SKELETON IN THE FLYING LEMURS. 157 



termination, each succeeding vertebra having its various apophyses disap- 

 pear, becoming lengthened and niucli simpler in character, straight and 

 subeylindrical, and with interarticulations of the most primitive kind. 

 As we come to the last four or five caudals they gradually shorten again, 

 being in this region represented by mere delicate straight rods, with 

 slightly enlarged articular ends, and finally temiinate in a minute osseous 

 tip, the smallest vertebra in the entire series, hardly worthy of the name 

 (figure 17). 



No neural spine occurs upon the atlas vertebra, although a very prom- 

 inent one, arising from the entire length of the neural arch is to be 

 found upon the axis. Its superior border is convex from before, backward. 

 The neural spine in the remaining five cervicals is very markedly reduced 

 in size, in each bone being represented by a mere stumpy process. In the 

 first dorsal it starts to increase again, being situated posteriorly and 

 directed backward. It then gradually increases in size; becomes nearly 

 vertical, quadrilateral in outline, and almost imperceptibly dwarfs once 

 more to include the eighth dorsal, when again its proportions increase, and 

 this continues as we pass through the lumbar region. In the mid-series 

 of the vertebraj in this latter locality in the Steere specimen, the neural 

 spine is a very conspicuous feature of the bone, being lofty, quadrilateral 

 in outline, and with a thickened superior border; it extends the entire 

 length of the neural arch, being about one-half the size on the last lumbar 

 vertebra. In other individuals these neural spines are not nearly so con- 

 spicuous (figure 17) as the ones just described (figure 7, Plate II). 



Neural spines are always present upon the sacrum, but here by fusion 

 they constitute a single plate of bone, being individualized only by the 

 thickened superior borders (figure 8). This plate is not so high f)Oste- 

 riorly as it is in front, while at the same time it extends the entire length 

 of the sacral neural arch. 



On the leading caudal vertebra we find a fairly well-developed neural 

 spine of about the same size as the one on the last dorsal vertebra. It is 

 centrally situated. This is the case with the next two succeeding caudals, 

 but in these the neural spine is becoming aborted, while in the fourth 

 caudal it may be only just evident, to entirely disappear in the vertebra 

 next following, and not be produced again for the balance of the vertebral 

 elements in the skeleton of the caudal appendage. 



A mere mesial raised line represents the liEemal process or hfcmapo- 

 physis, in the axis and the third cervical vertebra; for the rest of the 

 column no such structure is present, while the caudal vertebrse appear to 

 be entirely lacking in chevron bones. These last are found in many 

 mammals, as, for example, among marsupials, the Edentata, Cetacefe, 

 many rodents, carnivores, and even in some Insectivora, as Rhyncliocyon, 

 where, according to Flower, they "are well developed and bifid." ^^ 



"Osteology of the Mammalia (1885), 73. 



