STRUCTURE AND AFFINITIES. These turtles dif fer widely from ordinary chelonians, and com petent herpetologists differ as to their history and probable line of development. The factors in the discussion, and the varying views, are briefly presented by Hans Gadow in vol. viii. of The t'ambridge Natural History (London. 1901I. Gadow himself, supported by Boulenger, Cope, and others, believes Sphargis to be the sole remnant of a primitive group quite independent of the other eheloniang, and constituting with its scantily known fossil ancestors an order, Atheca'. opposed to all remaining turtles (order Theeo phora). (See TURTLE.) The opposite view is that the genus is a specialized offshoot from the typical Chelonia. and separable only as a family. The structure of till,: turtle is very peculiar, especially as to its 'shell.' This is not formed as in other turtles by an outgrowth of the spine, for it. is nowhere in contact with the internal skeleton. except by a timbal bone; but is a real
integument. continuous all around time body. and forming a jacket. This jacket consists of a dense leathery skin. in which are deeply im bedded a mosaic of many hundreds of little polygonal bony plates fitted closely together, and at intervals rising into twelve longitudinal ridges —seven dorsal, and five lateral and ventral. In young specimens the entire shell is soft. but ossi fication proceeds with growth, and when mature the integument is almost rigid, though thin. Such an integument more closely resembles that of a crocodile than that of a true turtle; but Sphargis has a plastron and neural plate.
BIBLIOGRAPHY. Consult Gadow (above cited) Bibliography. Consult Gadow (above cited) Boulenger, Catalogue of Chelonians in the British Museum (London. 1SS9) ; Case, Journal of Morphology, vol. xv. (London, 1897) ; Hay, merican Yaturalist, Vol. xxxii. ( Ph iladel ph i a. 189S).