for animals to take pleasure in practising whatever instinct they follow at other times for some real good.  These manifestations are primarily for the sake of producing sexual tumescence, and could not well have been developed to the height they have reached unless they were connected closely with propagation.  That they may incidentally serve to express “gladness” one need not feel called upon to question.

Another observer of birds, Mr. E. Selous, has made observations which are of interest in this connection.  He finds that all bird-dances are not nuptial, but that some birds—­the stone-curlew (or great plover), for example—­have different kinds of dances.  Among these birds he has made the observation, very significant from our present point of view, that the nuptial dances, taken part in by both of the pair, are immediately followed by intercourse.  In spring “all such runnings and chasings are, at this time, but a part of the business of pairing, and one divines at once that such attitudes are of a sexual character....  Here we have a bird with distinct nuptial (sexual) and social (non-sexual) forms of display or antics, and the former as well as the latter are equally indulged in by both sexes.” (E.  Selous, Bird Watching, pp. 15-20.)

The same author (ibid., pp. 79, 94) argues that in the fights of two males for one female—­with violent emotion on one side and interested curiosity on the other—­the attitude of the former “might gradually come to be a display made entirely for the female, and of the latter a greater or less degree of pleasurable excitement raised by it, with a choice in accordance.”  On this view the interest of the female would first have been directed, not to the plumage, but to the frenzied actions and antics of the male.  From these antics in undecorated birds would gradually develop the interest in waving plumes and fluttering wings.  Such a dance might come to be of a quite formal and non-courting nature.

Last, we owe to Professor Haecker what may fairly be regarded, in all main outlines, as an almost final statement of the matter.  In his Gesang der Voegel (1900) he gives a very clear account of the evolution of bird-song, which he regards as the most essential element in all this group of manifestations, furnishing the key also to the dancing and other antics.  Originally the song consists only of call-cries and recognition-notes.  Under the parallel influence of natural selection and sexual selection they become at the pairing season reflexes of excitement and thus develop into methods of producing excitement, in the male by the muscular energy required, and in the female through the ear; finally they become play, though here also it is probable that use is not excluded.  Thus, so far as the male bird is concerned, bird-song possesses a primary prenuptial significance in attracting the female, a secondary nuptial significance in producing