720 



SCIENCE 



[N. S. Vol. XXVI. No. 673 



typus as a new name for one of the species 

 lie included in it when founding it. In both 

 groups the type designated by the founder 

 " shall be accepted as type regardless of any 

 other considerations." (c) A genus proposed 

 with a single species takes that species as its 

 type, {d) Any genus founded without a type 

 being provided for it under one or the other 

 of the above conditions, but which " contains 

 among its original species one possessing the 

 generic name as its specific or subspecific 

 name, either as valid name or synonym, that 

 species or subspecies becomes ipso facto type 

 of the genus." 



It is safe to claim that 70 per cent, of all 

 generic names in ornithology, and probably 

 in vertebrate zoology, are determinable upon 

 the original basis of publication, by one or 

 the other of the methods above prescribed^ 

 methods, too, which everybody accepts. The 

 other 30 per cent, are provided for by rules 

 e to g, rule e designating the conditions upon 

 which rules f and g must rest — namely, that 

 the species alone available are those that were 

 included in the genus when it was originally 

 published, of which, however, none is avail- 

 able that was indicated by the author as, from 

 his standpoint, either of doubtful status or 

 of doubtful pertinency to the genus. With 

 this useless rubbish cleared away, rule f pro- 

 vides : 



In case a generic name without originally desig- 

 nated type is proposed as a substitute for another 

 generic name, with or without type, the type of 

 either, when established, becomes ipso facto the 

 type of the other. 



This is a sensible innovation that may now 

 and then prove extremely useful. But the 

 grand stroke is rule g, as follows: 



If an author, in publishing a genus with more 

 than one V,alid species, fails to designate (see a) 

 or to indicate (see 6, d) its type, any subsequent 

 author may select the type, and such designation 

 is not subject to change. 



This last rule, as old as the B. A. Code, 

 completes the rules for type determination, 

 and provides essentially only four methods, 

 which are designated: 



(1) " Type by oi-iginal designation " (rules 

 a and V) ; (2) " Monotypical genera " (rule 



c) ; (3) " Type by absolute tautonomy " (rule 

 (^) ; (4) " Type by subsequent designation " 

 (rules e-g). By a wise stroke of diplomacy, 

 the word " elimination " is not mentioned ; 

 yet elimination is the basis and the method, 

 and necessarily always has been, of any sound 

 work by a first reviser. 

 To rule g is added: 



The meaning of the expression " select a type " 

 is to be rigidly construed. Mention of a species 

 as an illustration or example of a genus does not 

 constitute a selection of a type. 



This seems explicit, but is far from being 

 so; while it will tide over some difiiculties, it 

 will open up others. Not only this, but must 

 the designation of a first reviser always be 

 accepted, right or wrong, or only when made 

 in accordance with fundamental rules of 

 nomenclature that have been extant in all 

 codes since the publication of the British 

 Association Code of 1842? 



One need not have had very extended ex- 

 perience with the work of " first revisers " to 

 have learned that it is of all grades of quality 

 from absolutely pernicious to unqualifiedly 

 beneficent, having been often done by sys- 

 tematists who knew nothing of rules of nom- 

 enclature, or else disregarded them. One 

 need not go very far back in the history of 

 even American ornithology — less than half a 

 century — to find that species have been taken 

 as types of genera that were not described till 

 long after the genera were founded; or that 

 genera have been taken from pre-Linnsean 

 authors when they became tenable only from 

 Linnaeus or from some much later author; 

 or that types thus designated for certain gen- 

 era had long before properly become the types 

 of other genera and were not available as types 

 of entirely different genera. 



That the new article 30 is not intended to 

 countenance such work is clearly indicated 

 by the first section of rule e, which states that 

 no species can be taken as the type of a genus 

 that was not included in it at the time of its 

 original publication. Again, if a reviser, ig- 

 norant of the literature of his subect, or mere- 

 ly neglectful of rules, chooses as the type of a 

 genus a tautonomic species of an earlier 



