434 University of California Publications in Zoology [Vol.12 



The outline of the nasals is distinctive. In Castor subauratus shas- 

 tensis the nasals do not taper so regularly or rapidly posteriorly as 

 in Castor subauratus subauratus. The lateral nasal outline is, in 

 shastensis, invaded by the postero-medial portion of backward- 

 extending tongue of the premaxilla. Consequently there is formed 

 a "bay" in the lateral outline of the nasals posteriorly, which is 

 lacking in subauratus. In one or two specimens of the latter there 

 is a slight indication of such a relation, but the diagnostic value of 

 the character as given, both in adult and young specimens of both 

 species, holds in available material. The nasals tend to maintain 

 their full breadth farther posteriorly in shastensis than in sub- 

 auratus. The temporal ridges tend to form a distinct sagittal crest 

 posteriorly and to show a higher degree of approximation anteri- 

 orly in shastensis, in specimens of the age which in subauratus 

 exhibits a weak development of the crest posteriorly and a lyrate 

 arrangement of the ridges anteriorly. Not only is there developed 

 a distinct sagittal crest, but also a much more distinct knob dorsally 

 on the crest located about 29 millimeters anterior of the posterior 

 border of the lambdoidal ridge. The size of the cranium and the 

 condition of certain sutures, whether open or closed, constitute the 

 chief bases for age determination. Specimen no. 51477 of shastensis, 

 which is considerably younger than specimen no. 12654 of sub- 

 auratus, presents the same general arrangement of the temporal 

 ridges. Shastensis no. 50979, which is certainly not older than 

 specimen no. 12654 of subauratus, exhibits the characteristically 

 distinct sagittal crest posteriorly and the higher degree of approx- 

 imation anteriorly. The lambdoidal ridge is also more strongly 

 developed in crania of nearly equal age. The use of these char- 

 acters as subspecifically differentiatory might be unjustified without 

 a mass of material, were they not correlated with others ; for they 

 undergo marked modification with age within the same subspecies. 

 However, a different degree of development for the same age 

 undoubtedly can be relied upon. In this case, although the available 

 material is not sufficient to prove, for instance, that very old examples 

 of subauratus would not show the distinct sagittal crest posteriorly, 

 the higher degree of approximation of temporal ridges anteriorly, 

 and the more strongly developed lambdoidal ridge, it does suffice to 

 indicate a difference at least in rate of progress, and this difference 

 is valid as a subspecific character. 



The interorbital constriction is broader in shastensis. This meas- 



