189076a0Nature18947581961010776760028-0836196110.1038/189076a0ukNatureNatureNATUREnatureNature is a weekly international journal publishing the finest peer-reviewed research in all fields of science and technology on the basis of its originality, importance, interdisciplinary interest, timeliness, accessibility, elegance and surprising conclusions. Nature also provides rapid, authoritative, insightful and arresting news and interpretation of topical and coming trends affecting science, scientists and the wider public./nature/journal/v189/n4758issueJournal homeArchiveCurrent issueAdvance online publicationPrivacy policySubscribeNature Publishing GroupCurrent issue189076a0Biochemical Classification of Yeasts
AU  - BARNETT, J. A.Low Temperature Research Station, Cambridge.Roberts and Thorne1 disagree with my proposals2 for improving the classification of yeasts, but themselves suggest no way of overcoming the difficulties (which they acknowledge exist) in using the present system3. Instead, they emphasize the desirability of constructing a phylogenetic classification ; though this is obviously impracticable.I advocated developing a classification of yeasts based mainly, not, as they claim, "exclusively", on the biochemistry of yeasts. Roberts and Thorne also imply that I hold the view that because morphological differences reflect chemical ones, morphology is neither interesting nor important. This argument is a non sequitur and I did not put it forward.
It is strange that Roberts and Thorne object to the idea that the most important thing about yeasts is their biochemical activity. Chemical processes determine the role of yeasts in brewing, wine-making, baking, nutrition, spoiling foodstuffs and as pathogens. Thus it is not surprising that of thirteen papers on yeast genetics published by Winge and Roberts4 between 1948 and 1958, at least nine were concerned with biochemical characteristics.
In the present lack of knowledge about the ancestral history of yeasts, it is self-deception to say that any classification of them is phylogenetic. The idea of chronology is implicit in any phylogenetic classification ; and this has little meaning except in terms of the age of ancestral forms5. Kudriavzev's6 and Zsolt's7 theories of evolutionary relationships are based on insufficient evidence ; and Wickerham's speculations8 on the phylogeny of the Hansenulae involve false assumptions9. Kudriavzev held that it was necessary for his classification to be phylogenetic. Because of the difficulty in developing a phylogenetic theory for the asexual forms (p. 80), he did not publish any data on the species of non-sporing yeasts, and this detracts enormously from the value of his book6.
The Dutch classification of yeasts3 is by far the most complete and therefore the most useful. As I have pointed out before2'9, it has, however, three major weaknesses. (1) Sporulation tests are given supreme importance. These involve a bewildering array of media (Phaff and Mrak11, list 18) and extensive searching for spores. Moreover, since a negative result is equivocal, these tests are very unsatisfactory. (2) The morphological criteria used, are ill-defined and difficult to specify precisely. (3) The biochemical tests are excessively crude and hence uninformative.
Roberts and Thorne do not say that these criticisms are invalid. Assuming their validity, it seems reasonable to develop techniques for classifying yeasts in which : (a) there are no Sporulation tests ; (6) the morphological criteria are given greater precision ; (c) biochemical tests are designed to give more information.
I do not suggest that studies of morphology or of Sporulation should stop ; but I do suggest that in classification biochemical criteria should be given more importance than at present. No embargo on evolutionary speculation has been proposed ; but there is no reason why such conjecture must be applied to yeast taxonomy. Perhaps it is worth noting van Kiel's striking volte-face : although in 1936 he held that bacterial classification had to be phylogenetic12, ten years later he wrote : "Bacterial taxonomy is far more similar to Linnaeus's original system of the plants. . . . Both kinds of system are useful if they permit the rapid assignment of a specimen to a restricted group, but neither can lay claim to representing phylogenetic relationships", and in another passage : "It is not hereby implied that such systems are inferior to the 'natural5 systems of the botanist and zoologist, but merely that they serve a different purpose, and are arbitrary"13.Roberts, , C., and Thorne, , R. S. W., Nature, 188, 872 (1960).ArticleISIBarnett, , J. A., Nature, 186, 449 (1960).ArticlePubMedISIChemPortLodder, , J., and Kreger-van Rij, , N. J. W., [ldquo]The Yeasts[rdquo] (North Holland Pub. Co., Amsterdam, 1952).Holter, , H., and Westergaard, , M., C.R. Trav. Lab. Carlsberg, Ser. Physiol., 26, 7 (1956). Winge, , O., and Roberts, , C., Nature, 177, 383 (1956); Genetica, 28, 489 (1957); C.R. Trav. Lab. Carlsberg, Ser. Physiol., 25, 420 (1957).Crowson, , R. A., [ldquo]Darwin and Classification[rdquo] in Barnett, S. A., [ldquo]A Century of Darwin[rdquo] (Heinemann, London, 1958).Kudriavzev, , V. I., [ldquo]The Systematics af Yeasts[rdquo] (Academy of Sciences of the U.S.S.R., Moscow, 1954).Zsolt, , J., Acta Botanica Acad. Sci. Hung., 5, 233 (1959).Wickerham, , L. J., Tech. Bull. U.S. Dept. Agric., 1029 (1951).Barnett, , J. A., Leeuwenhoek ned. Tijdschr., 23, 1 (1957).ChemPortBarnett, , J. A., Experientia, 15, 99 (1959).ArticlePubMedISIChemPortPhaff, , H. J., and Mrak, , E. M., Wallerstein Labs. Commun., 12, 29 (1949).ChemPortKluyver, , A. J., and van Niel, , C. B., Zbl. Bakt. (2 Abt.), 94, 369 (1936).van Niel, , C. B., Cold Spring Harb. Symp. Quant. Biol., 11, 285 (1946).