Anolis Paper British Virgin Islands.Pdf

870 OTES

The Influence of Temperature and Humidity on Activity Patterns of the Lizards Anolis stratulus and Ameiva exsul in the British Virgin Islands

KERRY L. NICHOLSON, SHA 0 M. TORRENCE, DA AM. GHIOCA, JOYDEEP BHATTACHARJEE, ADRIAN E. ANDREI, JENNIFER OWEN, NIKKI Io A. RADKE, AND GAD PERRY. Department of Range, Wildlife, and Fisheries Management, Box 42125, Texas Tech University, Lubbock, Texas 79409. Corresponding author: Gad Perry, email: [email protected] 1 i ABSTRACT.-Many organisms modify their behav- t ior to reduce exposure to unfavorable abiotic conditions, but detailed information is available for only a few species. We studied the diurnal activity patterns of AI101is stratulus and Ameiva exsul on Guana Is- land, British Virgin Islands, in order to determine how they are affected by temperature and humidity. We surveyed transects on foot between 0730 and

ms. received December 10, 2004; accepted Septem- ber 20, 2005 NOTES 871

1700 h, scanned the ground and vegetation for vis- Guana Island. Activity patterns and their ible lizards and recorded temperature and relative correlates have not been previously re- humidity. Lizard activity patterns were influenced ported for either species, although R. Pow- by ambient conditions and body size. We found an ell and R.W.Henderson (unpublished data) inverse relationship between daily activity patterns observed that activity of A. exsul on Guana and temperatures for juvenile A. stratulus; the cooler the temperature, the more juveniles were present. Island, BVI, primarily occurred between Adult A. stratulus did not show any significant cor- 0930 and 1400hours. relations with temperature and time of day. Tem- Guana Island is a privately-owned wild- perature and abundance were strongly positively life sanctuary characterized by tempera- correlated for A. exsul; the higher the temperature, tures ranging from 27-33 °C, a relative hu- the more abundant A. exsul became. Activity was midity of 60-90% (Dmi'el et al. 1997), and strongly significantly correlated with humidity. Be- an annual rainfall of about 900 mm (G. cause temperature and relative humidity were sig- Perry, unpublished data). The island has an nificantly inversely correlated, we cannot identify area of approximately 300 ha and a maxi- which parameter most impacted lizard activity. mum elevation of 240m (Lazell1996, 2005). Additional details regarding the study is- KEYWORDS.-body size, temperature, relative hu- land are in Lazell (1996,2005). midity, water loss, Guana Island To investigate the activity patterns of A. stratulus and A. exsul, we established Thermoregulation is a primary activity of twelve 50 m transects along existing trails ectotherms. Many reptiles such as lizards during October 2003.A single observer re- act to reduce exposure to unfavorable abi- peatedly surveyed the entire length of each otic conditions by simply moving out of di- transect on foot between 0730 and 1700h. rect sunlight or by increasing their water During each hourly pass, the observer intake. Thus, understanding the relation- would sample the same six stations, located ship between abiotic conditions and activ- 10 m apart. At each station, the observer ity patterns can be important in under- examined the ground and vegetation for standing the ecology of such animals. Here lizards, taking one minute to complete a we report on the daily activity patterns of 180° scan of the vegetation within a 2 m two lizard species in the British Virgin Is- radius. The count was repeated five con- lands (BVI),and the effects of temperature secutive times at each station, for a total of and humidity on those activity patterns. 5 minutes at each location, and the number The two were chosen as representatives of of lizards seen during the most productive the genera Anolis and Ameiva which consti- one-minute count was recorded. Each tran- tute a large percentage of the diurnally ac- sect required 30 min to complete, and a 30 tive lizards in the West Indies. min pause separated consecutive passes, Lizards have often served as model or- resulting in 10 sampling cycles per day. ganisms in ecological studies (e.g.,Milstead Twelve daily samples were used in calcu- 1967; Vitt and Pianka 1994). Because of lating the total numbers of animals ob- their abundance and visible nature, Anolis served. lizards have been especially extensively in- Perry et al. (unpublished MS) reported vestigated (Roughgarden 1995; Reagan that smaller body size makes individual liz- 1996). Anolis stratulus (adult mass approxi- ards more prone to water loss in Anolis cris- mately 1.7 g; Butler and Losos 2002) has tatellus and A. exsul. Based on this work, the received considerable attention, with stud- authors predicted that water conservation ies ranging from habitat use (Dial et al. requirements will force smaller lizards to 1994)to molecular systematics (Jackman et be active during cooler and more humid al. 1999).Ameiva exsul (adult mass approxi- times of day. They suggested that adults mately 40 g; Lewis and Saliva 1987)is also may be able to retain water more efficiently a highly visible and well-studied species than juveniles, who have a larger surface- (Lewis and Saliva 1987). Both species are area-to-body-size ratio. Juveniles should common on the islands of the Greater therefore decrease activity during the hot- Puerto Rico Bank, including our study site, test part of the day to avoid water loss, es- 872 NOTES

