1910.] Genetic Relations <if Chrysochloroidca and Centeloidea: Smuniarij. 2G7 



stress upon the similarity in brain structure (1907, p. 121), while Dobson 

 (1SS3) calls attention to important agreements in the musculature. Certain 

 of these similarities may well have been developed independently in the two 

 groups, but taken collectively they leave no doubt of a community of origin. 

 The jNIiocene Dimylidte of Schlosser, known chiefly from the dentition, 

 appear to be a specialized offshoot of the Erinaceid stem (c/. Zittel 1891-93, 

 p. 568). 



Chrysochloroidea. This superfamily (pp. 255, 259) is shown to be an off- 

 shoot of the stem of the Centetoidea by the peculiar characters of the dentition, 

 the general architecture of the skull, shape of the pelvis, etc. This inference is 

 reinforced by the evidence furnished by the tympanic region, the peculiarities 

 of W'hich are readily derived from the conditions preserved in MicrogaJc 

 (Leche). These further tend to disprove the suggestion of any near affinity 

 with the Talpida?: for, in the Chrysochloridiie the true tympanic bone be- 

 comes hemispherical and is encircled at its base by a continuous horizontal 

 ring, formed from tympanic branches of the petrosal, basi- and alis})henoids ; 

 and, although in the Talpidfe the same elements contribute to the complex 

 bulla, and thereby reveal the subordinal relationship of the two families, 

 yet here the proportions of the several elements are very diftejent, the basi- 

 sphenoid wing being very large, w^hile the true tympanic remains as a broad 

 subvertical zone with a large meatus. The malleus and incus of Chri/so- 

 chloris are extremely specialized and peculiar and throw no light upon its 

 relationships (Doran, 1879, p. 438, pi. 62, figs. 9-10). The stapes however 

 is of the widely open stirrup type w'hich is common among Insectivora. 



The Patagonian Miocene genus Necrolestes (p. 259) suggests as the com- 

 mon stem form of the Chrysochloridse and NecrolesticUie a brachycephalic 

 but long snouted Zalambdodont. x-Vs all known Zalambdodonts except 

 these two families are markedly dolichocephalic it is evident that the gap 

 between the Chrysochloroidea and the Centetoidea has not yet been bridged 

 over by palfpontological discovery. 



Centetoidea. Perhaps the dominant featur(> of the Centetidse is the 

 progressive lengthening of the skull especially of the middle and facial parts. 

 Correlated w'ith this and perhaps also with the large size of the olfactory 

 lobes of the brain, is the loss of the postorbital constriction and the resulting 

 cylindrical sha])e of the anterior part of the brain case. The malars are 

 absent, the nasals coalesce very early, the sagittal crest is very long and low, 

 the posterior mental foramen lies beneath m^. V\ are lost; the upper molars 

 consist of a high internal V-shaped cusp, reduced external cusps and a small 

 "pseudoprotocone" and hypocone. The lower molars have a very small 

 talonid without a distinct hypoconid. The dorsolumbar vertebrae (21-24) 

 are more numerous than in the Erinaceoidea (20). 



