During the three years of study, we caught 164 animals in total, including 92 from Klis (49 females, 40 males, and three juveniles) and 72 from Cres Island (34 females, 37 males, and one juvenile). This species exhibited marked differences in coloration between juveniles and adults: juveniles were gray with black patches and spots, and had coarse scales, whereas adults were uniformly brown with smooth and shiny scales. As we did not catch animals intermediate in color/pattern, differentiation between juveniles and adults was considered straightforward.
Individuals from Cres were generally larger than individuals from Klis, which was the case for all morphometric traits (Table 1). Morphometric data analysis showed an influence of site (MANCOVA; F=9.20, P<0.05) and sex (MANCOVA; F=23.12, P<0.05) on almost all examined variables. The only variables not influenced by sex were TLL (ANCOVA; F=0.11, P=0.74) and BWM (ANCOVA; F=2.39, P=0.12) and not influenced by site were TLL (ANCOVA; F=2.05, P=0.16) and IND (ANCOVA; F=2.08, P=0.15). Similarly, significant interactions between site and sex (MANCOVA; F=2.18, P<0.05) were detected, indicating that males and females differed in morphology depending on the site. Variables not affected by this interaction were HH (ANCOVA; F=3.59, P=0.06), ML (ANCOVA; F=2.88, P=0.09), IOD (ANCOVA; F=0.10, P=0.76), BWM (ANCOVA; F=1.64, P=0.20), BWC (ANCOVA; F=3.23, P=0.08), and TLL (ANCOVA; F= 0.68, P=0.41). All morphometric variables were influenced by SVL as a covariate (MANCOVA; F=25.71, P<0.05), except for TLL (ANCOVA; F=0.89, P=0.35). We excluded TL from this analysis as it was strongly correlated to TLL and SVL as a covariate. Log-linear model analysis of tail condition against site and sex showed that the interaction between site and tail condition, as well as site and sex, was not significant, but the interaction between tail condition and sex was highly significant (Table 2; df=1; P=0.001). The HL/HW ratio (Table 3) showed no differences between the sites, but was slightly lower in males than in females, though not significantly (P>0.05).
Cres Klis MANCOVA site×sex MANCOVA sex MANCOVA site F (n=34) M (n=37) F (n=49) M (n=40) TL (cm) 89.01±5.36 (79.10–98.00) 91.80±5.95 (78.50–107.40) 84.07±6.95 (55.60–95.7) 82.96±6.78 (64.50–97.60) – – – SVL (cm) 34.99±2.60 (30.50–40.50) 37.34±2.54 (31.30–41.9) 32.74±2.85 (24.00–39.90) 32.87±2.11 (28.30–36.50) – – – HL (mm) 40.45±4.40 (25.77–52.30) 48.58±5.73 (36.32–69.62) 37.44±3.67 (26.32–43.37) 41.15±3.47 (34.97–49.83) P<0.05 P<0.05 P<0.05 HW (mm) 22.40±1.91 (19.65–29.82) 27.23±2.89 (19.50–31.20) 20.27±1.33 (16.39–23.03) 27.23±2.89 (18.59–27.14) P<0.05 P<0.05 P<0.05 HH (mm) 21.34±1.89 (18.99–27.72) 25.76±2.87 (19.87–33.57) 20.27±1.68 (16.54–24.26) 22.65±1.81 (18.66–26.38) n.s. P<0.05 P<0.05 ML (mm) 20.53±3.11 (17.18–29.71) 24.93±3.51 (20.63–38.29) 18.16±1.96 (13.71–23.20) 20.17±2.05 (15.70–25.51) n.s. P<0.05 P<0.05 MW (mm) 18.75±1.92 (16.03–25.81) 22.37±1.98 (17.64–25.24) 17.24±1.66 (13.73–20.82) 18.67±1.72 (15.21–22.70) P<0.05 P<0.05 P<0.05 IOD (mm) 15.03±2.21 (7.86–19.52) 17.00±2.40 (8.86–21.66) 13.49±1.26 (11.36–16.78) 14.58±1.31 (10.74–17.45) n.s. P<0.05 P<0.05 IND (mm) 6.81±0.68 (5.80–8.37) 8.03±0.88 (6.10–9.34) 6.59±0.55 (5.25–7.61) 7.02±0.61 (5.71–8.33) P<0.05 P<0.05 n.s. BWH (mm) 21.36±1.83 (18.47–25.43) 24.48±2.69 (18.18–29.96) 19.29±1.86 (16.12–25.11) 20.30±1.90 (16.72–26.95) P<0.05 P<0.05 P<0.05 BWM (mm) 26.78±3.17 (20.96–32.38) 29.40±3.91 (22.61–36.65) 23.44±3.35 (18.24–31.44) 23.64±2.17 (18.76–30.2) n.s. n.s. P<0.05 BWC (mm) 22.14±1.76 (19.35–28.39) 24.13±1.88 (20.20–29.48) 19.56±1.52 (16.16–23.44) 19.95±1.11 (17.40–22.86) n.s. P<0.05 P<0.05 TLL (cm) 52.99±4.68 (44.10–59.70) 52.11±6.58 (19.90–57.50) 49.55±9.43 (0.70–60.50) 50.00±5.96 (33.40–63.80) n.s. n.s. n.s. BM (g) 325.29±63.87 (220–490) 404.11±79.54 (220–552) 240.45±57.24 (105–400) 243.48±36.09 (150–315) P<0.05 P<0.05 P<0.05 F: Female; M: Male. n: Number of individuals. TL: Total length; SVL (co-variate): Snout-vent length; HL: Head length; HW: Head width; HH: Head height; ML: Mouth length; MW: Mouth width; IOD: Interorbital distance; IND: Internarial distance; BWH: Body width near head; BWM: Body width at mid-body; BWC: Body width near cloaca; TLL: Tail length; BM: Body mass. –: Not available; n.s.: Not significant.