pecially in water-restricted locations such tivity levels (Fig. 1B) were negatively re- as Guana. To test this prediction, we cat- lated to temperature (rho = -0.684, N = 10, egorized A. stratulus into two size classes, P = 0.029)and positively related to relative juvenile and adult, based on lack of dewlap humidity (rho = 0.720, N = 10, P = 0.019). development in juveniles (G. Perry and Temperature and relative humidity were K. LeVering, unpublished). Size classes of also negatively correlated to one another A. exsul, which is less common on Guana, (rho = 0.848, N = 10, P = 0.002),making it were combined to obtain an adequate difficult to distinguish which factor lizard sample size. To avoid artifacts caused by activity patterns were responding. aberrant climatic conditions such as heavy We encountered 43 A. exsul of all sizes, rainfall, counts were conducted only dur- and their activity pattern was different than ing days with climatic conditions condu- that observed in Anolis. Number of sight- cive to lizard activity. We recorded ambient ings was positively correlated with tem- temperatures (0C) and relative humidities perature for A. exsul (Fig. IC: rho = 0.800; (%) at five minute intervals by placing a N = 10, P = 0.005), and negatively corre- datalogger at the starting point of each lated with relative humidity (rho = -0.804; transect. To avoid making assumptions N = 10, P = 0.005).Few individual Ameiva about data distribution, we used Spear- were active during the cooler parts of the man's non-parametric test of correlation for day and most observations occurred dur- all analyses of relationship between climac- ing the warmer, drier surveys. Activity tic conditions and lizard activity. peaked between 1430h and 1500h.R. Pow- We recorded a total of 140 A. stratulus ell and R.W. Henderson (unpublished) re- sightings during our surveys, of which '56 port similar activity times in A. exsul from were adults and 84 juveniles. Both adult Guana. Similarly, activity in Ameiva eryth- and juvenile A. stratulus showed variation rocephala begins at 0930,peaks midday, and in activity levels during the day. However, ceases by 1600h (Kerr et al. 2005).Thus, the the activity levels of adults (Fig, 1A) were activity pattern we describe, although not not significantly correlated with either tem- previously documented in detail, is not sur- perature (Fig. 1D; rho = 0.534,N = 10, P = prising. 0.112) or relative humidity (rho = -0.222, Although both species are diurnal, A. ex- N = 10, P = 0.537).In contrast, juvenile ac- sul and A. stratulus are most active at different times of the day. The physiological

"ii 12 C differences between the two species may E ~~12~ 9 ~ 9 partially explain this patter: members of the '0 '0 genus AnoZis are typically thermoconform- ~ 6 .8 6 ."0 Ameiva E 3 E 3 ers, whereas teiid genera such as ~ Z" are normally heliothermic (Hertz 1992; Rivera-Velez and Lewis 1994; Rogowitz "* 12 ,8 E [;31 '0,.. D 73~ 2003). However, this does not explain the .~ 9 i \ 70 ~ ". E '0 1Q30 I ,~67:::J intraspecific difference within A. siratulus. ~ 6 8. ~ ...., ,l ~ Juvenile A. stratulusare active during the E 3 529 ''''\/\. /t 64~ Z" •.... 'v' \,~!" 61 ~ coolest, most humid parts of the day, adult 7:45 10:45 13:45 16:45 7.45 10.45 13.45 16.45 A. straiulus are active throughout the day, Time of day and the much larger A. exsul are preferen-

