Thermal niches and activity periods in syntopic Phymaturus and Liolaemus lizards from the Andes, Argentina

Abstract

Explanations for differences in thermal biology within and between species of lizards employ concepts of phylogenetic inertia and plasticity. We compared the thermal biology of three liolaemid species in the Andean highlands in Argentina: two allopatric congeners (Phymaturus williamsi and P. aguanegra) each in syntopy with Liolaemus parvus. We predicted intra and inter-generic differences in ecophysiological traits and periods of activity at both sites, ecotypic differences between the (labile) Liolaemus populations, but predicted no interspecific differences between the (putatively conservative) Phymaturus. We determined the operative temperatures (T _e), field body temperatures (T _b), preferred temperatures (T _pref), effectiveness of thermoregulation (E), and activity periods. As expected, P. williamsi differed from L. parvus in T _b, T _pref, and activity periods, likely as result of niche segregation. Contrary to predictions, the Phymaturus populations exhibited differentiation in T _b and T _pref, while L. parvus populations differed in T _pref and E. Accordingly, Phymaturus species tend to be effective thermoregulators whereas L. parvus populations behave as good thermoregulators or thermoconformers depending on thermal conditions in fluctuating habitats. Phymaturus may be less evolutionarily conservative than previously suggested. The suite of co-evolving traits affecting thermal ecology may not be collectively conservative nor labile but rather a continuum between both evolutionary paths.

Key words
Phymaturus; Liolaemus; thermal niche; activity periods; Andes; Argentina

INTRODUCTION

Body temperatures (T _b) in non-avian reptiles influence physiological and behavioral processes, and consequently may have a profound impact on survival and fitness (Avery 1982, Huey 1982). Since T _b is also directly influenced by varying environmental temperatures (Angilletta 2009), many lizards reduce variation in their T _b through behavioral adjustments (Gvoždík 2012). In this sense, lizards have evolved a variety of mechanisms to control T _b (Cowles & Bogert 1944, Nelson et al. 1984, Stevenson 1985, Avery 1978) resulting in differences in the effectiveness of thermoregulation among species and populations (Hertz et al. 1993).

The achievement and maintenance of T _b within a range or near thermal optima that allows activity and enhances physiological performance depends on both the availability of suitable microhabitats and the effectiveness of thermoregulation (E, sensu Hertz et al. 1993, Beaupre 1995, Huey et al. 2003). However, the opportunity to bask at suitable temperatures will also depend on the risks and costs of thermoregulation, such as a high vulnerability to predation and the allocation of time for basking to the detriment of time for feeding and social activity (Adolph 1990, Blouin-Demers & Weatherhead 2001a, Gvoždík 2002). Maintenance of high T _b during the activity period and the overt usage of basking are common in diurnal lizards, but even more remarkable in those inhabiting climates with a wide diurnal fluctuation of ambient temperatures and a marked seasonality such as those present along the Andes range (Veblen et al. 2007). Under such an energetically challenging scenario, lizards that experience thermal extremes are expected to exhibit variation in their ability to thermoregulate (Veloso et al. 2007). Some authors pointed out that thermal physiology in closely-related taxa may be relatively similar as a result of a co-evolutionary process despite their subsequent ecological divergence even in allopatry (Hertz et al. 1983, Angilletta et al. 2002). However, differences in the effectiveness of thermoregulation in syntopic species subjected to similar thermal regimes would suggest different physiological requirements or resource use (Smith & French 2017).

Thermal preferences may differ among sex or age groups in each population (Ortega et al. 2016), because resource partitioning would occur according to a hierarchic system in which males would have access to better resources than females or juveniles (e.g., suitable microhabitats for basking; sensu Vidal et al. 2010, Gómez Ales et al. 2017). In addition, since the males are larger than females in many species (Stamps 1983), it is possible that males achieved a higher accuracy and effectiveness of thermoregulation (Sagonas et al. 2013). Larger size and body mass affects the rate of heat exchange in the individual-environment interaction (Tracy 1982). Although, a more reliable individual-environment interaction indicator could be the body condition (scaled mass index, M_i, sensu Peig & Green 2009), which is an estimate of an individual’s fitness (Green 2001). Hence, the variation of body condition can be directly influenced by local factors (e.g., intraspecific competition, resource partitioning and predation pressure) with pronounced effects on several natural history traits including the thermal preferences (Siliceo-Cantero & García 2014).

Lizard thermal physiology can be evolutionarily conservative such that closely related taxa are expected to show similar thermal biological responses even when living in distinct habitats (Bogert 1949, Huey 1982, Díaz de la Vega-Pérez et al. 2013). This is, for example, the case for the parthenogenetic Aspidoscelis cozumela that inhabits a wide range of environmental conditions, including areas where the parental species A. angusticeps and A. deppii are absent, yet all three species exhibit similar thermal preferences (Díaz de la Vega-Pérez et al. 2013). Moreover, thermal conservatism in thermal tolerances has been reported for North American Sceloporus species across a wide elevational range (Buckley et al. 2015). However, studies of tropical Anolis argue that some taxa have readily adapted to differences in their environments showing interspecific differences in T _b, T _pref and T _max and such ecotypic differences have been observed even in closely related species (Huey 1982, Hertz et al. 1983). Nevertheless, in sibling species inhabiting contrasting habitats, the mismatches between the thermal environments and thermal optimum may lead to less activity, less energy acquisition and a sub-optimal energy budget and a consequent fitness decrease betraying slow adaptation to changes in the environments (Logan et al. 2012, 2015).

Liolaemidae is one the most diverse families of lizards, comprising more than 300 species distributed in three genera, Ctenoblepharys (1 sp.), Phymaturus (54 spp.) and Liolaemus (> 285 spp.), inhabiting a broad variety of environments in southern South America (Etheridge 1995, Abdala & Quinteros 2014, Troncoso-Palacios et al. 2018, Esquerré et al. 2019, Abdala et al. 2021, Lobo et al. 2022). This family is an appealing model to test hypotheses on the evolution of interspecific and ecotypic differences in thermoregulatory effectiveness. At present, the thermal biology studies in Liolaemidae point out that Phymaturus is conservative in its thermal traits (Cruz et al. 2009, Gómez Ales et al. 2017, Duran et al. 2018) while observations of the genus Liolaemus supports the labile hypothesis as they exhibit greater flexibility that may vary according to lineage specialization or habitat characteristics (Rodríguez-Serrano et al. 2009, Bonino et al. 2011, Medina et al. 2012, Moreno-Azócar et al. 2013). Indeed, Phymaturus lizards are more selective in their habitat use, being strictly herbivorous and viviparous, and occur only on rocky promontories in mountain habitat (Cei 1986, Habit & Ortiz 1994, Espinoza et al. 2004, Laspiur 2010, Castro et al. 2013, Boretto et al. 2014, 2018). Due to these features and restrictive discontinuous distribution, all Phymaturus species are considered “vulnerable” in the Argentinean Red List (Abdala et al. 2012). In contrast, the genus Liolaemus, exhibits a substantial variation in natural history traits such as diet, habitat, and reproductive mode (oviparous and viviparous), and has developed a broad range of adaptive responses to different thermal environments (Cei 1986, Scolaro 2005, Medina et al. 2011). Species of Phymaturus and Liolaemus often occur in syntopy on latitudinal gradients along the Andean mountain range in Argentina and Chile (Díaz-Gómez 2009) offering an opportunity to study different strategies for thermoregulation under similar environmental conditions (Rodríguez-Serrano et al. 2009, Vitt & Caldwell 2014). Lizards competing in syntopy tend to select distinctive thermal microhabitats (sensu Hertz et al. 1993, Scheers & Van Damme 2002) as a result of niche segregation and it is presumed that syntopic species will experience different thermal qualities of the environment (Magnusson et al. 1979, Paterson & Blouin-Demers 2016).

Herein, we describe aspects of thermal biology, activity periods, and effectiveness of thermoregulation within each of two pairs of syntopic Phymaturus and Liolaemus populations occurring in harsh environmental conditions in the highlands of the Andes, Argentina. We address three classical concepts on thermal biology and ecology. First, syntopic species exhibit niche segregation in thermal traits, manifest in their use of heat sources (thigmothermy and heliothermy), effectiveness of thermoregulation, and the timing of activity periods to reduce competitive interactions. Secondly, Phymaturus aguanegra and P. williamsi are more conservative than Liolaemus parvus populations in their thermal biology based on previous studies (Cruz et al. 2009, Rodríguez-Serrano et al. 2009, Bonino et al. 2011) and we predict that the T _b, T _pref, effectiveness of thermoregulation and thermoregulatory behaviour (thigmothermy and heliothermy) will differ between populations of L. parvus but not between Phymaturus species. Finally, we consider that the T _b and T _pref are explained by M_i and affects the accuracy (d_b) and the effectiveness of thermoregulation (E). Since these lizards could exhibit a degree of sexual size dimorphism biased toward larger males (Cabezas-Cartes et al. 2010, Castillo et al. 2011, Castro et al. 2011), and because body condition is a coefficient of the relative size of energy stores compared with structural body components (Peig & Green 2009), we expect that males and those showing a greater body condition may have higher T _b and T _pref and would be more accurate and effective thermoregulators. Because thermoregulation directly affects activity in lizards, this multi-species approach may allow greater understanding of the mechanisms involved in the resource partitioning in terms of temperature and activity in coexisting species. It will allow the exploration of the thermoregulatory strategies that lizards can adopt under different conditions in montane environments in the central Andes.

