Evaluation of the Combined Temperature and Relative Humidity Preferences of the Colombian Terrestrial Salamander Bolitoglossa Ramosi (Amphibia: Plethodontidae)

Canadian Journal of Zoology

Evaluation of the combined temperature and relative humidity preferences of the Colombian terrestrial salamander Bolitoglossa ramosi (Amphibia: Plethodontidae)

Journal: Canadian Journal of Zoology

Manuscript ID cjz-2017-0330.R2

Manuscript Type: Article

Date Submitted by the Author: 14-Mar-2018

Complete List of Authors: Galindo-Martínez, Carlos; Universidad del Tolima, Departamento de Biología Cruz-Rodríguez,Draft Erika; Universidad del Tolima, Departamento de Biologia Bernal, Manuel; Universidad del Tolima, Departamento de Biologia

Is your manuscript invited for consideration in a Special Not applicable (regular submission) Issue?:

Amphibian; Bolitoglossa ramosi; Hydric environment; Plethodontid Keyword: salamander; Thermal selection; Urodela

https://mc06.manuscriptcentral.com/cjz-pubs Page 1 of 31 Canadian Journal of Zoology

1 Evaluation of the combined temperature and relative humidity preferences of

2 the Colombian terrestrial salamander Bolitoglossa ramosi (Amphibia:

3 Plethodontidae)

5 C.A. Galindoa, E.X. Cruzb and M.H. Bernalc.

7 a Grupo de Herpetología, Eco-Fisiología & Etología, Universidad del Tolima.

8 Ibagué, Colombia. [email protected]

9 b Grupo de Herpetología, Eco-Fisiología & Etología, Universidad del Tolima. 10 Ibagué, Colombia. [email protected] 11 c Grupo de Herpetología, Eco-Fisiología & Etología, Universidad del Tolima.

12 Ibagué, Colombia. [email protected]

14 Corresponding author: Manuel Hernando Bernal ([email protected]).

16 Grupo de Herpetología, Eco-Fisiología & Etología, Universidad del Tolima. Ibagué,

17 Colombia.

18 [email protected]

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 2 of 31

23 Evaluation of the combined temperature and relative humidity preferences of

24 the Colombian terrestrial salamander Bolitoglossa ramosi (Amphibia:

25 Plethodontidae)

27 C.A. Galindoa, E.X. Cruzb and M.H. Bernalc.

29 Abstract

30 31 Temperature and humidity are criticalDraft factors for terrestrial lungless salamanders, 32 as their body temperatures are largely determined by the environmental

33 temperature and require moisture to sustain cutaneous respiration. Herein, we

34 evaluated the preference of Bolitoglossa ramosi (Brame and Wake 1972) between

35 a high temperature and a high relative humidity, the influence of temperature on

36 the relative humidity preferences, and the influence of the relative humidity on the

37 thermal preferences. This study was performed in a field location in the

38 municipality of Líbano, Tolima, Colombia. There, on different nights, we collected

39 84 adult salamanders and carried out the preference experiments, using aluminum

40 troughs with different thermal and relative humidity gradients. We found that

41 between the high temperature and high relative humidity, salamanders preferred

42 the high relative humidity. However, B. ramosi selected high temperatures when

43 the gradient had a high relative humidity, and low temperatures when the gradient

44 had a low relative humidity. These results show that B. ramosi is able to

https://mc06.manuscriptcentral.com/cjz-pubs Page 3 of 31 Canadian Journal of Zoology

45 thermoregulate and hydroregulate. Nevertheless, hydrorregulation seems to be

46 more important than thermoregulation as these salamanders always selected the

47 high relative humidity gradients while their thermal selection relied on the hydric

48 environment.

49 Key words: Amphibian; Bolitoglossa ramosi; Hydric environment; Plethodontid

50 salamander; Thermal selection; Urodela.

53 Draft 54

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 4 of 31

64 Introduction

65 Due to their ectothermy and permeable skin, environmental temperature

66 and humidity are critical factors for the survival of terrestrial amphibians. Ambient

67 temperature largely determines their body temperature (Navas et al. 2008; Köhler

68 et al. 2011), whereas humidity is important to avoid the evaporative water loss

69 through the skin and to keep their hydric state (Shoemaker et al. 1992). Both

70 conditions, body temperature and hydric state, are known to strongly affect the

71 physiological and behavioral performance of amphibians (Jørgensen 1997; Navas

72 et al. 2008; Anderson and Andrade 2017). Amphibians are able to thermoregulate

73 behaviorally to maintain their body temperatures into a preferred range where the

74 biological activities can be carriedDraft out successfully (Hutchison and Dupré 1992;

75 O’Connor and Tracy 1992; Raske et al. 2012; Balogová and Gvoždík 2015;

76 Strickland et al. 2016). On the other hand, they avoid excessive water loss by

77 selecting microhabitats that allow the animals to balance water loss with water

78 uptake, or to limit activity to periods with favorable conditions (Wells 2007; Winters

79 and Gifford 2013). Therefore, terrestrial amphibians should prefer habitats with an

80 optimal combination of temperature and humidity to maximize the intake of energy

81 and regulate their hydration state (Preest and Pough 1989; Titon et al. 2010;

82 Köhler et al. 2011; Mitchell and Bergmann 2016), and because the interaction

83 between these two factors influences the regulation of each of them (Anderson

84 and Andrade 2017). For example, high environmental temperatures result in

85 higher evaporative water loss in terrestrial amphibians, and physiological

86 performance decreases when an organism becomes dehydrated (Preest and

https://mc06.manuscriptcentral.com/cjz-pubs Page 5 of 31 Canadian Journal of Zoology

87 Pough 1989, 2003; Rogowitz et al.1999; Anderson and Andrade 2017). As a

88 consequence, salamanders increase their resistance to water loss at warm

89 temperatures and, conversely, decrease it under cool temperatures (Winters and

90 Gifford 2013; Riddell and Sears 2015). Despite the importance of these

91 interactions in amphibians, the study of thermal and humidity preferences has

92 generally considered them separately (Spotila 1972; Hutchison and Dupré 1992;

93 Wells 2007). Recently, Mitchell and Bergmann (2016) tested for interactions

94 between environmental temperature, moisture and organismal hydration on

95 temperature and moisture preferences in the green frog Lithobates clamitans. In

96 this work, they found that frogs selected environmental conditions that minimized

97 cutaneous evaporative water loss,Draft hydroregulating more stringently than

98 thermoregulating.

