Great Basin Naturalist

Volume 45 Number 4 Article 6

10-31-1985

Sexual selection and variation in anuran amphibians of the Arizona–Sonoran Desert

Brian K. Sullivan Arizona State University, Tempe

Follow this and additional works at: https://scholarsarchive.byu.edu/gbn

Recommended Citation Sullivan, Brian K. (1985) " and mating system variation in anuran amphibians of the Arizona–Sonoran Desert," Great Basin Naturalist: Vol. 45 : No. 4 , Article 6. Available at: https://scholarsarchive.byu.edu/gbn/vol45/iss4/6

This Article is brought to you for free and open access by the Western North American Naturalist Publications at BYU ScholarsArchive. It has been accepted for inclusion in Great Basin Naturalist by an authorized editor of BYU ScholarsArchive. For more information, please contact [email protected], [email protected]. SEXUAL SELECTION AND MATING SYSTEM VARIATION IN ANURAN AMPHIBIANS OF THE ARIZONA-SONORAN DESERT

Brian K. Sulli\an'

Abstract—Mating system variation in anuran amphibians of the Arizona-Sonoran Desert was reviewed. Male density and breeding period duration were negatively correlated in seven bufonids and pelobatids. Variation in male mating behavior and ability of females to freely select their mates unhindered by active-searching males also was related directly to male density. These observations support hypotheses relating ecological factors to mating system organization. It is suggested that male calling behavior, and anuran sytems in general, may be significantly influenced by predation on vocalizing males.

The Arizona-Sonoran Desert boa.sts a surpris- ence opportunities for female mate selection ing diversity of frogs and toads: seven bu- and, hence, male behavior. Wells (1977) ar- fonids, three pelobatids, three ranids, two gued that for explosive-breeding anurans fe- hylids, one microhyhd, and one lepto- male choice should be restricted as a result of: dactyhd. As expected, the breeding biology of (1) temporal constraints limiting time avail- these anurans is significantly influenced by able for mate selection and (2) high male den- the xeric climate. Most of these species have sities and corresponding alternative male short, or "explosive," (Wells 1977) breeding mating tactics that typically reduce the ability periods during the summer rainy season, al- of females to move freely in breeding aggrega- though a few have relatively extended, or tions. Conversely, in species with prolonged "prolonged" breeding periods during the breeding periods and lower male densities, spring (Sullivan 1982a, 1983a, 1984, Sullivan females should be capable of freely selecting and Sullivan 1985). As suggested by Wells their mates and constitute a potentialK- signif- (1977), the mating behavior of anurans such as icant evolutionary force (e.g., Arnold 1983). those of the Arizona-Sonoran Desert appears The variation in breeding biology of anurans related to breeding period duration. For ex- of Arizona allows direct test of these hypothe-

ample, Woodhouse's toad {Biifo woodhousei) ses. Here I summarize recent studies of sexual forms low-density breeding aggregations at selection in pelobatids and bufonids of Ari- sources of permanent water throughout the zona and, following Emlen and Oring (1977), spring, and all males within these choruses document the existence of predictable rela- call to attract females (Sullivan 1982a). In con- tionships between the ecology, mating sys-| trast, the Great Plains toad (B. cog,natus) terns, and reproductive behavior of these spe-

forms high density aggregations at temporary cies. I also examine species recognition and rain pools for only a few nights each summer. acoustic competition among these anurans.

Within these short-lived choruses some males Lastly, I discuss the evolution of anuran lek adopt satellite positions near vocalizing males mating sytems. and attempt to intercept females approaching the calling males (Sullivan 1982b). General Methods This diversity in breeding period duration and reproductive behavior provides an excel- Methods employed in the studies reviewed lent opportunity for evaluating hypotheses of here ha\ e been described in detail elsewhere sexual selection and mating svstem theorv (Sullivan 1983c, 1984, Sullixan and Sullivan (Trivers 1972, Emlen and Oring 1977, Wells 1985); a brief simimary follows. Observations 1977). It is generally hypothesized that diua- at anuran breeding aggregations in southern tion of breeding period should directly influ- Arizona were made with a six-\ olt headlamp

Uepartini'iit of Zoolo sity, Tcnipc. Arizona H5'2H7 78712.

