LINNEAJiT SOCIETY OF LONDON. 1 9 



doubt will offer, by its completeness, an opportunity of revising 

 the subject in all its details. 



The contributions which, in my opinion, mark the most im- 

 portant advance, aud which throw a flood of light on the origin 

 of this Chelonian type and its distribution in space and time, 

 are tho3e referring to its occurrence in Madagascar. Indeed, 

 (xrandidier, as tar back as 1868, had mentioned his discovery of 

 Chelonian bones; but their true inter()retation was left to Vaillant 

 and Boulenger, wbo sbowed that they were true Testudo, and, 

 moreover, that they belonged to the Aldabran section. Vaillant 

 referred them to two species ; but additional materials, more 

 recently received in Ed gland, t^eem to indicate that the species 

 were more numerous, as we might have anticipated from the 

 wide distribution of the type in Madagascar, to which I have 

 referred already. 



The second important discovery to which I would draw your 

 attention is published in a paper by Dr. Gradow, who described 

 a number of Chelonian remains that were found in the Mare-aux- 

 Songes about 1890; this is the same locality from which I bad 

 previously described two species. Of the live '•'' forms''^ distin- 

 guished by Dr. Gadow, one claims our particular attention, 

 although it is known from fragments of the sternum only. It 

 differs from all the other Mascarene Tortoises, not only in having 

 a divided gular shield, but by having the anterior part of the 

 sternum bifurcated, reminding us of a simihir modification of the 

 corresponding part in Testudn atlas irom the Sivaliks *. I do not 

 think that this discovery affects my division of the Gigantic Land- 

 Tortoises into tbree sections ; I have no doubt that when we shall 

 have more information, notably of the form and scuteJlation of 

 the head, tlierewill be found sufficiently strong ground to establish 

 a fourth section for this new Mascarene species. 



Nor can we be certain that this Tortoise (or, perhaps, another 

 of those found in the Mare-aux-Songes) was really indigenous to 

 Mauritius, as Dr Gadow assumes. The Mare-aux-Songes is 

 a swamp close to the sea, separated from it by a ridge, and 

 situated near Grand Port, a place founded by the early Dutch 

 visitors, where, of course, in the course of succeeding years many 

 ships landed. Therefore, the possibility of some of the bones 

 from this locality being thuse of imported individuals cannot be 

 excluded; and this will apply more to the forms occurring in 

 small numbers than to the more common ones, although no one 

 will suggest any other origin but from some part of the Mascarene 

 region. 



It must appear a matter of some surprise that zoologists who 

 have studied the distribution of animals in space and time have 



* Is the bifurcated sternum a sexual character? Mr. Lydekker seems lo 

 think so, as far as Testudo atlas is concerned. It may be so in one species, 

 and yet be absent in both sexes in another. The large specimen with double 

 gulars in the military barracks at Port Louis is a male, aud shows no sign of 

 bifurcation. 



c2 