100 Plethodontid salamanders are a very interesting biological model to study

101 the interaction between temperature and humidity preferences due to their

102 particular physiology. They are strongly dependent on the ambient temperature, as

103 their body temperature is equal to the air or substrate temperature

104 (thermoconforms) (Brattstrom 1963; Catenazzi 2016; Lunghi et al. 2016), and they

105 are among the tetrapods with the lowest metabolic rate (Chong and Mueller 2012).

106 Moreover, they are unique among terrestrial vertebrates in having completely lost

107 their lungs relying mainly on the skin surface for gas exchange, which causes high

108 rates of desiccation principally when they are in air with a lower vapor pressure

109 than that of their skin (Feder and Londos 1984; Feder and Burggren 1985; Gifford

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 6 of 31

110 2016). Consequently, salamanders require moisture to sustain cutaneous

111 respiration (Spotila 1972; Gatz et al. 1975; Feder 1983), and their activities are

112 generally nocturnal and limited to moist habitats to minimize water loss (Feder

113 1983; Peterman and Semlitsch 2014; Gifford 2016). Additionally, terrestrial

114 salamanders of the family Plethodontidae are generally philopatric (Kleeberger

115 and Werner 1982), exhibiting minimal dispersal (Liebgold et al. 2011; Basile et al.

116 2017), which makes them extremely dependent upon suitable temperature and

117 moisture microclimates to successfully forage and meet their energy requirements

118 (Fraser 1976). Although it has been found that through a season or year, cave

119 salamanders are able to actively search habitats with suitable conditions of

120 temperature and moisture (LunghiDraft et al. 2015), while terrestrial salamanders are

121 often more abundant in areas of dense cover that retain moisture (O'Donnell et al.

122 2014).

123

124 Bolitoglossa ramosi (Brame and Wake 1972) is a plethodontid salamander

125 endemic to Colombia, with a distribution restricted to the eastern slope of the

126 Central Cordillera, between 1 200 and 2 000 m altitude (Palacio et al. 2006). This

127 species is listed as Least Concern by the IUCN (International Union for

128 Conservation of Nature) because it is common and with presumed large

129 populations (Castro et al. 2004). B. ramosi is mainly arboreal, as are the majority

130 of tropical plethodontids (McEntire 2016), nocturnal, with direct development and

131 limb regeneration (Arenas et al. 2017), and commonly found on top of leaves,

132 small trees or on the ground, mostly during cloudy nights or after rain events when

https://mc06.manuscriptcentral.com/cjz-pubs Page 7 of 31 Canadian Journal of Zoology

133 the air relative humidity is high and the environmental temperature is low.

134 However, there are few studies on its biology, and particularly on its thermal and

135 hydric relations. Among them, Cruz et al. (2016) found a positive and significant

136 relationship between the body temperature and the substrate and air

137 temperatures, demonstrating that this species is a thermoconformer, as it has

138 been also reported in other plethodontid terrestrial species of the genus

139 Hydromantes (Lunghi et al. 2016). They also reported that the body temperature

140 showed a higher thermal dependence on the substrate than the air temperature,

141 indicating that B. ramosi presents a thigmothermic regulation.

142 In this study we tested if: (1) Bolitoglossa ramosi would prefer a high

143 relative humidity rather than a highDraft temperature, in a combined temperature and

144 relative humidity gradient; (2) the temperature would have any effect on the

145 relative humidity preferences; and (3) the relative humidity would have any effect

146 on the thermal preferences. We hypothesized that between these two

147 environmental variables, salamanders would prefer the high relative humidity

148 because they depend on moisture to gas exchange, and are particularly abundant

149 at a high relative humidity in the field. In addition, we also hypothesized that under

150 an upper uniform experimental temperature or relative humidity, salamanders

151 would select the high relative humidity to avoid water loss, and the high

152 temperature to increase their physiological performances, respectively.

153

154 Materials and methods

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 8 of 31

155 Experimental animals

156 The present study was carried out with 84 adult salamanders collected by

157 hand (mean of snout-vent-length= 32.47 ± 1.58 mm, range: 29.10 - 35.70 mm),

158 between November and December 2015, from different places of a pre-mountain

159 forest patch in the municipality of Líbano (04º54’36.4’’ N, 75º03’43.7’’ W, 1 750 m

160 altitude) Tolima, Colombia. This region has an annual mean temperature of

161 20.5°C, a precipitation of 2 262 mm, and an air relative humidity between 85 and

162 100% (Cortolima 2009). After capturing the animals, from 18:00 to 20:00 hours,

163 they were taken to a field location near the collection sites (approximately 1 000

164 meters straight-line distance), where the experiments were conducted during the

165 same night, between 20:00 and 24:00Draft hours. Salamanders were used only once

166 and later released at the site of capture. We studied adults (males and females,

167 indistinctly) because they were abundant and easy to collect in several distant

168 places, which decreased the possibility of recapturing the same salamander for

169 the experiments, as they have low dispersal (Liebgold et al. 2011; Basile et al.

170 2017).