688 October 1985 SULLIVAN: SONORAN AMPHIBIANS 689

Table 1. Mean mule density (number of males) and brcedinj; period duration (days) for seven anurans (Bufo and Scap/^iopws) of southern Arizona (N ^ number of breeding aggregations). 690 Great Basin Naturalist Vol. 45, No. 4

Table 2. Relationships between male size and mating success five anurans (Bufo and Scaphiopus) of southern Arizona.

Snout-vent length (mm) Mating males Nonmating males Positive assortative mating Source

Species X N .X N P r N P

S. coiichi 71.0 70.0 6 >.05* 0.07* 12 =.50 Sullivan and Sullivan, 1985 74.1 71.1 18 >.05 0.07* 20 =.43 74.4 69.2 18 >.05=* -0.40* 10 =.24

S. multiplicatus 50.9 51.3 28 >.05 0.05 21 >.05 Sullivan and Sullivan, 1985 51.2 50.0 15 >.05 48.0 48.0 11 >.05* B. cognatus 78.6 77.2 150 >.05 0.09 24 >.05 Sullivan, 1983a B. debilis 41.3 42.3 16 >.05* 0.20* 6 >.05 Sullivan, 1984 B. woodhousei 84.1 85.4 43 >.05 0.23 47 >.05 Sullivan, 1983b

*For small samples (< 10) a Mann-Whitney U test or Kendal's tau was calculated

anurans should select large males as mates if sei in Arizona male mating success is weakly

size is an indicator of fitness (Wilbur et al. correlated (R" = — 25%) with participation in 1978, Fairchild 1981). Others have argued chorus activity (Fig. 1). However, this result that females might benefit from selecting a is expected since many males are only ob- male similar in size to themselves if fertiliza- served at a chorus for one night: a correlation

tion success is related to the size difference occurs because any male who mates two or between mating males and females (Licht three times must be present at the chorus for 1976, Davies and Halliday 1979). Analysis of at least as many nights. Further, many males male size and mating success in four explo- who are unattractive as mates (low call rate sive-breeding anurans of Arizona has not sup- males) are active at choruses for many nights ported either of these hypotheses (Table 2.) without obtaining any matings. Hence, for

Male size is positively correlated with mat- this anuran male persistance in chorus activity ing success in some explosive-breeding bu- is not an important determinant of mating fonids and pelobatids of the United States success. (Gatz 1981, Woodward 1982). However, in Female mate selection has been observed these forms large males might achieve greater in breeding aggregations of the three Arizona mating success as a result of size advantages bufonids with low-densitv choruses (Sullivan during direct struggles for possession of fe- 1983a, 1984). In all of these forms, females males (Sullivan 1983a). Davies and Halliday initiate amplexus by making physical contact (1979) and Lamb (1985) determined that large with a calling male and are able to move freely male bufonids are more successful in taking through breeding choruses to select their over females during male-male disputes. For mates from among the calling males. In B.

these species a mating advantage for large woodhousei male mating success is correlated males may be independent of any female pref- with call rate, suggesting that females prefer erence. Woodward (1982) documented a high call rate males as mates (Sullivan 1983a). large-male mating advantage in one of four S. Males within a chorus retain a consistent, rel- couchi and one of two S. multiplicatus breed- ative ranking in call rate within and between

ing aggregations in central New Mexico. This nights, and there is significantly more varia- interspecific variation suggests that a number tion in call rate between than within males of variables such as density and alternative (Sullivan 1983a, Sullivan and Leek in re\iew). mating tactics of males influence the outcome Females prefer speakers liroadcasting calls at of sexual selection in nature. the higher rate in simple two-speaker discrim- Male mating success in prolonged-breed- ination experiments, further supporting the ing anurans is often correlated with the num- female preference hypothesis. Proximately, ber of nights a male participates in chorus female choice of high call rate males could be activity (Kluge 1981, Woodward 1981, Ryan passive if they merely locate such males more

