414 On the Laws of the Distribution of Life and of Rocks. 
viously known (see § 21). There is no evidence that any fauna 
or flora ever was universal; physical considerations such as 
those detailed show that such a thing is impossible. 
31. The life on the earth is conveniently divided into a num- 
ber of groups called provinces, which have been produced by 
changes in physical geography modifying the natural distribu- 
tion in zones of varying organization. They are the evidence of 
these physical changes and the means by which they may be 
discovered. Every life-province, whether on land or in the sea, 
is only a geological fauna or flora which has not become fos- 
silized. Species have not been rooted to the area where found 
since their creation. Every geological fauna is only a life-pro- 
vince of the old sea in which the rock-material accumulated. If 
there is a change of life between two deposits, it indicates that 
a new life-province has migrated over the old one, and not ne- 
cessarily that there has been any denudation or any break in 
time. 
32. Since no life-province extends over a large area of the 
earth, it is impossible to identify distant geological formations 
by the similarity of their fossils. 
33. Since much disturbance may occur in an area adjacent to 
that where deposition is going on, it may happen that two or 
three groups of life succeed one another in one place, while, near 
by, the first group of the three remains stationary. This is one 
of the many difficulties that render it impossible to get any 
definite results from percentages in fixing the age of beds. 
Nothing is known of the duration in time of either recent or 
extinct species; and for the percentage method to give accu- 
rate results, it must be assumed that every species has exactly 
the same duration in time, stopping at a given point ; whereas 
it seems, from the case of some species sent to the antipodes for 
instance, that under changed conditions they may flourish better 
than ever. It is probable, too, that the distinctions at present 
in use between species add to these difficulties. And it is worth 
note that many genera survive from old paleozoic times; so 
that those periods, in a truer sense than that usually given to 
the term, might be called Eocene. 
34. Paleontology is the zoology of past times. But strati- 
graphical geology is, as it seems to me, the only means by which 
either the past or the present distribution of life can be under- 
stood. Both these powers for research need to be used to dis- 
cover the past mutations of the earth’s physical geography, by 
coordinating which changes it is possible to correlate strata 
over wide areas, and to obtain materials for their classification. 
35. No satisfactory classification of rocks can ever be made 
by fossils, for the reasons which have been given. Nor cana 
