THE ORDERS OF TELEOSTOMOUS FISHES 439 



In this classification which has been the basis of all subsequent 

 work of the American school, precision, classicism, and a strict 

 adherence to the canons of nomenclature reinforced a keen 

 analysis and a judicious weighing of taxonomic values. The 

 attempt was made to readjust these values so that they might 

 express more nearly the various degrees of affinity, and to intro- 

 duce more uniformity in the value assigned to the same taxo- 

 nomic grade in different groups of vertebrates. Many currently 

 recognized families were variously divided, the component parts 

 being elevated to the rank of separate families, while many 

 groups were labeled "of uncertain position." "The dictum 

 that ' ' analysis must precede synthesis ' ' was consistently followed , 

 and a great increase in the number of ordinal, subordinal, and 

 family divisions was deemed preferable to the premature group- 

 ings of the traditional classification. Attention was in this way 

 directed to the very numerous families and groups which were 

 really of uncertain affinities, but which had always been thrown 

 in with other divisions by the conservatism which resents the 

 introduction of new groups and new names. An important 

 synthetic step was the frequent use of the superfamily. 



In England and on the Continent the Guntherian system was 

 gradually found inadequate, and the importance of the skeleton 

 in classification became recognized as ichthyology and especially 

 palseichthyology developed. Dr. A. S. Woodward adopted the 

 broad features of Cope's classification, which he improved in 

 many respects, but the older system still remained in general use. 

 The new and very notable classification of Dr. Boulenger, re- 

 ferred to above, is the first since that of Gunther to gain general 

 acceptance in England. Dr. Boulenger refers 1 to the classifica- 

 tion of Gunther as being to a " great extent based on physiological 

 principles," whereas his new classification "aims at being 

 phylogenetic." It is based upon his studies of the rapidly grow- 

 ing collection of fish skeletons in the British Museum ; it reflects 

 also the labors of Cope, Gill, Sagemehl, A. S. Woodward, of 

 Jordan and his co-workers, and thus represents the most com- 

 prehensive analysis of osteological characters which has yet 

 appeared. 



1 Op. cit., p. 542. 