171

172 Temperature and relative humidity gradients

173 We built a combined thermal and relative humidity gradient for each of two

174 rectangular aluminum troughs of 200 cm (length) x 7 cm (width) x 5 cm (depth),

175 placed in parallel inside a greater aluminum channel of 280 cm (length) x 15 cm

176 (width) x 15 cm (depth) which was partially filled with dechlorinated water. We

https://mc06.manuscriptcentral.com/cjz-pubs Page 9 of 31 Canadian Journal of Zoology

177 used dry ice and a 200-W conventional aquarium thermostat heater (ReSun –

178 Sunlike-200) at opposing ends of the greater aluminum channel, in order to obtain

179 a gradient from 15 to 30°C on the surface of the two experimental troughs.

180 Additionally, along the experimental troughs, we positioned a uniform organic

181 substrate with great capacity to retain water (“peat”), which was from partially to

182 totally hydrated (with dechlorinated tap water), so as to provide a relative humidity

183 gradient on the substrate surface between 65% and 95%. After each trial, we

184 moved and sprayed the substrate with dechlorinated tap water to ensure the levels

185 of relative humidity required. Seven sensors of digital thermo-hygrometers (Extech

186 44550, precision of ±1ºC) were placed on the surface of the substrate at intervals

187 of 2°C and 4% RH through the gradients.Draft We selected these thermal and relative

188 humidity gradients according to the daily environmental ranges recorded in the

189 microhabitat of the salamanders studied (diurnal hours: 18-23°C, 60-89% RH;

190 night hours: 17.1-21.2°C, 73-99% RH).

191

192 Thermal and relative humidity preference experiments

193 In order to assess the temperature and relative humidity preferences of the

194 salamanders, and our three hypotheses, we carried out three experiments. In the

195 first, we simultaneously placed 9 salamanders in each of two combined thermal

196 and relative humidity gradients, where the end of the first gradient was 15°C-65%

197 RH and the opposite end was 30°C-95% RH, while the end of the second gradient

198 was 15°C-95% RH and the opposite end was 30°C-65% RH. We repeated this

199 experiment the next night with another 9 individuals per gradient (collected before

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 10 of 31

200 releasing the animals previously studied to avoid pseudoreplication), for a total of

201 36 different salamanders (9 X 2 X 2). Later on, in the second experiment, we used

202 6 salamanders (at the same time) for each of two gradients of relative humidity

203 between 65% and 95%, one of them at a uniform temperature of 22 ± 0.5°C and

204 the other at 25 ± 0.5°C. We also repeated the same experiment the following night

205 with 6 other salamanders per gradient (collected before releasing the animals

206 previously studied), for a total of 24 (6 X 2 X 2). Finally, in the third experiment, we

207 placed 6 salamanders simultaneously in each of two thermal gradients from 15°C

208 to 30°C, one of them at a uniform relative humidity of 75 ± 1.5% and the other at

209 95 ± 1%. As in the previous experiments, the third experiment was repeated the

210 next night with another 6 individualsDraft per gradient, for a total of 24 different

211 salamanders (6 X 2 X 2). We did not test individual salamanders in the gradient so

212 as to increase the number of data obtained per trial, and because previous essays

213 indicated that salamanders always began to individually explore their new

214 environment and later stopped anywhere.

215 Prior to the trials, we kept the salamanders for 30-45 minutes in plastic

216 containers at the environmental temperature, with wet soil and leaf litter from the

217 capture site, which were sprayed with dechlorinated water to ensure the

218 salamanders were hydrated. Then, we randomly placed three groups, with the

219 same number of salamanders, in the middle and at each end of the gradient, and

220 allowed them to explore and become habituated for 30 minutes before we began

221 the recordings. We kept the gradient in a dark room to eliminate possible light

222 cues, as Bolitoglossa ramosi is primarily nocturnal. However, a dim red light from a

https://mc06.manuscriptcentral.com/cjz-pubs Page 11 of 31 Canadian Journal of Zoology

223 headlamp allowed for continuous visibility of each salamander. We used the scan

224 method (Lehner 1996) to record the temperature and relative humidity at the point

225 where each salamander (at the middle of the body length) was found in the

226 gradient every 30 minutes (three measurements per salamander).

227

228 Statistical analyses

229 For all statistical analyses, we used the mean of the three data points

230 collected for each salamander in the gradient to eliminate pseudoreplication. The

231 intra-individual coefficient of variation for these data was less than 14% in almost 232 all cases (97.6%), indicating a lowDraft dispersion around the mean (Zar 1999). Then, 233 we separately compared the temperature and relative humidity selected by the

234 salamanders between the two parallel gradients, using the Mann–Whitney U test

235 as data were not normally distributed (Shapiro-Will test, P < 0.01) and variances

236 were not homogenous (F test, P < 0.01). All statistical analyses were conducted in

237 SPSS Statistic 19.0 (Inc., Chicago, Illinois).

238

239 Results

240 In experiment 1, in the gradient from 15°C-65% RH to 30°C-95% RH, the

241 salamanders selected a place with a mean temperature of 28.4°C ± 1.02°C

242 (median = 28.7°C), and 92% ± 2.14% of relative humidity (median = 93.7%) (n =

243 18). In the parallel gradient, from 15°C-95% RH to 30°C-65% RH, the

244 salamanders preferred a mean temperature of 17.2°C ± 1.17°C (median =

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 12 of 31

245 17.2°C), and 93.6% ± 2.32% of relative humidity (median = 92.7%) (n = 18) (Fig.

246 1). Between the two experimental gradients, salamanders showed a significant

247 difference in their preferences for temperature (Mann–Whitney U = 324, P = 0.001,

248 n = 36), but not for relative humidity (Mann–Whitney U = 119.5, P = 0.181, n = 36)

249 as it was higher than 90% in both gradients (Table 1).

