Selected body temperature in Mexican lizard species

Lizards are vertebrate ectotherms, which like other animals maintain their body temperature (T_b) within a relatively narrow range in order to carry out crucial physiological processes during their life cycle. The preferred body temperature (T_s) that a lizard voluntarily selecT_s in a laboratory thermal gradient provides a reasonable estimate of what a lizard would attain in the wild with a minimum of associate cosT_s in absence of constrainT_s for thermoregulation. In this study we evaluated accuracy of modified iButtons to estimate T_b and T_s of three lizard species (Sceloporus poinsettii and Sceloporus jarrovii in northeastern Durango, and Ctenosaura oaxacana in southern Oaxaca, Mexico). We used linear regression models to obtain equations for predicting T_b and T_s of these species from iButtons. ResulT_s from regression models showed that T_iButton is a good indicator of T_b for S. jarrovii and S. poinsettii during calibration process. In the same way, T_iButton is a good indicator of T_s for S. poinsettii, S. jarrovii and C. oaxacana through experimental gradient. Thus, external measuremenT_s using modified iButtons provided an accurate measurement of T_b for S. jarrovii and S. poinsettii and T_s for three species of lizards in this study. In laboratory, body temperature (T_b), and preferred temperature (T_s) obtained from S. jarrovii and S. poinsettii fell within the range of T_bs of other lizards in the family Phrynosomatidae. T_s measured for S. jarrovii, S. poinsettii, and C. oaxacana are within the range observed for lizards. Therefore, thermal preferences appear more phylogenetic that certain environmental factors present in each population of lizards of this group.

Abbreviations

T_s: Preferred body temperature; T_b: Body temperature; T_iButton: Temperature measured by modified dataloggers.

Introduction

The preferred body temperature (T_s) that a lizard voluntarily selecT_s in a laboratory thermal gradient (indoor enclosure with negligible wind and specific radiant heating regime) provides a reasonable estimate of what a lizard would attain in the wild with a minimum of associate cosT_s in absence of constrainT_s (biotic and abiotic factors) for thermoregulation [1,2].

The recent accessibility of small dataloggers has potential to revolutionize collection of thermal data in small animals [3-5]. The DS1921 Thermochron iButton (Dallas Semiconductor, TX USA) is an ideal device to quantify temperatures [6]. An iButton is a small lightweight datalogger (~14 mm diameter, 3 mm wide, and 1.5 g) that can store an enormous amount of measuremenT_s and their dates (2048 consecutive temperature readings) facilitating long term studies. In addition, iButtons are easily programmed and have a simple interface. Robert and Thompson [6], describe a technique for modifying an iButton, which involves removing the stainless-steel housing exposing the electronic board and apply a coating of plastic (Plasti Dip Spray, Performix, USA) to reduce the size and weight of iButtons. The modified iButton can then be glued to the skin of small reptiles. Given the relatively low thermal capacity of iButtons, it is possible to obtain a good estimate of T_b. Although, Robert and Thompson [6], did not verify the accuracy of iButtons to predict T_b of animals.

In this study we evaluated accuracy of modified iButtons to estimate T_b and T_s of three lizard species (Sceloporus poinsettii, Sceloporus jarrovii, and Ctenosaura oaxacana). We used linear regression models to obtain equations for predicting T_b and T_s of these species.

Materials and Methods

Study Sites and Lizard Species

The study of sympatric species S. jarrovii and S. poinsettii was conducted at Las Piedras Encimadas canyon (25°38’47”N, 103°38’40”W), 25 km northwest of Gomez Palacio, Durango, Mexico (elevation 1,425 m). The climate at the site is seasonal, with the highest temperature and rainfall occurring in spring and summer, respectively. Mean annual precipitation is 239 mm and mean annual temperature is 21°C [7]. Vegetation is within the Lechuguilla Scrub biome (Matorral Xerófilo with Agave lechuguilla [8].

Sceloporus jarrovii is a lizard species of montane foresT_s and some isolated mountain habitaT_s in the Chihuahuan and Sonoran DeserT_s with extensive rock exposure and crevices, at elevations between 1370-3550 m [9,10]. Sceloporus poinsettii is largely saxicolous, inhabiT_s vegetated areas in canyons in arid and semiarid regions within Chihuahuan Desert, as well as forested slopes of Sierra Madre Occidental [11,9].

