An Isotopic Technique to Marie:Mid-Sized Vertebrates Non-Lnvaslvelv

Journal of Zoology. Print ISSN 0952-8369

An isotopic technique to marie:mid-sized vertebrates non-lnvaslvelv

1,2 ,2 1,2 3 J. N. Pauli , M. Ben-David1 , S. W. Buskirk , J. E. DePue2 & W. P. Smith

1 Prograrn in Ecology, University of Wyorning, Laramie, WY, USA 2 Department of Zoology and Physiology, University of Wyoming, Laramie, WY, USA 3 USDA Forest Service, Pacific Northwest Research Station, Forestry and Range Sciences Laboratory, La Grande, OR, USA

Keywords Abstract carbon; deuterium; nitrogen; mesocarnivore; movement; ranging. Although dispersal is an important attribute of animal population ecology, knowledge of dispersal rates or distances for many wide-ranging species is lacking. Correspondence Current methods require capturing and restraining animals, which can be cost- Jonathan N. Pauli, Department of Zoology prohibitive, fail to collect sufficient samples or change animal ranging behaviors. and Physiology, University of Wyoming, Herein, we describe a novel, cost-effective and non-invasive method, using bait 1000 E. University Avenue, Laramie, enriched with stable isotopes to mark the hair of American martens Martes WY82071, USA. americana. Captive martens that consumed isotopically labeled glycine exhibited Email: [email protected] significant and progressive enrichment in the isotopic signature of 13C,lsN and 2H in both whole blood and hair. A distinct mark in hair, >2 standard deviations Editor: Nigel Bennett above natural abundance, occurred within 14 days of the second dose. The rate of isotopic labeling of hair was higher in spring, possibly because labeled amino acids Received 7 November 2008; revised 31 became diluted among the many hairs growing during the autumn. Because hair January 2009; accepted 3 February 2009 and feathers can be collected non-invasively from large geographic areas without capturing animals, this labeling method can be used to mark and study the doi:10.1111/j.1469-7998.2009.00562.x movement and dispersal rates of animals across landscapes efficiently.

Introduction bats Lasiurus cinereus can disperse > 2000: km. Similarly, using isotopic signatures of feathers, Hobson & Wassenaar Dispersal is a fundamental aspect of animal life history and (1997) determined the breeding origin of wintering Neotro- ecology; knowledge of it is necessary to develop effective pical migrants. These studies have largely relied on natural conservation strategies (Macdonald & Johnson, 2001). abundances of deuterium 2H) across latitudinal gradients, However, quantifying dispersal rates and distances for which does not exhibit sufficient variation to determine the wide-ranging species is challenging. Traditionally, methods dispersal of animals moving short or intermediate distances such as mark-recapture, radio-telemetry and, more recently, « 100 km; Wunder & Norris, 2008). Previous researchers DNA genotyping have been used to assess movements over have, however, used organic and inorganic compounds, long distances (Nathan et al., 2003). Although generally artificially enriched with stable isotopes, to track the fates accurate, these approaches require large investments of of substances in various ecosystem (Hall & Tank, 2003) and time, money and personnel, limiting the number of study physiological (Boutton, 1991; Hirons, Schell & St. Aubin, animals and duration that each can be monitored. Such 2001) processes. Animal ecologists recently have applied constraints reduce the likelihood of detecting rare, but these enriched isotopes as organismal markers: by dripping important dispersal events (Nathan et al., 2003). These enriched nitrogen 15N) into streams and ponds, Caudill methods also require capture, restraint and often anestheti- (2003) and Macneale, Peckarsky & Likens (2004) marked zation of animals, which can be stressful (Harper & Austad, and subsequently recaptured developing invertebrate larvae 2000) and alter subsequent behaviors (e.g. Clinchy, Krebs & to estimate dispersal distances; similarly, Wanner et at. Jarman, 2001). Therefore, for wide-ranging vertebrates, (2006) used 44Ca to mark and track the dispersal and particularly endangered ones, reliable estimates of dispersal foraging activity of the parasitoid wasp Cotesia glomerata. are lacking, impeding our ability to parameterize popula- Vertebrate ecologists are increasingly using tissues col- tion-based models for conservation (Macdonald & Johnson, lected non-invasively and DNA-based analyses to estimate 2001). dispersal and gene flow (DeYoung & Honeycutt, 2005). Naturally occurring stable isotopes have been used suc- However, this approach can be limited by degradation of cessfully to document long-distance movements of volant samples and related genotyping errors (Mills, 2007), which animals (Rubenstein & Hobson, 2004). For example, Cryan can increase bias of population estimates (Lukacs & Burn- et al. (2004) used stable isotopes in hair to show that hoary ham, 2005). Such analytical errors could be avoided with the

