In a general view of the social habit in animals certain outstanding facts are readily to be observed. It is of wide distribution and sporadic occurrence, it varies much in the completeness of its development, and there seems to be an inverse relation between its completeness and the brain power of the animal concerned. From the wideness of its distribution the social habit may be supposed to represent a forward step in complexity which comes about readily. It has the appearance of being upon a path which species have a natural tendency to follow, a line of evolution which is perhaps rendered possible by constantly occurring small variations common to all animals and taken advantage of only under certain circumstances of pressure or increase. It seems not to depend on any sudden large variation of type, and such is not necessary to account for it. It differs from many other modifications which we know animal life to have undergone in being immediately useful to the species from its very beginning and in its least perfect forms. Once started, however imperfectly, the new habit will have a natural tendency to progress towards fuller forms of sociality by reason of special selective forces which it inevitably sets going. The fact that it is valuable to the species in which it develops even in its most larval forms, {103} combined with its tendency to progress, no doubt accounts for the wonderful series of all degrees of gregariousness which the field of natural history presents. I have pointed out elsewhere that the fundamental biological meaning of gregariousness is that it allows of an indefinite enlargement of the unit upon which the undifferentiated influence of natural selection is allowed to act, so that the individual merged in the larger unit is shielded from the immediate effects of natural selection and is exposed directly only to the special form of selection which obtains within the new unit. There seems little doubt that this sheltering of the individual allows him to vary and to undergo modifications with a freedom which would have been dangerous to him as an isolated being, but is safe under the new conditions and valuable to the new unit of which he now is a part. In essence the significance of the passage from the solitary to the gregarious seems to be closely similar to that of the passage from the unicellular to the multicellular organism—an enlargement of the unit exposed to natural selection, a shielding of the individual cell from that pressure, an endowment of it with freedom to vary and specialize in safety. Nature has thus made two great experiments of the same type, and if one be reasonably careful to avoid arguing from analogy, it is possible to use one case to illuminate the other by furnishing hints as to what mechanisms may be looked for and in what directions inquiry may profitably be pursued.