no A CENTURY OF PROGRESS IN THE NATURAL SCIENCES 



implies that bacteria with doubtful phylogeny must then be treated likewise, 

 there seems to be much in favor of abandoning the practice. If and when the 

 natural relationships of a large number of bacteria have been unambiguously 

 established, it would become advisable to consider the construction of a system 

 of classification based on phylogeny. 



As long as this remains a pious hope for the future, one might do well to 

 approach the problem of the classification of bacteria and bluegreen algae in the 

 manner suggested by Winogradsky's latest recommendations. Substitution of 

 "biotypes" for genera and species, and the use of common names, such as "sulfur 

 bacteria," "photosynthetic bacteria," "chemoautotrophic bacteria," "denitrifying 

 bacteria," "nitrogen-fixing bacteria," etc., instead of the Latin names represent- 

 ing taxonomic units with definite phylogenetic implications, would permit the 

 development of more rational arrangements for the rapid identification and com- 

 parison of the organisms. This problem calls for an elaborate system of cross- 

 indexing of their properties, and the present organization, based on the Linnean 

 approach, not only is unjustifiably pretentious, but also impedes the best utiliza- 

 tion of established characteristics because they are employed for the construction 

 of mutually exclusive combinations. While much can be done to remedy the re- 

 sulting situation through the preparation of mechanical keys, such as the emi- 

 nently useful one developed by Skerman (1949), a more radical departure from 

 accepted procedure remains desirable in the opinion of the writer. 



In this connection attention should be called to the ideas recently expressed 

 by C. H. Andrewes concerning the classification and nomenclature of viruses 

 (1952, p. 136) : 



The nomenclature of plants and animals has been the subject of much controversy 

 and change, owing largely to the fact that the earlier names were bestowed without 

 understanding of the principles of taxonomy as we now know it, often without reference 

 to type material, and on the basis of very inadequate descriptions. In the reviewer's 

 opinion, such troubles would be avoided in the virus field by dating valid nomenclature 

 in this group not from the time of Linnaeus 200 years ago, but from a date to be de- 

 cided upon in the future. . . . 



A very few descriptions of viruses published hitherto would satisfy those who are 

 seriously considering the matter today. Binomials are not in common use for any viruses, 

 and there seems therefore everything to be gained by starting with a clean sheet. . . . 

 Such virus names already published as seem suitable would also be validated, but virus 

 nomenclature need not be forever overlaid by the dead hand of bad naming, linked to 

 descriptions which are hard to interpret and are based on unsuitable guiding principles. 

 If, however, students of viruses take thought in time and base their classification and 

 nomenclature on solid foundations with reference from the very beginning to type mate- 

 rial, they can forever be free from the nightmares of change and contentiousness which 

 bedevil nomenclature in other fields. 



In contrast to the quotation at the start of this paper, the above, with a few 

 minor modifications, seems eminently applicable to the problems presented by 

 the classification of the bacteria and bluegreen algae. 



REFERENCES 

 Andrewes, C. H. 



1952. Classification and nomenclature of viruses. Ann. Rev. Microbiol., 6:119-138. 



Andrewes, F. W. 



1930. The nomenclature and classification of micro-organisms. In A System of Bac- 

 teriology, 1:292-310. London: Med. Res. Council. 



