Like mother, like daughter: inheritance of nest-site

Like mother, like daughter: inheritance of nest-site location in snakes Gregory P. Brown and Richard Shine* School of Biological Sciences A0, University of Sydney, NSW 00, Australia *Author for correspondence (rics@bio.usyd.edu.au). 0 Manuscript for consideration in Biology Letters October 0, 00

SUMMARY 0 A trait can be passed from parents to offspring even if it has no genetic basis. For example, if daughters return to reproduce at the same sites where they were hatched themselves, nest-site location is consistent within matrilineages. Most cases of natal homing (nest-site philopatry) across generations have been inferred from molecular evidence rather than directly demonstrated, and involve species with low dispersal abilities. However, some animals disperse long distances but then return to their own place of birth to reproduce, based on cues imprinted early in their own development. Famously, adult salmon follow subtle chemical cues to relocate the same stream in which they were themselves hatched. Our field studies reveal a similar ability in tropical natricine snakes. Keywords: heritability; nesting; philopatry; reptile

0 0. INTRODUCTION Offspring frequently resemble their parents in biologically significant traits. The mechanisms generating such a resemblance may involve either genetic transmission (i.e. heritability in the sense of quantitative genetics) or environmental effects, or some interplay between the two. One interesting case involves culturally-inherited trait transmission. Many human societies possess complex "traditional knowledge" which is handed down from one generation to the next, either in written form (such as this journal) or by oral tradition. Other species similarly transmit fitness-relevant information from one generation to the next, and do so via a diverse set of mechanisms. In species with prolonged contact between parent and offspring, information (about foraging and social tactics, predator recognition, location of critical resources and so forth) can be passed from one generation to the next by observational learning. Through such a route, parents and offspring may resemble each other (and differ from conspecifics) in a wide range of behaviours and in the ways in which those behaviours are elicited by specific external cues. The mechanisms that generate parent-offspring resemblance in such cases presumably involve an interaction between genetic factors (e.g. reflecting selection on learning ability, and for adopting particular types of behavioural "rules") and non-genetic factors (e.g. the specific cues to which offspring are taught to respond in particular ways). Importantly, such "cultural" transmission of information can occur even in species without parent-offspring contact. Perhaps the most striking cases involve nest-site philopatry (natal homing), where offspring disperse during their juvenile life but return to their own site of birth as adults, when they are ready to reproduce.

Most cases of natal homing have been inferred from molecular evidence rather than directly demonstrated (Freedberg et al. 00) and frequently involve species with low dispersal abilities (so that return to the natal site may not require extensive migration). However, some aquatic animals (salmon, sea turtles) are known to disperse long distances but then return to their own place of birth to reproduce (Lohmann et al. ; Stewart et al. 00). Our field studies on tropical snakes reveal that nesting females return to the area where their own natal clutch had been deposited, suggesting that hatchling snakes imprint on specific sites that they encounter immediately posthatching. 0 0. MATERIAL AND METHODS Keelbacks (Tropidonophis mairii) are small (to 0. m) nonvenomous snakes from the same lineage as North American garter snakes and European grass snakes (Brown & Shine 00a). On the floodplain of the Adelaide River in tropical Australia, female keelbacks mature at about months of age, and produce clutches of eggs each year (Brown & Shine 00a). As in most snakes, there is no parental care of eggs or hatchlings: females leave the nest immediately after oviposition (Somma 00). Adult keelbacks are highly vagile: females that we radiotracked for two months had home ranges up to km in diameter, and moved up to 00 m in a single night (Brown & Shine 00, unpublished data). Immediately before laying their eggs, females typically migrate to higher ground to find well-drained nesting sites (Brown & Shine 00a). We hand-captured gravid females along a. km stretch of earthen dam wall that constitutes one of the only areas of high ground in our study site and hence, acts as a major focus for oviposition. Following capture, we kept female snakes in captivity until they