MATERIALS AND METHODS

Study areas, climate, and lizards

We compared thermoregulatory behaviour and effectiveness of thermoregulation among four populations: two allopatric species of Phymaturus at two sites, each syntopic with a local population of Liolaemus parvus. Specifically, we studied Phymaturus aguanegra at Paso Agua Negra (Iglesia Department; -30°23’S; -69°34’W, 3000 m asl) and P. williamsi at Quebrada Vallecito (Calingasta Department; -31°11’S -69°42’W, 3000 m asl), two microendemics separated by ~100 km (Lobo et al. 2013, Figure 1) and embedded in the highlands of San Juan Province, Argentina. Each Phymaturus species is syntopic with one of two separate populations of L. parvus, a species widely distributed from central-west La Rioja Province to northwest Mendoza Province (Quinteros et al. 2008). The three species are viviparous, but the Phymaturus are larger (snout-vent length: SVL ≤ 115 mm, Lobo et al. 2013) than L. parvus (SVL ≤ 74 mm, Quinteros et al. 2008).

Figure 1
Capture sites of Phymaturus aguanegra, Paso Agua Negra, Iglesia Department (white dot) and Phymaturus williamsi, Quebrada Vallecito, Calingasta Department (white star), both species in syntopy with Liolaemus parvus, San Juan Province, west-central Argentina.

These study areas belong to the Puna phytogeographic region characterized by xerophyllous plants such as the shrubs Adesmia pinnifolia and Ephedra multiflora, the cactus Lobivia formosa, and the dwarf shrub Artemisia mendozana (Cabrera & Willink 1973). The climate at both capture sites corresponds to cold arid desert (BWk, Köeppen 1948, Peel et al. 2007) with rainfall ocurring mainly in winter, monthly mean temperatures < 18°C even in the warmest season, and prolonged snowfalls and mud slides in spring (Figure 2).

Figure 2
Monthly mean of rainfall (mm, gray bars) and temperatures (°C, polygons) at Paso Agua Negra and Quebrada Vallecito during the study (2005 – 2010). Greater rainfall during cooler winter months results in greater moisture availability. Climatic data provided by 3CN database (Centro de Investigaciones del Mar y Atmósfera).

Lizards were captured during the activity season from late spring to early autumn of the southern hemisphere. All specimens were caught by hand or noose between 10:00 and 19:00 h, corresponding to the entire daily activity period of the three species (Laspiur 2010, Ausas 2011). Phymaturus aguanegra (N = 60; 25 males, 19 females, 16 juveniles) and Liolaemus parvus (N = 48; 24 males, 18 females, 6 juveniles) were captured at Paso Agua Negra during November 2005, December 2005, February 2006, April 2007 and March 2011. Meanwhile at Quebrada Vallecito, samples of Phymaturus williamsi (N = 84; 27 males, 40 females, 17 juveniles) and Liolaemus parvus (N = 87; 40 males, 37 females, 10 juveniles) were captured during October 2008, December 2008, and April 2009. Capturing and handling were conducted in accordance with international standards on animal welfare, being compliant with Argentinian regulations. All individuals were collected under permits Exp. N° 1300-4047 granted to JCA. After laboratory experiments, the captured specimens were euthanized using intraperitoneal injection of sodium pentobarbital Euthanyle ©, fixed in formalin (10%), preserved in ethanol (70%) and deposited into the herpetological collection housed in Centro Regional Universitario Bariloche (CRUB), Universidad Nacional del Comahue.

Field data

The following microenvironmental temperatures were recorded at the capture site for each individual lizard in order to determine the main heat source used: substratum temperature on rocks, bare soil or beneath dwarf shrubs (T _s, TES TP-K03 substrate probe), and air temperature 1 cm above the ground (T _a, TES TP-K02 gas probe). Each instrument was connected to a TES 1302 thermometer (TES ® Electrical Electronic Corp., Taipei, Taiwan, ± 0.01°C).

The operative temperatures (T _e, sensu Bakken 1992) were obtained using copper models set in the most heterogeneous and representative microhabitats used by the species in each field site. At Agua Negra, four common microhabitats were used by P. aguanegra and L. parvus (basaltic rock, bare soil, shade beneath shrubs and exposed crevices): 16 models were used during sampling, 8 models for Phymaturus were localized in the 4 microhabitats simultaneously in two different zones of the field site (4 microhabitats x 2 replicates) and the other 8 models for Liolaemus using the same methodology (4 microhabitats x 2 replicates). At Quebrada Vallecito, we identified the six representative microhabitats used by P. williamsi (feldspathic rock, basaltic rock, shade beneath shrubs, bare soil, crevices, and at crevice opening with the model partially exposed), whereas L. parvus commonly used seven (basaltic rocks, leaf litter, shade beneath shrubs, mossy rock, bare soil, weathered small rock, and expose crevices): 39 models were used during sampling, 18 models for Phymaturus (6 microhabitats x 3 replicates) and 21 for Liolaemus (7 microhabitats x 3 replicates). The models were designed using simultaneous comparisons of the T _b of live Phymaturus williamsi and L. parvus with copper models of varying sizes and colours. During the comparisons, catheter probes TES TP-K01 (TES Electrical Electronic Corp., Taipei, Taiwan ± 0.01°C) were used simultaneously to register the temperatures of both the models and the lizards. The calibration experiments were conducted during three consecutive hours and we calculated the animal-model correlation for each type of model. The best-fit correlation was observed for a flat-black hollow copper cylinder with dimensions 90 mm length x 20 mm diameter x 10 mm thickness for Phymaturus (Spearman Rank correlation, T _b vs. model, R = 0.93, N = 122, P < 0.001) and 60 mm length x 10 mm diameter x 8 mm thickness for Liolaemus (Spearman Rank correlation, T _b vs. model, R = 0.99, N = 122, P < 0.0001). Measures of T _e were recorded during lizard activity from 10:00 to 19:00 h. Mean temperatures of each model were used to estimate the temperature that thermoconforming lizards would achieve at each field site (T _e).

Body temperatures (T _b, sensu Pough & Gans 1982) were taken only in active lizards using a catheter probe TES TP-K01 (1.62 mm diameter) introduced ca. 5 mm into the cloaca. Individuals were handled by the head to avoid heat transfer and temperature was recorded within 20 s of handling. Time of day at capture was also recorded. Snout-vent length (SVL, vernier caliper ± 0.02 mm) and body mass (BW, 10 – 50 g Pesola ® spring scale ± 0.5 g) were also registered.

Laboratory experiments

Experiments were performed 3 to 7 days after capture on a random subsample of lizards from each site: P. aguanegra (N = 44; 20 males, 14 females, 10 juveniles) and L. parvus (N = 41; 21 males, 17 females, 3 juveniles) from Agua Negra; P. williamsi (N = 34; 14 males, 18 females, 2 juveniles) and L. parvus (N = 30; 15 males, 15 females) from Quebrada Vallecito. Lizards were placed individually in open-top terraria (200 cm length, 15 cm width, 30 cm height) with a sand substrate and a thermal gradient produced by a length-wise row of four infrared lamps (250 W, 150 W, 150 W, 100 W). The heights of the lamps above the sand were adjusted to make a linear gradient from 69° to 15°C on the substrate for lizards to thermoregulate. White fluorescent lamps along the sides of each terrarium homogenized illumination. Body temperatures were taken using ultra-thin (1 mm) catheter thermocouples located approximately 5 mm inside the cloaca and taped at the base of the lizard’s tail to prevent the thermocouple from being dislodged during the experiment. The temperature of each lizard was obtained every 10 min for 5 hr by connecting the thermocouple to a TES 1302 thermometer (TES ® Electrical Electronic Corp., Taipei, Taiwan, ± 0.01 °C) to avoid interference with their normal activities. The mean preferred body temperature (T _pref) and the inferior and superior set-point range temperatures (T _set, the interquartile range) were obtained for each lizard. In order to measure the accuracy of thermoregulation of these in their natural environment, the mean of the absolute values obtained from the deviations of T _b-i from T _set-i of each individual was calculated (individual deviation; d_b-i). The index of the average thermal quality of the habitat from the organism’s mean (d_e) was calculated as the absolute values from the deviations of mean T _e for each population with respect to set-point range (T _set) of each population and species. The effectiveness of temperature regulation, E, was calculated as 1 - (mean d_b / mean d_e) for both populations at each site (Hertz et al. 1993). The values of E range from -1 to 1, and E-values near 0 represent thermoconformers, values near 0.5 moderate thermoregulators, and values near 1 effective thermoregulators. Negative values of E occur when lizards avoid thermally high-quality habitats with T _e within the range of T _pref (Christian & Weavers 1996, Blouin-Demers & Weatherhead 2001b, Gutiérrez et al. 2010).

Statistical analyses

For comparative purposes, mean, standard error, variance, and range of temperatures were indicated. The dependence between variables was analyzed by simple or multiple stepwise regression. An analysis of variance (one-way repeated measures ANOVA) was used to determine the main heat sources used by lizards comparing the T _b with the T _s and T _a. Paired t-tests or Wilcoxon Signed Rank tests were used to detect differences between T _b and T _pref. When ANOVA effects were significant (P < 0.05), multiple post hoc comparisons were performed using the Holm-Šídák method.

