LAND-TOETOISES AND AN EXTINCT LIZARD FEOM MAURITIUS. 317 



3. The fourth to seventh marginals, which connected the plastron with the three 

 middle costal plates, are very much steeper, and almost flat vertically instead of being 

 convex. 



Plastron B. T. triserrata. Typical. Eepresented by the two disconnected anterior 

 and posterior two-fifths of a male specimen. 



Plastron C. T. triserrata. The anterior half of a large male plastron ; greatest 

 width of fragment 42 cm. 



Plastra D, E, F, G. T. sumeirei. When Dr. Giinther wrote his Monograph he 

 could state categorically : — 



1. That the specimens with a nuchal plate, and with a double gular, came from 

 Aldabra. 



2. That the specimens without nuchal, and with a single gular, came from the 

 Mascarenes. 



3. That the specimens without nuchal, and with a double gular, are Galapagos 

 Tortoises. 



Now this statement cannot be upheld any longer, because among the materials 

 brought by Mr. Sauzier from the Mare aux Songes are the anterior portions of four 

 very large plastra, which diff"er from all the others previously received from Mauritius 

 and Rodriguez in the following points ^ : — 



1. The anterior lobe of the plastron is very much elongated. 



2. It ends in a fork instead of being rounded off. 



3. There were two gular shields, a right and a left, as indicated by the deep 

 impressions left upon the bones. 



Another difference is exhibited by the posterior portion of the plastron (PI. XLII. 

 fig. 8), which, from its size, thickness, and colour, I suppose to belong to the same 

 Tortoise as the anterior portion of the plastron (fig. 6). 



The posterior margin of this specimen ends ventrally in a much swollen and rugose 

 tuberosity ; dorsally it possesses a somewhat triangular, very strong tuberosity, which 

 seems to have fitted upon the ischiadic symphysial tuberosity of the pelvis, and which, 

 to judge from its roughness, seems partly to have been anchylosed with the pelvis. 

 None of the Mauritian specimens, hitherto known, show any such tuberosities ; but 

 they exist in some of the Aldabran forms, namely in T. elephantina, T. daudini, and 

 T. hololissa, not, however, in T. ponderosa, which latter has, by the way, been recognized 

 by Boulenger as the female of T. elephantina. 



In the configuration of the pectoral impressions, and in the whole shape of the 

 anterior lobe, the plastra D, E, F, G agree mostly with T. daudini. 



^ There are also five precisely similar specimeas of anterior plastral portions in the Cambridge Museum, 

 which had probably been received together with those Tortoise-remains from Mauritius which Professor 

 Haddon has catalogued and described in Trans. Linn. Soc. ser. ii., Zoology, vol. ii. (1879) pp. 155-163, pi. 13. 

 They have, however, remained undetermined and do not seem to have been mentioned. 