The lizard Ctenosaura oaxacana inhabiT_s at Montecillo Santa Cruz in southern Oaxaca (16°22’05”N, 94°35’15.4”W) within Isthmus of Tehuantepec. The study site has tropical climate with a rainfall of 800 mm, and annual average temperature of 27.6 °C. The vegetation is mainly early successional and composed of shrubs and bushes [12-14]. The habitaT_s used by this species are dry forest, Nanchal (Byrsonima crassifolia), grassland, riparian vegetation, and mangrove [15].

Calibration

External data loggers generally provide accurate estimates of internal body temperatures in small ecthoterms that have limited physiological capacity to control rates of heat exchange [16].

In order to observe relationship between T_b and temperature measured by modified dataloggers (T_iButton) of adult lizards, three S. poinsettii and two S. jarrovii were collected by noosing in October 2012 and fitted with modified iButtons fixed dorsally on the back of lizards with cyanoacrylate glue [6]. Furthermore, in order to measure T_b, we inserted a K-type thermocouple into the cloaca of each lizard attached to a digital thermometer. We used masking tape to keep the thermocouple within the cloaca.

Thermochron iButtons were programmed to measure T_iButton per minute and simultaneously T_b of lizards was measured. Lizards were placed within a chamber of approximately 40 x 25 x 30 cm, equipped with incandescent light bulbs controlled by a thermostat for adjusting temperature. The temperature of each of five lizards was measured during one hour by incremenT_s of about 5 °C (within chamber) for 20-minute intervals. The experiment began with 25 °C, after 20 min of measurement increased to 30 °C, and finally at 35 °C.

Calibration temperature analysis

Analysis of covariance (ANCOVA) was performed to observe relationship between a change in T_b and T_iButton, and to determine whether there is a difference between those lizard species. Within analysis, T_b and T_iButton were assigned as covariates and the fixed factor was the species. Then, we developed a regression model for each species. The models met assumptions of normality and homoscedasticity of residuals. All data analysis and figure were performed using statistical package R [17].

Selected body temperature

AdulT_s of S. poinsettii and S. jarrovii were collected by noosing in October 2012 and separated into two categories based on individuals with presence or absence of iButtons attached on the back of both lizard species. Preferred body temperatures (T_s) and iButton temperatures (T_iButton) of captive lizards were measured in a thermal gradient (S. poinsettii: n = 9 lizards without iButtons and n = 8 lizards with iButtons; and S. jarrovii: n = 10 lizards without iButtons and n = 6 lizards with iButtons). The thermal gradient consisted of a 1 × 1.5 × 0.5 m wooden enclosure and divided lengthways into six separate compartmenT_s. Heat was provided at one end of each compartment by a 250 W infrared lamp. The laboratory climate was adjusted so that at one end of thermal gradient reached 50 °C and 20 °C at other end, thus forming a continuous gradient [18]. We collected lizards four days before T_s were measured to allow lizards to acclimate for at least 18 h prior to testing. T_s of all lizards was measured every hour from 9:00 until 18:00 hours (the normal activity period of each species). We inserted a K-type thermocouple in the cloaca using the same digital thermometer to measure T_b. Simultaneously; T_iButton was measured with iButtons attached on the back of both lizard species. Preferred temperature ranges were generated with upper and lower bounds established by using central 50% of body temperatures per lizard [18]. By using central 50%, the outliers resulting from nonthermoregulatory behavior (e.g., escape or exploratory behavior) were eliminated [18].

Moreover, T_sel for C. oaxacana (n = 7 lizards) was also measured in a thermal gradient inside a wooden box with similar dimensions to those described above and with same range of temperatures between 20 °C and 50 °C. Individuals were picked by noosing each hour to record simultaneously both temperatures, one automatically registered by modified iButton (T_iButton), and the other (T_s) by a cloacal thermometer Miller and Weber®.