The periods of activity in this study were analysed using a 4 x 8 contingency table combining frequencies calculated for each population (N = 4), and for each activity period (N = 8, see Table IV). Pearson’s Chi-square test (χ²) was performed to evaluate the significance of differences between observed versus expected frequencies under assumptions of homogeneity of site use by lizards during the activity period (null hypothesis). Post hoc analyses were performed from the adjusted residuals to obtain z-scores for all cases. Subsequently, P-values were calculated from the transformation of Chi-square values derived from multiple tests (Beasley & Schumacker 1995, García Pérez & Nuñez Antón 2003). A Bonferroni Correction was applied by dividing the theoretical P-level of significance (α = 0.05) by the number of multiple tests (N = 32). The adjusted α –level obtained for the whole model was P < 0.001.

Table I
Results of linear regressions of scaled mass index of body condition (M_i) with body temperature (T _b), preferred temperatures (T _pref), absolute values obtained from the individual deviation of T _b from T _set (d_b) and individual effectiveness of thermoregulation (E_i) of Phymaturus aguanegra and Liolaemus parvus at Paso Agua Negra, and P. williamsi and L. parvus at Quebrada Vallecito. Significant relationships are indicated in bold.

Species	Temperature	F (N)	t	P
P. aguanegra	T b	0.091 (40)	0.303	0.76
	T pref	3.073 (40)	-1.753	0.08
	db	0.755 (40)	-0.869	0.39
	Ei	0.284 (40)	0.553	0.59
L. parvus	T b	0.119 (12)	0.345	0.73
	T pref	4.974 (12)	2.230	< 0.05
	db	0.097 (12)	0.312	0.76
	Ei	0.001 (12)	-0.004	0.99
P. williamsi	T b	3.501 (33)	-1.871	0.07
	T pref	2.341 (33)	-1.530	0.13
	db	2.245 (33)	1.498	0.14
	Ei	2.449 (33)	-1.565	0.12
L. parvus	T b	0.362 (28)	-0.602	0.55
	T pref	1.262 (28)	1.123	0.27
	db	0.035 (28)	0.188	0.85
	Ei	0.772 (28)	-0.879	0.388

Table IV
Summary of Pearson’s Chi-Square test. Frequencies of activity of lizards by hour are indicated as frequencies and percentage (%). Post hoc analysis are indicated by their adjusted z-scores (Adj. z-score) and probabilities (P). Bonferroni’s correction (P > 0.001), and total sample sizes, and percentages are also indicated. Significant differences are in bold (*).

Hour of the day (h; GMT-3)		Pearson Chi-Square test ( 21 ) = 70.73, P < 0.0001				Total
		Phymaturus aguanegra (Agua Negra)	Phymaturus williamsi (Qa. Vallecito)	Liolaemus parvus (Agua Negra)	Liolaemus parvus (Qa. Vallecito)
10:00 ‒ 11:00	Frequency	3	0	0	2	5
	% within hour	60.0%	0.0%	0.0%	40.0%	100.0%
	Adj. z-score	2.11	-1.48	-1.03	0.43
	P < 0.001	0.034	0.138	0.303	0.667
11:00 ‒ 12:00	Frequency	18	14	14	2	48
	% within hour	37.5%	29.2%	29.2%	4.2%	100.0%
	Adj. z-score	2.96	-0.16	2.41	-4.44
	P < 0.001	0.003	0.872	0.015	0.00001*
12:00 ‒ 13:00	Frequency	12	25	15	19	71
	% within hour	16.9%	35.2%	21.1%	26.8%	100.0%
	Adj. z-score	-1.09	1.09	1.01	-0.93
	P < 0.001	0.275	0.275	0.312	0.352
13:00 ‒ 14:00	Frequency	14	4	9	21	48
	% within hour	29.2%	8.3%	18.8%	43.8%	100.0%
	Adj. z-score	1.42	-3.61	0.31	2.07
	P < 0.001	0.155	0.0001*	0.756	0.038
14:00 ‒ 15:00	Frequency	8	14	6	22	50
	% within hour	16.0%	28.0%	12.0%	44.0%	100.0%
	Adj. z-score	-1.05	-0.36	-1.08	2.16
	P < 0.001	0.293	0.718	0.280	0.030
15:00 ‒ 16:00	Frequency	5	17	4	11	37
	% within hour	13.5%	45.9%	10.8%	29.7%	100.0%
	Adj. z-score	-1.27	2.26	-1.11	-0.20
	P < 0.001	0.204	0.023	0.266	0.841
16:00 ‒ 17:00	Frequency	0	10	0	6	16
	% within hour	0.0%	62.5%	0.0%	37.5%	100.0%
	Adj. z-score	-2.16	2.91	-1.88	0.56
	P < 0.001	0.030	0.003	0.060	0.575
17:00 ‒ 18:00	Frequency	0	0	0	4	4
	% within hour	0.0%	0.0%	0.0%	100.0%	100.0%
	Adj. z-score	-1.05	-1.32	-0.92	2.99
	P < 0.001	0.293	0.186	0.357	0.002
Total	Frequency	60	84	48	87	279
	%	21.5%	30.1%	17.2%	31.2%	100.0%

Snout-vent length (SVL) and body mass (BW) were included in the scaled mass index of body condition in each individual (M_i , sensu Peig & Green 2009) to determine the scaled mass index of condition as an indicator of the health or quality assumed to be related to fitness. The scaled mass index was calculated as (M_i) = M_i * [SVL_o / SVL_i] ^ bSMA; where M_i and SVL_i are the BW and SVL of each individual, SVL₀ is the arithmetic mean SVL of each population, and bSMA is the standardized major axis slope from the regression of ln(BW) on ln(SVL) for each population (Peig & Green 2009). The scaling bSMA exponent was calculated directly using the software RMA v. 1.21 (Bohonak & van der Linde 2004). We then tested for the influence of M_i on thermal characteristics (T _b, T _pref, d_b and E) according to sex and age classes (males, females, and juveniles) in each population.

Variability in thermo-biological variables was described using descriptive statistics (mean ± standard deviation, minimum and maximum). Normality and variance homogeneity assumptions were tested using Kolmogorov-Smirnov’s test and Levene’s test, respectively. When normality or variance homogeneity assumptions were not met, non-parametric correlation, Mann–Whitney U-tests, and Kruskal-Wallis rank sum tests were used (Sokal & Rohlf 1969). Data were analyzed using Sigma Plot ® version 14.0 (Systat Software Inc., San José, CA), SPSS ® version 20.0 (IBM, SPSS Statistics for Windows, Armonk, NY), and figures were produced using Statistica ® version 10.0 (Statsoft Inc., Tulsa, OK).

RESULTS

Effects of body condition (M_i) on thermal traits and thermoregulation indices

Thermal traits and thermoregulation were unrelated to body condition: body temperature (T _b), accuracy (d_b) and effectiveness of thermoregulation (E_i) were not significantly correlated with (M_i) in any of the four populations (Table I). The only exception was that T _pref increased with M_i in L. parvus from Paso Agua Negra (Table I).

Comparison of T _b and T_pref among males, females and juveniles within species and populations

Paso Agua Negra T _b did not differ among males, females, and juveniles in either population (Kruskal-Wallis, H _P. _aguanegra (₂; ₆₀) = 3.167, P > 0.2; H_L. _parvus (₂; ₄₈) = 4.486, P > 0.1, Table II). Similarly, T _pref means were not different among those groups in both populations (ANOVA, F _P. _aguanegra (₂; ₄₄) = 2.338, P > 0.1; Kruskal-Wallis, H _L. _parvus (₂; ₄₁) = 4.116, P > 0.1, Table II).

Table II
Table II. Mean ± SD of body mass (BM, g), snout-vent length (SVL, mm), scaled mass index of body condition (M_i), body temperatures (T _b, °C), preferred body temperatures (T _pref, °C), absolute values obtained from the individual deviation of T _b from T _set (d_b), index of the average thermal quality of habitats (d_e), and index of effectiveness of thermoregulation (E) between sexes and juveniles in each studied species/populations. The sample sizes N and range (between parentheses) are also indicated.