FIG. 1. Numbers of adult (A) and juvenile (B)Anolis tially active during the hottest part of the straiulus and all Ameiva exsul (C) recorded between day. We believe that this difference is re- 0730-1700h on Guana Island, British Virgin Islands, lated to differences in body size and the Bars represent the combined totals from twelve tran- effects they have on water loss rates (Perry sects. Ambient conditions (relative humidity and tem- et al. submitted). Our results thus support perature) measured during a typical study day are Perry's et al. (submitted) conclusion that presented in D. Adult A. stratulus showed similar activity levels throughout the day, whereas juveniles smaller body size makes individual lizards were less active during the hot and dry midday. In more prone to water loss, and that this will contrast, A. exsul are primarily active during midday. affect their ecology, especially in water- NOTES 873

restricted locations such as Guana Island. Hertz, P. E. 1992. Temperature regulation in Puerto Unfortunately, the need to lump data for A. Rican Anolis lizards: a field test using null hypoth- exsul of different sizes may obscure size- eses. Ecology 73(4):1405-1417. Jackman, T. R, A. Larson, K. de Quiroz, and j. B. related differences in activity patterns Losos. 1999. Phylogenetic relationships and tempo within that species as well. We predict that of early diversification in Ana/is lizards. Syst. Bioi. a larger sample size would show that adult 48(2):254-285. A. exsul are active at warmer parts of the Kerr, A. M., R. Powell, and J. S. Parmerlee. 2005. day than are juveniles. In fact, Kerr et al. Ameiva erythrocepha/a (Teiidae) on Sint Eustatius, (2005)cite studies in which bimodal activ- Netherlands Antilles: Baseline data on a small population in a severely altered habitat. Caribb. J. ity patterns were seen in other Ameiva. Sci 41(1): 162-169. It has long been known that different Lazell, J. 1996. Guana Island. A Natural History Guide. species are active at different times of day, Jamestown, Rhode Island: The Conservation and that this has profound consequences Agency. for their ecology at both the autecological Lazell, J. 2005. Island: Fact and Theory in Nature. Berke- and community ecology level. The effects ley, California: University of California Press. of body size at multiple levels are also well 416 pp. Lewis, A R, and j. E. Saliva. 1987. Effects of sex and known (e.g., Perry and Garland 2002 for size on home range dominance, and activity bud- home range size). Our study demonstrates gets in Ameiva exsul (Lacertilia: Teiidae). Herpeto- that the two factors interact in ecologically logica 43(3):374-383. important ways. Further, it indicates that Milstead, W. W., ed. 1967. Lizard Ec%gJI A Sympo- further research needs to consider this in- sium. Columbia, Missouri: University of Missouri teraction when studying variability in ac- Press. Perry, G., and T. Garland, [r. 2002. Lizard home ranges tivity patterns. revisited: effects of sex, body size, diet, habitat, and phylogeny. Ec%gJ) 83(7):1870-1885. Acknowledgments.-We thank James La- Reagan, D. P. 1996. Anoline lizards. In The Food Web of zell, Kate LeVering, and the Guana Island n Tropical Rain-Forest, eds. D. P. Reagan and R B. staff for their assistance, and Henry and Wade, 321-345. Chicago, Illinois and London: Uni- Gloria Jarecki for access to Guana Island. versity of Chicago Press. Rivera-Velez, N., and A. R Lewis. 1994. Threshold This project was funded by The Conserva- temperatures and the thermal cycle of a heliother- tion Agency through a grant from the mic lizard. J. Herpeto/. 28(1):1-6. Falconwood Foundation, by financial assis- Rogowitz, G. L. 2003. Analysis of energy expenditure tance from the Department of Range, Wild- of Anolis lizards in relation to thermal and struc- life, and Fisheries Management, and by tural niches: phylogenetically independent com- Study Abroad Competitive Scholarships parisons. J. Herpelol. 37(1):82-91. Roughgarden, J. 1995. Ana/is lizards of the Caribbean: from the Office of International Affairs at Ecologi], Evolution, and Plate Tectonics. New York, Texas Tech University. This is manuscript New York: Oxford University Press. T-9-1040of the College of Agricultural Sci- Vitt, L. Land E. R. Pianka, eds. 1994. Lizard Ecology: , ences and Natural Resources, Texas Tech Historical and Experimental Perspectives. Princeton, University. New Jersey: Princeton University Press.

LITERATURE CITED

Butler, M. A, and J. B. Losos. 2002. Multivariate sexual dimorphism, sexual selection, and adaptation in Greater Antillean Anolis lizards. Eco!. Monog. 72(4): 541-559 Dial, R, J. Roughgarden, and S. C. Tobin. 1994. Notes on the absolute abundance of canopy anoles, Ano- lis cuuieri, A. siratulus, and A. eoermanni (Lacertilia: Polychridae) in the Luquillo forest, Puerto Rico. Caribb. J. Sci 30(3-4):278-279. Drni'el, R, G. Perry, and J. Lazell. 1997. Evaporative water loss in nine insular populations of the Anolis cristatellus group in the British Virgin Islands. Bio- tropica 29(1):111-116.