use of an independent organismal marker. We reasoned Biology Laboratory at the University of Wyoming. Here, that isotopically enriched compounds could also be used captive martens were exposed to ambient changes in photo- to mark free-ranging vertebrates. With artificially enriched period and temperature, minimizing interference with their baits and non-invasive sampling methods, researchers could moult cycle and fur growth. They were fed ad libitum mark animals and subsequently collect hair, feather or other (ExclusiveT cat food, PMI Nutrition's Henderson, CO, tissues to track the movements of many individuals. For USA) and had continuous access to water. Pens were mammals and birds, in which most amino acids incorpo- furnished with PVC tubes, nest boxes, branches and small rated into hair and feathers are derived from recently trees as environmental enrichment. Martens were accli- consumed food (Ayliffe et al., 2004), isotopic enrichment of mated to captivity and the cat food diet for 5 weeks. In the ingested amino acids could rapidly mark keratinous tissues. second week of May (coinciding with the moult; Soutiere & No previous research has tested whether this approach can Steventon, 1981), we administered a unique combination l3 15 2 produce an unambiguous mark in tissues that can later be of enriched (98-99 at. %) isotopes ( C, N and H; ISO- collected non-invasively. TEC™, Miamisburg, OH, USA) to the food of each animal l3 2 15 As part of a project assessing the effects of fragmentation (n = 2 for C and H, and n = 3 for N); animals that did l3 15 2 on the forest-associated mesocarnivore, the American mar- not receive an enriched diet for C, N or H were used as 13 2 ten Martes americana, we required a cost-effective and non- controls for that isotope (n = 3 for Cand H, and n = 2 for 15 invasive method to investigate movement of individuals N). For example, an animal that received a diet enriched in 2 13 15 between habitat patches. By providing bait enriched with H served as a control for C and N. Isotopic labels were different isotopic labels to martens inhabiting isolated forest provided in the form of glycine (Boutton, 1991), an amino fragments and subsequently collecting hair samples non- acid that is incorporated into both blood and hair (Hirons invasively from a large area over multiple sampling periods, et al., 2001). We estimated the isotopic dose required to we intended to estimate dispersal rates and distances. Before produce an isotopic label in the hair and blood of martens, using isotopically labeled baits in the field, we conducted assuming no elemental routing to different tissue types (i.e, feeding trials on captive martens to ascertain that consump- complete mixing; Table 1). Specifically, we calculated the ΕN tion of an isotopically enriched diet would produce a reliable mass of an isotope ( Ag)needed to elicit a significant mark in the hair. Our specific objectives were to: (1) measure increase in the signature of tissue with the equation: the dose-specific assimilation of isotopically labeled amino acids into marten blood and hair; (2) determine minimum ENA = [ENAxNA x A ]- [A xNA] (1) dosages needed to label wild martens; (3) identify the season g g g in which isotopically enriched bait would provide the most reliable mark. where eN A is the isotopic enrichment desired (the proportion of labeled isotope above natural abundance). For example, our desired enrichment was 1.05 times natural abundance Materials and methods for 15N, 1.04 times for l3C and, because roughly 20% of During the last week of March 2006, we captured five hydrogen within hair exchanges with the atmosphere 2 N martens along snowmobile routes and logging roads in the (Cryan et al., 2004), 1.50 times for H. A is the natural Snowy Range Mountains of south-eastern Wyoming abundance of the heavy isotope, and Ag is the mass of the (4104'N 106°9'W). We anesthetized captured martens with element in the species (Table 1). To determine the mass ketamine hydrochloride (l5.0mgkg-1 body weight) and of enriched glycine to feed martens, we multiplied the xylazine hydrochloride (1.6 mg kg-1 body weight; Ben- grams of enriched element to achieve the necessary mark

