216 MICHIGAN ACADEMY OF SCIENCE. 



THE LIMITS OF DIFFERENCE IN SPECIFIC AND SUBSFE- 



CIFIC DISTINCTIONS. 



HUBERT LYMAN CI-ARK. 



It was my misfortune last summer to feel called upon to criticise some 

 recent ornitholooical work in which the process of recognizing subspecies 

 had been carried to the extreme, and my opinions were published in 

 Science, August 8, 1902, under the heading "So-called Species and Sub- 

 species." In the same journal, September 5, Dr. J. A. Allen, the well 

 known zoologist, criticised my opinions as those of a layman, and em- 

 phatically denied two of my main contentions. As nothing is gained 

 by newspaper controversy, I made no reply, but the questions involved 

 are extremely important and after six months further consideration of 

 them. I have decided to set forth what seem to me some of the funda- 

 mental rules, which ought to govern work in systematic zoology. First, 

 however, as Dr. Allen has challenged my right to opinions on the subject, 

 it is only fair to say that, although I have never described a new or 

 supposedly new bird, I have had occasion to examine carefully several 

 thousand specimens of echinoderms, and have been under the necessity 

 of naming a number of new species in that group, so that I am not an 

 entire stranger to the numerous perplexities of the systematist, to which 

 Dr. Allen refers. Now I freely admit that from the systematist's point 

 of view, birds are more perplexing than echinoderms, and that Dr. Allen, 

 both because of his naturally judicious temperament and by his many 

 years of experience amid exceptional opportunities, is far better quali- 

 fied to discuss this subject than am T. Yet I do feel, that whether the 

 animal be a bird, a fish, a worm or an infusorian, the essential principles 

 of systematic zoology ought to be the same in all cases, and that any 

 zoologist who has wrestled honestly with the knotty problem of specific 

 distinctions is entitled to opinions on the subject. I therefore venture 

 to state some of these essential principles as they appear to me. 



1. Characters icJiich are 7iot sufficiently cotu^picnous, so that they can 

 be stated in huiffKarje or figures of some sort, ought not to he made the 

 basis of a new name. 



This principle appears so axiomatic that an apology would be made 

 for stating it here, if it had not been seriously questioned by Dr. Allen. 

 He says: "In ornithology, and especially in mammalogy, perfectly 'good 

 species' are often so similar in size and color that even the expert cannot 

 satisfactorily identify them from descriptions, and hence, almost from 

 time immemorial, direct comparison with authentic material has been 

 necessary in order to settle such difficult cases. As all experts in this 

 line of study well know, forms that may be indistinguishable by descrip- 

 tions are, when brought together, and especially when series are compared, 

 so noticeably different Ihat there is no trouble in distinguishing them at 

 a glance." Now I must confess that after giving these words careful 