1983). For prolonged-breeding B. woodhou- readily, or active if genetically determined October 1985 SULLIVAN: SONOHAN AMPHIBIANS 691

1981-1 . ,

692 Great Basin Naturalist Vol. 45, No. 4 would be determined by the heritability of patic species (Gerhardt 1982, Rose and male call rate and its relation to offspring sur- Capranica 1983, Sullivan and Leek, in re- vivorship and mating success. In any event, view). Discrimination experiments have re- mating is clearly nonrandom with respect to vealed that female tree frogs (Hylidae) can male phenotype and female behavior in B. discriminate between two acoustic stimuli dif- woodhousei fering only in pulse rate, and they prefer stim- These observations on male behavior and uli similar to advertisement calls of con- mate selection by females reveal that the pro- specific males (Loftus-Hills and Littlejohn longed-breeding species {B. piinctatiis and B. 1971, Straughan 1975). ivoodhotisei), and at least some of the explo- All pelobatids and most bufonids occurring sive-breeding forms (B. cognatiis and B. de- in southern Arizona have been observed in bilis), have lek-like mating systems (Sullivan mixed-species breeding aggregations, provid- 1983c). Females typically select their mates ing the potential for heterospecific interac- freely from among aggregations of displaying tions (Fig. 2). The ranges in pulse rate and males, and males provide only sperm to their dominant frequency of advertisement calls of mates. Females mate disproportionately of- these anurans are given in Figure 3. As ex- ten with males possessing high call rates in pected, advertisement calls of the closely re- both B. cognatus and B. woodhousei (Sullivan lated forms, S. boynbifrons and S . mitltiplica-

1983c); for the other species it is unknown tiis, and B. mic rosea pJuis and B. woodhousei whether females discriminate between con- are clearly dissimilar in pulse rate but not specific males on the basis of male phenotype frequency. In the pelobatids, hybrid offspring (Sullivan 1984). The explosive-breeding pelo- are sterile; hence, there is presumably strong batids exhibit a typical male-dominance mat- selection against females who select het- ing system in which mate selection by females erospecific mates. The divergence in adver- is somewhat limited (Emlen and Oring 1977). tisement calls of these two species in south- Males may obtain multiple matings within a eastern Arizona suggests that species breeding season if they participate in breed- recognition has been a significant selective ing aggregations following every rainstorm force acting on female . Hy- each summer (Sullivan 1983b). Although gen- bridization has also been documented be- erally consistent with the hypothesis of Emlen tween the two bufonids, but nothing is known and Oring (1977) concerning the significance concerning the evolutionary significance of of breeding period duration to mating system this interaction (Sullivan 1986b). Sympatri- structure, these observations indicate clearly cally breeding bufonids in Texas have also

that male density, through its impact on fe- been found to differ in pulse rate and fre- male choice and regardless of breeding period quency (Blair 1956). Advertisement call varia-

length, influences sexual selection and the tion in two other Arizona anurans is also com- organization of the mating systems of these patible with the species recognition hypo- desert anurans. thesis: the closely related but allopatric bu- fonids B. debilis and B. retifonnis do not differ dramatically in pulse rate. Presumably, selec- Species Recognition tion has not favored divergence in advertise- AND Acoustic Competition ment call structure in the absence of an oppor- An important component of female mate tunity for heterospecific interactions. selection concerns proper species recogni- Aspects of intra- and interspecific acoustic

tion. For anurans it is generally assumed that competition between anuran males have re- interspecific call differences allow sympatri- cently received considerable attention (re- cally breeding species to avoid hvbridization viewed by Sullivan, 1985). A number of work- (e.g., Hodl 1977, Drewry and Rand 1983, ers have argued that male anurans increase Duelhnan and Pyles 1983). For example, their locatability for selecting females by among hylid and bufonid forms, rate of ampli- calling asynchronously with neighboring con- tude modulation (pulse rate) of the male ad- specifics as well as heterospecifics. In Arizona vertisement call is apparently the primary call B. woodhousei follows this pattern: males re- parameter separating closely related sym- duce their calling activity in response to play- October 1985 SULLIVAN: SONOIUN AMPHIBIANS 693

