446 J. T. WILSON AND J. P. HILL. 
Dasyurus in the presence in the adult upper jaw (though 
only occasionally, according to Baldwin Spencer, 23, pp. 23 
and 26) of the successional last premolar as a “ minute and 
tubercular ” tooth (i.e. ‘‘p. 4,” cf. 22, p. 276). This tooth is, 
however, absent from the adult lower jaw. 
Thylacinus offers a still more advanced phase of develop- 
ment, for in it the successional premolar is now a well- 
developed tooth, larger than the second premolar (“p. 3” of 
Thomas), and is preceded by a milk premolar, though that is 
still only a minute and rudimentary one, and is shed during 
infancy (22, p. 255). 
In these respects Perameles may be regarded as standing 
just above Thylacinus, for while the last permanent pre- 
molar is well developed in both forms, its deciduous predecessor 
in Perameles is a well-formed though still relatively a small 
tooth, and is not extruded by the eruption of p. 3 (‘Thomas’s 
“py, 4”) until the animal has attained about three fourths 
adult size. The condition in Perameles may thus be re- 
garded as itself intermediate between that in Thylacinus 
and that in Didelphys, where the deciduous premolar is a 
very large and multicuspidate molariform tooth. 
It is beyond question that these several conditions repre- 
sent stages in a process of reduction affecting first the 
deciduous, and then the successional premolar, until the 
Dasyurinz condition is reached. 
Is it not in the highest degree probable that in this series 
we have to recognise, as it were going on under our eyes, the 
same process of reduction which, at an earlier epoch, has 
brought the other antemolar milk-teeth down to the condition 
of mere calcified vestigial structures—the so-called “ prelac- 
teal”? rudiments? In other words, the deciduous premolar is 
simply the hindmost member (so far as we know) of a dental 
series exhibiting various stages of a retrogressive process, some 
of the more anterior members of the series being represented 
by various ‘ prelacteal ” rudiments, 
