'mo 



SCIENCE. 



[N. S. Vol. XV. No 382. 



cies, and a generic name, whenever and 

 ■wherever used, will have a fixed point of 

 attachment to nature. Progress in the 

 science of systematic biology must still 

 «ompel endless modifications of the sup- 

 posed limits of genera, but the method of 

 types affords a complete and ideal solution 

 of all the attendant difficulties which can 

 be correctly assigned to the province of 

 nomenclature. By the simple expedient 

 of treating a generic name as inseparably 

 attached to its original species as its no- 

 menclatorial type, the whole maze of defini- 

 tions, history, casuistry, confusion and con- 

 tention is resolved into definite elements 

 capable of rational and permanent adjust- 

 ment. Of two or more generic names es- 

 tablished on the same species only the old- 

 est should be used, no matter how much 

 the original definitions may have differed, 

 while genera founded on species belonging 

 to distinct natural groups will never be the 

 same, no matter how closely the definitions 

 may have approximated. 



HOMONYMS. 



The adjustment of the claims of com- 

 peting generic propositions by reference 

 to types rather than to concepts has many 

 practical advantages. It becomes, for ex- 

 ample, more obvious than before that ge- 

 neric synonyms are of several kinds, the 

 nomenelatorial standings of which are 

 very different. The first recognition of 

 such distinctions is to be found in the so- 

 called ' law of homonyms, ' to the effect 

 that the same name should be used only 

 once in the plant or animal series. It has 

 been held by some systematists that a 

 homonym or second use of the same name 

 might hold where the first had for any 

 reason miscarried, but the impossibility of 

 ■establishing the fate of any particular 

 name under the method of concepts and 

 elimination has rendered it obviously un- 

 wise to risk the confusion attendant on a 



resurrection of the supposedly defunct old- 

 er genus, and the rule or principle ' once 

 a synonym always a synonjrm ' is receiving 

 general recognition. And yet this apho- 

 rism is very misleading, since all synonyms 

 are not homonyms, and the restoration of 

 other kinds of synonyms is a very conunon 

 occurrence. ' Once a homonym always a 

 homonym ' or ' once a homonym always a 

 synonjon' would be a much more correct 

 statement, though in these forms the idea 

 becomes a mere truism. 



TTPONTMS. 



Another class of sjnaonyms hopelessly 

 invalid from the beginning is the typonym, 

 a generic name based on a species which 

 has already been used as the type of a 

 genus. Even in dealing with a genus con- 

 taining but a single species variety in defi- 

 nitions has often led systematists to con- 

 tinue the multiplication of names. Thus 

 although Eostafinski found that the names 

 Strongylium fuUginoides, Dermodium in- 

 quinans and Lachnoholus cribrosus had 

 all been used for a single species of Myx- 

 omycetes, which he treated as representing 

 a monotypic genus, he again redefined the 

 same genus and rechristened it with a 

 fourth name. The only possibility of res- 

 urrection for a typonym is in the event of 

 the previous name being found to be a 

 homonym, as in the present instance where 

 Strongylium was preoccupied for a lichen, 

 so that the correct name for the Rostafin- 

 skian genus Amaurochatm appears to be 

 Dermodium* 



* Tlie binomial Dermodium atrum (A. & S.) 

 would have been used by Eostafinski if the prin- 

 ciple of priority had been observed, in spite of 

 the fact that Dermodium is usually treated as 

 a synonym of the unrelated genus Lycogala. 

 Neither can the name Lachnoholus be used in the 

 sense in which Rostafinski and subsequent writers 

 have employed it, since it was originally estab- 

 lished as monotypic and included only L. crib- 

 rosus as above. Fries had already in 1849 ap- 

 plied the name Nassula to the species which 



