274 ARTICLES

of the mole (). Can. J. Zool. Tilman, D., J. Knops, D. Wedin, P. Reich, M. Ritchie, and E. Siemann. 1997. 59:315–322. The influence of functional diversity and composition on ecosys- ———. 1983. Burrowing ability and behavior of of the tem processes. Science 277:1300–1302. Ambystoma. Can. J. Zool. 61:616–620. Urian, K., A. Gorgone, A. Read, B. Balmer, R. Wells, et al. 2014. Rec- ———. 1998. Biological delineation of terrestrial buffer zones for ommendations for photo-identification methods used in capture- pond-breeding salamanders. Conserv. Biol. 12:1113–1119. recapture models with cetaceans. Mar. Mamm. Sci. 31:298–321. Simon, T. P., R. Jankowski, and C. Morris. 2000. Modification of an index Van Horne, B. 1983. Density as a misleading indicator of habitat qual- of biotic integrity for assessing vernal ponds and small palustrine ity. J. Wildl. Manag. 47:893–901. wetlands using fish, crayfish, and assemblages along Vanni, M. J. 2002. Nutrient cycling by in freshwater ecosys- southern Lake Michigan. Aquat. Ecosyst. Health Manag. 3:407– tems. Ann. Rev. Ecol. Syst. 33:341–370. 418. Wake, D. B. 1991. Declining amphibian populations. Science 253:860. Smith, C. K., and J. W. Petranka. 2000. Monitoring terrestrial salaman- Walker, B. H. 1992. Biodiversity and ecological redundancy. Conserv. ders: Repeatability and validity of area-constrained cover object Biol. 6:18–23. searches. J. Herpetol. 34:547–557. Wang, Y., and D. K. Moskovits. 2001. Tracking fragmentation of natural Speybroeck, J., and K. Steenhoudt. 2017. A pattern-based tool for long- communities and changes in land cover: Applications of Landsat term, large-sample capture-mark-recapture studies of fire sala- data for conservation in an urban landscape (Chicago Wilder- manders Salamandra (Amphibia: Urodela: Salamandri- ness). Conserv. Biol. 15:835–843. dae). Acta Herpetol.12:55–63. Warren, R. J., A. Labatore, and M. Candeias. 2017. Allelopathic invasive Spyreas, G. 2016. Scale and sampling effects on floristic quality. PLoS tree (Rhamnus cathartica) alters native plant communities. Plant ONE 11:e0160693. Ecol. 218:1233–1241. Sullivan, J. 2003. An Atlas of Biodiversity. Chicago Wilderness, Chi- Werner, P. A., I. K. Bradbury, and R. S. Gross. 1980. The biology of Cana- cago, Illinois. 66 pp. dian weeds. 45. Solidago canadensis L. Can. J. Plant Sci. 60:1393– Swink, F., and G. Wilhelm. 1994. Plants of the Chicago Region. 4th ed. 1409. Indiana Academy of Science, Indianapolis. 936 pp. Whitlock. M. C., and D. Schluter. 2009. The Analysis of Biological Taft, J., G. Wilhelm, D. Ladd, and L. Masters. 1997. Floristic quality as- Data. Roberts and Company Publishers, Greenwood Village, Colo- sessment for vegetation in Illinois: A method for assessing vegeta- rado. 704 pp. tion integrity. Erigenia 1977:3–95. Tallamy, D. W. 2004. Do alien plants reduce insect biomass? Conserv. Biol. 18:1689–1692.

Herpetological Review, 2021, 52(2), 274–278. © 2021 by Society for the Study of and Reptiles An Unanticipated Ecological Trap: Entrapment of the California in Technogenic Structures as a Confounding Stressor for a Threatened Species

