Biotype, nomenclature and taxonomic diagnostic characters of Testudo hermanni hermanni GMELIN 1789 in Southern France with preliminary notes on comparative egg morphology with T. h. boettgeri MOJSISOVICS 1889.
A C Highfield
Distribution
The distribution of Testudo hermanni hermanni GMELIN 1789 in France is
restricted to the area of Provence and the Massif des Maures, between
Toulon and Saint-Raphael on the coast to Draguignan, Le Luc and Sollies
Pont inland. The main concentration of tortoises occurs in sites
containing suitable habitat between Collobrieres, La Garde Freinet,
Gonfaron and Gogolin. Few tortoises are seen close to human
habitation, even where the habitat appears suitable although there are
some localised exceptions to this. The majority inhabit the relatively
inaccessible hillsides and mountains which are a feature of this region
at a preferred altitude of between 500 to 700 metres.
Biotype
Two principal habitats are preferred, densely wooded hillsides and
coarsely vegetated gentle slopes. On an expedition in August we found
tortoises in both situations. Tortoises were found frequently in heavy
scrub and bramble thicket - making the location of specimens for study
a somewhat uncomfortable experience. At this time of year the
vegetation on the lower slopes is extremely dry and even in the more
densely vegetated areas humidity was minimal. Earlier in the year
green vegetation is more abundant and the tortoises are much more
active. In the very hot and dry conditions of mid-summer much of the
day is spent in retreat. Maximum activity occurs early in the morning
or in late afternoon.
Nomenclature
For some years this tortoise (along with those of Spain and the
Balearic islands) has been identified as Testudo hermanni
robertmertensi WERMUTH 1952. The eastern population of hermanni was
referred to T. hermanni hermanni GMELIN 1789. Recently however work by
Roger Bour (1987) has demonstrated that these attributions are
erroneous, and that the nominotypical species is actually that of
southern France, previously referred to T. h. robertmertensi.
Diagnostic characters
1. Supracaudal
It is frequently alleged that it is possible to distinguish between
the southern and eastern populations of T. hermanni by determining if
the supracaudal shield is divided or undivided. It is also sometimes
claimed that T. hermanni can be distinguished from T. graeca using the
same criteria (Devaux, 1988 p.22 fig. 1). In fact, neither is true and
this character is of no value in specific determination (Highfield,
1990a).
2. Plastron
The plastral markings of Testudo hermanni hermanni GMELIN 1789 are
characteristically formed of two almost solid dark bands running
longitudinally down the plastron (plate 2). Every specimen of this
sub-species examined by the author (several hundred, in situ) have
possessed this feature.
3. Dimensions
There appears to be considerable divergence in maximum size between the
eastern and southern forms, the former consistently attaining a greater
maximum observed size than those of the latter. Heron (1968) examined
over 400 French specimens and found that large females attained only
circa 155mm and males only circa 130mm. The largest animal of this
sample measured only 160mm. More recent observations indicate that the
largest size attained by females is 190mm and males 168mm (see Devaux,
1988 also Cheylan, 1981). The specimens observed during field studies
in France by the present author all measured less than 170mm.
By comparison the eastern populations of Testudo hermanni boettgeri
MOJSISOVICS 1889 can attain remarkable sizes; the author has previously
described a specimen which measured 264mm long and weighed 3,420g.
(Highfield, 1988) - believed to be a record for the species. Numerous
other specimens measuring over 200mm have also been observed.
4. Colouration & carapace markings
The groundcolour of T. h. hermanni is typically a bright golden yellow.
This feature was very obvious in all of the specimens observed in
France; it contrasts sharply with most specimens of the eastern T. h.
boettgeri where the groundcolour could best be described as a
greenish-yellow. Similarly, the carapace markings of the French
population seem to be unusually clear and well defined compared to most
eastern specimens. One feature particularly noted in all of the French
animals examined was a sharply defined "keyhole" marking in bright
yellow on the lower central posterior vertebral scute (just above the
supracaudal). We have not observed this character so sharply defined in
eastern specimens.