David, Schell & Flynn, 1997) and transported them to by the molecular mass (Mr) and divided by the atomic individual outdoor pens at Red Buttes Environmental mass (ma) of that isotope:

Marten body mass was estimated as 1000g, or 300g dry mass (Buskirk & Harlow, 1989). Estimates of elemental and isotopic abundances are found in Rutherford & Hawk (1907) and Karasov & Martinez del Rio (2007). "Values that are species-specific for American marten.

proportional growth of tail hair through time (Rose et al., 1984; Zullinger et al., 1984). Because the onset of hair Finally, to calculate the dose of glycine that should be replacement is delayed on the tail relative to other parts of delivered to the animal, we multiplied this value by the the body (Bassett & Llewellyn, 1949),we excluded the two fraction of animal dry mass (300g) over its wet mass (1000g). measurements before 7 September from the curve fitting We anesthetized each marten, plucked 10-20newlygrowing because they reflected the length of hair grown in the hairs from the neck and rump, and collected 0.5mL of blood previous year (Supporting information Fig. S1). We used from the jugular vein fivetimes during spring (10May-5 July) estimates of hair growth rates derived from the Gompertz and six times during autumn (10August-19 October). After curve to provide a temporal scale to the segments of cross- the martens recovered from anesthesia they were fed the sectioned guard hairs. labeled food. Therefore, each marten received a dose of labeled glycine every 2 weeks, except between 5July and Results 10August. In autumn, to monitor hair growth, we also measured the length of the longest hair (cm) on the back, During both periods (spring and autumn) whole blood hind quarters, and middle of the tail on each anesthetized of martens fed isotopically enriched diets exhibited pro- marten. During the final sample collection (19October), we gressive isotopic enrichment, whereas the blood of control l3 plucked 40-75 dorsal guard hairs from the tip of the tail of animals did not (Fig. 1; RCM: C-F1,41 = 20.6, P 1.5 times higher during analysis. All samples were stored frozen until prepared for spring than during autumn (Table 2). Following a 4-week mass spectrometry. lapse in isotopic dosing and sample collection (between Samples of whole blood were thawed at room tempera- 5July and 10August), enrichment of blood with 2H 15 l3 ture, dried for 72h at 60oC and homogenized in a ball mill returned to pre-dose levels; N and C remained only (Mixer Mill MM200, Retsch Inc., Newtown, PA, USA; slightly elevated (Fig. 1). MacAvoy, Macko & Arneson, 2005). Hair samples were Isotopic enrichment in plucked hair was similar to whole rinsed three times with 2:1 chloroform:ethanol solution to blood; martens fed enriched glycine exhibited elevated iso- remove surface oils (Cryan et al., 2004), dried for 72h at topic signatures compared with control animals, which 60oCand homogenized with surgical scissors. Samples were demonstrated variable and non-substantive isotopic l3 15 2 l3 weighed, placed in tin ( C and N)or silver H) capsules changes (Fig. 1; Table 2; RCM: C- F1 1 = 11.0, P = 15 2 ,4 and submitted to the Stable Isotope Facility at the Univer- 0.002, N- F1,41 = 8.1, P = 0.007 and H- F1,41 = 40.5, l3 15 sity of Wyoming. Analysis of C and N levels was P<0.001). In contrast to blood, plucked hair exhibited conducted witha Costech 4010 elemental analyzer (Costech delayed incorporation of isotopes (Fig. 1). Hair samples Analytical Technologies, Valencia, CA, USA) and 2H with a were not significantly enriched relative to controls following Finnigan TC/EA (High Temperature Conversion Elemental the first dose; an enrichment of 2 SD above control values Analyzer) attached to a Thermo Finnigan DeltaPLUS Xl' was observed only after the second dose (58% for l3C, 75% 15 2 Continuous Flow Isotope Ratio Mass Spectrometer (Ther- for N and 82% for H). The isotopic signature of hair did mo Fisher Scientific, Inc., Waltham, MA, USA). Results are not, however, change substantively following the third dose provided as per mil (parts per thousand [%o]) ratios relative (Fig. 1). Body hair demonstrated the greatest rate of enrich- to the international standards of Peedee Belemnite (PDB; ment during the spring; slope coefficients were higher by 4.6- 3 15 l3 15 2 δ1 C), atmospheric nitrogen (AIR; δ N)and Standard fold for C, 2.5-fold for N and 3.9-fold for H (Table 2). 2 Mean Ocean Water (SMOW; δ H)with calibrated internal In contrast to blood, hair remained significantly enriched in laboratory standards. Because all samples were collected, both l3C and 15Nduring autumn from isotopic dosing that processed and analyzed in the same geographic location we occurred the previous spring. The enrichment oflabeled hair 2 did not correct the raw H data for atmospheric exchange with deuterium declined by c. 20%o between spring and (Wassenaar & Hobson, 2000). We considered hair and autumn (Fig. 1), presumably because some keratin-bound blood samples to be siguificantly enriched when values were hydrogen exchanged with the atmosphere. elevated > 2 SD above those of control animals. Based on a Gompertz function of hair-growth (r2 = To account for the lack of independence and to improve 0.606, F2,18 = 12.3; P<0.001; Fig. 2), the oldest segment at the precision of parameter estimates, we analyzed changes in the tip of the hair shaft, plucked from the tail, was grown the isotopic signature of blood and plucked body hair with during the period 29June-27 July (28 days), which is within random coefficient models (RCM). We treated time, season the range of initiation dates reported for closely related taxa (spring or autumn) and treatment status (receiving isotopi- (Rose et al., 1984; Maurel, Coutant & Boissin, 1987). The cally labeled or unlabeled diet) as fixed effects, and indivi- remaining segments were grown during 27July-9 August dual identity as a random effect (Littell et al., 1996). To (13days), 9-21 August (12days), 21August-3 September model the growth of hair, we fitted a Gompertz curve to the (13days), 3-21 September (18 days) and 21 September to

