R. S. Lull — Cranial Musculature in Dinosaurs. 395 



part of the fenestra if it were present and which probably rep- 

 resents the point of final closure of that aperture.* 



There is a very precise analogy between the crest of a cha- 

 meleon and that of the Ceratopsia, as it is primarily, in each 

 instance, merely a backward extension of the parietal segment 

 of the skull to obtain a greater area for the origin of the tem- 

 poral muscles. This backward extension of the median region 

 especially gave greater area for the eomplexus major muscles, 

 which could then extend from their old insertion on the supra- 

 occipital backward along the interior surface of the median 

 parietal bar (interparietal), giving greatly increased leverage in 

 wielding the head. 



Fig. 6. Triceratops serratus. 



There seems to be an interesting correlation between the 

 development of the squamosal elements of the frill and that 

 of the paired horns. The broadening squamosals evidently 

 increased in size to allow for the extension of the great lateral 

 muscles of the neck from their original insertion on the exoc- 

 cipitals (vide supra, p. 391). This not only provides for larger 

 and more powerful muscles, but also gives greater leverage in 

 wielding the supraorbital horns. Centrosaurus apertus Lambe 

 (1904) (fig. 5) had a straight, powerful nasal horn, evidently 

 its chief weapon, and extremely small squamosals, as shown by 

 the very short parieto-squamosal suture, the major part of the 

 crest being composed of the parietal elements. In the Judith 



* Triceratops and Torosaurus it must be remembered, while contempo- 

 raneous, are not directly related, but represent parallel races derived 

 independently from Judith river ancestry. In Torosaurus the parietal 

 fenestra? were persistent. (Lull, 1908, p. 101.) 



