September 28, 1900.] 



SCIENCE. 



479 



royal palm and its allies, which have never 

 been designated by a correct generic name,* 

 whether the difficulty be adjusted by the 

 method of elimination or by the method of 

 types. Of course it is not necessary that 

 the types of phanerogams should be fixed 

 by the same method as in the other groups, 

 but all phanerogamists are not likely to re- 

 main contented with an illogical and faulty 

 method, and it is scarcely to be expected 

 that the Committee on Types appointed at 

 the Buffalo meeting, will bring in recom- 

 mendations for a variety of usage in a 

 matter of so much importance. 



In the incorporation of the desired legis- 

 lation into the Rochester Code a large vari- 

 ety of courses might be followed, but for 

 present purposes it may be sufficient to 

 point out that these lie between two gen- 

 eral policies, either of which may be devel- 

 oped in such form as to be both logical and 

 practical. If we adhere strictly to the bi- 

 nomial system, to 1753, and to the ' Species 

 Plantarum,' we must reconcile ourselves 

 to the misapplication of the pre-Linn£ean 

 names or treat them as exceptions and pro- 

 vide for the assignment of types by a com- 

 mittee or a congress, thus disposing at once 

 of many bibliographic complications. This 

 would be in accordance with the argument 

 advanced by some of the advocates of the 

 Rochester Code, that the process of revi- 

 sion of cryptogamic as well as of phanero- 

 gamic genera would be greatly simplified 

 by relief from the incubus of the pre-Lin- 

 nsean and non-binomial literature, an ex- 

 pectation which undoubtedly influenced 

 many in favor of that legislation. It tran- 

 spired, however, that instead of adhering to 

 the logical consequences of the adoption of 

 a nomenclature of genera and species based 



*A new genus Boysionea is proposed, differing 

 from Oreodoxa in the solitary growth, the douhle 

 spathe and other characters. The type is E. regia 

 (HBK), Nov. Gen. et Sp. 1 : 305, originally de- 

 scribed from Cuba. 



on the binomial system with the ' Species 

 Plantarum ' as a starting point, the very 

 committee which had framed the rules fell 

 into the practice of interpreting Linnseus 

 through the works of his predecessors in- 

 stead of establishing the usage and identifi- 

 cations of his followers, thus rendering the 

 date 1753 merely an arbitrary limit for 

 citations, and virtually abandoning all the 

 advantages which might have been secured 

 by a consistent adherence to the original 

 import of the Rochester Code, as far as it 

 affected the taxonomy of genera. More- 

 over, in addition to the re-introduction of 

 this complication, there was unearthed a 

 large body of irrelevant, non-binomial lit- 

 erature issued subsequent to 1758, much of 

 which had rested in merited oblivion for 

 upward of a century. To accept as taxo- 

 nomic literature such writings as those of 

 Adanson, while refusing to cite Tournefort 

 and Micheli, destroys every rational or 

 practical effect of the intended reform and 

 reduces the result of the Rochester legisla- 

 tion, as far as genera are concerned, to the 

 empty absurdity of requiring the false cita- 

 tion of Linnteus and Adanson as the au- 

 thors of genera which they knew only as 

 compilers from the works of older and bet- 

 ter botanists. 



It is plain, therefore, that any argument 

 which might have been drawn from the fact 

 of previous legislation, if it had been 

 logically carried out in this respect, has 

 been lost by the apparently unconscious 

 surrender of the Rochester Code reformers 

 to Professor Greene's contention for the 

 recognition of the pre-Linneean authors, 

 and we may thus without prejudice con- 

 sider the second of the available alternatives 

 for the enactment of a law for fixing generic 

 names by tj'pes. To abandon 1753 as the 

 initial date for generic nomenclature is but 

 frankly to admit what is already an ac- 

 complished fact, and to cease to quote Lin- 

 nseus, Adanson and others as the authors 