SYMPATRICALLY BREEDING ANURANS IN ARIZONA

Sb Sc Sm Ba Be Bd Bm Bp Br Bw

Sb -

Sc . -

Sm • •

Ba •

Be •

Bd • • • Bm

Bp •

Br • •

Bw • •

Fig. 2. Anurans (Biifo and Scaphioptis) observed breeding synipatrieally (•) in southern Arizona (Sb = S. = = = = bombifrons , Sc = S. couchi , Sm = S. miiltipUcatus, Ba B. alvaiius. Be B. cognatus, Bd B. debilis, Bm B.

=^ microscaphus , Bp = B. piinctatus, Br B. rctifonnis. Bw = B. icoodhousei).

back of conspecific advertisement calls as well to reduce the risk of predation; predators who as a variety of synthetic stimuli (Sullivan 1985, locate anurans acoustically have difficulty cap- Sullivan and Leek, in review). Proximately, turing synchronously chorusing forms (Ryan males avoid calling during playback of any and Tuttle 1982). Below I develop the hypoth- acoustic stimuli with a freqency close to the esis that predation has shaped the evolution of mean frequency of the species-typical call anuran lek mating systems in general. (Sullivan and Leek, in review). However, male B. piinctatus do not avoid overlap with Evolution of Anuran Lek Mating Systems broadcast of conspecific advertisement calls (Sullivan 1985). These bufonids possess rela- The evolution of lek mating systems has tively long calls (x = 7.0 sec) and appear to recently received much attention (Bradbury prefer to call synchronously with nearby 1981, Loiselle and Barlow 1978, Wrangham calling males. Such interspecific variation 1980). Lek mating systems, in which females

suggests that male calling behavior is influ- select mates based on male phenotype or enced by factors other than those associated dominance position on a display arena, have with locatability for females alone. It may be been described for a variety of organisms, that males of some species call synchronously including two Arizona bufonids (Sullivan 694 Great Basin Natur.\list Vol. 45, No. 4

r2 October 1985 SULLIVAN: SONORAN AMPHIBIANS 695

different sizes since for both species there course, female choice of individual males were always from 1 to 4 choruses simulta- within aggregations would remain an impor- neously active near the study chorus (<0.5 tant aspect of sexual selection in species with km) on the night of observation (some females lek mating systems.

even visited more than one chorus). It is im- Observations on mate selection by females portant to note that in this analysis the as- also suggest that the spatial distribution of sumption of independence of variables is vio- males within a chorus is unaffected by female lated since OSR is in part derived from chorus behavior. For example, some workers have size; however, the results indicate clearly that suggested that in lek breeding forms females for these aggregations OSR does not increase prefer males positioned at the center of the with chorus size. Hence, females do not display area. The linear, shoreline choruses of prefer to visit larger choruses when selecting B. woodhousei can be readily partitioned into their mates. Observations on the intensity of three equal areas with respect to the distribu- sexual selection in B. woodJiousei further sup- tion of males within the aggregations to test port the hypothesis that anuran females do not this hypothesis. Analysis of mating success prefer to visit larger aggregations of males and chorus location for all males on 18 nights when selecting a mate. Sullivan (1986a) docu- ofchorus activity reveals there was no position mented that the intensity of sexual selection effect: centrally located males did not achieve increases with chorus size (male density) in B. a disproportionate number of matings (X^ = woodhoiisei . That is, the proportional num- 0.33, P > 0.05, df = 2). Thus, within choruses ber of females decreases as male density in- females do not appear to favor the clustering creases: proportionally fewer males obtain of males. matings in larger choruses. Under the extremely arid conditions of the This result suggests that females do not Arizona-Sonoran Desert the lek- like mating derive any benefits by selecting their mate systems and acoustic courtship behaviors em- from among larger groups ofdisplaying males. ployed by anurans seem efficient means of I suggest that benefits accruing to individual rapidly bringing the sexes together. How- males who display communally are the pri- ever, this review indicates that some diversity mary factors influencing anuran lek behavior. in mating systems can occur under these For example, many authors have suggested severe ecological constraints. It is clear that that males who participate in aggregations suf- environmental factors influencing breeding fer less predation than males who display in period duration shape mating system organi- isolation (see reviews in Alexander 1975, zation as a result of corresponding variation in Bradbury 1981, Loiselle and Barlow 1978, male density and behavior. Continued study Wrangham 1980). Recently Ryan et al. (1981) will be necessary to further elucidate the pre- substantiated this claim by showing that in a cise action of sexual selection in relation to neotropical leptodactylid frog individual pre- mating system variation in breeding aggrega- dation risk decreases for males as chorus size tions of these desert anurans. increases. Ryan et al. (1981) found that at least three predators located males acoustically; hence males might benefit greatly by displaying Acknowledgments communally. Predation may only be important to males since they participate in choruses night This work was supported by grants from after night: most anuran females visit choruses Sigma Xi, the American Museum of Natural only on the night they oviposit and are inconspic- History, and the Zoology Department, Ari- uous while present in breeding aggregations zona State University. (e.g., Sullivan 1982a, Wells 1977).