	Class	BM	SVL	Mi	T b	T pref	db	de	E
P. aguanegra (Paso Agua Negra)	Males	28.27 ± 6.83 (12.0 ‒ 35.5) N = 20	95.15 ± 5.87 (82.7 ‒ 104.5) N = 25	22.80 ± 4.04 (14.4 ‒ 30.8) N = 20	30.22 ± 4.21 (14.9 ‒ 35.3) N = 25	37.03 ± 1.32 (34.4 ‒ 38.9) N = 20	4.73	10.98	0.56
	Females	22.53 ± 5.93 (11.0 ‒ 30.5) N = 13	93.06 ± 9.41 (70.9 ‒ 105.0) N = 19	20.70 ± 4.10 (14.2 ‒ 26.7) N = 13	27.00 ± 6.56 (13.2 ‒ 34.0) N = 19	37.81 ± 1.03 (35.8 ‒ 39.4) N = 14	7.97	11.51	0.30
	Juveniles	10.77 ± 6.42 (2.5 ‒ 20.9) N = 8	69.37 ± 11.13 (52.7 ‒ 89.1) N = 11	22.72 ± 3.58 (18.8 ‒ 27.7) N = 7	29.64 ± 4.70 (22.0 ‒ 35.6) N = 16	37.67 ± 1.28 (34.7 ‒ 39.7) N = 10	8.12	11.45	0.29
	Overall	23.03 ± 9.15 (2.5 ‒ 35.5) N = 41	89.27 ± 13.02 (52.7 ‒ 105.0) N = 55	22.14 ± 4.02 (14.2 ‒ 30.8) N = 40	29.05 ± 5.29 (13.2 ‒ 35.6) N = 60	37.43 ± 1.26 (34.4 ‒ 39.7) N = 44	6.53	11.26	0.41
L. parvus (Paso Agua Negra)	Males	6.30 ± 1.29 (3.5 ‒ 8.2) N = 23	65.42 ± 4.99 (55.0 ‒ 69.0) N = 7	5.76 ± 0.47 (5.1 ‒ 6.4) N = 7	33.26 ± 1.74 (29.4 ‒ 36.0) N = 24	36.72 ± 1.32 (33.3 ‒ 38.3) N = 21	2.59	8.93	0.70
	Females	5.44 ± 1.54 (3.5 ‒ 9.7) N = 17	59.16 ± 4.07 (54.0 ‒ 66.0) N = 6	5.90 ± 0.43 (5.3 ‒ 6.3) N = 4	31.15 ± 3.47 (23.2 ‒ 35.6) N = 18	37.32 ± 0.84 (35.0 ‒ 38.5) N = 17	5.29	10.27	0.48
	Juveniles	2.72 ± 1.86 (0.7 ‒ 5.0) N = 5	‒	‒	32.25 ± 3.40 (28.0 ‒ 35.5) N = 6	35.91 ± 1.24 (35.0 ‒ 37.3) N = 3	1.25	4.93	0.74
	Overall	5.58 ± 1.79 (0.7 ‒ 9.7) N = 45	62.28 ± 5.34 (54.0 ‒ 69.0) N = 14	5.75 ± 0.49 (4.8 ‒ 6.4) N = 12	32.19 ± 2.86 (23.2 ‒ 36.0) N = 48	36.91 ± 1.18 (33.3 ‒ 38.5) N = 41	3.61	9.19	0.60
P. williamsi (Q. Vallecito)	Males	35.80 ± 3.96 (27.0 ‒ 42.0) N = 15	101.40 ± 9.32 (71.0 ‒ 123.0) N = 19	33.31 ± 3.05 (29.3 ‒ 39.1) N = 14	29.35 ± 5.38 (15.8 ‒ 37.3) N = 27	35.90 ± 1.66 (32.8 ‒ 39.0) N = 15	4.91	8.86	0.44
	Females	34.88 ± 3.55 (28.0 ‒ 39.8) N = 17	101.77 ± 7.47 (76.0 ‒ 115.0) N = 36	32.88 ± 4.23 (25.4 ‒ 40.9) N = 17	31.21 ± 4.35 (20.1 ‒ 38.0) N = 40	36.50 ± 1.63 (32.2 ‒ 38.8) N = 17	5.08	9.12	0.44
	Juveniles	12.42 ± 1.52 (11.3 ‒ 13.5) N = 2	67.72 ± 10.74 (54.0 ‒ 92.0) N = 14	23.88 ± 8.03 (18.2 ‒ 29.5) N = 2	33.11 ± 5.91 (14.6 ‒ 40.0) N = 17	36.30 ± 1.11 (35.5 ‒ 7.1) N = 2	‒	‒	‒
	Overall	33.99 ± 6.56 (11.3 ‒ 42.0) N = 34	95.26 ± 16.22 (54.0 ‒123.0) N = 73	32.52 ± 4.44 (18.2 ‒ 40.9) N = 34	31.00 ± 5.15 (14.6 ‒ 40.0) N = 84	36.23 ± 1.60 (32.2 ‒ 39.0) N = 34	4.47	9.06	0.50
L. parvus (Q. Vallecito)	Males	7.05 ± 1.26 (4.5 ‒ 9.9) N = 15	64.50 ± 2.42 (60.0 ‒ 69.0) N = 15	7.08 ± 1.04 (4.5 ‒ 8.2) N = 14	33.34 ± 3.59 (24.5 ‒ 39.4) N = 40	35.08 ± 1.53 (32.2 ‒ 37.7) N = 15	2.42	2.31	-0.04
	Females	6.70 ± 1.87 (4.8 ‒ 10.5) N = 15	65.91 ± 4.05 (60.0 ‒ 74.0) N = 23	6.20 ± 1.56 (3.2 ‒ 8.8) N = 14	32.45 ± 3.75 (25.8 ‒ 41.1) N = 37	34.69 ± 1.68 (31.6 ‒ 37.7) N = 15	3.85	2.82	-0.36
	Juveniles	‒	51.87 ± 7.01 (37.0 ‒ 59) N = 8	‒	34.84 ± 3.72 (27.1 ‒ 40.1) N = 10	‒	‒	‒	‒
	Overall	6.88 ± 1.57 (4.5 ‒ 10.5) N = 30	63.24 ± 6.25 (37.0 ‒ 74.0) N = 54	6.64 ± 1.37 (3.2 ‒ 8.8) N = 28	33.14 ± 3.70 (24.5 ‒ 41.1) N = 87	34.89 ± 1.59 (31.6 ‒ 37.7) N = 30	3.13	2.57	-0.22

Quebrada Vallecito

T_b values of P. williamsi were significantly higher in juveniles than in males (Kruskal-Wallis, H (₂; ₈₄) = 8.445, P < 0.01; Dunn’s test, Q _{juveniles-males} = 2.892, P < 0.05; Q _{juveniles-females} = 2.111, P > 0.05; Q _{males-females} = 1.141, P > 0.05, Table II), whereas mean T _bs of L. parvus were not different among males, females and juveniles (ANOVA, F (₂; ₈₇) = 1.773, P < 0.17, Table II). Meanwhile, T _pref means were not different among males, females and juveniles in P. williamsi (ANOVA, F (₂; ₃₄) = 0.500, P > 0.5, Table II) or between females and males in L. parvus (Student’s t-Test, t (₁; ₃₀) = -0.654, P > 0.5, Table II). Juveniles were not analyzed in L. parvus.

Determination of the main heat sources for thermoregulation

Paso Agua Negra

Phymaturus aguanegra and L. parvus were found basking with T _b and T _s higher than T _a, while T _b and T _s were not different (Table III). T _bs of P. aguanegra increased only with T _ss (Stepwise Regression, F _Tb-Ts (₅₇) = 27.79, P < 0.001; F _Tb-Ta (₅₇) = 0.148, P > 0.70), while in L. parvus, T _bs did not depend on T _ss or T _as (Stepwise Regression, F _Tb-Ts (₄₁) = 6.83, P > 0.40; F _Tb-Ta (₄₁) = 3.48, P > 0.06). In both species, T _bs follows the fluctuation of T _ss throughout the day, whereas T _as and T _es were lower than T _bs (Figure 3a and 3b).

Figure 3
Variation during activity in mean body temperature (°C; T _b, black triangles with white foreground), micro-environmental temperatures: substrate (C°, T _s, white diamond), air (°C; T _a, white squares), and operative temperatures (°C; T _e grey triangles with black foreground) vs. time (h) of Phymaturus aguanegra (panel a) and Liolaemus parvus (panel b) at Paso Agua Negra, and Phymaturus williamsi (panel c) and Liolaemus parvus (panel d) at Quebrada Vallecito. Polygons connect the mean ± SE of the variables through the activity period. Grey bars are the mean number of active lizards per hour.

Table III
Intraspecific comparisons among field body (T _b, °C), substrate (T _s, °C) and air (T _a, °C) temperatures of Phymaturus aguanegra, P. williamsi and Liolaemus parvus populations. Shown are means ± SD. The One-way Repeated Measures ANOVA and the Holm-Šídák t-Test (only for significant differences in the main factors) are also indicated. Asterisks indicate statistical significance (P < 0.05).

Paso Agua Negra
P. aguanegra		T s	T a
	Tb = 29.05 ± 5.29	t = 0.154	t = 8.57*
	Ts = 29.14 ± 7.24	-	t = 8.73*
	Ta = 23.65 ± 5.17	-	-
	ANOVA, F2,59 = 49.92, P < 0.001
L. parvus		T s	T a
	Tb = 32.34 ± 2.84	t = 0.78	t = 6.76*
	Ts = 32.92 ± 7.11	-	t = 7.46*
	Ta = 26.92 ± 4.31	-	-
	ANOVA, F2,46 = 34.25, P < 0.001
Quebrada Vallecito
P. williamsi		T s	T a
	Tb = 31.00 ± 5.15	t = 2.18*	t =14.58*
	Ts = 29.86 ± 6.76	-	t = 12.39*
	Ta = 23.41 ± 4.82	-	-
	ANOVA, F2,83 = 123.75, P < 0.001
L. parvus		T s	T a
	Tb = 33.14 ± 4.46	t = 2.90*	t = 17.90*
	Ts = 31.78 ± 6.14	-	t = 14.99*
	Ta = 24.76 ± 5.24	-	-
	ANOVA, F(2, 86) = 184.66, P < 0.001

Quebrada Vallecito

In Phymaturus williamsi and L. parvus, the T _b values were significantly higher than the T _s and T _a values (Table III). The T _bs of P. williamsi increased only with T _ss (Stepwise Regression, F _Tb-Ts (₈₁) = 47.47, P < 0.001; F _Tb-Ta, (₈₁) = 0.622, P > 0.434). Also, the T _bs and T _ss were similar through the activity period, and were higher than T _a and T _e (Figure 3c). The T _bs of L. parvus increased with both T _ss and T _as (Stepwise Regression, F _Tb-Ts (₈₄) = 37.65, P < 0.001; F _Tb-Ta (₈₄) = 7.339, P < 0.006). T _bs followed the substrate temperature (T _ss) through the day, and it was higher than T _a, but T _e was higher than T _b during the warmer hours 13:00 – 15:00 h, coinciding with the highest frequency of lizard activity (Figure 3d).