Figure 1 Mean isotopic values (± 1 SD ) of whole blood (circles; a-c) and plucked hair samples (squares; d-f) from captive martens, Laramie, 13 15 Wyoming, 2006. Treatment animals (solid) were fed glycine isotopically labeled with C, deuterium (2H) or N.Control animals (hollow) received an isotopically normal diet.

Table 2 Parameter estimates (slope and y-intercept; ± 1 SE) from a random coefficient model for isotopic enrichment of blood and hair of captive American martens Martes Americana in spring (May-June) and autumn (August-October), 2007

13 15 2 Labeled martens were fed a dose of the amino acid glycine enriched with C, N or H; control animals did not receive an enriched diet for that isotope.

Figure 2 Gompertz function describing the growth rate of hair (following Rose et al., 1984; Maurel et al., 1987), based on measurements of guard hairs from the tails of five captive martens. This function was used to provide a chronology for growth of tail hair that was cross- sectioned and analyzed for isotopic signature (Fig. 3). Diamonds along the y-axis delineate the six segments of cross-sectioned tail hairs and the dotted lines provide location on the Gompertz function to estimate date of segment growth. Open symbols denote different individual animals. One observation (7 September, 4.7 cm) was excluded from modeling because it was an outlier. the time of collection of hair in October (29 days). Enrich- ment of cross-sectioned guard hairs from marten tails was highest in the middle segments of the hair shaft for all isotopes. The greatest enrichment was detected in the third segment for l3C and in the fourth segment for 15N and 2H (Fig. 3). Thus, the highest enrichment in tail hair was obtained during August-early September (Fig. 3), which Figure 3 Mean isotopic values (± 1 SD) of dorsal guard hairs plucked from captive martens and cross-sectioned into six equal lengths, coincided with a short growth period (12 and 13 days) Laramie, Wyoming, 2006. Values shown are for martens fed glycine immediately following a dosing (10 and 24 August). The 13 2 isotopically labeled with C, 15N or deuterium ( H). Using a Gompertz isotopic signature of the tip segments of hair from the tail function of tail growth (Fig. 2) we calculated the dates that each was similar to those of body hair at the end of the spring segment represented (1= 13July-2 August; 2 =2-13 August; 3 = trials (Figs 1 and 3), likely reflecting incorporation of the last 13-24 August; 4 =24 August-4 September; 5= 4-20 September; and dose administered on 5 July. In contrast, the isotopic values 6 = 20 September to the end of hair growth in October. of the two segments near the base were lower than those of body hair at the end of the autumn sampling, likely because during May-June, during a phase of predominantly hair the enriched glycine pool was diluted by the newly growing loss, reflects the incorporation of the label via maintenance underfur of the body. replacement of hair. During the peak of hair growth in autumn, when over half of the amino acids devoted to hair Discussion development are derived from recently consumed pools (Ayliffe et al., 2004), the quality of the mark should have Captive martens fed glycine enriched with l3C, 15N and 2H been the strongest. Potentially, the labeled amino acids were exhibited substantive enrichment in their blood and hair diluted by the many growing hairs in autumn and thus compared with control animals. Although consumption of isotopic labeling during peak hair growth was not optimal. labeled amino acids resulted in elevation in isotopic values Our results indicate that for mammals with moult patterns in body hair and blood during both seasons, we found that similar to martens, the best period to offer isotopically the rate of enrichment was higher in May-June for both enriched bait is summer (June-July). For other species, such tissues. In North America, martens begin moulting in late as birds with different moult chronologies (Holmgren & April and grow their winter coat during July-October Hedenstrom, 1995; Barta et al., 2006), the incorporation of (Markley & Bassett, 1942; Soutiere & Steventon, 1981; Rose isotopically labeled amino acids into keratinous tissue may et al., 1984). It is likely that the high enrichment we observed follow a different pattern.

