Identification of a new form of Testudo sp. of Greece from two individuals in captivity: Testudo graeca microtera mihi

___        

       In 1989 I started rearing two male turtles from the neighborhoods of Xanthi, in the Greek region of Thrace, but I realized that they were “anomalous” only after few years. In fact, other turtles, reared under the same environmental and feeding conditions, showed regular and considerable growth rates; conversely, in 4 to 5 years the two individuals studied, very similar to each other in size and morphology, attained body sizes clearly close to the asymptotic value, i.e. close to the asymptotic value of the logistic growth curve and much smaller than the normal sizes of the most closely related subspecies, Testudo graeca ibera. The size has remained unchanged until now (2002) in one individual and did not vary until its death (1997) in the other.

        Figure 1 shows the body weight growth pattern in the two individuals studied (bottom curves), compared with that of two individuals of Testudo graeca ibera (top curves). Moreover, it reports data on two additional individuals of the same subspecies:

          reference                     sex                    place of origin           

              85                          m              Kimmèria (Xanthi, Greece)

              86                          m                     “

              75                          m              Canakkale (Turkey)

             125                          f                near Istanbul (Turkey)

               30                          f                Arèti (Salonika, Greece)

               31                          f                    “  

Figure 2, taken from M.Cheylan (1981), shows the two localities of Kimmèria and Arèti. The French author raises some doubts about the presence of other forms of turtles, besides Testudo h. hermanni, in this area.

        As regards figure 1, it is evident that the two individuals studied had attained maturity since 1996-97 with a weight of about 900 grams, much lower than the values of the other individuals compared, which, in the graph, fall between 3000 and 4000 grams, with the possibility that some of them (no.75) may still grow significantly. It must be pointed out that the fluctuations in the values detected depend on the inaccuracy of the instruments of measurement, as well as on the different periods of the year the measurements were taken.

       Similar trends are shown by the other important measures of the carapace. Their maximum values in mm are summarized in the following table:

       Kimmèria                   160                    125                      80 

          Turkey                   270                    200                    130 

          Arèti                      240                    185                    115 

       Moreover, when the two individuals were collected, their body sizes were such as to allowthe estimation of their minimal age in the order of 7 to 8 years, assuming a regular and sufficient body growth since their birth. Under less favourable environmental conditions,the estimated age should obviously increase. Moreover, the logistic growth curve of these two individuals had already its peak when they were collected, since the trend of the growth curve concerning the four parameters examined (maximal weight, length, width and height of carapace) showed the typical longer life-cycle phase of the progressive slowing down of body growth, which occurs when the curve starts declining.

      Morphologically, the two individuals showed a hump on the plastron in correspondence of the median contact area between the two anal scutes and adhering to the posterior margin of the femoral scutes, which were not covered by the hump. Moreover, observed from above, the anterior portion of their carapace appeared less straitened.

     Some data and characters of the two individuals are reported in detail in the following:

      Individual no.86. Specimen still alive and in a very good condition. Periodically it attempts to copulate with a Testudo graeca graeca female. Gular plates greatly sunken with respect to the humeral ones. Concavity of plastron starting slightly from humeral plates to become deep in the region between abdominal and femoral plates; it steeply stops on the anterior edge of anal plates, making the hump of anal plates more evident. Pectoral scales very narrow in the central region of the plastron. Nuchal scale narrow. Sizes on April 30 2002: weight, 850 g; max. length, 162 mm; max. width: 125 mm; max. height: 77 mm.      

      Individual no.85. Dead in 1997. Gular plates sunken posteriorly. Concavity of plastron as in no.86 but less marked. Hump less marked but slightly shifted towards the posterior edge of the femoral plates. Pectoral plates as in no.86.Nuchal plate narrow. Ten marginal plates each side instead of 11. Sizes of carapace: max.length, 150 mm; max.width: 124 mm; max.height: 78 mm.

       Figure 3 is a dorsal representation of individual no. 86 together with no.75.

      Discussion.    

       The two males studied are suggestive of a subspecific small-sized form of Testudo graeca rather than of an isolated or local case of dwarfism. In fact, the locality of origin of the two specimens is situated in the western border area of Testudo graeca ibera, and therefore a geographical separation might have occurred in a relatively distant past. Moreover, besides size, there are some significant differential characters, among which the differently marked abdominal concavity. This hypothesis is also supported by the fact that of the two specimens from Arèti no.31 might be a female of this new form. In fact, it is clearly an “old” adult specimen due to the conditions of the carapace and the striation of the horny plates, with a slight concavity of the plastron, evident hum in the contact area between the anal scales and absence of the nuchal scale. The sizes of this specimen, preserved in alcohol, are the following: max. length, 190 mm; max.width: 140 mm; max.height: 91 mm. As already mentioned, the nuchal scale is absent, since the first two marginal scales touch each other by two-thirds, whereas the remaining third shows a V-shaped opening between the end of the marginal plates.

       Hybridization between Testudo graeca and Testudo hermanni should be ruled out due to the absence of characters of the latter in the two specimens studied. Moreover, an effective reproductive isolation between these two species is well known in literature in the areas where they coexist. The marked concavity of the plastron in males might suggest hybridization between Testudo graeca and Testudo marginata; however, the distribution of the latter species does not attain the territory included between Thrace and eastern Macedonia.

    Conclusions.

      The subject no doubt requires a deeper insight on the field and possibly in the laboratory. I provisionally assign the two males from Kimmèria examined to a new subspecies of Testudo graeca, indicating the still living specimen no.86 as the holotype and no.85 as the paratype. As soon as possible, this material will be deposited in the herpetological collection of the “La Specola” Museum of Florence.

      For its sizes, clearly smaller than those of Testudo graeca ibera, I name the new subspecies Testudo graeca microtera (Greek: “smaller”).