Preferred body temperatures and effectiveness of thermoregulation

Paso Agua Negra

The T _pref values in P. aguanegra and in L. parvus were higher than T _b (Wilcoxon signed-rank test, P. aguanegra: W = 988, z = 5.76, P < 0.001, N = 44; L. parvus: W = 859, z = 5.56, P < 0.001, N = 41; Table II).

In P. aguanegra the lower and upper T _set limits were 36.22°C ± 1.90 and 38.96°C ± 1.00, respectively (Figure 4a). Most of the T _bs (N = 40, 91%) were lower than the minimum T _set., and 9% (N = 4) were included within the T _set. Likewise, in L. parvus the lower and upper T _set limits were 35.71 ± 1.93°C and 38.46 ± 0.84°C, respectively (Figure 4b). Most of the T _bs (N = 37, 90%) were lower than the minimum T _set., and 10% (N = 4) were included within the T _set.

Figure 4
Distribution of body temperatures (T _b) at capture of Phymaturus aguanegra and Liolaemus parvus at Paso Agua Negra and P. williamsi and L. parvus at Quebrada Vallecito. The white arrows indicate median operative temperatures (T _e), black arrows indicate median T _b values and the dark shaded column indicates the lower and upper set-point ranges of selected temperatures (T _set).

The T _set values exceeded T _e in all P. aguanegra individuals and in all but one individual of L. parvus showed T _set lower than T _e. Males of P. aguanegra and juveniles of L. parvus exhibited the highest E-values compared with the other demographic classes in each population (Table II).

Quebrada Vallecito

The T _pref values of P. williamsi and L. parvus were higher than the T _bs (Paired t-test, P. williamsi: t (₃₃) = -8.189, P < 0.001; L. parvus: t (₂₉) = -2.61, P < 0.01; Table II). In P. williamsi lower and upper T _sets were 35.02 ± 1.71°C and 37.74 ± 1.68°C, respectively (Figure 4c). A majority (N = 28, 82%) of the T _bs were lower than the minimum T _set, 15% (N = 5) were included within T _set, and 3% (N = 1) were higher than the maximum T _set. In L. parvus, the lower and upper T _set were 33.47°C ± 1.79 and 36.65°C ± 1.46°C, respectively (Figure 4d). About half (N = 16, 53%) of the T _bs of L. parvus were lower than the minimum T _set, 17% (N = 6) of T _bs were included within the T _set, and 30% (N = 9) were higher than the maximum T _set.

The T _set of P. williamsi (100%) was higher than the T _e. In L. parvus, the 37% (N = 11) of the individuals, the minimum T _set was lower than T _e, the 7% (N = 2) the maximum T _set were lower than T _e (7%, N = 2), and the in remainder (56%, N = 17) the T _sets were higher than T _es.

Males and females of P. williamsi showed similar E-values, being both moderate thermoregulators, whereas in L. parvus, both the males and females exhibited negative E-values indicating there are non-thermal constraints or risks that prevent them from thermoregulating within their set-points of T _pref (Table II).

Comparison of T _b, and T _pref between sites for Phymaturus and Liolaemus

The T _b of P. williamsi was higher than that of P. aguanegra (Mann-Whitney U-test, U _Tb (₁) = 1923, P < 0.01, N _P. _aguanegra = 60, N _P. _williamsi = 84). Conversely, the T _pref of P. aguanegra was higher than those in P. williamsi (Student t-test, t _Tpref (₇₆) = -3.69, P < 0.0001, Table II).

Liolaemus parvus populations did not differ in T _b between Agua Negra and Quebrada Vallecito (Mann-Whitney U-test, U (₁) = 1630.500, P > 0.11, N _L. _{parvus Agua Negra} = 45, N _L. _{parvus Quebrada Vallecito} = 87), but T _pref values of L. parvus from Agua Negra were higher than those in L. parvus from Quebrada Vallecito (Student t-test, t (₆₉) = -6.147, P < 0.0001, Table II).

Periods of activity

Paso Agua Negra

Phymaturus aguanegra were active for about 6 hours (10:00 ‒ 16:00 h), whereas L. parvus was active for about 5 hours (11:00 ‒ 16:00 h). In both species, the observed frequencies were not different than expected on an hour-by-hour basis (Table IV). Both species show a unimodal pattern of activity, with the highest number of lizards active 11:00 to 14:00 (Figures 3a and 3b).

Quebrada Vallecito

Phymaturus williamsi was active during 7 hours (11:00 ‒ 18:00 h), and the observed frequencies of individuals were lower than expected during the 13:00 ‒ 14:00 h interval (Pearson Chi-Square test, χ² (₂₁) = 13.06, Multiple comparison, z = -3.61, P < 0.0001, Table IV). P. williamsi showed a bimodal pattern of activity with the most active individuals during late morning (12:00 – 14:00 h) and afternoon (15:00 – 18:00 h; Figure 3c). Likewise, L. parvus was active daily for 8 hours (10:00 ‒ 18:00 h) and the observed frequencies were lower than expected during the 11:00 ‒ 12:00 h interval (Pearson Chi-Square test, χ² (₃₁) = 19.72, Multiple comparison, z = -4.44, P < 0.00001, Table IV) but the population showed a unimodal pattern of activity (Figure 3d).

DISCUSSION

Phymaturus species are exposed to a broad range of climatic factors and thermal environments along the high cordilleras on both slopes of the southern Andes and eastward to the Patagonian steppe. The field body temperatures observed in P. aguanegra (mean = 29.05°C) and P. williamsi (mean = 31.00°C) varied according to the environmental conditions those populations experienced in their natural habitats, typical of the Puna biogeographic region (Fig. 2). The habitat of P. aguanegra is characterized by steep slopes with extensive debris aprons of basaltic rocks with interspersed dwarf shrubs at the base. The habitat of P. williamsi, is different since rocky outcrops are distributed in a gentler open-gully landscape leading to differences in sun and wind exposure between both locations with remarkable consequences for the thermal heterogeneity (Paso Agua Negra, Δ T _e = 9.62 °C vs. Quebrada Vallecito, Δ T _e = 6.42 °C; Figures 2 and 4). Thus, lower T _b values in P. aguanegra than P. williamsi could have resulted from a greater thermal restriction (d_{e P}. _aguanegra = 11.26; d_{e P}. _williamsi = 9.06). The activity patterns observed in Phymaturus species also corroborates the difference in thermal restriction with shorter periods of activity in P. aguanegra and a longer, but bimodal pattern in P. williamsi (Table IV; Figure 3). In contrast, the T _b values of L. parvus from Paso Agua Negra (mean = 32.19 °C) and in Quebrada Vallecito (mean = 33.14 °C) did not show differences, and were similar to those reported in another L. parvus population from a high-altitude site in the Puna region (Gómez Ales et al. 2017). Indeed, the L. parvus activity period was shorter in Paso Agua Negra (5 h) than in the milder environment of Quebrada Vallecito (8 h).

In general, the variability in T _b may be adaptive phenotypic plasticity favoured by the use of multiple and changing microenvironments (Kingsolver & Buckley 2017, Huey et al. 2003). The microenvironments used by the populations of P. aguanegra, P. williamsi and L. parvus (from Paso Agua Negra) varied substantially according to differences in the type of substrate, exposure to radiation, wind or cloudiness (Laspiur 2010). The strong similarity and dependence of T _b on T _s in P. aguanegra suggests that they heat mainly by thigmothermy during most hours of activity while, in P. williamsi there is a significant relationship between T _b and T _s, but T _b is higher than T _s suggesting they use both heat sources, thigmothermy and heliothermy. These results support our observation of Phymaturus lizards flattening the ventral surface of their body to the rock, maximizing heat transfer while orienting their body dorsum perpendicularly to incident solar radiation, as reported for P. palluma (Vicenzi et al. 2019). In contrast, L. parvus from Paso Agua Negra were seen occupying exposed homogeneous microsites composed of small weathered rocks beneath dwarf shrubs and they showed an opportunistic thermoregulatory behaviour. Liolaemus parvus is heliothermic when T _s substantially exceeds the T _b, but turns to thigmothermy during the colder hours in the late afternoon (See Figure 3, Panel B). Moreover, the finding of higher T _bs in L. parvus from Agua Negra compared to P. aguanegra was probably caused by behavioral selection of warm microenvironments with markedly greater exposure to radiation and sheltering from winds than those selected by P. aguanegra. Conversely, in Quebrada Vallecito, T _bs of L. parvus and P. williamsi have shown a strong relationship with T _ss, and T _bs were higher than T _ss and T _as suggesting that both species alternate between thigmothermy and heliothermy for thermoregulation. Furthermore, it would be interesting to study whether there is competition for thermal micro-environments within the rocky outcrops they use for thermoregulation and, if so, that would explain the lower T _b found in P. williamsi compared to L. parvus. The bimodality in activity period likely represents temporal niche partitioning (Fig. 3c and 3d) and could be explained by territorial or temporal segregation of both species related to thermal restrictions. In this sense, Liolaemus parvus exhibited 8 hours of activity, being more active during the warmest hours when the micro-environmental and operative temperatures reach their maxima, and depicting a unimodal pattern activity (Figure 3d, Table IV). In contrast, P. williamsi was active earlier and for only 6 h with a significant decrease in activity during the warmest periods (14-15 h). This resulted in a bimodal pattern of activity, probably as a way to avoid exposure to high environmental temperatures (Figure 3c, Table IV) and to take advantage of warm microenvironments early in the morning before L. parvus starts activity. A similar bimodal daily activity pattern, with cessation of activity at ~15:00 h, also occurs in P. vociferator from Laguna del Laja, central Chile (Vidal et al. 2010). In contrast, P. aguanegra showed a unimodal pattern like that reported for other species in the P. palluma group: P. palluma (Vicenzi et al. 2019), P. roigorum (Corbalán & Debandi 2014), and for the northernmost species in the P. patagonicus group, P. payuniae (Corbalán & Debandi 2014). Thus, environmental temperature has a strong effect on the activity of lizards during the day, and pattern differences may reflect different restraints for thermoregulation between coexisting species. In Quebrada Vallecito, our results support the prediction that syntopic species exhibit differences in T _b, T _pref, and activity periods as a result of niche segregation.