Numerous wildlife species have been captured or collected JEFF A. ALVAREZ in natural and manmade ecological traps throughout the The Wildlife Project, PO Box 188888 Sacramento, California 95818, USA world (Hawken 1951; Brattstrom 1953; Delibes et al. 2001; JAYMEE T. MARTY Villa et al. 2018). The Rancho La Brea tar pits in Los Angeles Marty Environmental Consulting, Inc., 8925 Lanier Way, County, California, provide a prime example of a natural trap, Sacramento, California 95826, USA where millions of specimens and approximately 600 species KIRSTEN CHRISTOPHERSON were trapped in naturally occurring tar pits and have been United States Air Force, Travis Installation Support Section, preserved (Brattstrom 1953). Sites like Rancho La Brea collect 550 Hickam Drive, Building 248, indiscriminately, and mortality to individuals is nearly certain Travis Air Force Base, California 94535, USA (Brattstrom 1953). These sites appear to represent natural PENN CRAIG Natural Resource Manager, United States Air Force, 411 Airman Drive, habitats (i.e., aquatic features, retreat sites, etc.) to which B-570, Travis Air Force Base, Fairfield, California 94535, USA species would normally be attracted (Lemon and Churcher DEANNE WEBER 1971; Churcher 1966; Campbell 1979). Delibes et al. (2001) Center for Environmental Management of Military Lands, characterized these natural sites as attractive sinks and believed Colorado State University, 110 University Ave., that this maladaptive selection of habitat or microhabitat is Fort Collins, Colorado 80523, USA likely rare or goes unreported for most species. A more recent CHRISTOPHER VANG suite of potential population sinks is the ecological trap, whereby 8480 Orchard Creek Way, Elk Grove, California 95624, USA an individual or species mistakenly selects suboptimal habitat for its preferred habitat (Hale and Swearer 2010; Sih 2013). They

Herpetological Review 52(2), 2021 ARTICLES 275

Fig. 1. A) Typical two-piece lid for communication vaults at Travis Air Force Base. Note the gap (up to 22 mm) between the two concrete lid structures. To open, lids slide to left and right. B) Open communication vault. Note that the vault has fiberglass interior that is straight sided, and approximately 1.5 m deep. C) Vault lid after the poly foam caulk saver was inserted into the gap and the patch and seal tape is applied. A small sign covered by clear patch and seal tape was added to remind maintenance staff that the vault should be resealed after opening. further include a subset of these traps as human-induced rapid at the surface may ultimately result in temporary or permanent environmental changes (Robertson et al. 2013; Sih 2013; Wong sequestration of individuals who find technogenic traps and Candolin 2015), herein referred to as technogenic traps. unintentionally (Delibes et al 2001). Ultimately, this may cause Since the beginning of the Holocene, artificial structures direct mortality, or a localized reduction in recruitment or built by humans have created technogenic traps for wildlife. Villa reproduction for species or individuals that choose maladaptive et al. (2018) reported that a 2,000-year-old Roman well collected microhabitat. reptiles and amphibians, acting more or less like a large pitfall In California there are myriad ground-level structures trap. At least 11 taxa have been identified from excavations that migrating amphibians may be required to navigate over within this well. Modern technogenic structures are also known or around, including temporary excavations, storm drains, to collect and trap wildlife. Doody et al. (2003) conducted daily roadside curbs, erosion control waddles, and numerous other inspections of 800 km of open pipeline trenches in Australia objects and obstructions. Herein we report on the observations over a two-year period and recorded 7,438 individuals from of a threatened amphibian—California Tiger Salamander 103 species that were trapped in the trenches. (Ambystoma californiense)—as well as numerous sympatric Many technogenic traps have been reported for herpetofauna, species that were observed entrapped in technogenic structures including wells, storm drains, railroad lines, construction on an Air Force Base in central California. trenches, and other structures (Enge et al. 1996; Stuart et al. 2001; Kornilev et al. 2006; Manning 2007; Garcia-Caretenete et al. 2014; Materials and Methods Villa et al. 2018). Unlike the single well site reported by Villa et al. (2018) and temporary excavation reported by Enge et al. (1996), Travis Air Force Base (Base) is an active military installation McInroy and Rose (2015) reported that roadside street drains, where cargo planes are stationed. The Base is located south of which are extremely numerous and distributed widely around Fairfield, California on flat terrain, with approximately 30 percent their site in the United Kingdom, were a long-term and common of the land on the Base covered by industrial, commercial, and technogenic traps for amphibians. residential structures. The remaining land is covered by non- In their daily or seasonal search for dark and thermally stable native annual grassland, seasonal and perennial wetlands microhabitat, reptiles and amphibians may find themselves and woodland habitat, with the southern portion of the Base entrapped (Lovich et al. 2011). This may become more common occupied by an airfield (i.e., two 4-km-long paved runways) among species that make mass migrations, like bi-phasic and associated infrastructure. Surrounding the runways is an amphibians, in that their seasonally abundant movements may expansive annual grassland with numerous vernal pools and offer more opportunity to encounter a technogenic trap. Frogs, other wetland features that extend well beyond the boundaries toads, and salamanders often make seasonal movements from of the Base. At least 14 special-status species utilize the Base for upland refuge habitat to reproductive sites (e.g., ponds and part or all of their natural history, including, but not limited to creeks), and then return to upland refugia (i.e., burrows, rock the State and federally threatened California Tiger Salamander, cracks, beneath logs, etc.) for the majority of the year (Stebbins federally threatened Fairy Shrimp (Brachinecta 1951). This above ground movement and search for refugia, lynchi), and a rare plant, Contra Costa Goldfields (Lasthenia may make amphibians especially susceptible to technogenic conjugens). traps. Furthermore, amphibians tend to have daily or seasonal Hundreds of utility vaults have been installed on the Base, behaviors that are saxicolous (use cracks within rocks) or including within and surrounding open grasslands. These fossorial (live underground) and may provide a susceptibility structures can be large, including units that are 2.5 m wide × to technogenic conditions that mimic more suitable habitat. 2.5 m long × 3 m deep, with a maintenance hole cover (formerly This natural tendency to seek out cracks, holes, or other cover known as a manhole cover). More numerous (i.e., 100s) are

