91 
being larger and fewer in number and having a more developed 
cingulum than those of Dryolestes. But the outer portion of the 
Dryolestes molar resembles greatly the whole Triconodon molar with 
its three cusps and two fangs, whereas the inner portion of the 
Dryolestes molar, the “protocone”, being supported by a special 
fang, is very different from the central cusp of the Triconodon 
molar; it seems rather to be a strongly developed heel-like cingu- 
lum-cusp, In the contemporary Dicrocynodon this heel is still more 
developed, with two rudimentary fangs. From these facts it may 
be concluded that the primary (reptilian) cone preserves its original 
place on the outer side of the upper molar. — The lower molars 
of Dryolestes and (in a lesser degree) of Paurodon are typically 
tuberculo-sectorial; in both of them the pr d ., me d . and pa d . seem 
to have appeared on the very place they now occupy, there is not 
a trace of rotation. Consequently the pr. and pr d ., pa. and pa d . 
etc. can not be really homologous cusps. — In the Upper Cretaceous 
Laramie forms we find partly the typical “trigonodont” tooth, partly 
the quadritubercular, but the hypocone seems to be of different 
origin; sometimes (Telacodon) it is clearly a cingulum-cusp, in other 
cases it is an outgrowth of the protocone ( Pantolambda ) or meta- 
cone. — Hence it follows that the tooth-types represented by Mul- 
tituberculata, Triconodon , Dryolestes and Dicrocynodon must have 
originated independently of one another frOm the reptilian tooth so 
that the tritubercular tooth shows only one of the different ways 
* 
in the development of the mammalian teeth —(certain rodents, e. g. 
Palæolagus, seem to have their tooth-form derived directly from. the 
*j 
triconodont tooth) — and even its apparently homologous cusps 
may have a different origin. 
Osborn and Gregory’s “Evolution of mammalian Molar 
Teeth” 1907 is essensially a reedition of Osborn’s most important. 
papers; in chapter IX some other theories (of Ameg hin o, Fleisch- 
mann and Mahn, Forsyth Major, M. F. Woodward, Marett 
Tim s, Wortman, G-idley) are summarised and partly refuted; 
the paper of Winge is alluded to on some few occasions — is 