Body temperature can also be affected by differences in activity patterns, use of microhabitats, and thermal preferences of males, females or juveniles (Bull et al. 1991, Butler et al. 2000, Lailvaux 2004, Ortega et al. 2016). Gender and reproductive condition can often cause differences in hierarchical social behaviour, physiology and ecology (Vitt and Cooper 1986, Irschick & Garland 2001, Lovern 2011, Sinervo & Miles 2011). In this sense, despite the differences in body size or mass existing among demographic groups (males, females and juveniles; Table II) the T _b or T _pref values were similar across groups as none showed a relationship with M_i in neither the L. parvus population nor in the P. aguanegra population. An exception was observed in P. williamsi from Quebrada Vallecito, where juveniles exhibited a higher mean T _b than adult males. Similarly, L. parvus juveniles at Agua Negra showed a higher E value than adults (Tables I and II). This could have been caused by smaller individuals with faster heating rates needing shorter exposures to maintain their temperature within the preferred range, while bigger body sizes resulted in higher thermal inertia. Hence, our results partially support our hypothesis that thermal traits would be different between juveniles and adults and affected by body condition. Probably, there are underlying and confounding factors such as thermal behaviour, predation pressure or competition that could also drive the thermal patterns within populations (Ortega et al. 2016).

Phymaturus aguanegra and L. parvus from Paso Agua Negra, and P. williamsi from Quebrada Vallecito are effective thermoregulators being able to maintain field body temperatures higher than T _e, despite the marked daily and seasonal fluctuations in air temperature and precipitation present in the highlands of the Andes (Aguado 1983, Borsdorf & Stadel 2015). Both Phymaturus species showed similar indices of effective thermoregulation (E _P. _aguanegra= 0.41 and E _P. _williamsi= 0.50), behaving as moderate thermoregulators. Liolaemus parvus E-values differed between sites, behaving as moderate-to-good thermoregulators in Paso Agua Negra (E _L. _parvus = 0.60) but as a thermoconformer in Quebrada Vallecito (E_L. _parvus = - 0.22). This difference could be attributed to the fact that the L. parvus population in Quebrada Vallecito needed less active thermoregulation as result of inhabiting a “less heterogeneous habitat” (sensu Basson et al. 2017). In fact, a high proportion of T _b were similar to T _e during the warmer hours. Liolaemus parvus pays the cost of T _bs that often exceed their T _set in Quebrada Vallecito, resulting in a negative E-value. Despite having suitable microenvironments available for thermoregulation, lizards may not use them because of other factors, such as the presence of greater predation risk, or lack of prey (Christian & Weavers 1996, Gutiérrez et al. 2010, Sinervo et al. 2010).

The existing difference in T _pref but similarity in T _b between L. parvus populations resulted in the difference in the effectiveness of thermoregulation, being medium to high at Paso Agua Negra and low (thermoconformer) at Quebrada Vallecito. So, even though L. parvus displays plasticity in thermoregulatory behavior, switching between heliothermy and thigmothermy, its thermoregulation appears to be constrained as only 17% of the population have values within the T _set. Even the index of the average of quality of the habitat from an organism’s perspective (d_e = |T _e - T _set|; sensu Hertz et al. 1993) is high (d_e = 2.57) in Quebrada Vallecito, there may be other restrictions, such as competition for microenvironments for thermoregulation with P. williamsi, that preclude them from effectively thermoregulating, resulting in low accuracy for thermoregulation (d_b = 3.13). However, L. parvus has the ability to tolerate harsh environmental conditions and to succeed in a wide geographic range in the Andes. The extent of the spatial distribution of L. parvus in the Andean and pre-Andean environments is a good indicator of the success of Liolaemus in comparison with the geographically restricted populations of Phymaturus (Quinteros et al. 2008, Díaz-Gómez 2009, Lobo et al. 2013).

Two contrasting views exist on thermal biology traits in lizards (Rodríguez Serrano et al. 2009). The first view holds that thermal biology traits are evolutionary conservative and they respond gradually to directional selection (Hertz & Huey 1981, Van Damme et al. 1990, Rodríguez-Serrano et al. 2009). The second view (labile view) asserts that thermal biology traits can rapidly adapt via directional selection to changing environmental pressures or changes within populations as experienced by some taxa (Huey & Kingsolver 1989, Van Damme et al. 1990, Rodríguez-Serrano et al. 2009, Sinervo et al. 2010). Our results indicate significant differences in T _b and T _pref between P. aguanegra and P. williamsi, and these species thermoregulate with high accuracy and effectiveness despite considerable site differences and low thermal qualities of their habitats in the Central Andes (d_{e P}. _aguanegra = 11.26; d_{e P}. _williamsi = 9.06). Thus, our results do not confirm the prediction of conservativism in Phymaturus contrasting with some previous studies (Cruz et al. 2009, Gómez Ales et al. 2017, Duran et al. 2018). On the other hand, the L. parvus populations studied here exhibited similar T _b values but differed in T _pref and E, suggesting that these lizards respond to short-term environmental temperature fluctuations. Hence, L. parvus exhibited lower variation in T _b but greater variation in T _pref, accuracy and effectiveness of thermoregulation and these results support an alternative to the labile-traits hypothesis (sensu Rodríguez-Serrano et al. 2009): some traits impacting thermal biology may be conserved within a population while others simultaneously show a combination of conservative and flexible-traits evolution (Bonino et al. 2011).

Phymaturus and Liolaemus occupy similar altitudinal ranges along the Andes mountains and therefore experience comparable environmental conditions and similar levels of availability of microhabitats. Nevertheless, differences in microhabitat selection among species may play a role in thermoregulation more so than gender, age or body condition. The Andean environments offer thermal opportunities for thermoregulation in which most lizards exhibit high thermoregulatory efficiency. High-elevation species suffer higher vulnerability in the face of a warming-climate scenario, and such vulnerability can be detected when T _e is closer to T _pref and particularly when species show T _b values greater than their T _set as observed for L. parvus from Quebrada Vallecito. Hence, future complementary studies on thermal tolerance and physiological plasticity are necessary to evaluate physiological thresholds to predict possible responses to climate change based on physiological models (Sinervo et al. 2018).

ACKNOWLEDGMENTS

We dedicate this work to Joel Gutiérrez and Mariano Ariza. We thank Jorgelina Boretto and Facundo Cabezas-Cartes for valuable help in laboratory and logistical support; Ana Naranjo, Verónica Blanco, and Laura Reus, Cristian Abdala and L.V. House members for their constant support. Mariano Ariza and Yanina Ripoll provided collection permits. Specimens were collected under permission of Secretaría de Medio Ambiente and Dirección de Conservación y Áreas Protegidas, Provincia de San Juan, Argentina. We thank CIMA/CONICET-UBA for providing climatic data of the 3ra. Comunicación Nacional sobre Cambio Climático. Financial support was received from a CONICET scholarship to AL and grants PICT 2017- 0553 to NRI and CICITCA E/1101 to AL.

