Teeth in the Centetidee. 531 
from it. To his question, “ Why may not Hyanodon and 
Pterodon be an extreme development of that type of the 
Insectivora which is at present represented by Centetes ?” 
Winge* returned the categorical answer, ‘ Because Hycwnodon 
and Pterodon have not in the slightest degree a special 
resemblance to Insectivora, least of all to Centetes ; there is 
not one single point in which Centetes does not stand in sharp 
contrast to Hyenodon and the Carnivora.” Winge then 
proceeds to develop this view by interpreting the upper 
molars of Centetes in the following manner :—“ '’he hindmost 
of the three outermost cusps has disappeared,” he writes, ‘ as 
also the posterior of the two that immediately follow these on 
the inner side; on the other hand, the anterior two of the 
original three outer cusps and the anterior of the two that 
immediately follow these are still present in a particularly 
well-developed form; we also observe remnants of a single 
talon, which in other Centetide is well developed.” Conse- 
quently in the interpretation of the homologies Winge agrees 
in the main with Mivart. 
I have shown in an accurately-drawn figure of an upper 
M.1 of Centetes ¢ that in a perfect condition this tooth, like 
that of Potamogale, possesses not less than five outer cusps ; 
it is true that two of these are more strongly developed than 
the rest. In the figure in question I distinguished the main 
cusp of the tooth, situated on the inner side of those just 
referred to, by the numbers 4+6, in order to indicate thereby 
that in Centetes there seems to have taken place a fusion of 
the inner talon—6 in Winge’s figures and diagrams—with 
the main cusp, Winge’s 4. 
Cope has recently { reasserted his former theory, that in 
the existing fauna the “ tritubercular”’? form of molar as 
exhibited by Centetide, Soricide, a few Lemurs, and the 
majority of the Carnivora is the primitive one. In Centetidee 
and Chrysochlorid the rudimentary “ low cingulum ”’ on the 
posterior base of lower molars and a “ posterior cingulum” 
on upper molars are regarded as the first step towards the 
complication of the tritubercular molars. 
M. F. Woodward § follows Mivart and Winge, since he 
considers the supposed “ paracone”’ and ‘ metacone” of 
Centetes and Chrysochloris to be homologous with the “ exe 
* Vidensk. Meddel. fra d. naturh. Foren. i Kjébenhavn, 1882, p. 58. 
+ “On Megaladapis madagascariensis &c.,” Phil, Trans. R. Soc. Lond. 
vol. 185, 1894, B, p. 28, fig. 4. 
{1 ‘The Primary Factors of Organic Evolution’ (Chicago, 1896), 
pp: 145, 335, } 
§ Proc. Zool. Soc. London, 1836, pp. 588, 589, and pl. xxvi. 
