January 25, 1907] 



SCIENCE 



149 



As an ar^ment against the first species 

 rule this has no weight, as it applies with 

 equal force to any method of fixing types. I 

 might say, for instance, " if the types of two 

 or more genera happen to be the same by 

 elimination the later genera become pure 

 synonyms of the earliest." Otogyps is sup- 

 pressed as a synonym of Sarcorhamphus by 

 this very method in Dr. Allen's paper. 



So much for my 'misleading statements.' 

 Turning now to Dr. Allen's elaborate discus- 

 sion of the types of the Vulturine genera, 

 which he gives as an example of how elimina- 

 tion should be practised and which we should 

 be very glad to see, as it gives us an actual 

 case or series of cases worked out by one who 

 is a recognized expert in this method of fixing 

 types. 



My chief objection to the method (i. e., 

 elimination) is that it will give different re- 

 sults in the hands of different workers owing 

 to the almost infinite variety of ways in which 

 it may be applied. Dr. Allen, far from re- 

 futing this claim, actually shows that two 

 different methods of elimination may (no 

 doubt unconseiouly) be used by the same 

 author in the same paper, thus emphasizing 

 the elasticity of the method and the impos- 

 sibility of formulating rules that will meet 

 all its varied requirements. 



Any one who has practised elimination 

 knows that there are two methods in use in 

 successively removing the species of a genus 

 which have been made the basis of subsequent 

 genera. 



(a) Some remove only the species which 

 has been made the type of a subsequent genus 

 at the date at which the genus was established. 



(6) Others remove along with the type any 

 other strictly congeneric species, and here 

 again there are two practises according as 

 we interpret congeneric to mean congeneric 

 from the standpoint of the author of the 

 genus, or congeneric from the standpoint of 

 the eliminator. 



Taking Dr. Allen's elimination of Sarco- 

 rhamphus at the top of p. Y76, he says : 



Sarcorhamphus, 1806; species gryphus, papa, 

 auricularis. The species papa was removed to 



Gathartes in 1811, gryphus to Oypagus in 1816, 

 leaving auricularis as the type of Sarcorhamphus. 



The species thus removed are not, accord- 

 ing to Dr. Allen's conclusions, the types of 

 the genera Gathartes and Gypagus, but they 

 were included in these genera by their authors 

 in 1811 and 1816, respectively. It will thus 

 be seen that Dr. Allen adopts method ' h ' 

 (above) in his elimination and interprets 

 ' congeneric ' to mean congeneric from the 

 standpoint of the original author, not from 

 that of the eliminator (or the usage of the 

 present day). Having fixed the types of the 

 four involved genera in this way, he next 

 proceeds to eliminate Vultur by removing the 

 species at the dates at which they hecame the 

 types of subsequent genera — i. e., according to 

 method ' a.' 



If Vultur were eliminated in the same way 

 as Sarcorhamphus the result would be as 

 follows : 



Vultur, 1758; species gryphus, harpy ja, papa, 

 aura, barbatus, percnopterus. The species bar- 

 batus was removed to Gypaetus in 1784, gryphus 

 and papa to Sarcorhamphus in 1806, percnopterus 

 to Neophron in 1808, aura to Gathartes in 1811, 

 leaving harpyja as the type of Vultur. 



If we do not trouble ourselves to ascertain 

 the types of Gathartes and Gypagus when, we 

 eliminate Sarcorhamphus, I fail to see why we 

 have to ascertain the types of the involved 

 genera when we eliminate Vultur. 



As a further example of the various ways in 

 which elimination may be practised, it will 

 be noticed that Dr. Allen pays no attention to 

 what may have been done to species prior to 

 the date of the genus that he is eliminating. 

 Under Gypagus, 1816, he says : " gryphus was 

 removed to the genus Gryphus in 1854," but 

 as a matter of fact it had already figured in 

 the establishment of the genus Sarcorham- 

 phus, 1806, and proves, according to Dr. 

 Allen's demonstration, to be the type of 

 Vultur, 1758. Here again very different 

 results may be obtained according as we con- 

 sider or ignore the work of authors prior to 

 the date of the genus we are eliminating. 



Dr. Allen truly says that elimination re- 

 quires ' a thorough knowledge of the litera- 

 ture of the cases involved ' and ' is therefore 



