NO. 1 SOOT-RYEN : THE FAMILY MYTILIDAE 5 



placed in the supraspecific group containing the "old" species and in due 

 time the whole group will be revised on the basis of new facts. A wrong 

 determination or allocation is the cause of many a too hastily made 

 hypothesis concerning zoogeography or geological history. 



These ideas seem to be those of a "splitter," but sound "splitting" is 

 far better than conservative "lumping," and can bring out new useful 

 information of various kinds, especially on the systematics of any par- 

 ticular group. 



When the knowledge of a family is insufficient, usually many species 

 are described and placed in different genera. As knowledge increases, 

 there will be fewer species but more subspecies, fewer genera but more 

 subgenera. The breaking down of vaguely circumscribed groups is neces- 

 sary for the building up of a more natural system based on increasing 

 knowledge. 



There is still another advantage with small species-groups. If a species 

 is listed as a Mytilus or a Modiolus with its specific name from a certain 

 region, or a certain fossiliferous layer, it tells us very little. But if the 

 narrower group, used either as genus or subgenus, is listed, even without 

 a specific name, it tells us much more about the characteristics of the 

 species, the past and recent distribution of the group, and makes it pos- 

 sible to use the record to add new facts to the picture of the whole group. 

 Nearly all species of mytilids, at least the littoral species, are variable in 

 outline and color. As most of the species described hitherto are character- 

 ized mainly by these variable features, several of them are nearly impos- 

 sible to recognize and many have to be considered synonymous. 



The first real attempt to arrange subgeneric groups of Alytilus s. lat. 

 and Modiolus s. lat. was made by von Ihering (1900) in his studies 

 "On the South American species of Mytilidae." Ihering had many good 

 ideas, but he used mainly the form and the sculpture and the more super- 

 ficial characters, and used the two "old" genera in a very broad sense. 

 The species he treated were those found on the east coast of South 

 America and some of his interpretations certainly are wrong. 



Jukes-Browne (1905) made a very valuable review of the mytilid 

 genera. He separated Brachidontes Swainson as a generic unit, but con- 

 tinued to use the four "old" genera for most of the other species. His 

 arrangement of the subgeneric groups and his diagnoses of them, how- 

 ever, laid a solid foundation for later studies of the family. 



Lamy's (1936-37) extensive paper on the recent mytilids of the 

 Museum National d'Histoire Naturelle de Paris is the only modern 

 revision of most of the species belonging to this family. He recognized 