250 In experiment 2, where the gradient had a uniform temperature of 22°C, the

251 salamanders were found at a relative humidity between 91% and 94%, with a

252 mean of 93% ± 1.12% RH (median = 93%) (n = 12). In the parallel gradient, at a

253 uniform temperature of 25°C, the salamanders selected a relative humidity from

254 90% to 93%, with a mean of 92.4% ± 0.95% (median = 92.7%) (n = 12) (Table 1)

255 (Fig. 2). Between the two gradientsDraft at the two different uniform temperatures,

256 there was not a significant difference in the relative humidity preferred by the

257 salamanders (Mann–Whitney U = 39.5, P = 0.60, n = 24).

258 Finally, in experiment 3, in the gradient with a uniform relative humidity of

259 95%, the salamanders selected a wide range of temperatures, between 18.2°C

260 and 29.2°C with a mean of 24.9°C ± 4.55°C (median = 27.2 °C) (n = 12). In the

261 parallel gradient, at a uniform relative humidity of 75%, the salamanders now

262 selected temperatures from 16.7°C to 20.6°C with a mean of 17.9°C ± 1.11°C

263 (median = 17.9°C) (n = 12) (Table 1), (Fig. 3). Herein, the salamanders showed

264 significant differences between the temperatures selected at the two different

265 relative humidity gradients (Mann–Whitney U = 127.5, P = 0.001, n = 24).

266

https://mc06.manuscriptcentral.com/cjz-pubs Page 13 of 31 Canadian Journal of Zoology

267 Discussion

268 Amphibians are highly sensitive to changes in environmental variables such

269 as temperature and humidity (Buckley and Jetz 2007; Wells 2007; Tracy et al.

270 2012). One mechanism by which amphibians can respond to these changes is

271 across behavioral regulation to optimize physiological processes associated with

272 survival activities (Strickland et al. 2016). In this study, the salamander

273 Bolitoglossa ramosi selected both a high temperature (28.4°C) and a high relative

274 humidity (92%) when this option was offered in the same gradient, which could be

275 explained because both high temperature and high relative humidity increase the

276 physiological performance in amphibians (Preest and Pough 2003; Kingsolver and

277 Huey 2008; Köhler et al. 2011; TitonDraft and Gomes 2017). However, in the parallel

278 gradient, where the salamanders were faced with a selection between a high

279 temperature with a low relative humidity, and a low temperature with a high

280 relative humidity, they now chose a humidity greater than 90% even at a low

281 temperature (17.2°C) (Fig. 1). Since the salamanders preferred a high relative

282 humidity in both experimental gradients, hydroregulation seems to be more

283 important than thermoregulation for B. ramosi, as has been found in other anurans

284 (Brattstrom 1979; Tracy et al. 1993; Mitchell and Bergmann 2016) and urodeles

285 (Spotila 1972; Feder 1982). This result was confirmed in experiment 2, where the

286 salamanders always chose a relative humidity greater than 90% in both gradients

287 at the two different constant temperatures (Fig. 2). A high hydric environment

288 could be principally preferred for plethodontid salamanders as they lose water at

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 14 of 31

289 accelerated rates and depend on skin surface for effective respiration (Feder

290 1983; Jørgensen 1997; Gifford 2016).

291 In the gradient with a high uniform relative humidity (experiment 3), we

292 found that Bolitoglossa ramosi selected temperatures greater than those in the low

293 relative humidity gradient (Fig. 3). This result agrees with some studies that report

294 that plethodontid salamanders are able to thermoregulate behaviourally along

295 experimental thermal gradients (Spotila 1972; Feder and Pough 1975; Sievert and

296 Andreadis 2002; Camp et al. 2014), but that the hydric requirements limit their

297 thermoregulation (Feder 1982; Camp et al. 2014). According to our field data, we

298 detected a slight thermal variation in the places where we observed the

299 salamanders at night (bromeliads,Draft upper sides of leaves, under leaf litter, moist

300 soil), from 17.1°C to 21.2°C. Then, we postulate that although B. ramosi might

301 thermorregulate and select warm temperatures (until 29.7°C), this salamander

302 must accept the available temperatures instead of showing thermal preferences

303 because of the limited thermal diversity in its microhabitat. In addition,

304 lunglessness forces plethodontid salamanders to remain moist and limit their

305 thermoregulation (Feder 1982). By contrast, in the experimental gradient with a

306 lower uniform relative humidity, the salamanders chose low temperatures (Fig 3),

307 which is defined in ectotherms as behavioural hypothermia and is explained as a

308 strategy to reduce evaporative water loss or energy consumption (Malvin and

309 Wood 1991; Tracy et al. 1993; Köhler et al. 2011). These preferences of B. ramosi

310 found in the thermal and humidity gradients are in agreement with the habitat

311 preferences showed for other terrestrial salamanders (O'Donnell et al. 2014;

https://mc06.manuscriptcentral.com/cjz-pubs Page 15 of 31 Canadian Journal of Zoology

312 Lunghi et al. 2015). Nevertheless, contrary to the limited thermal variation found in

313 the field for this tropical species, temperate species show strong seasonal

314 variation in microhabitat selection, such as preferences to relatively cold and

315 humid sectors in summer but not during winter (Lunghi et al. 2015).