Selected body temperature analysis

To test whether there was significant difference between T_s (measured through a digital thermometer) of S. poinsettii and T_s of S. jarrovii, and between T_iButton and T_s in both species, were compared with a Student t test when data met requiremenT_s for use of parametric tesT_s (normality and homogeneity of variances, test Kolmogorov-Smirnov and Levene, respectively), and if not we used nonparametric Mann-Whitney test. We also performed a multiple regression analysis to determine relationship of T_iButton with T_s and compared slope for each species. Data analysis and graphics processing programs were performed with Systat 13®. The tesT_s were performed with P ≤ 0.05 and means are expressed ± 1 SE, n and total range.

Moreover, for C. oaxacana a Wilcoxon two-sample test was performed in order to test whether there was significant difference between T_s measured through a cloacal thermometer and T_iButton measured by modified iButtons.

All lizards used in this study were marked with a nontoxic paint to identify lizards that had already been measured so that repeat measuremenT_s were not made and then released at the exact site of capture.

ResulT_s

Calibration

The mean T_b and mean T_iButton of three individuals of S. poinsettii was 29.5 ± 0.31 °C (range, 23.1 - 36.5 °C, n = 118) and 31.1 ± 0.33 °C (range, 23.5 – 38.0 °C, n = 120), respectively. Moreover in two individuals of S. jarrovii, mean T_b and mean T_iButton was 31.1 ± 0.32 °C (range, 23.7 – 37.0 °C, n = 118) and 31.9 ± 0.33 °C (range, 24.0 – 39.0 °C, n = 120), respectively.

The ANCOVA analysis showed a significant difference between two species (F1, 297 = 70.71, P < 0.0001), so we proceeded to perform a regression model for each species (Figure 1). The model met assumption of normality of residuals and homoscedasticity. The linear equation obtained for S. poinsettii was T_b = 1.29 + 0.91 x T_iButton (n = 180, R2 = 0.95, P < 0.0001), and S. jarrovii was T_b = 1.38 + 0.93 x T_iButton (n = 120, R2 = 0.95, P < 0.0001).

Figure 1: Relationship between iButton temperature (T_iButton) and body temperature (T_b) of Sceloporus jarrovii and Sceloporus poinsettii. The observation of calibration curve for S. jarrovii is shown as open circles with fitted regression as a dashed line, and a solid circle with fitted regression as a solid line for S. poinsettii.

Selected body temperature

The mean T_s and mean T_iButton of S. poinsettii was 32.3 ± 0.60 °C (n = 76, range, 19.8 – 40.0 °C) and 33.0 ± 0.66 °C (n = 76, range, 19.0 – 45.0 °C), respectively. Moreover for S. jarrovii, mean T_s and mean T_iButton was 32.7 ± 0.44 °C (n = 57, range, 21.7 – 39.8 °C) and 33.0 ± 0.59 °C (n = 57, range, 20.5 – 41.0 °C), respectively.

A Mann-Whitney U-test on T_s between S. poinsettii and S. jarrovii revealed no significant differences (W = 5,842, P = 0.41). Furthermore, a Student t-test revealed no significant differences between T_s and T_iButton for S. poinsettii (t = 0.67, df = 150, P = 0.501), and a Mann-Whitney U-test revealed no significant differences between T_s and T_iButton for S. jarrovii (W = 5,247.5, P = 0.48).

Both S. poinsettii as S. jarrovii T_s was positively related with T_iButton (Figure 2). The linear equation obtained for S. poinsettii was T_s = 1.08 x T_iButton – 2.20 (n = 76, R2 = 0.98, P < 0.0001), and S. jarrovii was T_s = 1.28 x T_iButton – 9.13 (n = 57, R2 = 0.93, P < 0.0001).

Figure 2: Relationship between iButton temperature (T_iButton) and selected temperature (T_s) of Sceloporus jarrovii and Sceloporus poinsettii. The observation of curve for S. poinsettii is shown as solid circles with fitted regression as a solid line, and an open triangle with fitted regression as a dashed line for S. jarrovii.

