MEMBRA CIDJE. 
107 
notwithstanding’ the extreme divergences of their fruits and flowers. Similarly we 
may wonder at the usual grouping into one family of the rose, the plum, the 
strawberry, and the almond tree. 
These remarks are introductory to the facts that great divergence of form in tlie 
insects of the family under consideration must be regarded as sports from accidental 
types, that is, that often such variations do not afford generic or even specific 
characters of importance. 
Professor Huxley said, “those who refuse to go beyond fact rarely get as far as 
fact.” Almost every great step in science has been in anticipation of nature through 
the invention of hypothesis, even though, in the career of usefulness, such hypothesis 
may have proved to be erroneous. 
Hypothesis becomes more valuable in proportion to the number of facts it 
embodies. The evidence of evolution, the same thinker says, is inductively proved 
in accordance with the facts connected with the Eocene horse, exampled by the 
Eohippus of America. 
Phenomena which appeal to our senses ought to be thought out descriptively. 
We imagine hypotheses through the contemplation of similarities, but theories must 
be used principally as helps in search of truth, and as such, they must not pre¬ 
occupy the mind as if they were the laws themselves, or truths, which we can regard 
as absolutely conclusive. 
Modern research has largely expanded Fairmaire’s original genus Darnis, in which 
he described twenty-three species. Stal and recent authors have grouped numerous 
insects together, under a third section or sub-family (Darninse) as indicated above. 
The members of this group, however, are far from being homogeneous, and they 
hardly represent a natural classification. The materials available are yet too scanty 
to allow of such. Although about twenty genera are comprehended in the sub¬ 
family, some genera are represented by only three species, and others at present have 
only a single representative to give the characters for species which perhaps may be 
subsequently discovered. 
Attempts have been made, with more or less success, to found characters on the 
wing-veining, coupled with the number of the cliscoidal cells. But authors are not 
agreed as to what should be the normal number of discoidal areas. Really the group 
is difficult to diagnose, particularly as Canon Fowler has shown that such characters 
as the position and distances of the ocelli are too obscure for satisfactory use, as has 
been proposed by Stal and other writers. 
Fowler, in a measure, separates the curious genus Heteronotus from the rest, 
and places it last in the series, as having the tegmina twice as long as the wings, 
and as having the pronotum both spinose and nodose. It will be seen that the 
