absence in the female in numerous insect species), or special sense organs (the much more expanded antennæ of male cecropia, promethea, polyphemus, and other bombycine moths, as compared with those of the female), or special structures for the care of the young (milk glands of female mammals, brood pouches of female marsupials, pits on the back of the male of the frog Pipa (Fig. 43), for carrying the eggs, etc.), or recognition marks (the eye spots, collars, wing bands, tail blotches, and such other conspicuous color spots and markings possessed by the males and wanting in the females of various bird species), or, finally, characters connected with special habits of one sex differing from those of the other (the pollen baskets and wax plates of the worker female honey bees, the winglessness of certain female parasitic insects, the males being nonparasitic and winged, etc.).

FIG. 42.-Fore leg of male water beetle, Dyticus, showing special suckerlike expansion of the leg. (After Miall.)

The special characters may be apparently for the purpose of attracting or exciting the other sex, as the brilliant colors, markings, and other ornamentation of many male birds, some mammals, and some reptiles and very many fishes, and the cries and songs, special odors, and curious antics or dancing of the males of various animals (mammals, birds, spiders, insects, etc.). In many of these cases the special secondary sexual characters appear only during the breeding season; in others they are persistent.

The characters may also be of the type called reciprocal, that is, organs which exist in functional condition in one sex, but in the other appear in rudimentary and often nonfunctional forms, as the reduced horns of female antelopes and goats, the undeveloped stridulating organs of female crickets and katydids, small spurs on the female pheasant, reduced mamma of male mammals, undeveloped mimicry of male butterflies, etc.

FIG. 44.-Male (A) and female (B) of the fly, Calotarsa insignia Ald., showing secondary sexual characteristics on the feet of the male. (After Aldrich.)

Finally the characters may be indifferent, that is, without any apparent utility; as the reduced wings of numerous female insects, the rudimentary alimentary canal of the male Rotatoria, absence of antlers of female deer, loss of wings in insect females, small differences in size and markings between males and females, slight differences in wing form in hummingbirds, dragon flies, and butterflies, differences in number of tarsal and antennal segments in insects, etc.

The explanation of these various differences between males and females plainly cannot be a single one. The extreme variety of the secondary sexual differences of itself makes it necessary to find more than one explanation for their existence. To take the most obvious case, it is apparent that the useful characters, such as the fighting antlers of the male deer, can be explained probably by natural selection. At least these characters fall readily into line with precisely that type of useful specialization for whose explanation we rely on natural selection. So practically all those secondary sexual characters of our first category, namely, those obviously useful to the possessor or to its young, such as organs of offense and defense, brood pouches, food-producing or gathering organs, special

means of locomotion, etc., may be considered to offer no special problem. Although indeed the reason why these useful characteristics should be possessed by but one sex is by no means always, or perhaps even often, plain to us.

But the real problem presented by secondary sexual characters is that thrust on us by the nonuseful and even apparently disadvantageous differences. Why the male bird of paradise should be decked out in a plumage certain to make it a conspicuous object to every enemy it has, and of a weight and difficulty of manipulation that must mean a constant demand on the strength and attention of the bird, is a question that demands a special answer. In the same case with the bird of paradise are the peacock, the gorgeous male pheasant (Fig. 45),

many hummingbirds (Fig. 40), etc. Now to explain these extraordinary secondary sexual differences the theory of sexual selection has been devised.

This theory, in few words, is that there is practically a competition or struggle for mating, and that those males are

successful in this struggle which are the strongest and best armed or equipped for battle among themselves, or which are most acceptable by reason of ornament or other attractiveness to the females. In the former case mating with a certain female depends upon overcoming in fight the other suitors, the female being the passive reward of the victor; in the second case the female is presumed to exercise a choice, this choice depending upon the attractiveness of the male (due to color, pattern, plumes, processes, odor, song, etc.). The actual fighting among males, and the winning of the females by the victor is an observed fact in the life of numerous animal species. But a special sexual selection theory is hardly necessary to explain the development of the fighting equipment, antlers, spurs, claws, tusks, etc. This fighting array of the male is simply a special phase of the already recognized intraspecific struggle; it is not a fight for room or food, but for the chance to mate. But this chance often depends on the issue of a life and death struggle. Natural selection would thus account for the development of the weapons for this purpose.

For the development, however, of such secondary equal characters as ornament, whether of special plumage, color, pattern, or processes, and song, and special odors, and "love dancing," the natural selection theory can in no way account; the theory of sexual selection was the logical and necessary auxiliary theory, and when first proposed it met with quick and wide acceptance. Wallace in particular took up the theory and applied it to explain many cases of remarkable plumage and pattern development among birds. Later, as he analyzed more carefully his cases, and those proposed by others, he became doubtful, and finally wholly skeptical as to the theory.

The theory as proposed by Darwin was based on the following general assumptions, for the proof of each of which various illustrations were adduced. First, many secondary sexuai characters are not explicable by natural selection; they are not useful in the struggle for life. Second, the males seek the females for the sake of pairing. Third, the males are more abundant than the females. Fourth, in many cases there is a struggle among the males for the possession of the females. Fifth, in many other cases the females choose, in general, those males specially distinguished by more brilliant colors, more conspicuous ornaments, or other attractive characters. Sixth,

many males sing, or dance, or otherwise draw to themselves the attention of the females. Seventh, the secondary sexual characters are especially variable. Darwin believed that he had observed certain other conditions to exist which helped make the sexual selection theory probable, but the conditions noted are sufficient if they are real.

Exposed to careful scrutiny and criticism, the theory of sexual selection has been relieved of all necessity of explaining any but two categories of secondary sexual characters; namely, the special weapons borne by males, and special ornaments and excitatory organs of the males and females. For examination has disclosed the fact that males are not alone in the possession of special characters of attraction or excitation. Regarding these two categories Plate in his able recent defense of Darwinism, says "the first part of this theory, the origin of the special defensive and offensive weapons of males through sexual selection, is nearly universally accepted. The second part of the theory, the origin of exciting organs, has given rise to much controversy. Undoubtedly the presumption that the females compare the males and then choose only those which have the most attractive colors, the finest song, or the most agreeable odor, presents great difficulties, but it is doubtful if it is possible. to replace this explanation by a better." Some of these difficulties may be briefly enumerated.

The theory can be applied only to species in which the males are markedly more numerous than the females, or in which the males are polygamous. In other cases there will be a female for each male whether he be ornamented or not; and the unornamented males can leave as many progeny as the ornamented ones, which would prevent any accumulation of ornamental variations by selection. As a matter of fact, in a majority of animal species, especially of the higher vertebrates, males and females exist in approximately equal numbers.

Observation shows that in most species the female is wholly passive in the matter of pairing, accepting the first male that offers. Note the cock and hens in the barnyard, or the fur seal in the rookeries.

Ornamental colors are as often a characteristic of males of kinds of animals in which there is no real pairing, as among those which pair. How explain by sexual selection the remarkable colors in the breeding season of many fishes, in which the