50 PROCEEDINGS OF THE ACADEMY OF [18*79. 



articulations to be the principal odontoinorpliic centres, as all 

 mandibular movements are regulated from them as axes. 



From the fact that the movements are from without inwards in 

 selenodonts generally, the outer series of cusps, in the advent of 

 such lateral movements, seem to be most compressed laterally 

 and their tips most deflected in an inward direction, and I shall 

 therefore consider the cause as acting from the outside instead of 

 from the inside as formerly. The paleontological evidence afforded 

 by symborodonts, where the selonodont or crescent-shaped cusp 

 first appears in the outer row of molar cusps, whilst the inner row 

 is still bunodont, is pretty conclusive. This singular combination 

 of cusp characters is also exhibited by Titanotherium, Palseosyops, 

 etc. Other series of ungulates show the almost synchronous de- 

 velopment of the crescent-shaped cusps in both outer and inner 

 rows, but earlier forms seem to indicate pretty unanimously that 

 the outer or buccal series are the first to be differentiated. In 

 Coryphodon, figured by Cope, 1 is exhibited the inward flexure and 

 flattening of the exterior rows of cusps of the upper molars, in a 

 perhaps unparalleled degree. 



The change of view is simply in regard to the manner in which 

 the modifying force is applied, and does not change the principle 

 involved, which assumes broadl}' that the lateral mandibular 

 movements produce lateral changes in dental forms, whilst on the 

 other hand reciprocating ones produce antero-posterior and pos- 

 tero-anterior ones. Having definitel}- concluded as to the direction 

 of the action of the forces, in evidence of which a great catalogue 

 of facts might he adduced, more indeed than my space and means 

 will afford, I will briefly consider that offered by the molar denti- 

 tion of Symborodon. 



In the diagram- which represents a rear molar of the right side 

 seen from in front, in which E is the external and I the internal 

 side; the simple anterior inner bunodont cusp B is seen to be but 

 little modified, whilst the outer selenodont cusp S is seen to be 

 much w^orn away, showing great disparity of development between 

 the two. The appearance of tubercle S in vertical median trans- 

 verse section, is indicated by the entire and dotted lines from a to a' 



' U. S. Geog. Surveys West of the 100th Meridian, Vol. IV. Vertebrate 

 Paleontology, 4to., by E. D. Cope. Washington, 1877. 



2 Prof. Cope has permitted me to make this sketcb from a specimen in 

 his superb collection of remains of these extinct animals. 



