80 
thesen der amerikanischen Palæontologen werden dadurch nicht be- 
riihrt.” (pag. 406). — But further the reptiliau teeth exhibit some- 
times complications to be interpreted as a clustering ( Scelidosaurus , 
Anthodon , Galesaurus a. o.), they must have been formed by the 
uniting of se veral selachian teeth and probably this may be the ' 
case with all reptiliau teeth. — The starting-point for the forma¬ 
tion of the teeth is the dentine tooth of the fishes; by coalescence 
of several of these is formed the reptilian tooth and through 
reiterated coalescence of these teeth the mammalian tooth is formed; 
the teeth of the three classes are thus not homologous, but repre¬ 
sent three stages in the tooth-evolution. „Damit ist zugleich der 
einfache mechanische Grund der allmåhligen Abnahme der Denti- 
tionen gegeben. u (pag. 408). 
In the numerous papers of Rose from 1892 onwards (“Zur 
Phylogenie des Såugetiergebisses” a. o.) the same theory is essen- 
tially set forth, supported by investigations of the embryonic devel- 
opment of the teeth. A multicuspid molar is not developed from 
a single dental papilla, but from just as many separate papillæ 
as the adult tooth has cusps; afterwards the enamel organ grows 
down like a cap around the summit of the papillæ and unites them: 
„Die Molaren der Saugetiere sind also entstanden durch Ver- 
wachsung mehrere einfacher, kegelformiger Zahne zu einen kom- 
plicierten, hochdifferenzierten Zahngebilde.” (Phyl. d. Såugetgeb. 
pag. 625). We have — it is true — no certain palæontological 
evidence of that; though it must be rernembered, „dass die Pala- 
ontologie lediglich in positiver Hinsicht beweiskraftig ist, niemals 
aber in negativer” (pag. 629); the assertion .of the palæontologists that 
the origin of a true cusp can be traced back to a small basal 
knob is of no great significance, for it can be shown in the mar- 
supials that each of the basal knobs originates from its own sepa¬ 
rate papilla. The embryological positive evidence must weigh more 
heavily than the palæontological negative. — Of the reptilian teeth 
the unicuspid are certainly homologous with the fish-teeth, the 
multicuspid are homologous with the thekodont reptilian teeth. 
