229 
of a positive nature. In so far as they are negative, there is 
doubtless room for much divergence of opinion : — 
] . The common phenomenon of closely-allied forms directly 
succeeding one another in time renders it a reasonable suppo- 
sition that in certain zoological groups many forms so distinct 
as to have been described by competent observers as distinct 
species , may have descended from a single primitive ancestral 
type. 
2. The evidence at present in our hands is opposed to the 
view that this production of groups of allied forms from as 
many primitive types has been effected solely or mainly by 
“natural selection”; though it is probable that this agency 
may have played a subordinate part in the process. 
3. New types of life are constantly making their appearance, 
without, so far as we know, being preceded by any closely-allied 
types ; and we have, therefore, no positive ground for believing 
that the origin of such types is due to evolution from pre- 
existent forms. 
4. Variability — even in the most variable groups — has never 
been shown to be indefinite ; but, on the contrary, appears to 
be confined within certain fixed limits for each species; in 
some cases wide, in others very narrow. Paleontology shows 
no instances in which we can positively assert that the varia- 
bility has been unlimited ; and though we meet with types 
connected bv intermediate links, we have also to account for 
the existence of a vast number of isolated forms, which, so far 
as our present knowledges goes, stand alone, and are not in- 
timately related to other forms. 
5. Even where we find types which may be regarded as 
strictly transitional or intermediate (as Hipparion in its rela- 
tion to Anchitherium on the one hand, and Equus on the other 
hand), we nevertheless are confronted with forms which are in 
themselves quite distinct, and which could not be confounded 
with the forms which they serve to connect. 
6. We cannot fairly have recourse to the “imperfection of 
the record,” as satisfactorily explaining the absence of the 
numerous intermediate types required by the Darwinian theory. 
Such imperfection admittedly exists, and is in some instances 
almost hopelessly great. On the other hand, we have had in 
other instances a fairly complete series of successive forms pre- 
served to us. This is the case with the Brachiopoda and 
Cephalopoda, for example, and it is by these and similarly well- 
preserved groups that any theory of the origin of species will 
have to be tested. 
7. The examination of such tolerably complete groups affords 