0 oviposited (mean =., SD =.) days later. The eggs were incubated in moist vermiculite in our laboratory, and the offspring individually marked before release at their mother's site of capture. We have recaptured many of these animals as reproductive adults about to lay their own eggs, enabling us to compare their oviposition sites with those selected by their mothers. Analyses in this paper are based only on females that were heavily gravid when encountered and hence, oviposited soon after capture (see above). Indeed, some animals were captured while they were in the process of laying their eggs. In nine cases where we released gravid females before they laid but then recaptured them on the following night while they were still gravid, most had moved <0 m and none had moved >0 m from the previous night's capture site. Thus, the places where we captured all of these heavily gravid females likely were close to their oviposition sites. To avoid pseudoreplication, our analyses are based only on the first recaptured reproductive daughter per family. 0. RESULTS AND DISCUSSION Like spawning salmon and nesting sea turtles (Lohmann et al. ; Stewart et al. 00), keelback snakes that are ready to lay their eggs return to the area where their mother reproduced. A linear regression of nest-site location (distance from the eastern end of the dam wall) for mothers versus daughters reveals a highly significant effect (n =, r = 0., p < 0.000; see figure ). Parent-offspring regressions of this type often are used in quantitative genetics analyses to estimate heritabilities of phenotypic traits (Lynch & Walsh ). Although in this case it is clear that the mechanism generating the correlation is not primarily genetic, the same is likely to be true to a greater or lesser extent for most field-based estimates of heritability and thus, the method gives us a basis for comparison with published data on heritabilities estimated in this way.

0 0 Applying this parent-offspring regression method to the data in figure generates a heritability estimate (twice the calculated regression slope) for nest-site location of 0. (n =, SE = 0.), higher than for almost any other reproductive trait in this snake population (Brown & Shine 00) and higher than most heritabilities estimated for lifehistory traits in field populations of other animal species (Mousseau & Roff ; Réale & Festa-Bianchet 00). The strong trend for females to oviposit close to their own natal sites cannot be dismissed as a result of low vagility; radio-tracked snakes moved long distances (above) and females that we released as hatchlings (see below) were found an average of 0 m from their release site when recaptured as juveniles, vs. m when recaptured as they about to oviposit (i.e., they returned to the natal nest-site, rather than remaining close to it). The potential selective advantage of natal homing is clear (i.e. this nesting area produced successful hatchlings previously, so is likely to do so again) but what mechanism drives this remarkable homing behaviour? The eggs in our study were laid and incubated in the laboratory, so any cues must come either from intrinsic factors (genetics, or maternally-deposited cues within the egg) or from the hatchlings' experience of their post-release environment. Six cases where we released hatchlings at the wrong nest-site provide an opportunity to test between these ideas. These animals later returned to nest close to the sites where we released them (n =, r = 0., p < 0.0) rather than to the sites where their mothers had been collected (n =, r = 0.0, p = 0.0). Thus, natal homing in keelbacks reflects spatial learning by hatchlings rather than an intrinsic attribute of the eggs that incorporates specific encoding of spatial information.

0 0 The most plausible interpretation of our results is that hatchling keelback snakes imprint on the first field area that they experience post-hatching, and return to that area as adults when they are ready to reproduce. The specific location of nests is affected by local weather conditions and annually variable flooding levels (Brown & Shine 00a) but female keelbacks prefer to oviposit in sites containing old eggshells (Brown & Shine 00b). This behaviour suggests that female keelbacks often may deposit their own eggs close to the shell of the egg that they hatched from themselves at least a year earlier. Extreme fidelity to natal nesting locations has strong implications. First, this behaviour confounds simple interpretations about inheritance: even in the absence of any contact between mothers and their daughters, offspring may resemble their parents because they encounter similar environments during early life, as well as because of genetic factors. In keelbacks as in many other reptiles, the thermal and hydric conditions inside a nest can substantially modify the phenotypic traits of offspring that emerge from that nest (Deeming 00; Brown & Shine 00a). Thus, any matrilineal consistency in nest-site location is likely to result in similarity in nest conditions and hence, in hatchling phenotypes also. As a result, we might expect morphological traits of offspring to resemble those of their mothers more than would be expected from genetic factors alone. Second, nest-site philopatry may affect the evolution of traits such as environmental sex determination, because of sex-specific fitness effects (Reinhold ). That is, mothers that select unusually good nest-sites (i.e. that produce highly viable offspring via incubation-mediated effects) may enhance their fitness by overproducing daughters (the sex that will benefit most from "inheriting" the maternal nest-site). In contrast, females that select unusually poor nest-sites would