Herpetological Review 52(2), 2021 276 ARTICLES

smaller valve boxes that are 15 cm diameter and 30 cm deep, all of which are located below ground, with a cover at ground level. Our work focused on moderate-sized units that were made of fiberglass, were approximately 1 m wide × 2 m long × 1.5 m deep, and included a concrete lid that was locked with up to four security bolts. The lid was separated into two equal parts (Fig. 1a) where one of both halves could be removed by lifting and sliding half of the lid to one side (Fig. 1b). These vaults were typically used to connect or maintain utilities (e.g., communication lines) on the Base and are essential to the Base mission. They were infrequently checked—typically when there was a need—and were generally maintained by various utility companies that operate on the Base. In April 2016, an anecdotal report was made of an adult California Tiger Salamander found within a large utility vault prompting surveys of additional, similar-sized vaults on the Base (Barbara Ely, City Light and Power, pers. comm.). We conducted herpetofaunal surveys in a subset of utility vaults (#26–#45)­ that were located adjacent to an off-site pond known to support breeding California Tiger Salamanders (Fig. 1C). During our surveys an observer used a flashlight and looked into the darkened areas of the vault to check for and identify species that were entrapped. All live animals that were encountered were removed and carried to a nearby release site (e.g., rodent burrow). Dead animals were removed from the vault and discarded, while dead California Tiger Salamanders, if encountered, were collected for deposit in the Museum of Vertebrate Zoology, Berkeley, California. Vault checks occurred approximately monthly, or more frequently if California Tiger Salamanders were known to be moving on the soil surface. Two years following the initial surveys to check vaults, any vault Fig. 2 General location of communication vaults along Perimeter known to trap California Tiger Salamanders was sealed. At each Road on Travis Air Force Base. Numbers and green diamonds indicate vault a 10-cm wide waterproof patch and seal tape made by general layout of vaults that may be encountered by California Tiger Gorilla® Glue Incorporated was applied to the open gap between Salamanders. Specific locations and details obscured for security. lid halves. This product comes in rolls that are 3 m long and can be cut to fit most structures. The product is waterproof, sticks in cold and warm conditions, and is flexible enough to be molded All but two of the 79 tiger salamanders found in vaults during to fit various structures. At many of the vaults, we encountered the 34-month survey period were recent post-metamorphic a significant gap in the lid opening that required sealing. Into individuals (i.e., only one adult), likely dispersing from a nearby these gaps we placed various widths of Poly Foam Caulk Saver natal pond (Fig. 1a). This assumption is based on a comparison (foam tube) made by Frost King® such that the gap was filled and of the size of animals collected from vaults with those animals any sticky surface from the waterproof tape would be attached to migrating from the pond at a similar time period. One adult the foam rather than exposed. This was a precautionary choice tiger salamander was found in a vault covered by 5-mm thick to avoid allowing climbing amphibians to locate exposed tape steel plate with 18-mm holes drilled into it (access holes; Fig. 3), and become attached. and the other adult was found under a steel plate that covered a culvert standing on end and buried in the ground (Fig. 4). The Results latter structure suggested that the individual California Tiger Salamander climbed up the vertical side of the vault and into the We conducted 362 vault checks at 67 unique vaults between covered area, where it presumably became entrapped. November 2017 and September 2020. During the vault checks we We found up to 12 live California Tiger Salamanders in a single found a total of 2275 , including 79 California Tiger vault during a check. Among the 86 salamanders collected, 5 Salamanders. In addition to the California Tiger Salamanders, (5.8%) were found dead. On three site visits we found dead post- our surveys also included at least seven additional species: Pacific metamorphic tiger salamanders, three of which were found when Chorus Frog (Hyliola regilla; N = 2135), Western Fence Lizard tiger salamanders were initially discovered entrapped in vaults (i.e., (Sceloporus occidentalis; N = 49), Western Skink (Plestiodon summer 2017). Based on condition and site conditions, we skiltonianus; N = 9), Western Terrestrial Gartersnake (Thamnophis surmised that the dead individuals likely became desiccated (i.e., elegans; N = 26), California Kingsnake (Lampropeltis californiae; those found dead in summer), or succumbed to cold (i.e., a single N = 3), Gophersnake (Pituophis catenifer; N = 2), and cricetid salamander found dead in a vault in late December). rodents (Peromyscus sp.; N = 8). We also noted that each vault had The general location of the vaults that entrapped tiger numerous live individuals or the dried exoskeletons of isopods salamanders appeared to be associated with proximity to the (order Isopoda), ground beetles (family Carabidae), Jerusalem pond, although this was not tested (Fig. 2). Over the 34-month crickets (family Stenopelmatidae), and other uncounted and survey period, 59 vaults had zero salamanders; 7 vaults had ≤ 5 unrecorded unidentified . individuals; 2 vaults had nine and ten each, and 2 vaults (#32,

