This conception altered the whole meaning of the classifica tory groups ; instead of regarding them as having a real existence, as permanent and sharply marked categories, zoologists realized that they were of human origin, representing little more than a convenient mode of expressing definite conceptions of relation ships.
The group names which are now in use are in order of ascending scope, species, genus, family, super-family, sub-order, order, class and phylum. It is believed by many zoologists that the smallest group, the species (q.v.), differs somewhat in its nature from the others. A species may be defined as a group of individual animals, whether alive or dead, which agree with one another in their structure, within close but variable limits, and are capable of breeding together. In the nature of the case it is impossible to give any general quantitative expression to the permissible amount of variation, and, in practice, the determination of the limits of a species depends on the judgment of the taxonomist. The second part of the definition, the possibility of breeding from any two individuals of the species of different sexes, cannot be extensively used because many animals will not, in fact, breed in captivity, although under natural conditions they would be fertile, and because the vast majority of species are only known from the dead specimens preserved in museums.
Darwin's conception of the mode of origin of species was that by some one or more of different processes, under the control of natural selection, a group of individuals of a species and their descendants gradually diverged from the normal type, until by the slow accumulation of changes they came to differ so greatly from it in their structure that they were, in most cases, no longer fertile with the original type of individual, and presented morphological characters sufficiently distinct to acquire specific status. On this hypothesis of its origin it is clear that a species is an artificial group, in that no two observers will draw the line which separates it from the parent species at the same point. At the same time, if only the living representatives of the new and the parent species be considered, the groups may appear completely discrete.
Modern work on genetics, the investigation of the nature of heredity (q.v.), has suggested that the initiation of a new species may depend, not on a continuous process of change, in which the stages may be of infinitely small magnitude, but on a discontinu ous variation, in which each change is a definite step, even if a very small one, and the individual which exhibits it differs from its parents in every cell of its body from the moment of its formation. Such definite steps may, in some cases, be found to depend on a modification, presumably of a chemical nature, in a particular structure, a gene, in a chromosome of one or both of the germ cells from whose fusion the new individual arose.
If such a view of the nature of evolution be true, it is clear that a species is a natural group, in that it begins at a definite place and time by a definite event ; that, like Athena, it sprang fully developed from its parent.
`'hat little evidence exists suggests, however, that few species are actually distinguished from their parent species by their possession of a single mutant character, that, in general, they must have arisen by the accumulation in one individual of a number of such mutations (q.v.). Indeed, in certain cases the differences between two species, although they may be supposed to arise from modification of the chromosomal mechanism, ap pear to depend, not on the change of individual genes, but on much grosser events, such as the fragmentation of the chromo somes themselves.
It is evident that the doubt as to the real existence of genera is of the same character as that which occurs in the case of species; indeed, in this case it seems certain that the single term covers two quite independent classes of groups. It may imply that the group of species included within the genus have all arisen from a single species, which itself presented the generic characters, having, by evolution, come to differ in structure from its own parents and from sister species to a greater extent than species within a genus usually do. It is probably this conception which is usually present in the minds of taxonomists. But palaeontol ogists in general hold a very different view of the nature of a genus. It is widely believed by them that evolution has been to a large extent orthogenetic, that is, that all the species which fall within a small group pursue parallel evolutionary courses, so that, as time goes on, their members exhibit changes of structure which are of similar nature in all, although they have been independently acquired. To the group made artificially to include all the allied species which are in the same stage of evolutionary advance it is customary to apply the term genus. A group so defined differs from that which is ordinarily understood by the name, in that its members have not arisen from a single species, itself capable of inclusion within the genus.