134 CORRELATION OF GEOLOGICAL FAUNAS. [bull. 210. 
We have thus demonstrated the lapping of the faunas. ' This is a 
perfectly legitimate conclusion on the presumption that each of the 
faunas is not the universally distributed marine life of a particular 
epoch, but the fauna of a particular environment of that epoch. We 
are perfectly familiar with this discordance in the limits of dynasties 
of different races of peoples in human history. 
The facts have also shown that migration — not of single species, 
but of the whole fauna, a shifting of the metropolis with the limits of 
distribution of the fauna as a corporate whole — has taken place. 
This has been expressed in relation to formations by a transgression 
of one fauna over another, thus calling for the assumption that the 
limits of a formation based upon sudden change in the fossil contents 
can not be regarded as synchronous for two parts of even the same 
province and, wherever they are thus sudden and sharp, can not be 
synchronous with the limits of either the earlier or later fauna in 
evidence. 
Nevertheless, with all this lapping, shifting, and incomplete expres- 
sion of the faunas, the statistics also demonstrate the intrinsic value 
of fossils for measuring and indicating time. The sediments, whether 
by their lithological constitution, their structural form, or their strati- 
graphical position, furnish no such positive evidence of points or 
durations of geological time. 
The bionic method of measurement of time relations, though in the 
present state of knowledge it can not be used as a substitute for the 
more apparent structure scale, will serve to make the imperfections 
of the present methods apparent. Our ignorance of the actual as 
well as relative life periods of the great majority of species of paleon- 
tology makes it impossible to reduce life periods to actual years or 
centuries. 
It is also to be said that for the practical purposes of geological 
mapping and the descriptions of geological structure the formations 
are the essential elements, and a chronological classification of them 
is a convenient rather than an essential one. 
Nevertheless, whenever the attempt is made to become accurate in 
establishing time equivalencies or correlations, it is in this direction 
we must turn. The collection of statistics along the lines here pro- 
posed will facilitate the formation of a definite time-scale for geology. 
It is by making our knowledge of the composition, the range, and 
the geographical distribution of fossil faunas more complete and more 
exact that our classification and correlation of geological formations 
is to be perfected. 
At present we know too little about fossil faunas to be able to pre- 
dict in what manner their actual time limits will be defined or dis- 
criminated, but enough light has already been thrown upon the 
matter to show that it will be by means of the history which organ- 
isms have expressed in their continuous life and evolution that we 
may expect ultimately to mark off the stages of geological time. 
