DIFFERENCE IN SPECIFIC AND SUBSPECIFIC DISTINCTIONS. 217 



thought I am unable to believe that the validity of my contention is 

 affected. I am utterly unable to conceive of two objects, which I could 

 "distinguish at a glance," the differences between which would be so in- 

 tangible that I could not state them "in language or figures of some sort." 

 As to the comparison of specimens with types or other authentic speci- 

 mens, "from time immemorial," surely it is well known that the neces- 

 shj for this is due to imperfect, inaccurate and erroneous descriptions, 

 and not to the fact that "perfectly 'good' species" cannot be distinguished 

 without comparison. If a character, whether in color, size, form, texture, 

 odor, notes, habit or anything else, cannot be detected by sight, touch, 

 smell, taste or hearing to such a degree as to admit its translation into 

 intelligible language or figures, it surely is not fit to be made the basis 

 of a new name. Of course I do not contend that the "language or figures" 

 must be intelligible to the "layman," for that unfortunately is not at 

 present feasible and probably never will be. 



2. Differences in dimensions, of less than five per cent., ought not to 

 he made the hasis of a new name. 



This principle is certainly not radical, jet it would shut out a large 

 number of recently described subspecies of birds, and perhaps other ani- 

 mals also. The reason for this rule is that individual variation in a 

 species is so much larger than was formerly supposed, no constant differ- 

 ence can be maintained between two forms which differ from each other 

 by less than five per cent, in size. I believe ten per cent. Avould be a safe 

 rule, but if five per cent, could be agreed on many ridiculous new names 

 would never see the light of day. In Dr. Allen's famous paper "On the 

 Mammals and \yinter Birds of East Florida" (Cambridge, 1871), he 

 says: "The facts of the' case show that a variation of from fifteen to 

 twenty per cent, in general size, and an equal degree of variation in the 

 relative size of different parts, may be ordinarily expected among speci- 

 mens of the same species and sex taken at the same locality, while in 

 some cases the variation is even greater than this." Such being the case 

 five per cent, is not a high standard to suggest. 



S. Characters which cannot he recognized without knowledge of the 

 geographical origin of the specimen ought not to he made the hasis of a 

 new name. 



This is a very essential principle if we agree that an important end 

 of systematic zoology is correct knowledge of the geographical distribu- 

 tion of animals. It seems to me axiomatic that characters which cannot 

 be recognized regardless of the locality where the specimens are collected 

 are worthless, yet Dr. Allen holds to the contrary, and regards my sup- 

 port of this principle as evidence of my writing without ])ossessing the 

 necessary familiarity with the facts. The horned lark from Mexico named 

 Otocoris aJpestris chn/solaema by Oberholser differs from the same au- 

 thor's subspecies actiu so slightly that he himself admits they are indis- 

 tinguishable, unless the locality where collected is knoAvn. I am unable 

 to see what possible gain there is in giving a name to such a form ; while 

 christening it may easily lead to serious errors in determining the geo- 

 graphical distribution of the real subspecies of horned larks. And in all 

 28 