REFERENCES

• ABDALA CS ET AL. 2012. Categorización del estado de conservación de los lagartos de la República Argentina. Cuad Herp 26: 215-248.
• ABDALA CS, LASPIUR A, SCROCCHI GJ, SEMHAN RV, LOBO F & VALLADARES P. 2021. Las lagartijas de la Familia Liolaemidae: Sistemática, distribución e historia natural de una de las familias de vertebrados más diversas de Sudamérica. Volumen 1. Santiago de Chile, RIL Editores, 350 p.
• ABDALA CS & QUINTEROS SA. 2014. Los últimos 30 años de estudios de la familia de lagartijas más diversa de Argentina. Actualización taxonómica y sistemática de Liolaemidae. Cuad Herp 28: 55-82.
• ADOLPH SC. 1990. Influence of behavioral thermoregulation on microhabitat use by two Sceloporus lizards. Ecology 71: 315-327.
• AGUADO C. 1983. Comparación del inventario de glaciares de la cuenca del Río de los Patos con otros inventarios de los Andes centrales de Argentina. An IANIGLA 4: 3-11.
• ANGILLETTA JR MJ. 2009. Thermal Adaptation: A Theoretical and Empirical Synthesis. Oxford: Oxford University Press, 289 p.
• ANGILLETTA JR MJ, NIEWIAROWSKI PH & NAVAS CA. 2002. The evolution of thermal physiology in ectotherms. J Therm Biol 27: 249-268.
• AUSAS S. 2011. Ecofisiología térmica del lagarto Phymaturus cf. palluma. Bsc Thesis. San Carlos de Bariloche: Universidad Nacional del Comahue, 55 p.
• AVERY RA. 1978. Lizards ‒ a study in thermoregulation. The Institute of Biology’s. studies in biology, n. 109. Baltimore: University Park Press, 56 p.
• AVERY RA. 1982. Field studies of body temperatures and thermoregulation. In: Gans C & Pough FH (Eds), Biology of the Reptilia, Vol. 12. New York: Academic Press, New York, USA, p. 93-166.
• BAKKEN GS. 1992. Measurements and application of operative and standard operative temperatures in ecology. Am Zool 32: 194-216.
• BASSON CH, LEVY O, ANGUILLETTA MJ & CRUSELA-TRULLAS S. 2017. Lizards paid a greater oportunity cost to thermoregulate in a less heterogeneous environment. Func Ecol 31: 856-865.
• BEASLEY TM & SCHUMACKER RE. 1995. Multiple regression approach to analyzing contingency tables: Post hoc and planned comparison procedures. J Exp Educ 64: 79-93.
• BEAUPRE SJ. 1995. Effects of geographically variable thermal environment on bioenergetics of mottled rock rattlesnakes, Crotalus lepidus. Ecology 76: 1655-1665.
• BLOUIN-DEMERS G & WEATHERHEAD P. 2001a. An experimental test of the link between foraging, habitat selection and thermoregulation in black rat snakes Elaphe obsoleta obsoleta. J Anim Ecol 70: 1006-1013.
• BLOUIN-DEMERS G & WEATHERHEAD P. 2001b. Thermal ecology of black rat snakes (Elaphe obsoleta) in a thermally challenging environment. Ecology 82: 3025-3043.
• BOGERT CM. 1949. Thermoregulation in reptiles, a factor in evolution. Evolution 3: 195-211.
• BOHONAK AJ & VAN DER LINDE K. 2004. RMA for JAVA V. 1.21: Software for Reduced Major Axis regression, Available at: http://www.kimvdlinde.com/professional/rma.html
• BONINO MF, MORENO-AZÓCAR DL, TULLI MJ, ABDALA CS, PEROTTI MG & CRUZ FB. 2011. Running in cold weather: Morphology, thermal biology, and performance in the southernmost lizard clade in the world (Liolaemus lineomaculatus section: Liolaemini: Iguania). J Exp Biol 315: 495-503.
• BORETTO JM, CABEZAS-CARTES F & IBARGÜENGOYTÍA NR. 2018. Slow life histories in lizards living in the highlands of the Andes Mountains. J Comp Physiol B 188: 491-503.
• BORETTO JM, CABEZAS-CARTES F, TAPPARI F, MÉNDEZ DE LA CRUZ F, SINERVO B, SCOLARO JA & IBARGÜENGOYTÍA NR. 2014. Reproductive biology of Phymaturus spectabilis (Liolaemidae): Females skip reproduction in cold and harsh environments of Patagonia, Argentina. Herpetol Conserv Biol 9: 170-180.
• BORSDORF A & STADEL C. 2015. The Andes. A geographical portrait. Springer Geography Series. Switzerland: Springer International Publishing AG Switzerland, 368 p.
• BUCKLEY LB, EHRENBERGER JC & ANGILLETA JR MJ. 2015. Thermoregulatory behavior limits local adaptation of thermal niches and confers sensitivity to climate change. Func Ecol 29: 1038-1047.
• BULL CM, MCNALLY A & DUBAS G. 1991. Asynchronous seasonal activity of male and female sleepy lizards, Tiliqua rugosa. J Herpetol 25: 436-441.
• BUTLER MA, SCHOENER TW & LOSOS JB. 2000. The relationship between sexual size dimorphism and habitat use in Greater Antillean Anolis lizards. Evolution 54: 259-272.
• CABEZAS-CARTES F, BORETTO JM, ACOSTA JC, JAHN G, BLANCO GM, LASPIUR A & IBARGÜENGOYTÍA NR. 2010. Morphological and histological study of the reproductive biology of the viviparous lizard Phymaturus palluma (Liolaemidae) from San Juan, Argentina. Herpetol Conserv Biol 5: 430-440.
• CABRERA A & WILLINK A. 1973. Biogeografía de América Latina, 2nd ed., Washington DC: Secretaría General de la Organización de los Estados Americanos, 120 p.
• CASTILLO GN, ACOSTA JC, LASPIUR A & BLANCO G. 2011. Morphometrical sexual differences in Liolaemus parvus (Liolaemidae) from San Juan. Biocell 35: A248.
• CASTRO SA, LASPIUR A & ACOSTA JC. 2013. Variación anual e intrapoblacional de la dieta de Phymaturus cf. palluma (Iguania: Liolaemidae) de los Andes Centrales en Argentina. Rev Mex Biodiv 84: 1258-1265.
• CASTRO SA, LASPIUR A, ACOSTA JC & NIEVA R. 2011. Sexual dimorphism in Phymaturus cf. palluma (Liolaemidae) from San Juan, Argentina. Biocell 35: A247.
• CEI JM. 1986. Reptiles del centro, centro‒oeste y sur de la Argentina. Herpetofauna de las zonas áridas y semiáridas, Monografie IV, Torino: Museo Regionale di Scienze Naturali Torino, 527 p.
• CHRISTIAN KA & WEAVERS BW. 1996. Thermoregulation of monitor lizards in Australia: An evaluation of methods in thermal biology. Ecol Monogr 66: 139-157.
• CORBALÁN V & DEBANDI G. 2014. Resource segregation in two herbivorous species of mountain lizards from Argentina. Herpetol J 24: 201-208.
• COWLES RB & BOGERT CM. 1944. A preliminary study of the thermal requirements of desert reptiles. Bull Am Mus Nat Hist 83: 265-296.
• CRUZ FB, BELVER L, ACOSTA JC, VILLAVICENCIO HJ, BLANCO G & CÁNOVAS MG. 2009. Thermal biology of Phymaturus lizards: Evolutionary constraints or lack of environmental variation? Zoology 112: 425-432.
• DÍAZ DE LA VEGA-PÉREZ AH, JIMÉNEZ-ARCOS VH, MANRÍQUEZ-MORÁN NL & MÉNDEZ DE LA CRUZ FR. 2013. Conservatism of thermal preferences between parthenogenetic Aspidoscelis cozumela complex (Squamata: Teiidae) and their parental species. Herpetol J 23: 93-104.
• DÍAZ-GÓMEZ JM. 2009. Historical biogeography of Phymaturus (Iguania: Liolaemidae) from Andean and Patagonian South America. Zool Scr 31: 1-7.
• DURAN F, KUBISCH E & BORETTO JM. 2018. Thermal physiology of three sympatric and syntopic Liolaemidae lizards in cold and arid environments of Patagonia (Argentina). J Comp Physiol B 188: 141-152.
• ESPINOZA RE, WIENS JJ & TRACY CR. 2004. Recurrent evolution of herbivory in small, cold‒climate lizards: breaking the ecophysiological rules of reptilian herbivory. Proc Natl Acad Sci 101: 16819-16824.
• ESQUERRÉ D, BRENNAN IG, CATULLO RA, TORRES-PÉREZ F & KEOGH JS. 2019. How mountains shape biodiversity: The role of the Andes in biogeography, diversification, and reproductive biology in South America’s most species-rich lizard radiation. Evolution 73: 214-230.
• ETHERIDGE RE. 1995. Redescription of Ctenoblepharys adspersa Tschudi, 1845, and the taxonomy of Liolaeminae (Reptilia: Squamata: Tropiduridae). Am Mus Nov 3142: 1-34.
• GARCÍA PÉREZ MA & NÚÑEZ ANTÓN V. 2003. Cellwise residual analysis in two-way contingency tables. Educ Psychol Meas 65: 825-839.
• GÓMEZ ALES R, ACOSTA JC & LASPIUR A. 2017. Thermal biology in two syntopic lizards, Phymaturus extrilidus and Liolaemus parvus, in the Puna of Argentina. J Therm Biol 68: 73-82.
• GREEN AJ. 2001. Mass/length residuals: measures of body condition or generators of spurious results? Ecology 82: 1473-1483.
• GUTIÉRREZ JA, KRENZ JD AND IBARGÜENGOYTÍA NR. 2010. Effect of altitude on thermal responses of Liolaemus pictus argentinus in Argentina. J Therm Biol 35: 332-337.
• GVOŽDÍK L. 2002. To heat or to save time? Thermoregulation in the lizard Zootoca vivipara (Squamata: Lacertidae) in different thermal environments along an altitudinal gradient. Can J Zool 80: 479-492.
• GVOŽDÍK L. 2012. Plasticity of preferred body temperatures as means of coping with climate change? Biol Lett 8: 262-265.
• HABIT EM & ORTIZ JC. 1994. Ámbito de hogar de Phymaturus flagellifer (Reptilia, Tropiduridae). Bol Soc Biol Concepción 65: 149-152.