316 The geographic distribution and habitat preferences of Bolitoglossa ramosi

317 have been poorly studied; however, this species has been reported in the Andean

318 mountains of Colombia, occupying restricted forest patches composed of tall trees,

319 epiphytes and variable vegetation on the forest floor that maintain good humidity

320 conditions in the environment. According to our observations, B. ramosi is

321 principally arboreal, which is a trait that is highly conserved or has evolved

322 repeatedly in plethodontid salamanders,Draft as it suggests adaptive benefits,

323 particularly in tropical forests where diverse plants (e.g. bromeliads) provide moist

324 microclimates for nesting, foraging opportunities and shelter (McEntire 2016). The

325 preference of B. ramosi for a high relative humidity, around 90% (Fig. 2), is

326 concordant with the high humidity available in the field during night hours when

327 they were more surface active and abundant (above 85%). This result also agrees

328 with the preferences of Bolitoglossa paraensis (Correa et al. 2012) and several

329 species of salamanders that selected a relative humidity of the air correlated with

330 that of the microhabitat chosen in nature (Spotila 1972; Wells 2007). On the

331 contrary, the thermal preference of B. ramosi for high temperatures is not linked to

332 its low and narrow microhabitat temperatures. It is unknown if this thermal

333 preference might be related to its physiological maxima, as reported in larval

334 stream salamanders (Strickland et al. 2016), or because species select

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 16 of 31

335 temperatures at which metabolic function is optimized when given the choice

336 (Huey and Bennett 1987; Angilletta et al. 2006). The information on thermal

337 physiology is increasingly used to understand the species responses to

338 environmental changes (Lunghi et al. 2016). However, due to the notable

339 importance of humidity for B. ramosi, and plethodontid salamanders, it is also

340 fundamental to know more about the hydric tolerances and hydroregulation of

341 these species, particularly in a changing world affected not only by the rising of

342 temperature, but also for the alteration of the hydric environment (Cruz-Piedrahita

343 et al. 2018).

344 In conclusion, we found that Bolitoglossa ramosi might thermoregulate and

345 hydroregulate, as the salamandersDraft were able to select between thermal and

346 relative humidity gradients. Particularly, with respect to our three hypotheses, we

347 found that: (1) B. ramosi prefers a high relative humidity rather than a high

348 temperature; (2) the temperature does not affect the preference of B. ramosi for a

349 high relative humidity; and (3) the relative humidity affects the thermal preferences

350 of these salamanders, as they selected high temperatures when there was a high

351 relative humidity, but low temperatures at the low relative humidity gradient. Little

352 is currently known on the ecology of neotropical salamanders (Correa et al. 2012),

353 and remains to be investigated to what extent the local distribution and abundance

354 of populations or species of salamanders are influenced by the reduction in

355 shadow and humidity refuges caused by deforestation or other anthropogenic

356 effects. Furthermore, there is also a lack of studies on the physiology of

https://mc06.manuscriptcentral.com/cjz-pubs Page 17 of 31 Canadian Journal of Zoology

357 salamanders (Gifford 2016), which are necessary to assess the impacts of local or

358 global threats to the conservation of these vertebrates.

359

360 Acknowledgments

361 This research was supported by Fondo de Investigaciones de la

362 Universidad del Tolima, project number 420214. The collection permit was

363 approved by Corporación Autónoma Regional del Tolima, CORTOLIMA

364 (resolution number 3052 from December 9, 2014). Protocols and procedures for

365 this research were approved by the Bioethics Committee of the Universidad del 366 Tolima. The authors thank TeófilaDraft María Triana, Liliana Marcela Henao, Jorge Luis 367 Turriago, Sigifredo Clavijo, James Herrán and David Salgado for their valuable

368 help in this research. Finally, we are very grateful to Antonia Quinn for correcting

369 the English version of the manuscript, and Beatriz Rubio and Armando Carbonel

370 for the authorization to perform the experimental field work in their farm “Villa

371 Beatriz”, in Líbano, Tolima, Colombia.

372

373

374

375

376

377

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 18 of 31

378 References

379 Anderson, R.C.O., and Andrade, D.V. 2017. Trading heat and hops for water:

380 dehydration effects on locomotor performance, thermal limits, and

381 thermoregulatory behavior of a terrestrial toad. Ecol. Evol. 7(21): 9066-

382 9075. doi: 10.1002/ece3.3219.

383 Angilletta Jr., M.J., Bennett, A.F., Guderley, H., Navas, C.A., Seebacher, F., and

384 Wilson, R.S. 2006. Coadaptation: a unifying principle in evolutionary

385 thermal biology. Physiol. Biochem. Zool. 79(2): 282-294. doi:

386 10.1086/499990.

387 Arenas, C.M., Gómez, A., Zapata, J.D., and Delgado, J.P. 2017. Limb

388 regeneration in a direct-developingDraft terrestrial salamander, Bolitoglossa

389 ramosi (Caudata: Plethodontidae). Regeneration, 4(4): 227-235. doi:

390 10.1002/reg2.93.

391 Balogová, M., and Gvoždík. L. 2015. Can newts cope with the heat? Disparate

392 thermoregulatory strategies of two sympatric species in water. PLoS ONE,

393 10(5): e0130918. doi: 10.1371/journal.pone.0130918.

394 Basile, M., Romano, A., Costa, A., Posillico, M., Roger, D.S., Crisci, A., Raimondi,

395 R., Altea, T., Garfì, V., Santopuoli, G., Marchetti, M., Salvidio, S., Cinti, B.,

396 and Matteucci, G. 2017. Seasonality and microhabitat selection in a forest-

397 dwelling salamander. Sci. Nat. 104(9-10): 80. doi 10.1007/s00114-017-

398 1500-6.

https://mc06.manuscriptcentral.com/cjz-pubs Page 19 of 31 Canadian Journal of Zoology

399 Brame Jr., A.H., and Wake, D.B. 1972. New species of salamanders (genus

400 Bolitoglossa) from Colombia, Ecuador and Panamá. Contributions in

401 Science, 219: 1-34.

402 Brattstrom, B.H. 1963. A preliminary review of the thermal requirements of

403 amphibians. Ecology, 44(2): 238-255. doi: 10.2307/1932171.

