4A GENEKAL EVOLUTION. 



from genera up (excepting, perhaps, families) have been evolved 

 under the first mode, combined with some intervention of the sec- 

 ond, and that specific characters or species have been evolved by a 

 combination of a lesser degree of the first with a greater degree of 

 the second mode. 



I propose to bring forward some facts and propositions in the 

 present essay illustrative of the first mode. 



I. ON THE RELATIONS OF NEARLY ALLIED GENERA. 



First. The writer's views of the relations of genera have al- 

 ready been given at the close of an "Essay on the Cyprinoid 

 Fishes of Pennsylvania."* It is easy enough to define isolated 

 genera which have few immediate affines ; but among extensive 

 series of related forms the case is different. One principle, how- 

 ever, pervades the conception and practice of all zoologists and 

 botanists, which few take pains to analyze or explain. It is 

 simply that they observe a successional relation of groups, by 

 which they pass from one type of structure to one or several other 

 types, and the presence or absence of the steps in this succession 

 they regard as definitions of the genera. 



It is true that the reader will often find introduced into diag- 

 noses of genera characters which indicate nothing of this sort. It 

 is often necessary, indeed, to introduce characters which are not 

 peculiar to the genus characterized, for the sake of distinguishing 

 it from similar ones of other series, but this only in an imi^erfect 

 state of the record. Moreover, the ability of the writer to distin- 

 guish genera being thus tested, he too often fails by introducing 

 family and sisecific characters, or by indulging in an unnecessary 

 redundancy. In general, it may be said that adjacent genera of 

 the same series differ from each other by but a single character ; 

 and, generally, that the more remote differ by characters as numer- 

 ous as the stages of their remove. 



It is precisely as, among the inorganic elements, we pass from 

 the electro-negative, non-oxidizing extreme of the halogens, with 

 fluorine as the extreme, to the electro-positive, violently oxidizing 

 extreme of the alkaline metals, whose extreme is potassium, by 

 steps whose relative position is measured or determined first by 

 these tests ; and as these steps have each their included, series of 



* " Trans. Amer. Philos. Soc," 1866, from " Proc. Acad. Nat. Sci.," Phil, 1859, 

 p. 332. 



