Evolution, 56(5), 2002, pp. 1045±1058 DWARFISM IN INSULAR SLOTHS: BIOGEOGRAPHY, SELECTION, AND EVOLUTIONARY RATE ROBERT P. ANDERSON1,2,3 AND CHARLES O. HANDLEY,JR.4,5 1Division of Vertebrate Zoology (Mammalogy), American Museum of Natural History, Central Park West at 79th Street, New York, New York 10024 2E-mail: [email protected] 3Natural History Museum and Department of Ecology & Evolutionary Biology, University of Kansas, Lawrence, Kansas 66045 4Division of Mammals, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560 Abstract. The islands of Bocas del Toro, Panama, were sequentially separated from the adjacent mainland by rising sea levels during the past 10,000 years. Three-toed sloths (Bradypus) from ®ve islands are smaller than their mainland counterparts, and the insular populations themselves vary in mean body size. We ®rst examine relationships between body size and physical characteristics of the islands, testing hypotheses regarding optimal body size, evolutionary equilibria, and the presence of dispersal in this system. To do so, we conduct linear regressions of body size onto island area, distance from the mainland, and island age. Second, we retroactively calculate two measures of the evolutionary rate of change in body size (haldanes and darwins) and the standardized linear selection differential, or selection intensity (i). We also test the observed morphological changes against models of evolution by genetic drift. The results indicate that mean body size decreases linearly with island age, explaining up to 97% of the variation among population means. Neither island area nor distance from the mainland is signi®cant in multiple regressions that include island age. Thus, we ®nd no evidence for differential optimal body size among islands, or for dispersal in the system. In contrast, the dependence of body size on island age suggests uniform directional selection for small body size in the insular populations. Although genetic drift cannot be discounted as the cause for this evolution in body size, the probability is small given the consistent direction of evolution (repeated dwar®sm). The insular sloths show a sustained rate of evolution similar to those measured in haldanes over tens of generations, appearing to unite micro- and macroevolutionary time scales. Furthermore, the magnitude and rate of this example of rapid differentiation fall within predictions of theoretical models from population genetics. However, the linearity of the relationship between body size and island age is not predicted, suggesting that either more factors are involved than those considered here, or that theoretical advances are necessary to explain constant evolutionary rates over long time spans in new selective environments. Key words. Bradypus, body size, darwins, haldanes, island, selection differential, selection intensity. Received November 6, 2000. Accepted January 14, 2002. Islands have long attracted biologists' attention as win- per 1998), follows a strikingly consistent pattern: Large spe- dows into evolution and community ecology (e.g., Darwin cies generally become dwarfed on islands, whereas small 1859; Wallace 1880; MacArthur and Wilson 1967; Brown mammals commonly evolve larger size. and Lomolino 1998). Usually relatively depauperate in spe- Reviews differ in which and how many mechanisms drive cies richness, insular faunas undergo a shuf¯ing of com- that change in body size and under what circumstances each munity composition as species are lost due to extinction or is relevant. Some (Lomolino 1985; Roth 1992; McNab 1994) gained through immigration and speciation (MacArthur and conclude that insular populations of small species become Wilson 1967; Simberloff 1974; Heaney 1986, 2000; Lomo- larger to use broader resource bases in the absence of former lino 1986, 2000). The community ecology of islands isolated competitors (competitive release of Lomolino 1985), and from continental areas by rising sea levelÐlandbridge is- large mammals are often resource-limited on islands, favor- landsÐis typically dominated by extinction (faunal relaxa- ing smaller size (resource limitation of Lomolino 1985). Re- tion), rather than by a more even equilibrium of extinction cently, Marquet and Taper (1998, p. 135) suggested that re- and colonization, as on oceanic islands (Lawlor 1986; Pat- source limitation may cause both dwar®sm in large mammals terson and Atmar 1986). and gigantism in small ones, due to changes in intraspeci®c Following isolation, many mammals surviving on land- competition that are related to home range size. In contrast, bridge islands undergo evolutionary changes in physiology, Adler and Levins (1994) put forward models invoking se- behavior, and morphological features, such as