404 Brattstrom, B.H. 1979. Amphibian temperature regulation studies in the field and

405 laboratory. Am. Zool. 19(1): 345-356. Available from

406 http://www.jstor.org/stable/3882441 [accessed 4 March 2017].

407 Buckley, L.B., and Jetz, W. 2007. Environmental and historical constraints on 408 global patterns of amphibianDraft richness. Proc. R. Soc. Lond. B Biol. Sci. 274: 409 1167-1173. doi:10.1098/rspb.2006.0436.

410 Camp, C.D., Wooten, J.A., Jensen, J.B., and Bartek, D.F. 2014. Role of

411 temperature in determining relative abundance in cave twilight zones by two

412 species of lungless salamander (family Plethodontidae). Can. J. Zool. 92(2):

413 119-127. doi:10.1139/cjz-2013-0178.

414 Castro, F., Herrera, M.S., and Acosta-Galvis, A. 2004. Bolitoglossa ramosi. The

415 IUCN red list threatened species: e.T59198A11886496. Available from

416 http://www.iucnredlist.org/details/59198/0 [accessed 22 May 2017].

417 Catenazzi, A. 2016. Ecological implications of metabolic compensation at low

418 temperatures in salamanders. PeerJ, 4: e2072. doi: 10.7717/peerj.2072.

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 20 of 31

419 Chong, R.A., and Mueller, R.L. 2012. Low metabolic rates in salamanders are

420 correlated with weak selective constraints on mitochondrial genes.

421 Evolution, 67(3): 894-899. doi:10.1111/j.1558-5646.2012.01830.x.

422 Correa, F.S., Neckel-Oliveira, S., and Rodrigues, L.C. 2012. Influence of climatic

423 variables on the abundance of a Brazilian salamander (Bolitoglossa

424 paraensis) in Santa Bárbara Do Pará, Brazil. S. Am. J. Herpetol. 7(1): 41-

425 46. doi:10.2994/057.007.0105.

426 Cortolima. 2009. Agenda ambiental del municipio de Líbano: documento técnico.

427 Available from 428 https://www.cortolima.gov.co/sites/default/files/images/stories/centro_docuDraft 429 mentos/estudios/agendas/2009_Agenda_Ambiental_del_Municipio_de_Lib

430 ano.pdf. [accessed 20 April 2017].

431 Cruz, E.X., Galindo, C.A., and Bernal, M.H. 2016. Dependencia térmica de la

432 salamandra endémica de Colombia Bolitoglossa ramosi (Caudata,

433 Plethodontidae). Iheringia, Sér. Zool. 106: e2016018. doi:10.1590/1678-

434 4766e2016018.

435 Cruz-Piedrahita, C., Navas, C.A., and Crawford, A.J. 2018. Life on the edge: a

436 comparative study of ecophysiological adaptations of frogs to tropical

437 semiarid environment. Physiol. Biochem. Zool. 91(1): 740-756. doi:

438 10.1086/695705.

439 Feder, M.E. 1982. Thermal ecology of neotropical lungless salamanders

440 (Amphibia: Plethodontidae): environmental temperatures and behavioral

441 responses. Ecology, 63(6): 1665-1674. doi:10.2307/1940108.

https://mc06.manuscriptcentral.com/cjz-pubs Page 21 of 31 Canadian Journal of Zoology

442 Feder, M.E. 1983. Integrating the ecology and physiology of plethodontid

443 salamanders. Herpetol. 39(3): 291-310. Available from

444 http://www.jstor.org/stable/3892572. [accessed 13 March 2017].

445 Feder, M.E., and Burggren, W.W. 1985. Cutaneous gas exchange in vertebrates:

446 design, patterns, control and implications. Biol. Rev. Camb. Philos. Soc.

447 60(1): 1-45. doi:10.1111/j.1469-185X.1985.tb00416.x.

448 Feder, M.E., and Londos, P.L. 1984. Hydric constrains upon foraging in a

449 terrestrial salamander, Desmognathus ochrophaeus (Amphibian:

450 Plethodontidae). Oecologia, 64(3): 413-418. doi:10.1007/BF00379141.

451 Feder, M.E., and Pough, F.H. 1975.Draft Temperature selection by the red-backed 452 salamander, Plethodon c. cinereus (Green) (Caudata: Plethodontidae).

453 Comp. Biochem. Physiol. A Comp. Physiol. 50(1): 91-98.

454 doi:10.1016/S0010-406X(75)80207-6.

455 Fraser, D.F. 1976. Empirical evaluation of the hypothesis of food competition in

456 salamanders of the genus Plethodon. Ecology, 57(3): 459-471. doi:

457 10.2307/1936431.

458 Gatz, R.N., Crawford Jr., E.C., and Piiper, J. 1975. Kinetics of inert gas

459 equilibration in an exclusively skin-breathing salamander, Desmognathus

460 fuscus. Respir. Physiol. 24(1): 15-29. doi:10.1016/0034-5687(75)90118-8.

461 Gifford, M.E. 2016. Physiology of plethodontid salamanders: a call for increased

462 efforts. Copeia, 104(1): 42-51. doi:10.1643/OT-14-223.

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 22 of 31

463 Huey, R.B., and Bennet, A.F. 1987. Phylogenetic studies of coadaptation:

464 preferred temperatures versus optimal performance temperatures of lizards.

465 Evolution, 41(5): 1098-1115. doi: 10.2307/2409194.

466 Hutchison, V., and Dupré, R. 1992. Thermoregulation. In Environmental

467 physiology of the amphibians. Edited by M.E. Feder and W.W. Burggren.

468 University of Chicago Press, Chicago. pp 206-249.

469 Jørgensen, C.B. 1997. 200 Years of amphibian water economy: from Robert

470 Townson to the present. Biol. Rev. Camb. Philos. Soc. 72(2): 153-237.

471 doi:10.1017/S0006323196004963.