5. External cranial characters
There is a marked difference in appearance between the heads of the
western and eastern populations of T. hermanni. The western
population have much more elongate and smoothly contoured heads which
are typically snake-like in general appearance. The eastern
populations have a shorter, more bulbous head of quite distinctive
construction. The internal characters of the skull of T. hermanni
hermanni have been described by Bour (1989).
Conclusions
The southern French population of Testudo hermanni hermanni GMELIN 1789
is clearly distinguished from the more easterly populations of T. h.
boettgeri MOJSISOVICS 1889 by a variety of diagnostic characters. Of
these, the relative difference in size between similarly aged adults of
the two populations and the unusually bright markings of the southern
population are particularly notable. Recent studies also reveal a major
difference in egg morphology between the French and eastern European
populations as shown in table 1. (Highfield, 1990b and in litt).
Acknowledgements:
The author would like to thank Roger Bour, Bernard Devaux and Eric
Gibson for their assistance during his visit to France.
References
Plant species sympatric with Testudo hermanni in southern France
include; Opuntia ficus-indica, Sedum reflexum, Sedum album, Lavandula
staechus, Rubus spp., Thymus vulgaris, Helichrysum staechus, Euphorbia
characias, Genista scorpius, Salix spp., Cistus monspeliensis and
Juniperus oxycedrus. Tortoises are rarely encountered in habitats
thickly populated with Beech trees or where the ground is heavily
covered by ferns, favouring a more open aspect. Preferred nesting
sites include sunny slopes, deserted terraced olive groves or woodland
openings. We found several vacated eggshells in such locations. Most
of these appeared to have been deposited by predators, although in one
location may have been the result of a successful hatching. Other
reptiles sympatric with Testudo hermanni include Emys orbicularis
(European pond terrapin), Natrix spp., Laecerta spp. and Vipera spp.
Predators of T. hermanni include badgers (Meles meles), beech martens
(Martes foina) and weasels (Mustela nivalis). Of these, badgers and
beech martens are particularly prevalent predators of nests, in some
areas attacking up to 90% of sites (Stubbs and Swingland, 1984).
Analysis of faeces samples from wild T. hermanni in France revealed
traces of grass, opuntia seed and snail shells in addition to
high concentrations of helminths; no external parasites were observed
(Bruekers, 1986).
Following the discovery in the collection of the Strasbourg Zoological
Museum of the holotype of T. hermanni GMELIN 1789 (specimen reference
MZUS 121) it became clear that this specimen was derived from the
French and not the eastern population of the species. The
nomenclature of the original publication therefore stands for this
population by virtue of priority over Wermuths later name
robertmertensi which is consequently invalid. Of the eastern
population, this should be referred to Testudo hermanni boettgeri
MOJSISOVICS 1889 (lectotype Testudo graeca var. boettgeri, specimen
reference Senckenberg Museum, Frankfurt No. SMF 7836).
The respective type localities of each population are assigned to
Collobrieres, France for T. hermanni hermanni GMELIN 1789 and to
Orsova, Romania for T. hermanni boettgeri MOJSISOVICS 1889.
The plastral markings of the eastern form, T. hermanni boettgeri
MOJSISOVICS 1889 appear to be somewhat more variable but the
characteristic form is far less dense and well defined than that of the
French population. Some specimens examined have possessed plastrons
with dense markings which almost approach that of T. h. hermanni
however, so this character should not employed in isolation to diagnose
speciation.
Measurements of Testudo hermanni hermanni from the Balearic islands and
Spain tend to confirm that these dimensions are fairly typical of the
western population as a whole (Vroom, 1983, Kramer & Vickers, 1983,
Pelaz, 1988).
Within the eastern populations of T. h. hermanni GMELIN 1789 there are
considerable local geographic variant forms. These may well eventually
prove to be further sub-species.