Seasonal differences in slope coefficients for isotopic The safety of using enriched stable isotopes to study enrichment, particularly 2H in whole blood, are difficult to physiological processes in animals has been demonstrated explain. Because animals were acclimated to captivity for (Koletzko, Sauerwald & Demmelmair, 1997). Unlike pre- 5 weeks before feeding trials, diet was constant during the viously developed markers, which provide only a temporary study (although we did not quantify the isotopic signature mark (e.g. rhodamine B, sulfadimethoxine) or require the of each batch of commercial feed), and none of the martens sacrifice of the animals for bone or tooth examination (e.g. exhibited substantial changes in body size (i.e. length or tetracycline; Southey, Sleeman & Gormley, 2002), isotopic mass), it is unlikely that differential isotopic incorporation marks persist within the keratinous tissues (e.g. feathers, in whole blood was due to incomplete isotopic mixing, hair, scales) until they are moulted, and can be collected variation in diet quality or animal growth (Martinez del non-invasively, Non-invasive collection techniques for hair Rio & Wolf, 2005). Potentially, seasonal differences in or feather (e.g. those shed or collected in hair traps and incorporation rates of the isotopic label into blood were snares) have been developed and successfully employed due to variation in isotopic values in either the drinking (Mills, 2007) to sample wide-ranging, elusive and rare water or food (Hobson, Atwell & Wassenaar, 1999). Indeed, species, where live-capture can be impractical. Such collec- differences in incorporation rates were most apparent for tion techniques, primarily developed for DNA-based ap- 2H, and control animals exhibited a change in blood values proaches, are cost-effective, and are well-suited for the l3 during autumn. However, isotopic values of C and 15N did method we describe. Isotopic labeling and analysis are not change as much in control animals, suggesting that further advantageous because isotopic composition of kera- seasonal differences were not the result of the isotopic tin, unlike DNA (Roon, Waits & Kendall, 2003), is resilient composition of the diet alone. That the incorporation slopes to environmental degradation (Macho et al., 1999). Finally, for all isotopes were similar within seasons suggests another our cross-sectional analyses of hair reveal that the timing of potential mechanism; the patterns we observed were caused bait consumption by the animal can be determined based on by seasonal changes in tissue turnover rates. Seasonal chan- the temporal sequence of hair growth. Past studies have ges in turnover rates have been reported for whole blood similarly shown that cross-sectional analyses of hairs can production in humans (Gunga et al., 2007) and muscle tissue provide a chronology for changes in animal diet and of hibernating bears (Lohuis, Harlow & Beck, 2007). For dispersal (Cerling et al., 2006). example, Lohuis et al. (2007) observed that turnover rate of Our experiment demonstrated that isotopically enriched muscle in black bears Ursus americanus declined between amino acids can be used to non-invasively and cost-effectively autumn and winter, which was accompanied by a decline mark mid-sized vertebrates. We are now applying this in the rate of change of 15N. Whether seasonal changes in method to a study with free-ranging animals. Researchers tissue turnover rates are common in non-hibernating verte- pursuing this technique for other vertebrate species will need brates is unclear and merits further study. to calculate the amount of dose required for desired enrich- Our results suggest that isotopically labeled glycine can ment levels and develop species-specific administration be a cost-effective method to label small- to mid-sized methods. Once such methods are established, isotopic label- vertebrates for studies of movement and dispersal. Indeed, ing could facilitate marking large numbers of free-ranging a single dose of enriched bait for one marten currently costs vertebrates and tracking their movements at broad spatial l3 approximately US$0,45 (2 H), $0.90 (15N) and $6.60 ( C), scales, including the effects of fragmentation and meta- and the expenses of isotopic analyses with mass spectrometry population dynamics (McCullough, 1996). are declining. It is likely that in the near future, additional amino acids will become available for administering isotopic labels. Notably, cysteine, a sulphur-containing amino acid Ac nowledgements and major component of keratinous tissues (Richards et al., k 2003), is routed to hair growth (O'Connell & Hedges, 2001; We thank R.D. Mares, J.P. Whiteman, E.A. Flaherty and Thomas et al., 2007) and, therefore, could provide another K.A. Greller for assisting with fieldwork. R.O. Hall assisted effective vehicle for isotopic enrichment. Currently, how- in dose calculations and C. Martinez del Rio provided ever, using this amino acid is cost-prohibitive. Still, the valuable discussions on physiology. S. DeVries and J. availability of multiple isotopes leads to the possibility of Bobbitt helped with captive martens, and R. Anderson- isotope combinations that can allow investigators to gener- Sprecher provided statistical advice. Financial support was ate site-specific baits and isotopic marks .. The number of provided by the Pacific Northwest Research Station (US unique permutations of isotopic marks will be 2n-1, where n Forest Service), and scholarships through the Program in is the number of isotopes of two forms used in the marking Ecology and the College of Arts and Sciences at the scheme. For example, if four elements were represented, University of Wyoming. Animal handling and housing each in two isotopic forms, 15 isotopic combinations, each permits were obtained through the Wyoming Game and including at least one form rare in nature, could be gener- Fish Department; all methods were approved by the Institu- ated. Such an approach could provide an effective means for tional Animal Care and Use Committee at the University of labeling multiple subpopulations and monitoring the move- Wyoming and adhered to the ethical guidelines for the use ment and dispersal of individual animals across different of mammals in research set forth by the American Society of habitats. Mammalogists.

