no REPTILES 



ficult to accept the latter explanation in the case of the extinct 

 group referred to above. In these reptiles, which occur in the 

 Permian strata of North America and constitute the family 

 ClepsydropidcB of the order Pelycosauria, the spines of the 

 vertebrae of the back are elongated to a height of about two 

 feet, whereas the bodies of the vertebrae themselves are not 

 more than an inch in diameter. In some cases, as in Naosaur- 

 ns claviger, these upright vertical spines carried horizontal 

 projections like yardarms. In life it is difficult to suppose that 

 these remarkable " masts " could have had any other function 

 than to support a large sail-like or fin-like membranous expan- 

 sion. Now such a sail-like structure, so far as we can see, 

 would have been a monstrous inconvenience and impediment 

 to a terrestrial animal ; and it can therefore only be concluded 

 that Naosaurus and Clepsydrops were, to a great extent at any 

 rate, aquatic, and that the sail-like structure had some con- 

 nection with the action of swimming. Possibly it may have 

 acted as a real sail, as is said to be the case with the huge 

 back-fin of the existing swordfishes of the genus Histiophoriis. 



A very large number of reptiles, dating from the Cretace- 

 ous onwards but apparently attaining their maximum at the 

 present day, have acquired an extremely elongated, slender, 

 and more or less nearly cylindrical form of body, with the 

 partial or complete loss of all external traces of both pairs of 

 limbs. In most cases this snake-like type of body has ap- 

 parently been acquired in the first instance as an adaptation to 

 a gliding mode of movement on the ground, a burrowing, 

 arboreal, or aquatic mode of life having apparently been some- 

 times developed from this original gliding type. In other 

 cases, however, as in the skinks, the adaptation has been 

 directly developed for a burrowing existence. 



Adaptation to the gliding or limbless burrowing type has 

 been independently acquired in at least six totally distinct 

 groups of reptiles. Firstly, we have the entire group of 

 snakes, constituting the suborder Ophidia of the order Squamata, 

 and represented by several families ; secondly, the family 

 Anguidcz, or snake-lizards, as represented by the English 

 slow- worm or blind-worm (Anguis fragilis), and the much larger 

 glass-snake or scheltopusik {Opliisaurns apus) of the Continent ; 

 thirdly, certain types allied to the Anguidce ; fourthly, the am- 



