AUSTRALIAN HYMENOPTERA CHALCIDOIDEA, IV.—GIRAULT. 
141 
an index of blood relationship. I believe that I have thus insured identification of species. 
As to the genera time and increased knowledge will decide their fate at the which I am not 
and need not, for the present, be greatly concerned. I am not proud of them, since they are 
not mine but rather nature’s. 
In work of this kind to escape error is impossible blit the enormous number of errors 
so far made are due largely to one fact, namely, to what may be called piece-work in taxonomy. 
A systematist scatters his energy and attention over a large field with the idea, though why 
so I know not, that in so doing he is showing greater ability, sagacity and broadness; he 
describes a few species here one time, a few there another with the result that, not having 
an intimate knowledge of genera, these species often become misplaced, the descriptions are 
illy made and unbalanced and in the end there comes as a consequence that feeling of helpless- 
ness in regard to specific identities. In dealing with large complexes and in the relationships 
of orders and higher groups extensive and broad study is necessary but in describing species 
intensive study is as equally necessary and if variety is needed or desired a group should be 
taken by itself, studied intensively, advanced as much as possible and when completed left; 
any other group may then be studied in the same manner as the first. In this way, specific 
identities will be insured. Some years ago it was the common practice for many systematists 
simultaneously to describe species in several orders of insects but if these men failed to insure 
identity, how can a modern systematist hope to do so? Of those men, none are more notable 
for failure in this respect than Francis Walker. Was he broadening his knowledge by incor- 
rectly describing species or was he pampering his vanity? He was certainly mistaken and 
unquestionably wasted a large amount of energy and caused an enormous amount of delay; 
his efforts resulted in impediments. If the idea in describing species is to make them known, 
intensive study seems necessary ; if, however, the motive is otherwise, no study at all is needed, 
since in the end the purposes are served by merely publishing the new name and designating 
a type. The descriptions are made for others and the future. 
The principles underlying taxonomy are the same for all groups. He who knows them 
for one group knows them for all but is necessarily unacquainted with the kind and scope of 
variation peculiar to every group of animals and because of the enormous number is incompe- 
tent to judge of species within them without that insight and knowledge gained from compara- 
tive and intensive study. A sincere desire to advance our knowledge of specific forms requires 
intensive study. A desire merely to describe as many forms as possible, whatever the motive, 
requires none at all. 
Systematists seem to fail in another respect, a fruitful cause of trouble to the future: 
too often they are not self-critical. If errors are inevitable, which they seem at present, many 
could be corrected by their agents. This does not seem the usual practice. An individual 
may- see at least some of his own mistakes much more clearly than others, more especially if 
h*‘ constantly on the alert for them and constantly revising and criticising his own work. 
An error uncorrected by its author may cause endless trouble in the future. Frankness' aud 
more explanation in regard to changes made, a feeling of greater responsibility to the future, 
greater sincerity, less egotism and vanity and much more thought for clear, intelligible and 
accurate work, these in themselves, will produce tremendous advances in our knowledge of 
specific forms. Systematic work is fundamental; the descriptive part of it by no means 
comprises its scope. 
ACKNOWLEDGMENT. 
In the preparation of this paper I have been greatly aided by my wife who has, in the 
absence of unobtainable clerical assistance, copied out nearly the entire manuscript. 