References Hirons, A.C., Schell, D.M. & St Aubin, D.J. (2001). Growth rates of vibrissae of harbor seals (Phoca vitulina) and Steller Ayliffe, LK., Cerling, TE., Robinson, T., West, AG., sea lions (Eumetopiasjubatus). Can. J. Zool. 79, 1053-1061. Sponheimer, M., Passey, B.H., Hammer, J., Roeder, B., Hobson, K.A, Atwell, L & Wassenaar, L.1. (1999). Influence Dearing, M.D. & Ehleringer, J.R. (2004). Turnover of of drinking water and diet on the stable-hydrogen isotope

carbon isotopes in tail hair and breath CO2 of horses fed an ratios of animal tissue. Proc, Nat!. Acad. Sci. USA 96, isotopically varied diet. Oecologia 139, 11-22. 8003-8006. Barta, Z., Houston, A.I., McNamara, J.M., Welham, R.K., Hobson, K.A. & Wassenaar, L1. (1997). Linking breeding Hedenstrom, A., Weber, T.P. & Fer6, O. (2006). Annual and wintering grounds ofneotropicalmigrant songbirds routines of non-migratory birds: optimal moult strategies. using stable hydrogen isotopic analysis of feathers. Oikos 112, 580-593. Oecologia 109,142-148. Bassett, C.F. & Llewellyn, LM. (1949). The molting and fur Holmgren, N. & Hedcnstrom, A. (1995). The scheduling of growth pattern in the adult mink. Am. Midl. Nat. 42, molt in migratory birds. Evol. Ecol. 9, 354--368. 751-756. Karasov, W.H. & Martinez del Rio, C. (2007). Physiological Ben-David, M., Schell, D.M. & Flynn, R.W. (1997). Annual ecology. Princeton: Princeton University Press. and seasonal changes in diets of martens: evidence from Ko1etzko, B., Sauerwa1d, T. & Demmelmair, H. (1997). Safety stable isotope analysis. Oecologia 111, 280--291. of stable isotope use. Eur. J. Pediatr. 156 (Supp!. 1), Boutton, T.W. (1991). Tracer studies with 13C-enriched S12-S17. substrates: humans and large animals. In Carbon isotope Littell, R.C., Milliken, G.A., Stroup, W.W. & Wolfinger, techniques: 219-242. Coleman, D.C. (Ed.). San Diego: R.D. (1996). SAS system for mixed models. Cary: SAS Academic Press Inc. Institute Inc. Buskirk, S.W. & Harlow, H.J. (1989). Body fat dynamics of Lohuis, T.D., Harlow, H.J. & Beck, TD.I. (2007). the American marten (Martes americana) in winter. Hibernating black bears (Ursus americanus) experience J. Mammal. 70, 191-193. skeletal muscle protein balance during winter anorexia. Caudill, c.c. (2003). Measuring dispersal in a metapopula- Compo Biochem. Physiol. B: Biocliem. Mol. Biol. 147, tion using stable isotope enrichment: high rates of 20--28. sex-biased dispersal between patches in a mayfly metapo- Lukacs, P.M. & Burnham, K.P. (2005). Estimating popula- pulation. Oikos 101, 624-630. tion size from DNA-based closed capture-recapture data Cerling, T.E., Wittemyer, G., Rasmussen, H.B., Vollrath, F., incorporating genotyping error. J. Wildl. Mgmt. 69, Cerling, C.E., Robinson, T.J. & Douglas-Hamilton, 1. 