GENERAL NOTES 207 



concept of the species is seen in the Corethron population. The genus Corethron is con- 

 sidered here as a monotypic genus, the type species being C. criophilum Castracane. 

 Apart from the type phase, four other phases of the one species are recognized, which 

 correspond to C. hystrix, C. hispidum, C. inerme and C. pelagicum. All of these names 

 I place in the synonymy of the type, and the phases are designated, for convenience, by 

 the specific epithets they replace. In this manner the unity of all these diverse forms is 

 recognized under one specific name, and the phase variation expressed in terms which 

 are well known to workers in the group. The advantages of such a method are obvious ; 

 while the recognition of specific unity amongst the various forms allows the whole 

 population to be expressed by one specific name, simplification in nomenclature is 

 obtained, and — of greater importance — the variations, which in Corethron are correlated 

 with geographical areas, may be specified without having to resort to the use of the 

 subspecific ranks, variety and form. In diatoms no one is yet in a position to say what 

 these subspecific ranks mean, how or in what respects a variety differs from a form or 

 what relation either has to the species. Where I have been able to find a series of inter- 

 mediate forms connecting a species with a so-called variety in such a way that it would 

 be impossible to tell where the variety began and where the species ended, I have re- 

 frained from using the varietal rank. My method of expressing the relationship of these 

 forms is more direct : to take my first example, we are dealing with one species, namely, 

 Rhizosolenia hebetata Bailey; as far as present-day research carries us, we recognize two 

 phases of the one species, and I find it undesirable to adopt any legal subspecific ranks 

 to designate them. Occasionally I have retained the varietal names used by authors, but 

 only because I have not had at my disposal a sufficient number of specimens that would 

 indicate unity with the respective types. That such a unity exists is to my mind beyond 

 question. 



It appears to me to be a matter of the utmost importance that every name should be 

 referred to a published description and illustration by which the organism named can 

 be identified. Whether the citation given be the original, or whether the name used be 

 technically correct is a matter of secondary importance. In dealing with the specific 

 names used in the following systematic account, care has been taken to use the specific 

 name under which the organism was first described, providing that name does not violate 

 the fundamental principle as set out above. Original references and synonyms are 

 quoted, and where illustrations and descriptions are alleged to be the originals of some 

 well-known species, but by reason of ambiguity offer reasonable grounds for doubt, 

 later combinations have been adopted. An example of this is seen in Licmophora Lyng- 

 byei. Licmophora abbreviate, (Hustedt, 1931), based on Echinella cuneata Lyngbye, has 

 been used for this organism. Personally, I think the description and illustrations 

 provided by Lyngbye are very vague, and I am very doubtful whether Lyngbye's 

 species is identical with what we call Licmophora Lyngbyei Grunow to-day. Conse- 

 quently I have adopted Grunow's combination, for there is no question whatever that 

 the organism so named is identical with that in the material examined, and the use of 

 Grunow's name for this organism fixes its identity. 



