1910.] Summary of the "Wedge Theory." 193 



(4) The evolution of the premolars in Tertiary inanimal.s may offer 

 an unsafe guide to the history of the molars in the early Jurassic, (p. 182), 

 and the theory (Wortman, Gidley) that the protocone is a "heel" develop- 

 ing pari passu with the talonid fails to explain the conditions in several 

 Jurassic Trituberculates with a high i)rotocone and very small talonid (p. 

 184). 



(5) The foregoing considerations lead to a new theory of the origin of 

 the trituberc-ular molar which may be designated as the " Wedge Theorij." 

 Its basal inferences are as follows: (a) The protocone in Jurassic Trituber- 

 culates is not a neomorph, developed as in the premolars of Tertiary mam- 

 mals (contrast Gidley, Wortman) but represents the "summit of the reptilian 

 crown" (Osborn) ; but this crown was not the "simj)le cone" conceived by 

 Cope and Osborn: {b) In the stage preceding the tritubercular stage, 

 before the acquisition of any secondary cusps the upper and lower molars 

 were not similar (contrast the Theory of Trituberculy) but quite dissimilar 

 in form, the upper molars being very wide transversely and somewhat larger 

 than the lower molars, which were short antero-posteriorly (Fig. 11). (c) 

 The relation of reversed wedges obtaining between the upper and lower 

 molars is not a secondary but a primary condition antedating the appearance 

 of the para- and metacones (contrast Trituberculy). (d) As regards the 

 Jurassic Trituberculates the secondary cusps arose approximately in the 

 locations in w^hich thev are observed without anv marked rotation, or migra- 

 tion of cusps (contrast Trituberculy). (e) The paraconid metaconid and 

 talonid arose from the basal cingulum; the paracone arose from the outer 

 part of the crown, the metacone budded off from the paracone; the parastyle, 

 metastyle and metaconule arose very early. 



(6) The Wedge Theory further differs from the Theory of Trituberculy 

 in the following particulars: 



(a) It looks backward to unknown Cynodontia with transversely 

 widened upper molars, rather than to the Protodonta and Triconodonta, for 

 an explanation of the peculiar spatial relations of the parts of the upper and 

 lower molars. 



{b) It centers attention as much upon the changes in the contour of 

 the croAvn as upon the development of cusps. 



(c) It devotes almost as much attention to interspaces, depressions and 

 valleys as it does to cusps. 



(d) It regards the principles of "change of function" and of "morpho- 

 logical correlation of parts" as of prime importance. 



(e) While recognizing what may be called the individuality and self- 

 initiative character of cusps, the present hypothesis nevertheless subordinates 

 the parts to the whole much more fully than does the Theory of Trituberculy. 