396-403. (2006). Stable isotopes in elephant hair document migra- MacAvoy, S.E., Macko, SA. & Arneson, LS. (2005). tion patterns and diet changes. Proc. Natl. Acad. Sci. USA Growth versus metabolic tissue replacement in mouse 103,371-373. tissues determined by stable carbon and nitrogen isotope Clinchy, M., Krebs, C.J, & Jarman, P.J. (2001). Dispersal analysis. Can. J. Zool. 83, 631-641. sinks and handling effects: interpreting the role of immi- Macdonald, D.W. & Johnson, D.D.P. (2001). Dispersal in gration in common brushtail possum populations. J. Anim. theory and practice: consequences for conservation Ecol. 70, 515-526. biology. In Dispersal: 358-372. Clobert, J., Danchin, E., Cryan, P.M., Bogan, M.A, Rye, R.O., Landis, G.P. & Dhondt, AA. & Nichols, J.D. (Eds). Oxford: Oxford Kestler, C.L (2004). Stable hydrogen isotope analysis of University Press. bat hair as evidence for seasonal molt and long-distance Macho, S.A., Engel, M.H., Andrusevich, V., Lubec, G., migration. J. Mammal. 85, 995-1001. O'Connell, T.C. & Hedges, R.E.M. (1999). Documenting DeYoung, R.W. & Honeycutt, R.L (2005). The molecular the diet in ancient human populations through stable toolbox: genetic techniques in wildlife ecology and isotope analysis of hair. Philos. Trans. Roy. Soc. Lond. Ser. management. J. Wildl. Mgmt. 69, 1362-1384. B 354,65-76. Gunga, H.-C., Kirsch, K.A., Roecker, L, Kohlberg, E., Macneale, K.H., Peckarsky, B.L & Likens, G.E. (2004). Tiedemann, J., Steinach, M. & Schobersberger, W. (2007). Contradictory results from different methods for Erythropoietin regulations in humans under different en- measuring direction of insect flight. Freshwater Biol. 49, vironmental and experimental conditions. Respir. Physiol. 1260--1268. Neurobiol, 158, 287-297. Markley, M.H. & Bassett, C.P. (1942). Habits of captive Hall, R.O. & Tank, J.L (2003). Ecosystem metabolism con- marten. Am. MidI. Nat. 28, 604-616. trols nitrogen uptake in streams in Grand Teton National Martinez del Rio, C. & Wolf, B.O. (2005). Mass-balance Park, Wyoming. Limnol. Oceanogr. 48, 1120--1128. models for animal isotopic ecology. In Physiological and Harper, J.M. & Austad, S.N. (2000). Fecal glucocorticoids: ecological adaptations tofeeding in vertebrates: 141-173. a noninvasive method of measuring adrenal activity in Starck, J.M. & Wang, T. (Eds). Enfield: Science Publishers. wild and captive rodents. Physiol. Biochem, Zool. 73, Maurel, D., Coutant, C. & Boissin, J. (1987). Thyroid and 12-22. gonadal regulation of hair growth during the seasonal molt