0 benefit by overproducing sons (the sex that would pay a lower cost in this respect, because they do not return to re-use the same nesting area: Reinhold ). Third, the ability of ectothermic vertebrates to retain specific early memories poses questions about sensory and neural mechanisms in these putatively "simple" creatures. Fourth, disturbance to traditional nest-sites - especially in taxa with communal oviposition - may have severe ecological consequences. But fifth, the modified nesting preferenda of our inadvertently relocated hatchlings suggests that wildlife managers may be able to exploit this imprinting to facilitate translocation programs for threatened species. In summary, our longterm field study of tropical snakes reveals an unexpectedly high level of similarity between mother and daughter for nest-site location. The mechanisms generating that resemblance remain unknown, but presumably reflect a combination of selection on genetic traits (such as the ability to remember and relocate sites that are encountered very early in life) and phenotypically plastic (learned) responses to specific types of spatially-variable environmental cues. The parallel to "traditional knowledge" in other species, including our own, suggests that social systems and reproductive behaviour in snakes are far more complex than has generally been assumed (Greene et al. 00). 0 We thank Cathy Shilton, Ben Phillips and the other members of TeamBufo for advice and encouragement, Melanie Elphick for help with manuscript preparation, the managers and staff of Beatrice Hill Farm for logistical assistance, the University of Sydney Animal Ethics Committee for approval, and the Australian Research Council for funding.

0 0 Brown, G. P. & Shine, R. 00a Female phenotype, life-history, and reproductive success in free-ranging snakes (Tropidonophis mairii). Ecology, 0. Brown, G. P. & Shine, R. 00b Nesting snakes (Tropidonophis mairii, Colubridae) selectively oviposit in sites that provide evidence of previous successful hatching. Can. J. Zool.,. Brown, G. P. & Shine, R. 00 Repeatability and heritability of reproductive traits in free-ranging snakes. J. Evol. Biol. in press. Deeming, D. C. 00 Post-hatching phenotypic effects of incubation on reptiles. In Reptilian incubation. Environment, evolution and behaviour (ed. D. C. Deeming), pp.. Nottingham, UK: Nottingham University Press. Freedberg, S., Ewert, M. A., Ridenhour, B. J., Neiman, M. & Nelson, C. E. 00 Nesting fidelity and molecular evidence for natal homing in the freshwater turtle, Graptemys kohnii. Proc. R. Soc. Lond. Ser. B, 0. Greene, H. W., May, P. G., Hardy, D. L. S., Sciturro, J. M. & Farrell, T. M. 00 Parental behavior by vipers. In Biology of the vipers (ed. G. W. Schuett, M. Hoggren, M. E. Douglas & H. W. Greene), pp. 0. Eagle Mountain, Utah: Eagle Mountain Publishing. Lohmann, K. J., Witherington, B. E., Lohmann, C. M. F. & Salmon, M. Orientation, navigation, and natal beach homing in sea turtles. In The biology of sea turtles (ed. P. Lutz & J. Musick), pp. 0. Boca Raton: CRC Press. Lynch, M. & Walsh, B. Genetics and analysis of quantitative traits. Massachusetts: Sinauer Associates. Mousseau, T. A. & Roff, D. A. Natural selection and the heritability of fitness components. Heredity,.

0 Réale, D. & Festa-Bianchet, M. 000 Quantitative genetics of life-history traits in a long-lived wild mammal. Heredity, 0. Reinhold, K. Nest-site philopatry and selection for environmental sex determination. Evol. Ecol., 0. Somma, L. A. 00 Parental behavior in Lepidosaurian and Testudinian reptiles. Florida: Krieger Publishing. Stewart, I. J., Quinn, T. P. & Bentzen, T. P. 00 Evidence for fine-scale natal homing among island beach spawning sockeye salmon, Oncorhynchus nerka. Environ. Biol. Fishes,. 0

Caption to Figure 0 Figure. Mother-to-daughter inheritance of nest-site locations in keelback snakes. Locations of nests are expressed in terms of distances along the earthen wall of Fogg Dam (beginning from the eastern end, so 00 = 00 m west of the eastern end of the dam wall) on the Adelaide River floodplain. When they reach adulthood, female keelbacks return to lay their eggs at sites close to where their mother was collected as she was about to nest (and hence, where they themselves were released as hatchlings). The red dots show data for clutches that we inadvertently released at a site distant from the mother's capture location (see text for details). Short title: Nest-site inheritance in snakes