d Spider reproductive behaviour: sects and Arachnids (J.C. Choe & B.J. Crespi, a review of Gerhardt's work from 1911±1933, eds.). Cambridge Univ. Press, U.K. with implications for sexual selection. Bull. Brit- Eberhard, W.G. & D. BricenÄo L. 1983. Chivalry in ish Arachnol. Soc., 11(3):81±91. pholcid spiders. Behav. Ecol. Sociobiol., 13: Huber, B.A. & W.G. Eberhard. 1997. Courtship, 189±195. copulation, and genital mechanics in Physocyclus Eberhard, W.G. & D. BricenÄo L. 1985. Behavior globosus (Araneae, Pholcidae). Canadian J. and ecology of four species of Modisimus and Zool., 74:905±918. Blechroscelis (Araneae, Pholcidae). Rev. Arach- Kaster, J.L. & E.M. Jakob. 1997. Last-male sperm nol., 6:29±36. priority in a haplogyne spider (Araneae: Pholci- Eberhard, W.G. & B.A. Huber. 1998. Possible links dae): correlations between female morphology between embryology, lack of innervation, and and patterns of sperm storage. Ann. Entomol. the evolution of male genitalia in spiders. Bull. Soc. America, 90:254±259. British Arachnol. Soc., 11(2):73±80. Mùller, A.P. & A. Pomiankowski. 1993. Fluctuat- HUBERÐSEXUAL SELECTION IN PHOLCID SPIDERS 141

ing asymmetry and sexual selection. Genetica, (Pholcidae; Araneae). Acta Zool., Stockholm, 89:267Ð279. 75:1±12. Ruhl, M. & W. Rathmayer. 1978. Die Beinmusku- Uhl, G., B.A. Huber & W. Rose. 1995. Male ped- latur und ihre Innervation bei der Vogelspinne ipalp morphology, and copulatory mechanism in Dugesiella hentzi (Ch.) (Araneae, Aviculariidae). Pholcus phalangioides (Fuesslin, 1775). Bull. Zoomorphologie, 89:33±46. British Arachnol. Soc., 10:1±9. Shapiro, A.M. & A.H. Porter. 1989. The lock-and- Watson, P.J. & Thornhill, R. 1994. Fluctuating key hypothesis: evolutionary and biosystematic asymmetry and sexual selection. TREE, 9:21±25. interpretation of insect genitalia. Ann. Rev. En- Yoward, P. 1998. Sperm competition in Pholcus tomol., 34:231±245. phalangoides (Fuesslin, 1775) (Araneae, Pholci- Uhl, G. 1993. Mating behaviour and female sperm dae) - Shorter second copulations gain a higher storage in Pholcus phalangioides (Fuesslin) (Ar- paternity reward than ®rst copulations. Pp. 167± aneae). Mem. Queensland Mus., 33(2):667±674. 170. In Proc. Arachnol. Congress, Edinburg. Uhl, G. 1994. Genital morphology and sperm stor- (P.A. Selden, ed.). age in Pholcus phalangioides (Fuesslin, 1775) Manuscript received 29 April 1998, revised 6 Sep- tember 1998.