in the male European badger, Meles meles L. Gen. Camp. cysteine, lysine, alanine, and leucine in wool follicles of Endocr. 65, 317-327. sheep. J. Anim, Sci. 85, 2205-2213. McCullough, D.R. (1996). Metapopulations and wildlife Wanner, H., Gu, H., Hattendorf, B., Gunther, D. & Dorn, S. conservation. Washington: Island Press. (2006). Using the stable isotope marker 44Ca to study Mills, L.M. (2007). Conservation of wildlife populations: dispersal and host-foraging activity in parasitoids. J. Appl. demography, genetics, and management. Malden: Blackwell Ecol. 43,1031-1039. Publishing. Wassenaar, L.I. & Hobson, K.A. (2000). Improved method Nathan, R., Perry, G., Cronin, J.T., Strand, A.E. & Cain, for determining the stable-hydrogen isotopic composition M.L. (2003). Methods for estimating long-distance disper- (3D) of complex organic materials of environmental inter- sal. Oikos 103, 261-273. est. Environ. Sci. Technol. 34, 2354-2360. O'Connell, T.C. & Hedges, R.E.M. (2001). Isotopic compar- Wunder, M.B. & Norris, D.R. (2008). Improved estimates of ison of hair, nail and bone: modern analyses. J. Archaeol. certainty in stable-isotope-based methods for tracking Sci. 28,1247-1255. migratory animals. Eco!. Appl, 8, 549-559. Richards, M.P., Fuller, B.T., Sponheimer, M., Robinson, T. Zullinger, E.M., Ricklefs, R.E., Redford, K.H. & Mace, G.M. & Ayliffe, L. (2003). Sulphur isotopes in paleodietary (1984). Fitting sigmoidal equations to mammalian growth studies: a review and results from a controlled feeding curves. J. Mammal. 65, 607-6361. experiment. Int. J. Osteoarchaeol. 13, 37--45. Roon, D.A., Waits, L.P. & Kendall, K.C. (2003). A quantita- tive evaluation of two methods for preserving hair samples. Mol. Ecol. Notes 3,163-166. Supporting Information Rose, J., Stormshak, F., Oldfield, J. & Adair, J. (1984). Additional Supporting Information may be found in the Induction of winter fur growth in mink (Mustela vison) online version of this article: with melatonin. J. Anim. Sci. 58, 57-61. Rubenstein, D.R. & Hobson, K.A. (2004). From birds to Figure Sl. Mean length (± 1 SD) of guard hairs from butterflies: animal movement patterns and stable isotopes. three locations on captive martens, Laramie, Wyoming, Trends Ecol. Evol. 19, 256--263. 2006. Prior to 7 September, measures of hair growth from Rutherford, T.A. & Hawk, P.B. (1907). A study of the the tail reflected the hair from the previous molt, making comparative chemical composition of the hair of different measurements of new growth uncertain. Therefore, these races. J. BioI. Chem. 3, 459--489. data were omitted from our model of hair growth with the Southey, A.K., Sleeman, D.P. & Gormley, E. (2002). Sulfa- Gompertz function (Fig. 2). These data suggest that hair methoxine and rhodamine B as oral biomarkers for replacement on the body precedes that of the tail as European badgers (Meles meles). J. Wildl. Dis. 38, previously reported (Bassett and Llewellyn 1949). 378-384. Soutiere, E.C. & Steventon, J.D. (1981). Seasonal pelage Please note: Wiley-Blackwell are not responsible for the change of marten (Martes americana) in Maine. Can. Field content or functionality of any supporting materials sup- Nat. 95, 356. plied by the authors. Any queries (other than missing Thomas, N., Tivey, D.R., Penno, N.M., Nattrass, G. & Hynd, material) should be directed to the corresponding author P.I. (2007). Characterization of transport systems for for the article.