Thus one sees how difficult it is to have knowledge of the breeding test, even in regard to large animals. It is obvious that the difficulty of obtaining it in regard to the thousands of kinds of minute creatures is much greater. Yet when they say, "This is a distinct species," naturalists do mean that it is not only marked off from other animals or plants most like to it by a certain shape, colour, or other quality or qualities, but that it breeds apart with its own kind and does not naturally hybridize with those other forms most like to it. Although the kind of naturalist called a "systematist" who makes it his business to accurately describe and record and distinguish from one another all the existing species of some one group–say, of antelopes, of mice, of flowering plants, of fishes, or of fleas–has only a knowledge in a few instances of the breeding of the organisms which he describes as "distinct species," he yet does know, in regard to some one or more of his species in most groups, the facts of pairing and reproduction, and what are the limits of variation in the markings and other characteristics of at least one or two species definitely submitted to the "breeding test," that is to say, ascertained to be "true physiological species," kept apart by deep-seated chemical differences in their blood and tissues. Hence it is legitimate for him, by careful balancing and consideration of all the facts, to determine–not absolutely, but by analogy–the value to be assigned (whether as indicating true species or merely varieties capable of pairing with the main stock) to points of difference among the specimens of a dead collection brought from some distant land or from some position in which it would be impossible to make observations with regard to "pairing" and "breeding true."