213 
justified in the view he has taken, at any rate within certain 
limits. On the other hand, it may reasonably be maintained 
that this poverty of our collections is greatly reduced when we 
take certain groups of animals, or take the entire faunre of 
certain formations. It may reasonably be maintained that the 
known collections, for example, of Silurian and Devonian fossils 
are not so fragmentary as to vitiate all the negative evidence 
drawn from them. In North America, at any rate, where the 
Devonian follows the Upper Silurian conformably and without 
any palaeontological break of a marked kind, and where both 
sets of rocks are richly fossiliferous, it cannot be said that the 
poverty of our collections is such that no value can be attached 
to the absence of intermediate forms between the species of 
successive formations. If the Brachiopoda of these forma- 
tions alone be taken, there are many species of which many 
thousands of perfect specimens have been collected ; and if 
evolution can ever be proved by palaeontology, we might fairly 
expect the proof here. Similarly, our collections of the fossils 
of various of the Secondary formations, as regards the marine 
animals, are sufficiently complete to render any negative evi- 
dence drawn from them of very decided value. Upon the 
whole, therefore, whilst the fragmentary nature of our palaeon- 
tological collections must be fully admitted, it remains certain 
that as regards the marine faunae of certain formations, and 
as regards certain groups of marine animals, this imperfection 
of our collections is not so great but that we may attach 
considerable importance to any negative evidence that they 
may afford. 
4. The Vastness of Unrepresented Time. — Every modern 
geologist, probably, admits that the great geological formations 
are separated by vast lapses of time, more or less completely 
unrepresented by any accumulation of sediment. It is also 
universally admitted that all unconformities, whether between 
two formations, or as occurring in the limits of a single forma- 
tion, similarly mark intervals of time not represented in the 
area where the want of conformity occurs by any stratified 
deposits. Every want of conformity, therefore, undoubtedly 
marks a time in which great biological changes may have 
taken place without our having any record of them now pre- 
served to us ; and it may be, as believed by some, that the 
periods unrepresented by any fossiliferous sediments are ac- 
tually much longer than those of which we have material 
record in the form of strata charged with the remains of extinct 
animals. It is certain, therefore, that we have here a very 
marked cause of the imperfection of the palmoutological record ; 