Herpetological Review 52(2), 2021 ARTICLES 277

Fig. 4. The vault is constructed of a culvert that is buried on end with a small, metal, half-round plate as a cover. The California Tiger Sala- mander on the lid was removed from inside and placed on the lid for scale. The inner portion of the vault is approximately 20 cm deep and not a climbable distance for the salamander, but it was accessible from the outside, which was only 5 cm high from standing soil level.

case of the steel-plate covered vault with 18-mm holes and the Fig. 3. One of several steel-plate-covered vaults. This lid has open maintenance hole covers with 26-mm holes, we speculated that holes that include seven (up to) 18 mm holes that allowed access to California Tiger Salamanders utilized the access holes as if it were eight California Tiger Salamanders, as well as fence lizards, a garter- a rodent burrow and were subsequently trapped following entry. snake, and a kingsnake. Lid is askew to indicate the style of opening We found no means by which a California Tiger Salamander (or and closing. Holes were later sealed with patch and seal tape. lizard or snake) could escape what was effectively a smooth- walled pitfall trap. In the majority of communication vaults, there 33) had 18 and 26 each. With the exception of the size of the gap is a recessed concrete lid that has only small joints that allow for between the two lids, there was no visible difference between the imperfect closure of concrete lids. These joints can expand and vaults. All known entrapped salamanders were collected in vaults contract slightly with changes in temperature but the largest one 26–45 during our survey period (Fig. 2) . Other vaults that were we found was 22 mm and the smallest was effectively sealed. surveyed (i.e., 1–25 and 46–65), did not contain California Tiger Nevertheless, California Tiger Salamanders, as well as six other Salamanders during the surveys conducted and reported here. herpetofaunal species, found the vaults and entered into this Our vault check frequency changed through the year and technogenic trap. among years. We did note that post-metamorphic salamanders It has been reported that temporary technogenic structures were typically collected between the months of June and August (e.g., excavations, trenches) can collect many (> 450) amphibians (N = 21) and November and January (N = 56), which followed and reptiles in slightly more than a three-week period (Enge our direct observations of above ground movement for this et al. 1996). These types of impacts can be significant if an size class. Based on concurrent pitfall trapping data at the Base, open excavation was accessible to migrating amphibians for post-metamorphic tiger salamanders (approximately 800–2000 long periods. Utility boxes and vaults are common across the annually) began to leave the natal pond approximately late May landscape, particularly in areas of human residential/industrial and complete their above ground movement by late July. development, but also in areas that appear open and natural (e.g., grasslands surrounding runways). Recently a colleague Discussion anecdotally reported a California Tiger Salamander in an irrigation valve box associated with a residential development Our findings were the result of reactive mitigation to reduce (J. Alvarez, unpubl. data). Whether in open spaces or residential any loss of a listed species. We were not able to assess or survey areas, development continues into wildlands where California vaults with equal effort or time. Our observations are therefore Tiger Salamanders and numerous other species are extant post facto, and include gaps in information, limitations to analysis, (Mitchell et al. 2008). These species can and do become trapped and include observational data and speculative conclusions that within structures associated with utilities, with some regularity, are based on our primary goal: to reduce entrapment. and likely also become trapped in storm drains and similar Our findings indicate that migrating adult and sub-adult structures (Hawken 1951; McInroy and Rose 2015). In areas California Tiger Salamanders, as well as sympatric Pacific Chorus where species may have special conservation status, some effort Frogs, encounter potential saxicolous or subterranean refugia should be made to mitigate this potential technogenic trap. This within upland habitats during the process of migration to and may mean blocking holes in lids that would otherwise allow from aquatic sites. During this time, they may find rodent burrows access. Alternatively, boxes can be set slightly above ground (≥ (see Loredo and Van Vuren 1996), but also find widely dispersed 15 cm) so that the majority of small herpetofauna and mammals technogenic structures when searching for refuge sites. In the will be excluded from access.