Table 1. Means±SD, including range (in parenthesis), and MANCOVA results for morphometric measurements at each site
Klis Cres Tail condition M F M F Original 9 17 6 20 Autotomized 31 32 31 14 In both populations, females had significantly more often intact tails than males (P<0.05). F: Female; M: Male.
Table 2. Number of individuals with intact and autotomized tails at both sites and for both sexes
Site M F Klis 1.81±0.14 1.85±0.15 Cres 1.78±0.19 1.82±0.14 M: Male; F: Female. t-test showed no significant differences between sexes or between sites. (P>0.05).
Table 3. Means±SD of head length (HL) to head width (HW) ratios (HL/HW) for studied populations
The SMI differed significantly between two populations, with the Cres population having higher values (K-W Test; P<0.05), indicating that the animals from Cres were larger than those from Klis. A within-population comparison of Cres individuals showed that males had higher SMI values (K-W Test; P<0.05) than females. The individuals from Klis, however, did not show such differences between the sexes (K-W Test; P>0.05; Table 4, Figure 2B).
Locality n Sex bSMA 95 % CI L0 SMI Cres 34 F 3.24 2.20, 4.28 89.01 327.98 34 M 3.89 2.85, 4.93 92.78 398.69 Klis 48 F 3.59 2.84, 4.34 84.66 239.38 39 M 2.08 1.49, 2.66 83.43 243.94 n: Number of individuals; F: Female; M: Male; bSMA: Scaling exponent; L0: Arithmetic mean of L (individual total length).
Table 4. Scaled mass index (SMI)
Mean recorded values of physical habitat data showed slight differences between the two sites, with Cres Island having lower average recorded temperature and humidity but higher air pressure than Klis (Table 5). Factor analysis of ecological and physical habitat data yielded three factors that explained 71.52% of the variability. Air and soil temperature showed significant factor loadings for factor 1 (0.81 and 0.84), relative humidity and air pressure for factor 2 (–0.74 and 0.77), and distance to hiding place for factor 3 (–0.77). MANOVA of the extracted factors showed differences between the two sites (F=31.40, P<0.05, Partial η2=0.49), but not between the sexes (F=0.82, P=0.49, Partial η2=0.03) nor the interaction between sex and site (F=0.42, P=0.74, Partial η2=0.02).
Variable Cres Klis Recorded air temperature (°C) 21.6±4.3 24.3±5.5 Annual air temperature (°C) 15.3* 16.1* Recorded soil temperature (°C) 19.4±4.0 19.5±2.5 Air movement 2.1±2.2 2.1±1.9 Relative air humidity at capture site (%) 43.0±10.5 56.1±13.5 Annual air humidity (%) 70.4* 58.1* Recorded air pressure (hPa) 1 010.7±4.0 999.9±6.3 Annual air pressure (hPa) 1 009.6* 1 000.4* Distance to hiding place 223.8±319.4 226.6±318.1 *: Taken from Meteorological and Hydrological Service of Croatia (Zaninović et al., 2008); standard deviation (SD) data not available.
Table 5. Measured parameters of physical and ecological habitat data with significant factor loadings from a factor analysis expressed as means±SD
Total vegetation, short vegetation up to 10 cm, herbaceous vegetation up to 50 cm, and tree vegetation were more abundant in Klis, whereas shrubby vegetation was more prominent on Cres. Factor analysis of the variables describing vegetation cover yielded two factors that explained 67.97% of the variability. Total and herbaceous vegetation showed significant factor loadings for factor 1 (0.91 and 0.78) and short vegetation up to 10 cm showed significant factor loadings for factor 2 (0.80). MANOVA of extracted factors showed differences between the two sites (F=18.98, P<0.05, Partial η2=0.20), but no differences between the sexes (F=1.12, P=0.33, Partial η2=0.01) nor the interaction between sex and site (F=1.55, P=0.22, Partial η2=0.02).
Morphological and ecological divergence in two populations of European glass lizard, Pseudopus apodus (Squamata: Anguidae)
- Received Date: 2019-09-20
- Publish Date: 2020-03-01
Abstract: The European glass lizard, Pseudopus apodus (Pallas, 1775), is a large, legless lizard with wide distribution across south-eastern Europe and eastern and central Asia. To date, morphological diversification among populations on a geographically small scale has not yet been reported in this lizard. Thus, we investigated the morphological variations and corresponding differences in habitat utilization between two populations of P. apodus inhabiting the same biogeographical zone within a relatively close geographic area. We hypothesized that minor differences in habitat could induce a significant level of morphological differentiation, thus indicating morphological plasticity in this species on a small geographical scale. We sampled 164 individuals (92 from the Croatian mainland and 72 from the island of Cres). Results showed that P. apodus indeed exhibited morphological differences between populations in the same biogeographical zone within a relatively close geographic area, with the Cres Island individuals being generally larger than the individuals from the mainland. Some ecological characteristics were similar in both populations (e.g., soil temperature, distance to hiding place), whereas others were distinct (e.g., air temperature and humidity). In addition, vegetation cover differed between the two sites, with more vegetation present on the mainland than on the island. Furthermore, the Cres Island population showed clear sexual dimorphism, which was absent in the mainland population.
|Citation:||Olga Jovanović Glavaš, Paula Počanić, Vanja Lovrić, Lorena Derežanin, Zoran Tadić, Duje Lisičić. Morphological and ecological divergence in two populations of European glass lizard, Pseudopus apodus (Squamata: Anguidae). Zoological Research, 2020, 41(2): 172-181. doi: 10.24272/j.issn.2095-8137.2020.025|