The Use of Models in Bacterial Classification

Abstract

SUMMARY:The making of models for taxonomic purposes and their use is described, and the fidelity with which they can represent the taxonomic relations between bacteria is discussed. These models are particularly useful for teaching purposes.The use of equally weighted features in creating taxa allows the comparison of organisms by simple mathematical methods in which the similarity between two strains can be expressed as a percentage; the strains can then be grouped into taxa whose members show high similarity with each other. This method was described by Sneath (1957a, b), who used it for a study of the genus Chromobacteriurn. It also permits solid models to be made of taxonomic groups, in which the spatial arrangements of the component parts of a model is an expression of the relation of organisms or groups of organisms within the taxonomic group. The preparation of such models is the subject of the present paper. METHODSThe meaning of the symbols used is given in Sneath (19573). Two groups of bacteria were used for the models illustrated here. The first was the family Enterobacteriaceae as it is a t present generally defined. The second was the group of chromobacteria studied by Sneath (1957 b).Enterobacteriaceae. In Table 1 are given the features currently used for the differentiation of a number of groups of the Enterobacteriaceae under names which are commonly given them. From Table 1 values of S were calculated by the method of Sneath (19573); these values are shown in Table 2. Since the data in Table 1 do not represent the features of single strains, a slightly different convention for obtaining S was used; +, ( +) and x were counted as positive, and d was counted as NC (no comparison). The data in Table 1 It should be noted that the tests used are selected, and therefore are biased; the use of these tests cannot show whether the groups are valid taxa, since this prejudges the question. They are used only as an illustration of the use of models, and the model is not a model of the ideal taxonomy of the family, but only of the current ideas about the taxonomy of the family. The model shown (PI. 1, figs 1, 2) was made with distance d proportional to 1 -8 (see below).(1 9 57).

Numerical Analysis of the Relationship Between Mycobacterium, Rhodochrous Group, and Nocardia by Use of Hypothetical Median Organisms

“…x loo%, where n, is the number of characters which show similar code symbols ( + + or --1 and nd is Relationships between the taxa were estimated by taxonomic distance (10). The taxonomic distance ( d ) was expressed by the following equation: d = (100 -xYx, where x is a mean M value (%I between the taxa.…”

Comparative Carbohydrate Metabolism and a Proposal for a Phylogenetic Relationship of the Acetic Acid Bacteria

“…To obtain this goal, a file-index with the complete properties of existing strains would be required, which is a task beyond practical possibilities for most investigators. Nevertheless, a first step in this direction has been made by the analysis with an electrical computing machine of many properties of Chromobacterium (Sneath, 1957 b; Lysenko & Sneath, 1959) and of many bacteria from the (British) National Collection of Type Cultures (Sneath & Cowan, 1958).…”

Pseudomonas--An attempt at a general classification

“…This method was found to be very useful for making a table to show a complex comparison of the strains. The relationship between the strains was also calculated by the method described by Sneath (1957) and expressed by models (Lysenko & Sneath, 1958). The final scheme was obtained on the basis of these methods.…”

“…The methods used by Sneath (1957) and Lysenko & Sneath (1958) were used to show the relations between species. The S values calculated from Tables 1 and 2 are given in Tables 3 and 4.…”