Furthermore, mean T_s and mean T_iButton of C. oaxacana was 32.1 ± 0.6 °C (n = 37, range, 13.1 – 39.5 °C) and 32.1 ± 1.0 °C (n = 37, range, 14.1 – 39.1 °C), respectively; which were not significantly different (Mann-Whitney U-test, W = 720, P = 0.7), and we proceeded to perform a regression model between T_s and T_iButton (Figure 3) and relationship was positive. The linear equation obtained for C. oaxacana was T_s = 0.75 x T_iButton + 7.16 (n = 37, R2 = 0.93, P = 0.02).

Figure 3: Relationship between iButton temperature (T_iButton) and selected body temperature (T_s) of Ctenosaura oaxacana. Observations of curve are shown as solid diamonds with fitted regression as solid line.

Discussion

In laboratory, body temperature (T_b), and preferred temperature (T_s) obtained from S. jarrovii and S. poinsettii fell within the range of T_bs of other lizards in the family Phrynosomatidae, with a mean T_b of 35.2 °C [19-21]. However, there is substantial variation in T_bs among species and values range from 26.8 and 41.5 °C. In the present study, T_b (calibration) for S. jarrovii and S. poinsettii was 31.1 and 29.5 °C, respectively; whereas T_iButton was 31.9 and 31.1 °C, respectively. Furthermore, T_s for S. jarrovii and S. poinsettii were 32.7 and 32.3 °C, respectively; whereas T_iButton was 33.0 and 33.0 ºC, respectively. Moreover, T_s and T_iButton for C. oaxacana were 32.1 and 32.1 °C, respectively. T_s measured for S. jarrovii, S. poinsettii, and C. oaxacana are within the range observed for lizards, which generally approaches 30 °C [22,23]. Therefore, thermal preferences appear more phylogenetic that certain environmental factors present in each population of lizards of this group.

In this study there was a significant relationship between T_iButton and T_b for S. jarrovii and S. poinsettii, as well between T_iButton and T_s for three species of lizards. ResulT_s from regression models showed that T_iButton is a good indicator of T_b for S. jarrovii and S. poinsettii during calibration process. In the same way, T_iButton is a good indicator of T_s for S. poinsettii, S. jarrovii and C. oaxacana through experiment gradient. External and internal temperature measuremenT_s were highly correlated. Thus, external measuremenT_s using modified iButtons provided an accurate measurement of T_b for S. jarrovii and S. poinsettii and T_s for three species of lizards in this study.

Our data confirm that iButtons must be calibrated prior to placing on the lizards. The failure to do so can result in the underestimation of body temperature by as much as 1.4 ºC [24]. Slopes and R2 data of these study suggest that a linear fit to calibration data is adequate for physiological temperature ranges. The modified iButtons are thus a convenient device for making these preliminary studies and should aid in improving the quality of selected temperature data and other metabolic data obtained in laboratory.

Funding for this project was provided to HG by Instituto de Ecología, A. C. (project: 20021/10050). Permit SEMARNAT-SPGA/DGVS/07946/08.