Herpetological Review 52(2), 2021 278 ARTICLES

We surmised that the vaults that contained salamanders ap- Disclaimer: The views expressed in this article are those of the peared to be those that had gaps or large cracks through which ti- authors and do not reflect the official policy or position of the U.S. ger salamanders could gain entrance. We also concluded, based on Government, the Department of Defense, or the Department of the an absence of texture to the interior side walls, lack of exposure to Air Force. the outside (no low-level gaps, cracks, or joints, or other potential opportunities for self-release), that once the herpetofaunal spe- Literature Cited cies were entrapped, they were not able to escape. The exception to this is the Pacific Chorus Frog, which readily climbed walls with Brattstrom, B. H. 1953. The amphibians and reptiles from Rancho La no external assistance. In an effort to reduce or eliminate vault in- Brea. Trans. San Diego Soc. Nat. Hist. 11:365–392. gress by California Tiger Salamanders and other herpetofauna, we Campbell, K. E., Jr. 1979. The non-passerine Pleistocene avifauna of tested different methods to seal the vaults. Specific conditions set the Talara tar seeps, northwestern Peru. Royal Ontario Mus. Life Sci. Contrib. 118:1–203. by the Base were required to be met for the vaults to be sealed, in- Churcher, C. S. 1966. The insect fauna from the Talara tar-seeps, Peru. cluding remaining accessible to communications contractors. We Can. J. Zool. 44:985–993. conducted ad hoc testing to determine what materials might work Delibes, M., P. Gaona, and P. Ferreras. 2001. Effects of an attractive sink to block gaps (presumed to be the access point) in the vault doors leading into maladaptive habitat selection. Am. Nat. 158:277–285. but many materials failed to meet the standards for airfield safety. Doody, J. S., P. West, J. Stapley, M. Welsh, A. Tucker, E. Guarino, M. Pauza, The resulting solution was waterproof patch and seal tape made N. Bishop, M. Head, S. Dennis, G. West, A. Pepper, and A. Jones. 2003. by Gorilla® Glue Incorporated. This mitigation resulted in a sealed Fauna by-catch in pipeline trenches: Conservation animal ethics, vault cover, to which we attached a small sign indicating that re- and current practices in Australia. Aust. Zool. 32:410–419. moval of the seal should be reported to a biologist who can re- Enge, K. M., D. T. Cobb, G. L. Sprandel, and D. L. Francis. 1996. Wildlife seal the vault (Fig. 6). We found that sealing the vaults with sealing captures in a pipeline trench in Gadsden County, Florida. Florida Sci. 59:1–11. tape proved effective at restricting California Tiger Salamanders Garcia-Cardenete, L., J. M. Pleguezuelos, J. C. Brito, F. Jimenez-Cazalla, from entering the vaults. Following the deployment of the sealing M. T. Perez-Garcia, and X. Santos. 2014. Water cisterns as death traps tape on vaults 26–44 the number of tiger salamanders captured for amphibians and reptiles in arid environments. Environ. Con- was reduced to zero during the monthly vault checks over a two- serv. 41:341–349. year period. We noted that the survival of individual California Ti- Hale, R., and S. E. Swearer. 2010. Ecological traps: Current evidence and ger Salamanders at this site was likely increased by frequent trap future directions. Proc. Royal Soc. B: Biol. Sci. 283:20152647. checking. Although this could not be tested on a threatened spe- Hawken, J. L. 1951. Water system acts as a reptile and amphibian trap. cies, we speculate that entrapped animals would likely die from Herpetologica 7:81–83. desiccation, starvation, exposure, or potential predation. Lemon, R. R. H., and C. S. Churcher. 1961. Pleistocene geology and pa- At Travis Air Force Base the variety and location of vaults on leontology of the Talara region, northwest Peru. Am. J. Sci. 259:410– 429. the Base makes checking every vault extremely difficult, if not Loredo, I., and D. Van Vuren. 