NUMBER AS DETERMINING FORM OF GROUP 159
over only the one colleague (which is very easy for the stronger nature among three) in order to dominate the other ; that is, in fact, to dominate the other two in the most legal form. On the whole, it may be said that unions in pairs, as contrasted with those of larger numbers, favor a relatively higher individuality of the participants, while, on the other hand, they presuppose that the restraint of peculiarity through the social articulation to an average level is lacking. If it for that reason is true that women are the less individual sex, that their differentiations vary less from the species type than is on the average the case with men, it would help to explain the further very general opinion that they are, as a rule, less accessible to friendship than are men. For friendship is a relationship entirely founded upon the individuality of the elements, perhaps even more than marriage, which, through its traditional forms, its social fixities, its real interests, includes much that is super-individual and independent of the peculiarity of the personalities. The fundamental differ- entiation upon which marriage rests is, in itself, not individual, but it pertains to the species ; friendship, however, rests upon a purely personal differentiation, and hence it is intelligible that in general real and permanent friendships are rare at the inferior levels of personal development, and that, on the other hand, the modern highly differentiated woman manifests notably enhanced capacity and inclination for friendships, alike with men and with women. The entirely individual differentiation has, in this case, attained decisiveness over that which pertains to the species, and we thus see the correlation formed between the sharpest individ- ualization and a relation that at this grade is absolutely limited to duality. This, of course, does not prevent the same person from forming at the same time various relations of friendship.
That combinations of two in general have, as such, specific traits is shown not merely by the fact that the entrance of a third modifies them entirely, but still more the variously observed fact that the still further extension to four or more by no means modifies the nature of the combination to a correspondingly wide degree. For example, a marriage union resulting in a single child has a quite different character from a childless union,
