210 On Root-Jivisw)) in the Molar Teeth q/ Trifyloclon. 



XXII. — If^ofe on Root-division in the Molar Teeth of 

 Tritylodoii. By Dr. Branislav Petronievics. 



The longitiulinal root-division in tlie molar teeth of Tri- 

 iylodon was derinitively established by the present writer in a 

 previous paper (comp. " On the Skull of Tritylodon longcevu.i/' 

 in Ann. & Mag. Xat. Hist. [8] vol.xx., 1917, p. 283). The 

 transverse root-division in the molar teeth of the similar type 

 Siereognathns, also established by me in another paper 

 (" Note on the Lower J^iw of Stereognathxis ooliticus" in 

 Ann. & Maj?. Nat. Hist. [9] vol. i., 1918, p- 67), suggested 

 the possibility of the same division in Tritylodon. This 

 suggtstion was confirmed by a new preparation of the 

 specimen (M. 1951) in the British Museum during my last 

 stay in London in 1920. 



The text-figure sliows the hinder side of the recently 



h 



Section of the penultimate molar of Tritylodon longavus, Owen, left side. 



Nat. size. 



prepared penultimate molar of Tritylodon situated on the 

 left side of the skull (comp. I. c. tooth 5 in text-fig. 2), the 

 same which shows the longitudinal root-division also (comp. 

 the photograpli, /. c. pi. x. fig. 2). But while in its longitu- 

 dinal direction it lias only two roots, its transverse direction 

 is characterized by three distinct roots («, h, y in the figure), 

 corre-sponding to the three longitudinal rows of cusps. The 

 middle root (6 in the figure) — the shortest of the three — is 

 the largest, is of triangular shape, and shows the closed pulp- 

 cavity in section. The inner root (a in the figure), closely 

 applied to the middle one, is of rectangular shape, wliile the 

 outer (y in the figure) is the longest, but in the present 

 state of the specimen unfortunately represented by a small 

 fragment only (in the photograph, I. c. pi. x. fig. 2, it was 

 still completf). Behind the penultimate molar in question 

 there was in tlie specimen a root-impression of the last tnolar 

 (which has been cut away by the new preparation), showing 

 only an imperfect transverse division of this root. 



It is, I think, not necessary to emphasize the importance 



