538 



SCIENCE. 



[N. S. Vol. XXI. Xo. 5.36. 



Embryology, by no means a purely mor- 

 phological science, but one daily assuming 

 a more physiological aspect, has come to 

 have a weighty voice in matters of classi- 

 fication. More recently chorology, or 

 biogeography — a distinctively ethological 

 science— has come to play an equally im- 

 portant part. And rightly, because the 

 organism may be said to seek, and in many 

 cases even to make, its own environment. 

 Every field naturalist knows that he is fre- 

 quently guided to the more delicate specific 

 and varietal distinctions, not so much by 

 the structural differences between the or- 

 ganisms he is observing, as by differences 

 in their habitat or behavior. Then closer 

 scrutiny may often, although not always, 

 reveal correlated structural differences. 

 When such structural differences are not 

 to be detected we speak of ethological 

 species, and the number of these is un- 

 doi;btedly much greater than was formerly 

 supposed.* 'The great reliance on geo- 

 graphical distribution in the more refined 

 taxonomy of certain groups of organisms, 

 like the birds, mammals and social insects, 

 shows an ever-deepening appreciation of 

 ethological characters. It is even jocosely 

 asserted that certain mammalogists are 

 quite unable to identify a specimen unless 

 they are first informed of the exact fence- 

 corner in which it was trapped. Then, and 

 not till then, are they able to perceive the 

 delicate specific or subspecific shade of 

 pelage which goes with life in that partic- 

 ular corner. 



The fact that the morphologist has so 

 consistently either neglected or opposed the 

 use of ethological characters in classifica- 

 tion shows very clearly that in his heart of 

 hearts he has never very earnestly con- 



* I liavc in mind a nnnihor of cases among in- 

 .sects, such as certain species of ants. There are 

 American forms of the genera Pheidole, Mi/rmica, 

 M yrmecocyntus, Formica, etc., which exhibit geo- 

 graphical (lifferenees in habits without perceptible 

 morphological diHercnces. 



cerned himself wdth the parallelism of 

 structure and function. He is inclined to 

 regard function, especially psychical func- 

 tion, as something utterly intangible and 

 capricious. For does it not seem to make 

 its appearance in the embryo or young 

 after structure has developed, and to de- 

 part at death before the dissolution of vis- 

 ible structure ? And are not our museums 

 largely mausoleums of animal and plant 

 structures which we can forever describe 

 and redescribe, tabulate and retabulate, ar- 

 range and rearrange, without troubling 

 ourselves in the least about anything so 

 volatile as function ? 



It is, indeed, not only conceivable, but 

 very desirable, that a taxonomy should be 

 developed in which the ethological will re- 

 ceive ample consideration, if they do not 

 actually take precedence of the morpholog- 

 ical characters. It is certainly quite as 

 rational to classify organisms as much by 

 what they do as by the number of their 

 spines and joints, the color of their hairs 

 and feathers, the course of their wing- 

 nervures, etc. To regard our existing 

 purely structi;ral classifications as anything 

 more than the most provisional of make- 

 shifts, is to ignore the fact that the vast 

 majority of organisms which they are de- 

 signed to cover are known only from a 

 few dead exuvije. There are, of course, 

 enormous difficulties in the way of con- 

 structing ethological classifications, quite 

 apart from the fact that our knowledge of 

 behavior is even more fragmentary than 

 that of structure, as any one will realize 

 who tries to write an ethological description 

 of some common animal or group of ani- 

 mals. In morphology the elements of de- 

 scription can be treated as parts of an 

 orderly and traditionally respected routine, 

 but in ethology we still lack the necessary 

 preliminary analysis of the more complex 

 instincts, and are therefore unable to con- 

 struct uniform and nmtually comparable 