If males derive substantial benefits by dis- playing communally, and females incur no (or Literature Cited insignificant) costs by selecting their mate Alexander, R. D. 1975. Natural selection and special- from among such groups, then it is conceiv- ized chorusing behavior in acoustical insects. able that females play no active role in the Pages 35-77 in D. Pimentel, ed., Insects, science evolutionary maintenance of lek behavior. Of and society. Academic Press, New York. .

696 Great Basin Natufl^list Vol. 45, No. 4

Arak, a. 1983. Male-male competition and mate choice Ryan, M J , M D Tlttle, and L K Taft 1981, The costs in anuran amphibians. Pages 181-210 in P. P. G. and benefits of frog chorusing behavior, Behav. Bateson, ed.. Mate choice. Cambridge University Ecol. Sociob. 8:27,3-278.

Press, Cambridge. Str.\ughan, I. R 1975. An analysis of the mechanisms of call Arnold, S. J. 1983. Sexual selection: the interface of mating discrimination in the frogs Hyla regilla theory and empiricism. Pages 67-107 in P. P. G. and H. cadaierina. Copei 1975:41.5-424.

Bateson, ed.. Mate choice. Cambridge University Sullivan. B K 1982a. Sexual selection in Woodhouse's

Press, Cambridge. toad (Bitfo woodhousci). I. Chorus organization.

leks. , .30:680-686. Bradbury, J. W. 1981. The evolution of Pages Anim. Beha\ 138-169 in R. D. Alexander and D. W. Tinkle, 1982b. Male mating behavior in the Great Plains eds.. Natural selection and social behavior: recent toad {Bufo cognatiis). Anim. Behav. 30:9.39-940. research and new ideas. Chiron Press, New York. 1982c. Significance of size, temperature and call Davies, N. B. and T R. Halliday. 1979. Competitive attributes to sexual selection in Bufo woodhousci

mate searching in male common toads, Bufo hitfo australis. J. Herpetologv' 16:10.3-106, Anim. Behav. 27:1253-1267. 1983a, Sexual selection in Woodhouse's toad Drewry, G. E. and a, S Rand 1983. Characteristics of an (Bufo woodJwusei). II, Female choice, Anim, Be- acoustic community: Puerto Rican frogs of the hav, 31:1011-1017.

genus Eletitherodactylus . Copeia 1983:941-953. 1983b, Sexual selection in the Great Plains toad Duellman, W. E. and R A Pyles. 1983. Acoustic re- {Bugo cognatus). Behaviour 85:58-64. source partitioning in anuran communities. 1983c. Sexual selection and mating system varia- Copeia 1983:639-649. tion in the Great Plains toad {Bufo cognatus Say)

Emlen. S. T. and L. W. Oring 1977. Ecology, sexual and Woodhouse's toad {Bufo woodhousei australis selection and the evolution ot mating svstems. Shannon and Lowe). Unpublished dissertation, Science 197:21.5-223. Arizona State University, Tempe.

