THE EVOLUTION OF THE MAMMALIA. 29 
tered bones. The proper association of the various parts of a single 
skeleton may require the laborious efforts of many workers for 
years. In this work of piecing together the dry bones ludicrous 
blunders have been made, even by competent men, because of the 
extraordinary and unexpected character of many of the extinct 
animals. In one instance, the various parts of a single species, 
found at different times and places, were referred to no less than 
three distinct mammalian orders. The reason for this was that, 
prior to experience, no one could have ventured to predict the 
association in a single animal of such apparently incompatible 
characteristics. This example is by no means an isolated one. 
It is on account of this incompleteness of preservation that 
much which we should wish to learn concerning the structure of 
extinct animals cannot be determined. The soft parts, including the 
muscles, the viscera, the brain, the nerves, and the blood-vessels, 
are entirely lost, and can be inferred only from indirect, and often 
insufficient evidence. There are several groups of very ancient 
fossils which are so entirely unlike anything now living that no 
one has succeeded in interpreting the parts which are preserved, 
or in comprehending the organisms or their place in the scheme 
of classification. 
In the third place, the apparent order of succession of animals 
in the geological scale must not be too implicitly and uncritically 
accepted as a matter beyond doubt or dispute. One assemblage of 
animals may be of undeniably later date than another found in 
the same region, and yet the newer series may not have been 
derived from the older or have any direct connection with them. 
Any species of animal distributes itself as widely as possible 
from its place of origin, until stopped by some impassable barrier. 
Changes of climate, or in the connections between continental 
masses, may open at any time a way to the extension of an old 
type to new regions, and a great migration may occur. These 
migrations have been in progress for countless ages past, and they 
sometimes greatly confuse the record, when we attempt to read 
it in terms of evolutionary descent. A newer species which 
appears to have descended from an older one of the same region, 
