May 19, 1921 mcateE: nomenclature 233 



In the search for earliest family names the following difficulties, 

 among others, will be encountered. Names based on those of genera 

 are understood in each case.^ 



1. Family nomenclature as such was not in general use until more 

 than 50 years after the Systema A^a/«ra; therefore what are now con- 

 sidered distinctive endings need not be looked for in the earlier names 

 of supergeneric groups; names of whatever endings will have to be 

 considered. 



2. Rank of supergeneric groups was not always indicated and cer- 

 tainly cannot be closely correlated as between diverse authors. 

 Families and genera of various authors equal in scope modern orders 

 and families. Names of every rank will have to be considered. Any 

 name for a group agreeing chiefly in composition with one that now is, 

 or shall in future be, considered of family rank is on the eligible list. 

 Opinion will enter largely into the decision as to whether there is a 

 sufficient degree of coincidence. 



3. It is impracticable to choose between names set forth in the 

 formal language of science and vernacular names. Van Duzee, who 

 is quoted as a disciple of priority in family names and of the type- 

 genus method, says: "It was my intention to follow Dr. Horvath's 

 scheme in all its details but as the work advanced it was found neces- 

 sary to abandon his plan to discard all names not strictly in latin 

 form. His method would have thrown out a large number of groups 

 founded by Amyot and Serville in their great work which is really 

 the foundation of all our modern classification of the Hemiptera."^° 

 Here it may be added that in numerous well-conceived systems the 

 family and higher groups were given vernacular names only; indeed, 

 in some cases only numbers were used. 



As between a family or subfamily name in classical form and prop- 

 erly used, a scientific-appearing name casually used, a vernacular name 

 appearing in an evidently good system, and a vernacular simply in 

 text, who is to be the arbiter.-^ and who will accept his judgment? 



4. It would appear that names of whatever form, whatever rank, 

 whatever language, wherever proposed, in fact every supergeneric 



' This restriction is necessary of course on account of the ex post facto rules in modern 

 codes relating to the formation of family names. Many names not based on generic 

 names have been used for supergeneric groups, which in other respects are as logical and 

 worthy of consideration as those embodying generic stems. Such names are available 

 for suborders and higher groups but for families, never. Our system of priority is indeed 

 garnished with many an exception. 



'" Check-list of the Hemiptera . . . of America north of Mexico, pp. vi-vii. 1916. 