size, cranial lection for larger body size in small species in response to and dental characteristics, and coloration (e.g., Heaney 1978; higher intraspeci®c competition (see also Crowell 1983). Fi- Lomolino 1985; Kalko and Handley 1994). Models from pop- ulation genetics can predict morphological shifts in quanti- nally, Heaney (1978) proposed that island area determines tative characters (e.g., such as body size) under changing which factor drives selection: resource limitation on small environmental conditions such as mainland versus insular islands and competitive release on large islands. General environments (Lande 1980, 1986; Kirkpatrick 1982). Dif- models for the evolution of body size in insular terrestrial ferentiation in body size, one of the most important predictors vertebrates have been proposed and a few notable exceptions of life-history characteristics in mammals (Marquet and Ta- investigated (e.g., Foster 1964; Case 1978; Lawlor 1982; Case and Schwaner 1993; Petren and Case 1997). Thus, while the overall causes of body size evolution in insular situations 5 Charles O. Handley, Jr. passed away on June 9, 2000. remain an open area of research, most reviews to date suggest 1045 q 2002 The Society for the Study of Evolution. All rights reserved. 1046 R. P. ANDERSON AND C. O. HANDLEY, JR. that selection for smaller size in large mammals on islands swamps. Using a composite curve with years before present is caused by resource limitation and strong intraspeci®c com- and depth below current sea level as axes, Handley and Varn petition (Heaney 1978; Lomolino 1985; Roth 1992). dated each of the bathymetric contours and thus estimated Body size is but one of many morphological traits that island ages from the dates of disappearance of land bridges have been examined in studies assessing evolutionary rates between each island and the mainland (Table 1). Even if their and selection intensities (Lynch 1990; Gingerich 1993; Hen- absolute dates err in one direction or the other, relative dates dry and Kinnison 1999; Hoekstra et al. 2001; Kingsolver et of island formation will be correct to the extent that sea-¯oor al. 2001). In studies where relative ®tness, w, is known, the contours in this region have remained constant through the linear selection gradient, b, is de®ned as the slope of a re- Holocene. gression of relative ®tness on a continuously varying (quan- Isla Escudo was the ®rst of the islands to be separated from titative) trait or character, z (Arnold and Wade 1984). In the mainland (Fig. 1). It fragmented from the eastern shore contrast, a selection differential measures evolutionary rate of the PenõÂnsula Valiente about 8900 years ago and is not and can be calculated without data regarding ®tness (see Ma- directly related to any of the other islands. To the northwest terials and Methods). The islands of Bocas del Toro, Panama, of the PenõÂnsula Valiente and Isla Escudo, the islands of the provide a superb empirical opportunity to investigate differ- Laguna de ChriquõÂ are younger (Fig. 1A). They fragmented entiation and rates of morphological evolution in mammals, sequentially from the PenõÂnsula Tierra Oscura, which was because the ages of the islands have been estimated, the se- once a long, J-shaped peninsula jutting out from the south- quence of their formation is known, and the mammalian fauna western shore of the Laguna de ChiriquõÂ (Fig. 1B). That pen- is well sampled. insula was formed by the opening of the Boca del Drago Pass at the western end of the laguna. The outermost islands of The Islands of Bocas del Toro the Laguna de ChriquõÂ are about 5000 years old: Isla ColoÂn, which was the ®rst to split off of the PenõÂnsula Tierra Oscura The province of Bocas del Toro is located on the Caribbean (ca. 5200 years ago), and Isla Bastimentos, which separated coast of northwestern Panama adjacent to Costa Rica (Fig. from the peninsula along with what currently is Cayo Nancy 1). Just off shore lies a group of landbridge islands that (ca. 4700 years ago). Cayo Agua became isolated from the formed during the Holocene as a result of postglacial events, adjacent mainland (presently part of Isla Popa) about 3400 including rising sea level and continental submergence (Ol- years ago. In the past 1000 years, Cayo Nancy split from Isla son 1993; Kalko and Handley 1994; Anderson and Handley Bastimentos proper, and Isla Popa and Isla CristoÂbal each 2001). Rising sea levels isolated hilltops and ridges, ®rst as separated from the mainland. peninsulae, and then eventually as islands. These islands vary Biological interest in the islands of Bocas del Toro in area, distance from the mainland, depth of surrounding emerged only relatively recently (Handley 1959, 1993; Olson water, and ageÐall contributing