In the study of the sexual question it is absolutely necessary to guard against subjectiveness and all preconceived theory, and to avoid sentimentalism as well as eroticism. These two dangers play a considerable part in the study of human sexual life. Presented in a conscientious and scientific way the history of marriage furnishes us the most trustworthy material for the study of the sexual relations of man in social life. It is from this material that we can learn the relative importance of the different psychological and psycho-pathological factors in social evolution. But, to furnish valid material, history must not only be based on trustworthy and veracious sources; it must also give a comparative study of the sexual relations which exist in most, if not all, of the peoples actually existing. The present savage tribes no doubt resemble more closely the primitive peoples than our hybrid agglomeration of the civilized world. Moreover, the modern study of ethnology gives us more certain information than the uncertain, incomplete and often fabulous statements of ancient documents. I am speaking here of primitive history, and not of the Greek and Roman civilizations. Unfortunately the correctness of ethnological observations, and especially their interpretation, still leave much to be desired. Edward Westermark, professor at Helsingfors, in his "History of Human Marriage," has given us a monumental work, which is remarkable, not only for the richness and exactness of its material, but also for the clearness and good sense of its criticism. I shall give a _résumé_ of Westermark's results, as the subject is beyond the domain of my special studies. The author has collected a great number of observations in order to avoid erroneous conclusions. He warns the reader against a hasty generalization, which attributes without proof certain customs of living savage tribes to our primitive ancestors. ORIGIN OF MARRIAGE In the previous chapter we have considered the phylogeny of love in general. We have seen that some of the lower animals, such as the ants and bees, give evidence of an instinctive social altruism much greater than that of man, while other animals, such as birds, are superior to us as regards monogamous conjugal fidelity. But it is a question here of analogies due to phenomena of convergence, and these animals are of interest to us only as remote objects of comparison. As regards marriage in primitive man, we can only compare ourselves with the living animals most closely allied to us, viz. the _anthropoid apes_.