FairCHILD, L. 1981. Mate selection and behavioral ther- 1984. Advertisement call variation and observa- moregulation in Fowler's toads. Science tions on breeding behavior oi Bufo debilis and B.

212:950-951. punctatus. J. Herpetology 18:406-411. Gerhardt, H. C. 1982. Sound pattern recognition in 1985. Male calling behavior in response to play- some North American treefrogs (Anura: Hylidae): back of conspecific ad\ertisement calls in two bu-

implications for mate choice. Amer. Zool. fonids, J. Herpetology 19:78-83. 22:581-595. 1986a. Intra-populational variation in the intensity

HODL, W. 1977. Call differences and call site segregation of sexual selection in breeding aggregations of in anuran species from central Amazonian floating Woodhouse's toad {Bufo woodhousei). J. Her-

meadows. Oecologia 28:351-363. petolog> , In press, KiRKPATRiCK, M. 1982. Sexual selection and the evolution 1986b, Hybridization between the southwestern of female choice. Evolution .35:1-12. toad {Bufo microscaphus) and Woodhouse's toad {Bufo woodhousei) in Arizona: morphological vari- Lambert, D M., P. D. Kingett, and E. Slooten. 1982. ation, Herpetology. In press. Intersexual selection: the problem and a discus- J, sion of the evidence. Evol. Theory 6:67-78. Sullivan, B. K. and E. A. Sullivan. 1985. Size-related variation in advertisement calls and breeding be- Lande, R. 1981. Models of speciation by sexual selection havior of spadefoot toads {Scaphiopus bombifrons. on polygenic traits. Proc. Nat. Acad. Sci. S. Couchi and S, multiphcatus). Southwestern 78:3721-3725. Natur. 30:349-355. Light, L. E. 1976. Sexual selection in toads (Biifo amcri- Sullivan, B, K, andM. R. Leek, In review. Acoustic com- camis). Canadian J. Zool. 54:1277-1284. munication in Woodhouse's toad {Bufo woodhou- LOFTUS-HlLLS, J. J. AND M. J. LiTTLEJOHN. 1971. Pulse sei). I, Response of calling males to variation in repetition rate as the basis for mating call discrimi- spectral and temporal components of advertise- nation bv two sympatric species oi Hyla. Copeia ment calls. Behaviour. 1971:154-156. Taylor, P. D. andG. C. Williams. 1982. The lek paradox Loiselle, p. V. AND G. VV Barlow 1978. Do fishes lek is not resolved. Theor. Pop. Biol. 22:392-409. like birds? Pages 31-75 in E. S. Reese and F. S. Tri\ers.K L 1972, Parental investment and sexual selec- Lightner, eds., Constrasts in behavior. J. Wiley & tion. Pages 136-179 in B, Campbell, (ed,). Sexual Sons, New York. selection and the descent of man: 1871-1971, Parker, G. A. 1982. Phenotype-limited evolutionarily sta- .\ldine Press, Chicago, ble strategies. Pages 173-201 in King's College Wi:lls, K D 1977, The social behaviour of anuran am- Sociobiology Group, ed.. Current problems in phibians, Anim. Behav. 25:666-693. sociobiology. Cambridge University Press, Cam-

bridge. Wilbur, H. M , D, L. Rubenstein, and L. Fairghild Rose. G.J and R. R. Capraniga 1984. Processing ampli- 1978. Sexual selection in toads: the roles of female tude-modulated sounds by the auditory midbrain choice and male body size. Evolution 32:264-270.

oftwo species of toads: matched tenijioral filters. J. Wrangham, R. W 1980. Female choice of least costly Comp. Physiol. A 154:211-219. males: a possible factor in the evolution of leks. Z. Ti(>rps>chol. 54:357-367.