472 Kingsolver, J.G., and Huey, R.B. 2008.Draft Size, temperature, and fitness: three rules. 473 Evol. Ecol. Res. 10: 251-268.

474 Kleeberger, S.R., and Werner. J.K. 1982. Home Range and homing behavior of

475 Plethodon cinereus in Northern Michigan. Copeia, 1982(2): 409-415.

476 doi:10.2307/1444622.

477 Köhler, A., Sadowska, J., Olszewska, J., Trzeciak, P., Berger-Tal O., and Tracy,

478 C.R. 2011. Staying warm or moist? Operative temperature and thermal

479 preferences of common frogs (Rana temporaria), and effects on locomotion.

480 Herpetol. J. 21: 17-26.

481 Lehner, P.N. 1996. Handbook of ethological methods. Cambridge University

482 Press, Cambridge.

https://mc06.manuscriptcentral.com/cjz-pubs Page 23 of 31 Canadian Journal of Zoology

483 Liebgold, E.B., Brodie, E.D., and Cabe, P.R. 2011. Female philopatry and male-

484 biased dispersal in a direct-developing salamander, Plethodon cinereus.

485 Mol. Ecol. 20(2): 249-257. doi: 10.1111/j.1365-294X.2010.04946.x.

486 Lunghi, E., Manenti, R., and Ficetola, G.F. 2015. Seasonal variation in

487 microhabitat of salamanders: environmental variation or shift of habitat

488 selection?. PeerJ, 3: e1122. doi: 10.7717/peerj.1122.

489 Lunghi, E., Manenti, R., Canciani, G., Scarì, G., Pennati, R., and Ficetola, G.F.

490 2016. Thermal equilibrium and temperature differences among body

491 regions in European plethodontid salamanders. J. Therm. Biol. 60: 79-85. 492 doi: 10.1016/j.jtherbio.2016.06.010.Draft 493 Malvin, G.M., and Wood, S.C. 1991. Behavioral thermoregulation of the toad, Bufo

494 marinus: effects of air humidity. J. Exp. Zool. 258(3): 322-326.

495 doi:10.1002/jez.1402580307.

496 McEntire, K.D. 2016. Arboreal ecology of Plethodontidae: a review. Copeia,

497 104(1): 124-131. doi: 10.1643/OT-14-214.

498 Mitchell. A., and Bergmann, P.J. 2016. Thermal and moisture habitat preferences

499 do not maximize jumping performance in frogs. Funct. Ecol. 30(5): 733-742.

500 doi:10.1111/1365-2435.12535.

501 Navas, C.A., Gomes, F.R., and Carvalho, J.E. 2008. Thermal relationships and

502 exercise physiology in anuran amphibians: Integration and evolutionary

503 implications. Comp. Biochem. Physiol. A Mol. Integr. Physiol. 151(3): 344-

504 362. doi:10.1016/j.cbpa.2007.07.003.

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 24 of 31

505 O’Connor, M.P., and Tracy, C.R. 1992. Thermoregulation by juvenile toads of Bufo

506 woodhousei in the field and in the laboratory. Copeia, 1992(3): 865-876.

507 doi:10.2307/1446164.

508 O’Donnell, K.M., Thompson, F.R., and Semlitsch, R.D. 2014. Predicting variation

509 in microhabitat utilization of terrestrial salamanders. Herpetol. 70(3): 259-

510 265. doi: 10.1655/HERPETOLOGICA-D-13-00036.

511 Palacio, J.A., Muñoz, E.M., Gallo, S.M., and Rivera, M. 2006. Guía de campo:

512 anfibios y reptiles del Valle de Aburrá. Editorial Zuluaga Ltda, Medellín,

513 Colombia.

514 Peterman, W.E., and Semlitsch, R.D.Draft 2014. Spatial variation in water loss predicts 515 terrestrial salamander distribution and population dynamics. Oecologia,

516 176(2): 357-369. doi:10.1007/s00442-014-3041-4.

517 Preest, M.R., and Pough, F.H. 1989. Interaction of temperature and hydration on

518 locomotion of toads. Funct. Ecol. 3(6): 693-699. doi:10.2307/2389501.

519 Preest, M.R., and Pough, F.H. 2003. Effects of body temperature and hydration

520 state on organismal performance of toads, Bufo americanus. Physiol.

521 Biochem. Zool. 76(2): 229-239. doi:10.1086/367948.

522 Raske, M., Lewbart, G.A., Dombrowski, D.S., Hale, P., Correa, M., and Christian,

523 L.S. 2012. Body temperatures of selected amphibian and reptile species. J.

524 Zoo Wildl. Med. 43(3): 517-521. doi: 10.1638/2011-0244R.1.

https://mc06.manuscriptcentral.com/cjz-pubs Page 25 of 31 Canadian Journal of Zoology

525 Riddell, E.A., and Sears, M.W. 2015. Geographic variation of resistance to water

526 loss within two species of lungless salamanders: implications for activity.

527 Ecosphere, 6(5): 1-16. doi: 10.1890/ES14-00360.1.

528 Rogowitz, G.L., Cortes-Rivera, M., and Nieves-Puigdoller, K. 1999. Water loss,

529 cutaneous resistance, and effects of dehydration on locomotion of

530 Eleutherodactylus frogs. J. Comp. Physiol. B Biochem. Syst. Environ.

531 Physiol. 169(3): 179-186. doi:10.1007/s003600050209.

532 Shoemaker, V.H., Hillman, S.S., Hillyard, S.D., Jackson, D.C., McClanahan, L.L.,

533 Withers, P.C., and Wygoda, M.L. 1992. Exchange of water, ions, and 534 respiratory gases in terrestrialDraft amphibians. In Environmental physiology of 535 the amphibians. Edited by M.E. Feder and W.W. Burggren. University of

536 Chicago Press, Chicago. pp 125-150.

537 Sievert, L.M., and Andreadis, P.T. 2002. Differing diel patterns of temperature

538 selection in two sympatric Desmognathus. Copeia, 2002(1): 62-66.