1. Huey RB, Slatkin M (1976) Costs and benefits of lizard thermoregulation. Q Rev Biol 51: 363-384. Link: https://goo.gl/o1uhYB
2. Pough FH, Gans C (1982) The vocabulary of reptilian thermoregulation. In: Gans, C. and F.H. Pough (eds) Biology of the Reptiles, Vol. 12: 17-23. Academic Press, London. Link: https://goo.gl/Hv7q5Q 
3. Muller JM, Rakestraw DL (1994) Evaluation of a new miniature temperature data logger. Herpetol Rev 26: 22-23.  Link: https://goo.gl/nDMGAf
4. Angilletta MJ Jr, Krochmal AR (2003) The Thermochron: A Truly Miniature and Inexpensive Temperature-Logger. Herpetol Rev 34: 31-32. Link: https://goo.gl/cacCDo
5. Angilletta MJ, Jr (2009) Thermal adaptation: A theoretical and empirical synthesis. Oxford University Press, USA. Link: https://goo.gl/1Cq458
6. Robert KA., Thompson MB (2003) Reconstructing thermochron iButtons to reduce size and weight as a new technique in the study of small animal thermal biology. Herpetol Rev 34: 130-132. Link: https://goo.gl/6sS51c
7. CONAGUA-SARH (2006) Dirección de hidrología- Departamento de cálculo hidrométrico y climatológico. Durango: Estación Lerdo.
8. Rzedowski J (1981) Vegetación de México. Limusa, México. Link: https://goo.gl/wukm2r
9. Lemos-Espinal JA, Smith HM (2007) Amphibians and reptiles of the state of Chihuahua, Mexico. Universidad Nacional Autónoma de México y Comisión Nacional para el Conocimiento y Uso de la Biodiversidad, México.
10. Gadsden H, Estrada-Rodríguez JL (2007) Ecology of the spiny lizard Sceloporus jarrovii in the central Chihuahuan desert. Southwest Nat 52: 600-608. Link: https://goo.gl/gmbsDJ
11. Gadsden H, Rodríguez-Romero FJ, Méndez-de la Cruz FR, Gil-Martínez R (2005) Ciclo reproductor de Sceloporus poinsettii Baird y Girard 1852 (Squamata: Phrynosomatidae) en el centro del desierto Chihuahuense, México. Acta Zool Mex (Nueva Ser) 21: 93-107. Link: https://goo.gl/kY5ZSX
12. Pérez-García E, Meave J, Gallardo C (2001) Vegetación y flora de la región de Nizanda, Itsmo de Tehuantepec, Oaxaca, México. Acta Bot Mex 56: 19-88. Link: https://goo.gl/4b9B5D
13. Farias V (2004) Spatio-temporal ecology and habitat selection of the critically endangered tropical hare (Lepus flavigularis) in Oaxaca, Mexicio. PhD Dissertation, Massachusetts University, USA. Link: https://goo.gl/nMCEf9
14. Sántiz LE (2005) Selección de hábitat y densidad de la liebre del Istmo Lepus flavigularis (Wagner 1844) en Oaxaca México. Master Dissertation, Instituto de Ecología, A.C. México.
15. Rioja T, Carrillo-Reyes A, Espinoza-Medinilla E, López-Mendoza S (2012) Basic ecology of the Oaxacan Spiny-tailed Iguana Ctenosaura oaxacana (Squamata: Iguanidae), in Oaxaca, Mexico. Rev Biol Trop 60: 1613-1619. Link: https://goo.gl/xM6XLx
16. Bakken GS (1992) Measurement and application of operative and standard temperatures in ecology. Amer Zool 32: 194-216. Link: https://goo.gl/fa41Hx
17. R Development Core Team (2012) R: a Language and environment for statistical computing. In, p., Vienna, Austria. Link: https://www.Rproject.org
18. Hertz PE, Huey BR, Stevenson RD (1993) Evaluating temperature regulation by fied-active ectotherms: the fallacy of the inappropriate questions. Amer Nat 142: 796-818. Link: https://goo.gl/ppRDzS
19. Sinervo B, Méndez-de la Cruz F, Miles DB, Heulin B, Bastiaans E, et al. (2010) Erosion of lizard diversity by climate change and altered thermal niches. Science 328: 894-899. Link: https://goo.gl/QvXczV
20. Gadsden H, Castañeda G, Huitrón-Ramírez RA, Zapata-Aguilera SA, Ruíz S, et al. (2018) Ecology of the Lagoon Spiny Lizard, Sceloporus gadsdeni, from the central Chihuahuan Desert, Mexico. Phyllomedusa17: 181-193.
21. Lara-Resendiz RA (2017) A quantitative analysis of the state of knowledge of the thermal ecophysiology of reptiles in Mexico. Árido-Ciencia 2: 36-47.
22. Angilletta Jr MJ (2009) Thermal adaptation: A theoretical and empirical synthesis. Oxford University Press, USA. Link: https://goo.gl/Fnvn3v
23. Gadsden H, Castañeda G, Huitrón-Ramírez RA (2015) Sceloporus cyanostictus. Field and preferred body temperature. Herpetol Rev 46:436–437. Link: https://goo.gl/xwLM9T
24. Lovegrove BG (2009) Modification and miniaturization of thermochron iButtons for surgical implantation into small animals. J Comp Physiol B 179: 451-458. Link: https://goo.gl/RL5VQL