• HERTZ PE & HUEY RB. 1981. Compensation for altitudinal changes in the thermal environment by some Anolis lizards on Hispaniola. Ecology 62: 515-521.
• HERTZ PE, HUEY RB & NEVO E. 1983. Homage to Santa Anita: Thermal sensitivity of sprint speed in Agamid lizards. Evolution 37: 1075-1084.
• HERTZ PE, HUEY RB & STEVENSON RD. 1993. Evaluating temperature regulation by field-active ectotherms: the fallacy of the inappropriate question. Am Nat 142: 796-818.
• HUEY RB. 1982. Temperature, physiology, and the ecology of reptiles. In: Gans C & Pough FH (Eds), Biology of the Reptilia, Vol. 12., New York: Academic Press, New York, USA, p. 25-91.
• HUEY RB, HERTZ PE & SINERVO B. 2003. Behavioral drive versus behavioral inertia in evolution: A null model approach. Am Nat 161: 357-366.
• HUEY RB & KINGSOLVER JG. 1989. Evolution of thermal sensitivity of ectotherm performance. Tree 4: 131-135.
• IRSCHICK DJ & GARLAND JR T. 2001. Integrating function and ecology in studies of adaptation: investigations of locomotor capacity as a model system. Annu Rev Ecol Syst 32: 367-396.
• KINGSOLVER JG & BUCKLEY LB. 2017. Evolution of plasticity and adaptative responses to climate change along climate gradients. Proc R Soc B 274: 20170386.
• KÖEPPEN W. 1948. Climatología. Con un estudio de los climas de la tierra, Mexico DF: Fondo de Cultura Económica, 478 p.
• LAILVAUX SP. 2004. Interactive effects of sex and temperature on locomotion in reptiles. Integr Comp Biol 47: 189-199.
• LASPIUR A. 2010. Termorregulación, actividad temporal y uso de microhábitats de Phymaturus cf. palluma (Iguania: Liolaemidae) de los Andes centrales de la provincia de San Juan. Bsc. Thesis. San Juan: Universidad Nacional de San Juan, 58 p.
• LOBO F, BARRASSO DA, VALDECANTOS S, GIRAUDO AR, DI PIETRO DO & BASSO NG. 2022. Exploring the morphological diversity of Patagonian clades of Phymaturus (Iguania: Liolaemidae). Integrative study and the description of two new species. Cuad Herp 36: 197-231.
• LOBO F, LASPIUR A & ACOSTA JC. 2013. Description of new andean species of the genus Phymaturus (Iguania: Liolaemidae) from northwestern Argentina. Zootaxa 3683: 117-132.
• LOGAN ML, FERNÁNDEZ SG & CALSBEEK R. 2015. Abiotic constrints on the activity of tropical lizards. Func Ecol 29: 694-700.
• LOGAN ML, MONTGOMERY CE, BOBACK SM, REED RN & CAMPBELL JA. 2012. Divergence in morphology, but not habitat use, despite low genetic differentiation among insular populations of the lizard Anolis lemurinus in Honduras. J Trop Ecol 28: 215-222.
• LOVERN MB. 2011. Hormones and reproductive cycles in lizards. In: Norris DO & López KH. (Eds), Hormones and reproduction of vertebrates. Vol. 3, London: Academic Press, United Kingdom, p. 321-353.
• MAGNUSSON JJ, CROWDER LB & MEDVICK PA. 1979. Temperature as an ecological resource. Am Zool 19: 331-343.
• MEDINA M, SCOLARO A, MÉNDEZ DE LA CRUZ F, SINERVO B, MILES DB & IBARGÜENGOYTÍA NR. 2012. Thermal biology of the genus Liolaemus: A phylogenetic approach reveals advantages of the genus to survive climate change. J Therm Biol 37: 579-585.
• MEDINA M, SCOLARO A, MÉNDEZ DE LA CRUZ FR, SINERVO B & IBARGÜENGOYTÍA NR. 2011. Thermal relationships between body temperature and environment condition set upper distributional limits on oviparous species. J Therm Biol 36: 527-534.
• MORENO-AZÓCAR DL, VANHOOYDONCK B, BONINO MF, PEROTTI MG, ABDALA CS, SCHULTE II JA & CRUZ FB. 2013. Chasing the Patagonian sun: comparative thermal biology of Liolaemus lizards. Oecologia 171: 773-788.
• NELSON DO, HEATH JE & PROSSER CL. 1984. Evolution of temperature regulatory mechanisms. Am Zool 24: 791-807.
• ORTEGA Z, MENCÍA A & PÉREZ-MELLADO V. 2016. Sexual differences in behavioral thermoregulation of the lizard Scelarcis perspicillata. J Therm Biol 61: 44-49.
• PATERSON JE & BLOUIN-DEMERS G. 2016. Do ectotherms partition thermal resources? We still do not known. Oecologia 183: 337-345.
• PEEL MC, FINLAYSON BL & MCMAHON TA. 2007. Updated world map of the Köppen-Gieger climate classification. Hydrol Earth Syst Sci 11: 1633-1644.
• PEIG J & GREEN AJ. 2009. New perspectives for estimating body condition from mass/length data: the sacled mass index as an alternative method. Oikos 118: 1883-1891.
• POUGH FH & GANS C. 1982. The vocabulary of reptilian thermoregulation. In: Gans C & Pough FH (Eds), Biology of the Reptilia, Vol. 12., New York: Academic Press, New York, USA, p. 17-23.
• QUINTEROS SA, ABDALA CS, DÍAZ-GÓMEZ JM & SCROCCHI GJ. 2008. Two new species of Liolaemus (Iguania: Liolaemidae) of central west Argentina. S Am J Herpetol 3: 101-111.
• RODRÍGUEZ-SERRANO E, NAVAS CA & BOZINOVIC F. 2009. The comparative field body temperature among Liolaemus lizards: Testing the static and the labile hypotheses. J Therm Biol 34: 306-309.
• SAGONAS K, MEIRI S, VALAKOS ED & PAFILIS P. 2013. The effect of body size on the thermoregulation lizard on hot dry Mediterranean islands. J Them Biol 38: 92-97.
• SCHEERS H & VAN DAMME R. 2002. Micro-scale differences in thermal habitat quality and a possible case of evolutionary flexibility in the thermal physiology of lacertid lizards. Oecologia 132: 323-331.
• SCOLARO JA. 2005. Reptiles patagónicos sur: Guía de campo, 1st ed., Trelew: Universidad Nacional de la Patagonia San Juan Bosco, 75 p.
• SILICEO-CANTERO HH & GARCÍA A. 2014 Differences in growth, body condition, habitat use ad food availability between island and mainland lizard populations of Anolis nebulosus in Jalisco, Mexico. J Trop Ecol 30: 493-501.
• SINERVO B & MILES DB. 2011. Hormones and behavior of reptiles. In: Norris DO & López KH (Eds), Hormones and reproduction of vertebrates. Vol. 3, London: Academic Press, United Kingdom, p. 215-246.
• SINERVO B ET AL. 2010. Erosion of lizard diversity by climate change and altered thermal niches. Science 328: 894-899.
• SINERVO B, MILES DB, WU Y, MÉNDEZ-DE LA CRUZ FR, KIRCHHOF S & QI Y. 2018. Climate change, thermal niches, extinction risk, and maternal-effect rescue of toad-headed lizards, Phrynocephalus, in thermal extremes of the Arabian Peninsula to the Qinghai‒Tibetan Plateau. Integr Zool 13: 450-470.
• SMITH GD & FRENCH SS. 2017. Physiological trade-offs in lizards: Cost for individuals and populations. Integr Comp Biol 57: 344-351.
• SOKAL RR & ROHLF FJ. 1969. Biometry: The Principles and practice of statistics in biological research. San Francisco: Freeman W.H., 778 p.
• STAMPS JA. 1983. Sexual selection, sexual dimorphism, and territoriality. In: Huey RB et al. (Eds), Lizard ecology: studies of a model organism, Massachusetts: Harvard University Press, Cambridge, Massachusetts, USA. p. 169-204.
• STEVENSON RD. 1985. The relative importance of behavioral and physiological adjustments controlling body temperature in terrestrial ectotherms. Am Nat 126: 362-386.
• TRACY CR. 1982. Biophysical modeling in reptilian physiology and ecology. In: Gans C & Pough FH (Eds), Biology of the Reptilia, Vol. 12, London: Academic Press, London, UK, p. 275–321.
• TRONCOSO-PALACIOS J, FERRI YAÑEZ F, LASPIUR A & AGUILAR C. 2018. An updated phylogeny and morphological study of the Phymaturus vociferator clade (Iguania: Liolaemidae). Zootaxa 4441: 447-466.
• VAN DAMME R, BAUWENS D & VERHEYEN R. 1990. Evolutionary rigidity of thermal physiology: the case of the cool temperate lizard Lacerta vivipara. Oikos 57: 61-67.
• VEBLEN TT, YOUNG KR & ORME AR. 2007. The physical geography of South America. Oxford: Oxford University Press, 361 p.
• VELOSO J, SEPÚLVEDA P, CANALS M & SABAT P. 2007. Thermal biology of Liolaemus lemniscatus (Iguanidae) from low and high altitude populations in Central Chile. Comp Biochem Physiol A 148: 138-141.
• VICENZI NP, IBARGÜENGOYTÍA NR & CORBALÁN V. 2019. Activity patterns and thermoregulatory behavior of the viviparous lizard Phymaturus palluma in Aconcagua Provincial Park, Argentine Andes. Herpetol Conserv Bio 14: 337-348.
• VIDAL MA, HABIT E, VICTORIANO P, GONZÁLEZ-GAJARDO A & ORTIZ JC. 2010. Thermoregulation and activity pattern of the high‒mountain lizard Phymaturus palluma (Tropiduridae) in Chile. Zoologia 27: 13-18.
• VITT LJ & CALDWELL JP. 2014. Herpetology: An introductory biology of amphibians and reptiles, 4th ed., Massachusetts: Academic Press, p. 757.
• VITT LJ & COOPER JR WE. 1986. Skink reproduction and sexual dimorphism: Eumeces fasciatus in the southeastern United States, with notes on Eumeces inexpectatus. J Herpetol 20: 65-76.

Publication Dates

• Publication in this collection
  12 Feb 2024
• Date of issue
  2024

History

• Received
  1 Oct 2019
• Accepted
  30 May 2020