RELATIONSHIPS OF THE AUSTRALIAN CAINOZOIC SYSTEM. 
accepted rule has been questioned. Kayser remarks*, “ Although 
the principle underlying this classification has in progress of time 
proved in the main accurate, the percentages of living species 
originally adopted by Lyell for the various groups have not remained 
firm. Thus for the Pliocene we must take 40-90 instead of 35-50, 
and for the Miocene 10-40 instead of 17.” 
That the percentage method of correlating strata in widely 
separated areas is attended with serious difficulties has been early 
recognised, as, for example, by so able an observer of faunal 
distribution in the past and present as the late Capt. Hutton, of 
Christchurch, whose words on the subject we cannot do better than 
quote. After discussing the relative ages of two beds in the Wanganui 
System of New Zealand, which, on account of the close percentage 
of living to extinct species, Hutton was inclined to think, overlapped, 
and advocating the use of this same means of discrimination in 
determining the relative ages of beds in the same general area, he 
speaks thus :—t 
“ But it does not follow that this method can be trusted for 
correlating with accuracy sets of beds in widely distant areas. On 
the contrary, different districts have undergone different physical 
changes, and we have therefore every reason to suppose that 
alterations in floras and faunas would proceed with unequal rapidity 
in different parts of the world. At the same time, as the replacement 
of a whole marine fauna can rarely be sudden, it follows that the 
percentage system has some value even here. But it must always 
be used in conjunction with a comparison of the specific forms of 
the two areas. And here, again, it is only the wide-ranging oceanic, 
or deep-sea species—such as sharks, cephalopods, and a few 
bivalves—which should be depended upon for evidence, but these 
wide-ranging forms are of the very greatest value in correlating 
strata all over the world.” 
Another difficulty which confronts us with regard to the com- 
~ putation of the actual percentage of living species is the variable 
estimation by different authors of the value of minor characters ; 
as, for example, of shell sculpture in the group of the mollusca, 
where slight differences are seen only after careful study, or it may be, 
a trivial variation in form, which, if sufficiently constant would be 
regarded by some as specific. Hence, critical work on any fauna 
will always tend to lower the percentage of living species in any given 
fossiliferous series, and consequently to increase us approximate age. 
It seems, therefore, that we, in the southern hemisphere, to use the 
percentage method, must gradually erect a standard of percentages 
which will generally accord with the evidence afforded by a study 
of the strata in this part of the world; never forgetting, however, 
to exercise a cautious spirit in regard to species-making in working 
froin this stand-point. 
* Text Book of Comparative Geology. English translation by P. Lake, 1895, p. 330. 
{ Truos. and Proc. New Zealand Inst., vol. xviii., 1886 (for 1885), p. 345. 
[11 ] 
