834: REPORT—1896. 
process known as develovment, and it is shown that the organs of the adult are 
not traceable back into definite areas of the ovum, or even blastula. So far as it 
oes, this is a most valuable kind of dissection; but it does not touch the question 
of how the hereditary powers of animals may be altered and so congenital inherit- 
able variations produced ; and this is the main problem of zoology. 
2. The study of individual variations. 
The drawbacks to this method are— 
(a) It is often quite impossible to distinguish a congenital variation from a 
variation produced in the particular individual examined by some accident in the 
environment. 
(b) Many of the most conspicuous variations are shown by a study of specific 
and generic characters to have had no part in the evolutionary process. © 
(c) It is not enough that a variation should occur ; it must occur in a sufficient 
number of individuals to prevent its being immediately swamped by intercrossing. 
3. The statistical study of individual variations or mathematical zoology. 
The drawbacks to this method are— 
(a) It is only capable of application to one character at a time, and a character 
is only a mental abstraction; natural selection acts on the balance of all the 
characters. 
(6) Even if we could establish that a certain value of a given character was 
accompanied by a low death rate, and that therefore this value was likely to 
become a specific character, the success of its possessors might be due to some 
obscure constitutional change associated with it. 
(c) But the only way it is possible to get such a result is to compare the 
variations with respect to a particular character of young and fully adult animals. 
To attribute the lesser number of deviations from the mean in the latter case to 
the death of individuals which had widely varied is to overlook the possibility of 
a self-regulating tendency in growth. 
The reason of the discontent with the morphological method is that it proves 
too much, z.c., the most contradictory conclusions may be drawn from the same 
premisses, for 
(a) Evolution is not only a progress from the simple to the complex ; degene- 
racy or simplification of structure plays an important part, and so also does 
homoplasy or parallel development, the evolution of similar structures in different 
animals independently. 
(5) It has been customary to postulate modifications as part of evolutionary 
history, the utility of which is to be taken on faith; and if this principle be 
admitted, the evolutionary theorist can, armed with progressive degeneracy, as 
well as progressive differentiation, derive any one animal from any other. 
Suggestions as to better ways of dealing with morphological facts :— 
1. There are many cases where the fact that a certain modification has taken 
place is doubted by no one; for instance, no one seriously doubts that Teredo and 
Pecten have been derived from the ordinary Lamellibranch type. 
The evolutionary changes which can be deduced from such cases as these are 
really the data the morphologist has to go on; if he departs from these he is on 
unsafe ground. It is possible that by a comparative study of such cases, ‘laws of 
evolution’ might be formulated. 
2. In reiation to the question of how degenerate and primitive structures are 
to be distinguished, we have to consider two subsidiary questions :— 
(az) Does the fact that an animal is obviously degenerate in some points 
invalidate any conclusions that may be arrived at as to its general primitive 
eharacter ? 
(>) Can an animal which has descended to a degenerate mode of life give rise to 
highly organised descendants P 
The answer to the first question is that all animals which in their general 
organisation are primitive are likewise degenerate, since they have by their 
degeneracy escaped competition with their more highly organised relatives. 
