‘OPENING ADDRESS.- *2 15 
some way; but I strongly suspect that in some cases the 
apparently useless and mysterious structural features of 
the adult animal may be explained as the remains of, or 
as having been caused by the presence of, useful larval 
structures. If that is the case, then we may have some 
specific characters which although not actually useful 
_ themselves are the representatives of the useful structures 
of a younger stage, and would therefore be maintained in 
each generation by the action of natural selection. This 
is a point to which I would especially direct the attention 
of those of our members who study the Crustacea. In 
that group there are many larval stages living under very 
varied conditions and possessing often remarkable larval 
structures which probably have some influence upon the 
form and other characteristics of the adult body. 
Dr. Romanes, in the paper with which he opened the 
discussion upon specific characters at the British Associa- 
tion, described the method which he had adopted in 
judging of the relative utility of the specific characters in 
the cases of some species of kangaroos. He placed the 
specific characters as given in the British Museum 
Catalogue of these animals in one of two columns, 
according as they seemed useful or indifferent in their 
nature, and he found that the column containing the 
indifferent characters was the longer one. But this 
method I hold will not give satisfactory results, because 
while the one column, that of the useful characters, gave 
positive results, the other column only gave negative 
results, i.e. it contained those characters in regard to 
which we do not yet know whether they are useful or not, 
and I would anticipate that as time and investigation went 
on, and we came to know the habits and the mode of life 
of the animals more minutely, character after character 
would have to be transferred from the negative to the 