1996. Reproductive ecology of a popula- prohibitive. Some vault styles include metal hinged doors that tion of the California tiger salamander. Copeia 1996:895–901. are constructed with gaps of up to 1 cm, which can allow nearly Lovich, J. E., J. R. Ennen, S. Madrak, and B. Grover. 2011. Turtles, culverts, every juvenile reptile and amphibian to enter. This poses a long- and alternative energy development: An unreported but potentially term concern, not only for species in decline, but also for lo- significant mortality threat to the desert tortoise (Gopherus agas- cally abundant species, like chorus frogs, that may be affected sizii). Chelon. Conserv. Biol. 10:124–129. by these numerous technogenic traps, due in part to their own Kornilev, Y. V., S. J. Price, and M. E. Dorcas. 2006. Between a rock and a natural history needs (e.g., seeking cover daily or seasonally, hard place: Response of the eastern box turtle (Terrapene carolina) moving seasonally from breeding habitat, etc.). We recommend when trapped between railroad tracks. Herpetol. Rev. 37:145–148. vault checks twice annually (July and February) to determine if Manning, G. J. 2007. Uta stansburiana (side-blotched lizard): mortality. Herpetol. Rev. 38:465. adjustments need to be made to lids and sealing mechanisms. McInroy, C., and T. A. Rose. 2015. Trialing amphibian ladders within These months are within the migration period for adults (Feb- roadside gullypots in Angus Scotland: 2014 impact study. Herpetol. ruary) and post-metamorphic young (July) at Travis Air Force Bull.132:15–19. Base and are within time frames in which salamanders were en- Mitchell, J. C., J. Brown, and B. Bartholomew (eds.). 2008. Urban Herpe- trapped at this location. Vaults that appear to be used in spite tology. Society for the Study of Amphibians and Reptiles, Salt Lake of being sealed should be checked more frequently (≤ monthly). City, Utah. 586 pp. Any cracked vaults should be repaired so that ground level gaps Robertson, B. A., J. S. Rehage, and A. Sih. 2013. Ecological novelty and and cracks do not allow access to vault interiors. Employing a the emergence of evolutionary traps. Trends Ecol. Evol. 28:552–560 mitigating technique to exclude threatened species, as well as Sih, A. 2013. Understanding variation in behavioural responses to common herpetofauna, can reduce or eliminate this confound- human-induced rapid environmental change: A conceptual over- view. Anim. Behav. 85:1077–1088 ing stressor to the California Tiger Salamander. Stuart, J. N., M. L. Watson, T. L. Brown, and C. Eustice. 2001. Plastic net- ting: entanglement hazard to snakes and other wildlife. Herpetol. Acknowledgments.— We are grateful for access to the Travis Air Rev. 32:162–164. Force Base and specifically to Bowhead (UIC Government Services) to Stebbins, R. C. 1951. Amphibians of Western North America. University access their vaults. We also thank Barbara Ely, City Light and Power, for of California Press, Berkeley, California. 539 pp. reporting a California Tiger Salamander that entered a maintenance Villa, A., M. Bon, and M. Defino. 2018. Trapped in a Roman well: Am- lid hole and became entrapped. We are grateful to S. M. Foster, S. phibians and reptiles from Tenuta Zuccarello near Marcon, Venice, George, R. Gaitan, and C. Reddin for assisting with vault checks. Edito- Italy. Hist. Biol. 32:55–70. rial comments provided by K. Cecala greatly improved the readability Wong B. B. M., and U. Candolin. 2015. Behavioral responses to chang- and clarity of this manuscript. Work was conducted under the Emer- ing environments. Behav. Ecol. 26:679–680. gency Biological Opinion (BO# 08ESMF00-2017-FE-2767) prepared by the U.S. Fish and Wildlife Service.

Herpetological Review 52(2), 2021