The Theory of Evolution \ 307 



specimens with the specimens in the editha cluster is .462. The Q 

 matrix and WVG diagram are answers in themselves, expressions of 

 the similarities of the individuals studied. There is no need for com- 

 plicated guesswork about interbreeding potentials to establish which 

 (if any) of the clusters should be named and considered "good" 

 species. Names (or numbers) may be given to clusters wherever con- 

 venient. In general, they may be considered analogous to floats 

 thrown into the ocean to be used in charting currents. Too many 

 floats would confuse the picture; too few would lose detail. 



Needless to say, an accurate description of patterns of variation is 

 of extreme importance in the study of evolutionary processes. Evo- 

 lutionary theory is an explanation of the origin of organic diversity, 

 both neontological and paleontological. Therefore, agreement on 

 the nature of this diversity and unambiguous methods of describing 

 it are obvious prerequisites to achieving maximum coherence in our 

 theory. Otherwise we shall always be somewhat in the position of 

 the blind men formulating a theory of elephant morphology. 



It is not being suggested here that human and cultural biases and 

 subjective taxonomy have conspired to give an entirely fallacious 

 picture of the structure of the biotic world. It is being suggested that 

 in several areas distorted ideas may be impeding progress toward a 

 more thorough understanding of evolutionary processes. For in- 

 stance, it seems very likely that the required taxonomic structure ( as 

 well as the cultural factors discussed) may have resulted in serious 

 misapprehensions about the transfer of genetic information among 

 groups of animals. Unfortunately, most taxonomic studies do not 

 provide the kinds of data that are helpful in establishing just what 

 patterns are present; all too often the results merely present nature 

 tied up in neat packages. This became especially obvious during the 

 writing of the chapter of this book dealing with differentiation of 

 populations. North American butterflies are considered by most tax- 

 onomists to consist of an array of some 400 "species" and many more 

 "subspecies." Unfortunately many (probably most) of these entities 

 exist only in the cabinets of collectors, not in nature. The appalling 

 complexity of some of our butterfly interrelationships is now slowly 

 being elucidated; that of many others can only be guessed. The in- 

 adequacy of current taxonomic systems for describing the observed 

 situations is very striking. It is in the area of description that nu- 

 merical taxonomy is superior to classical taxonomy. Phylogenetic or 

 other interpretation of the results is done by the taxonomist rather 

 than the computer. 



It would, of course, be folly to assume that the situation in butter- 

 flies ( or birds, or any other group ) can be considered representative 



