2 20 POPULAR SCIENCE MONTHLY. 



saiirs, and the fact tliat marine sanrians had scale-covered skins. The 

 paddle was formerly a matter of conjecture, and, in the absence of 

 such remains, the sanrians were supposed to have had marine turtle- 

 like flippers. Prof. O. C. Marsh, of Yale College, was the first natural- 

 ist to discover sufiicient of the missing parts of skeletons to determine 

 that marine sanrians propelled themselves with paddles rather than 

 flipjDers. As to the scales and skin found perfectly preserved by 

 Snow, they do not differ materially from those of the Old World 

 lizards, the monitors, existing to-day. The paddles, skin, and scales 

 are very delicate functions, and it is remarkable that they should 

 have been preserved through millions of years. Williston says of the 

 paddles: "The specimen figured by Chancellor F. H. Snow, of the 

 University of Kansas, has been thoroughly cleaned from the matrix, 

 enabling an accurate drawing to be made, also a photographic repro- 

 duction as it lies on a chalk slab. The parts concealed beneath the 

 ribs and vertebrae have been carefully laid bare from the opposite 

 side and their position shown. The position of the paddle is a 

 natural one, and the fact is of interest as showing the general expan- 

 sion and curvature of the digits." The limb is very flexible, with 

 considerable space between the bones, which were but partly filled 

 out with cartilage, and must have had very free articulations. The 

 remains of the skin were found between the bones, indicating a thin, 

 pliable membrane, and extending fully between the fingers to their 

 tips. Small scutelike scales extended as far as the metacarjials. The 

 fifth finger is long. The paddles are slenderer, more flexible, and rela- 

 tively longer than in other genera, which, with other characteristics, 

 would show that TyJosaurns was the least lizardlike of the Python o- 

 morplia (Cope). As to the structure of the hind paddle, it is of 

 interest in having five functional toes, although Williston thinks that 

 the fifth toe was undergoing reduction, and that the first toe was not 

 as long as in the front paddle. He concedes five toes to the hind 

 paddle of Platecar-pus (Cope), but doubts, in the absence of a com- 

 plete skeleton, if Clidastes had more than four functional toes, as in 

 Mosasaurus. Upon this character, together with the absence of a 

 sternum, he has established two families, Tylosauridce and Mosa- 

 sauridce, and the two typical genera, representing the extremes of 

 development of this order of reptiles. 



Mosasaurs are known to have existed in many parts of the world, 

 New Zealand, ISTorth and South America, and Europe, the oldest 

 being regarded as the jSTew Zealand types, Liodon and Taniwha- 

 saurus. Dollo thinks that ISTew Zealand was the center of their 

 irradiation, where they appeared in the end of the Cenomanian,* to 

 appear in America in the Turonian,* whence they migrated to 



* Subdivisions of the Cretaceous formation. 



