1907.] ELEPHANT REMAINS FROM CRETE, 239 



Admii-al Spratt *, both of which have since been visited by the 

 present writer f. 



Signor Simonelli, whose geological researches in Crete were 

 carried on in 1893, was seemingly the first to obtain Elephant 

 remains, which he procured from caves near Retymno, which is 

 situated on the north coast and lies between Khania and Oandia. 

 In a paper published in 1894 t he identifies the specimens, which 

 included a perfect mandibular ramus, as those of Elephas priscus 

 {■=E. cmtiqtous §), though no detailed descriptions or figures are 

 given, and I have been able to find no other published reference 

 to them. 



Unfortunately the collection of teeth and limb-bones which 

 forms the subject of the present paper consists of a small amount 

 of material only, many of the specimens being but imperfectly 

 preserved. JSTotwithstanding this they fall naturally into two 

 groups representing animals of different sizes, the smaller of 

 which agrees in this respect with F. melitensis, whilst a number 

 of limb-bones indicate an Elephant superior in dimensions to the 

 average Indian species {E. maxinius) of the present day, biit not 

 equalling the gigantic proportions at times attained by E. meri- 

 dionalis and E. antiqitus, though, as will be shown later, it is to 

 this last that they must evidently be referred. Thus at first sight 

 it seemed probable that the collection included remains of both a 

 dwarfed race and the parent form from which it had sprung. 

 However, before remarking on the relationships suggested by a 

 study of the specimens obtained, it will perhaps be more con- 

 venient first to briefly describe them, commencing with those by 

 which the pigmy form is represented. 



I. Elephas creticus, sp. n. 



The remains of the smallest of the Cretan Elephants were all 

 obtained from a much damaged and weathered cave-deposit in the 

 limestone cliffs near Cape Maleka in the west of the island, which 

 has ah-eady been described ||, and where only some teeth and 

 limb-bones of small rodents were found besides those under dis- 

 cussion. These latter include nine imperfect molars and a few 

 fragments, among which are a portion of an incisor and the 

 dorsal half of a vertebra. As this small race differs from those 

 of other Mediterranean islands, and its minute proportions being 

 seemingly the result of specialisation due to isolation in Crete, it 

 is suggested that it may be known by the above specific name 

 denoting its island habitat. 



Incisors. — Of the milk-incisors no specimen was procured, 

 while, as already remarked, only a fragment of the proximal end 

 of a permanent tusk (M. 9375) was obtained. From the limited 



* Op. cit. vol. ii. pp. 194-5. 



t Geol. Mag. n. s. dec. v. vol. ii. (1905) pp. 194-6. 



+ Atti R. Accad. Lincei, 5" ser. Rendiconti, vol. iii. 1894, Sem. 2, pp. 265, 268; 

 also ' Candia ' (Parma, ] 897), pp. 171-2. 



§ Cat. Foss. Mamm. Brit. Mus. part iv. (London, 1886) p. 122. 

 il Geol. Mag. n. s. dec. v. vol. ii. (1905) p. 195. 



17* 



