FOSSIL VARA NID AE AND MÉGALANIDAE. 
403 
the other hand, compared with V. griseus Daud., to which it is zoogeo- 
graphically nearest related, not only presents striking differences in size, 
but important morphological dissimilarities as well. However contrarily to 
the species above mentioned, the reptile nearest related to our fossil, viz. 
V. griseus Daud., presents a different geographical distribution, and since 
Varanidae are now extinct in Europe, the connexion was relaxed by a 
further reaching geographical différenciation, which resulted in the deve¬ 
lopment of particularities of specific value. On this subject Bolkay writes 
as follows (op. cit. p. 226—227): «Varanus ..deserticolus has a relative in 
Varanus griseus Daud. . . .» which statement appears quite natural, the only 
question to be settled being how this connexion may have taken place. 
Regarding the mode of life (biology) of this reptile not much can be 
said. Whether the synonym denomination of «deserticola» is justifiable 
from an œcological point of view for our fossil Varanus, cannot yet be estab¬ 
lished ; a point to the affirmative could reside in the fact of the Steppe- 
type being represented, according to dr. Kormos’s verbal communication, 
by the Præglacial fauna of Beremend and Csarnóta. However the question 
remains whether V . marathonensis was in reality, strictly speaking, a desert 
form, such as V. griseus Daud., its recent relative, or may it be admitted to have 
lived in the vicinity of water, as' V. niloticus L. or V. salvator Laur. Palaeon¬ 
tological data have as yet given no clue to'the question ; as the biology of 
V. salvator Laur. has taught us, this reptile in swimming uses its bilaterally 
compressed tail just is the custom of the recent water Lacertilia or as did 
the marine Sauria. The caudal vertebrae of such reptiles present strongly 
developed, long proc. spinosi, which character I have not been able to 
observe on V. griseus inhabiting dry land ; in this species the proc. spinosi 
are not only shorter — a natural consequence of other proportional dif¬ 
ferences, but rather bent backwards, thus not appropriated for sup¬ 
porting a larger lateral surface. Contrarily to this V. marathonensis shows 
remarkably long proc. spinosi, — as has been mentioned in the description 
— protruding perpendicularly, from the upper arch, just 
as or perhaps yet more vertically erected, than in V. salvator Laur. or 
V. varius Shaw ; whether this particularity is to be. considered as an ana¬ 
logous character, admitting of a conclusion to aquatic habits in V . maratho¬ 
nensis, remains to be established. 
On the other hand the presence of cértáiri dsteológical characters 
cannot be denied (on the posterior surface of the proc. obi. posteriores 
of dorsal vertebrae, the conformation of proc. obi. posteriores on the 
caudal vertebrae, the position of proc. transversi on the caudal vertebrae), 
which would prove to some, at least convergent, affinities between our 
fossil reptile and the Indian V. salvator Laur. and, hereby, with the Aus- 
26 * 
