Search

The processes underlying macroecological gradients in body size are widely debated, in part because their intraspecific variability remains poorly described even in well-studied taxa such as vertebrates. In this study, we investigated how climate, habitat, genetic lineage and sex explain body size variations in French populations of the European pond turtle (Emys orbicularis). We measured 7016 adult individuals captured in 41 populations, covering most of the species’ distribution in metropolitan France, including Corsica. Body size variation in our sample was wide and comparable to that found across the species’ worldwide range. Variation was similar in magnitude at regional and local levels, suggesting that body size is influenced by local factors as much as by regional factors such as climate or genetic lineage. Smaller sizes were associated with Mediterranean or altered oceanic climates, and with two lineages (E. o. galloitalica and E. o. galloitalica/E. o. orbicularis), while larger sizes were associated with northern environments and the orbicularis lineage. Body size variations recorded at local level reflect an adaptive response to environmental constraints, suggesting that habitat is also an important factor in understanding size variation.

Abstract Over a period of two years we used radiology to investigate breeding traits in females of two western populations of the threatened Hermann's tortoise Testudo hermanni hermanni. The main purpose of the study was to see if reproductive traits – clutch size, clutch frequency, annual egg production and egg width – varied between populations in Corsica (France) and the Ebro Delta (Spain), and if these traits were affected by female body size and amounts of spring rainfall. All the breeding traits analysed were greater in the Corsican population than in the Ebro Delta population. These differences were also significant when we used female body size as a covariable. In both populations the amount of spring rainfall affected breeding traits but not clutch size. Nevertheless, only in the Corsican population did clutch frequency, annual egg production and egg width increase significantly in the year with a wet spring. Contrary to the predictions of Optimal Propagule Size theory, we found that egg size and clutch size increased with female size, and that both variables were independent. On the other hand, clutch frequency was not related to female body size. Thus, in the Hermann's tortoise, as clutch size did not change between years and clutch frequency was independent of female size, all females are able to increase their reproductive output in years with favourable conditions by increasing their clutch frequency. Nevertheless, other factors besides female size and rainfall may influence in a highly complex way variability in breeding traits among populations.

Growth ring counts on the shell have been widely used for age estimation in Chelonians. However, as stated by Wilson et al. (2003), most studies have applied this method without assessing its reliability by proving that 1 : 1 ratio between number of rings and real age exists. In the present study, the reliability of this method is analysed for a population of Hermann's Tortoise, Testudo hermanni introduced in the Ebro Delta (NE Spain). Age estimations were obtained from direct observations of tortoises in the field in 2000 (n = 82) together with those from photographs of the same and other individuals of the population taken between 1991 and 2001 (n= 356). A second photograph was taken at one or more years after the first one for 101 individuals. Results of Model II linear regression analysis indicated that the method was reliable only for tortoises between 0 and 7 years old, whereas tended to underestimate age for those between 8 and 11 years. Since, sexual maturity in this population is attained around 8 years (mean for both sexes), ring counts are only reliable for juveniles and subadults. Finally, it is noteworthy that the results coming for photographs were equivalent to those coming from direct observations in the field.

Abstract For the first time, inguinal palpation effectively detected calcified oviductal eggs in a tortoise species. Inguinal palpations for 36 Hermann's tortoises, from six samples during spring 1999 and spring 2000, were verified with X-ray radiography. Palpation results did not differ significantly from radiography results and did not differ between two independent observers. A logistic regression analysis indicated that palpation success depended on the presence of oviductal eggs, and not on observer, year (1999 or 2000) or tortoise age class (young adult or old adult). The probability of correctly classifying females as gravid or nongravid through palpation was 78.4%. Although inguinal palpation did not allow detection of all clutches, it is a simple, inexpensive and effective means to determine the breeding status of Hermann's tortoise populations.