Page:Bergey's manual of determinative bacteriology.djvu/43

 to the differentiation of larger natural groups or genera should, in addition to constancy, show some correlation with other constant characteristics. The presence of this relationship or correlation between characters for the division of genera indicates that the groupings are being made along natural rather than artificial lines."

Type culture. It is quite evident that when a new species of bacterium is described, it must include the particular culture from which the species description was made. This original culture is termed the type culture. One may develop a definition as follows: A species of bacterium is the type culture or specimen together with all other cultures or specimens regarded by an investigator as sufficiently like the type (or sufficiently closely related to it) to be grouped with it. It is self-evident that different investigators may not draw the same boundaries for a given species. There are some practical difficulties, but no better definition has been evolved.

(2) Genus (plural genera). A genus is a group of related species. In some cases a genus may include only a single species (is said to be monotypic); in most cases several to many species are included in a genus. The quite pertinent question should be asked: How close must be the resemblances (how close the relationships) among the species of a group to entitle them to inclusion in the same genus? In other words, how is it possible to delimit accurately the boundaries of a genus? This is a matter on which there is no agreement, and probably can be none. Much of the confusion in modern bacteriological terminology is to be attributed to this fact. Nevertheless, in the course of time experience tends to delimit many genera with reasonable accuracy. As stated by Hitchcock (Descriptive Systematic Botany, New York, 1925, p. 9): "Convenience may play a role in determining generic lines. Extremely large groups may be broken up on the basis of differences of smaller degree not common to a group of closely allied species, than if the group consisted of a few species. In general, the botanist, in delimiting genera, keeps in mind two important rec^uirements, that of showing natural aflftnities and that of aiding correct identification."

However, a genus may be defined helpfully in another way. One of the species described as belonging to a genus is designated as the type species; a genus may therefore be defined as including this type species together with such other species as the investigator (or taxonomist) regards as sufficiently closely related. It is apparent that some authors may draw the lines narrowly, others broadly. Some early authors, for example, recognize only two genera of rod-shaped bacteria, one for those without endospores(Bacterium), and one for those producing endospores (Bacillus). These genera thus defined are very large, each containing hundreds, perhaps thousands, of species. Other students break up these large genera into many smaller ones. There is not much point to the question as to which is right and which is wrong. A better question is, which is the more convenient, better represents relationships, better facilitates diagnosis, and proves most useful.

(3) Family. A family in taxonomy is a group of related genera one of which