August 13, 1920] 



SCIENCE 



145 



which provides for a permanent tj^e genus, 

 since this rule corresponds almost exactly to 

 the method of determining the type species 

 of a genus. 



A demonstration of the advantage of the 

 "permanent type genus" rule is to be found 

 in the case of the family Bubonidse, to which 

 the writer has elsewhere already called atten- 

 tion.* The generic name 8trix Linnaaus has 

 been, by the mutations of nomenclature, 

 transferred from the barn owls, family 

 Strigidse, to the horned owls, family Bubo- 

 nida, and instated there as the proper name 

 for the genus formerly known as Syrnium. 

 It thus becomes the oldest generic name in 

 the family Bubonidse, and by the " oldest 

 genus " rule would require the change of the 

 name Bubonidse to Strigidse. By the third 

 method above discussed, the genus Bubo, 

 from which the family name Bubonidas is 

 formed, continues as the type genus, and no 

 change in the name of the family Bubonidse, 

 into which the generic name Strix is intro- 

 duced, is necessary. The family name Stri- 

 gidse would, in this case, disappear entirely, 

 for the generic term Strix, removed from the 

 former family Strigidse, necessitates a change 

 in this name Strigidse to Tytonidse, based on 

 Tyto, the new name of its type geniis 

 formerly known as Strix. Thus, the same 

 generic group in each of these families would 

 continue to remain the type genus, just as a 

 species, whatever its name hecmnes, reinains 

 the type of a genus. This metlrod of a per- 

 manent type genus has been recently endorsed 

 in print, at least inferentially, in the Ento- 

 mological Code;^ by Dalla Torre, as an ex- 

 amination of his " Catalogus Hymenoptorum " 

 clearly shows; by Dr. C. W. Richmond in the 

 case of the family Threskiornithidse;^ and 

 definitely by Mr. E. P. Van Duzee'' and Dr. 

 Witmer Stone.* Furthermore, the following 



4 Froc. V. S. Nat. Mus., LIT., February 8, 1917, 

 p. 190. 



5 Entomological Code, May, 1912, Rule 114, p. 22. 



6 Proc. V. 8. Nat. Mm., LIII., August 16, 1917, 

 p. 636. 



7 Ann. Entom. Soc. Amer., IX., 1916, pp. 89-91. 



8 Aulc, XXXIV., No. 2, April, 1917, p. 228. 



specialists in various groups, many of whom 

 have personally furnished valuable sugges- 

 tions, have given their approval to the prin- 

 ciples and rules here presented: 



Dr. T. S. Palmer; Dr. Witmer Stone; Mr. 

 J. A. G. Eehn; Dr. C. W. Richmond; Dr. 

 W. H. Dall; Dr. P. Bartsch; Dr. O. P. Hay; 

 i¥r. G. S. Miller; Mr. N. liollister; Mr. J. W. 

 Gidley; Mr. A. N. Caudell; Major E. A. 

 Goldman; and Dr. W. H. Osgood. 



Since some rule for the determination of 

 the type genus is evidently necessary in order 

 to stabilize family and subfamily names in 

 zoology, the adoption of the third and last 

 method above discussed, i. e., that providing 

 for a permanent concept of the type genus, is 

 now advocated. 



For the sake of completeness it seems worth 

 while to formulate the following tentative 

 nomenclatural rules for the determination and 

 treatment of family and subfamily names. 

 These embody all the above provisions in 

 modern codes, with some additions, including 

 that for the type genus just mentioned, and 

 provide for the most important contingencies 

 that may arise. 



RULES FOR FAMILY AND SUBFAMILY NAMES 



1. The name of a family is to be formed by 

 adding the ending idm to the stem of the ten- 

 able name of its type genus. 



2. The name of a subfamily is to be formed 

 by adding the ending inte to the stem of the 

 tenable name of its tjrpe genus. 



3. Subfamily names shall for purposes of 

 nomenclature be accorded the same treatment 

 as family names. 



4. The type genus of a family or subfamily 

 must be one of its included genera. 



5. The type genus of a family or subfamily 

 is the included generic group from the name 

 of which the family or subfamily name was 

 originally formed, and is to remain the type 

 genus irrespective of changes in its name. 



6. A family or subfamily name formed 

 from the name of an included genus is valid 

 whether or not originally accompanied by a 

 diagnosis, or by specific mention of the type 

 genus. 