539 doi:10.1643/0045-8511(2002)002[0062:DDPOTS]2.0.CO;2.

540 Spotila, J.R. 1972. Role of temperature and water in the ecology of lungless

541 salamanders. Ecol. Monogr. 42(1): 95-125. doi:10.2307/1942232.

542 Strickland, J.C., Pinheiro, A.P., Cecala, K.K., and Dorcas, M.E. 2016. Relationship

543 between behavioral thermoregulation and physiological function in larval

544 stream salamanders. J. Herpetol. 50(2): 239-244. doi:10.1670/15-028.

545 Titon Jr., B., and Gomes, F.R. 2017. Associations of water balance and thermal

546 sensitivity of toads with macroclimatic characteristics of geographical

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 26 of 31

547 distribution. Comp. Biochem. Physiol. A Mol. Integr. Physiol. 208: 54-60.

548 doi: 10.1016/j.cbpa.2017.03.012.

549 Titon Jr., B., Navas, C.A., Jim, J., and Gomes, F.R. 2010. Water balance and

550 locomotor performance in three species of neotropical toads that differ in

551 geographical distribution. Comp. Biochem. Physiol. A Mol. Integr. Physiol.

552 156: 129-135. doi:10.1016/j.cbpa.2010.01.009.

553 Tracy, C.R., Christian, K.A., O’Connor, M.P., and Tracy, C.R. 1993. Behavioral

554 thermoregulation by Bufo americanus: the importance of the hydric

555 environment. Herpetol. 49(3): 375-382. Available from 556 http://www.jstor.org/stable/3892756.Draft [accessed 15 April 2017]. 557 Tracy, C.R., Christian, K.A., Burnip, N., Austin, B.J., Cornall, A., Iglesias, S.,

558 Reynolds, S.J., Tixier, T., and Le Noëne, C. 2012. Thermal and hydric

559 implications of diurnal activity by a small tropical frog during the dry season.

560 Austral Ecol. 38(4): 476-483. doi:10.1111/j.1442-9993.2012.02416.x a.

561 Wells, K.D. 2007. The ecology and behavior of amphibians. The University of

562 Chicago Press, London.

563 Winters, A., and Gifford, M.E. 2013. Geographic variation in the water economy of

564 a lungless salamander. Herpetol. Conserv. Biol. 8(3):741-747. Available

565 from

566 http://www.herpconbio.org/Volume_8/Issue_3/Winters_Gifford_2013.pdf

567 [accessed 24 January 2018].

568 Zar, J.H. 1999. Biostatistical analysis. Prentice Hall, New Jersey, USA.

https://mc06.manuscriptcentral.com/cjz-pubs Page 27 of 31 Canadian Journal of Zoology

569 Figure legends 570 571 Figure 1. Box plot of thermal (white boxes) and relative humidity (gray boxes)

572 preferences of the salamander (Bolitoglossa ramosi) in the experiment 1. Box

573 limits indicate lower and upper quartiles (Q1, Q3), whiskers indicate the range of

574 the distribution excluding outliers, and the solid line indicates the median.

575

576 Figure 2. Box plot of relative humidity preferences of the salamander (Bolitoglossa

577 ramosi) in two gradients with a uniform temperature of 22°C and 25°C. Box limits

578 indicate lower and upper quartiles (Q1, Q3), whiskers indicate the range of the

579 distribution excluding outliers, andDraft the solid line indicates the median.

580

581 Figure 3. Box plot of temperature preferences of the salamander (Bolitoglossa

582 ramosi) in two gradients with a uniform relative humidity of 75% and 95%. Box

583 limits indicate lower and upper quartiles (Q1, Q3), whiskers indicate the range of

584 the distribution, and the solid line indicates the median.

585

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 28 of 31

Draft

Fig. 1.

44x36mm (300 x 300 DPI)

https://mc06.manuscriptcentral.com/cjz-pubs Page 29 of 31 Canadian Journal of Zoology

Draft

Fig. 2.

44x35mm (300 x 300 DPI)

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 30 of 31

Draft

Fig. 3.

44x36mm (300 x 300 DPI)

https://mc06.manuscriptcentral.com/cjz-pubs - - - - at 95% at 95% Gradient 2Gradient RH RH - - 0 66.7 at 75% at 75% Experiment 3 at uniform 3uniform at Experiment Gradient 1Gradient - - 8.3 8.3 - - - - 91.7 25 0 at 25°C at 25°C Gradient 2Gradient at uniform T T at uniform Experiment 2 2 Experiment ------0 at 22°C at 22°C Gradient 1Gradient that selected different ranges of relative humidity (RH) and/or humidity (RH) relative of that different ranges selected Draft Canadian Journal of Zoology 0 0 0 to to 100 100 100 100 100 100 - https://mc06.manuscriptcentral.com/cjz-pubs Gradient 2: 2: Gradient 15°C95%RH 15°C95%RH 30°C65%RH - Data were not collected under these conditions conditions these under collected not were Data - Bolitoglossa ramosi Bolitoglossa Note: 0 to 100 100 Gradient 1:Gradient Combined T and RH gradient gradient and RH TCombined 15°C65%RH 15°C65%RH 30°C95%RH

75.1-85% 75.1-85% 85.1-95% 20.1-25°C 0 100 25.1-30°C 65-75% 65-75% 15-20°C 0 0 0 0 0 - - Percentage of individuals of individuals of Percentage Table 1. Range of RH (%) Range Tof (°C) temperature (T) in the three study experiments. experiments. study in three the (T) temperature 1 Experiment Page 31 of