Bower Ornamentation versus Plumage Characters in Bower-Birds

Abstract

450 tvo•. •3 GENERAL NOTES Bower Ornamentation versus Plumage Gharacters in Bower-birds.-Bower-birds and birds of paradise are believed to be related though they are often placed in different families. The courtship behavior, while elaborate in both groups, is very different. All but the most primitive of the birds of paradise display their spectacular plumes. The bower-birds, on the other hand, have little in the way of morphological ornamentation, though a few species are brightly colored and some have a crest. Their courtship behavior is unique, at least among birds, in that it is conducted in or near a "bower" constructed by the male bird (except in two primitive species). These bowers are often very elaborate stick structures which fall into four categories: 1) the maypole type (a column of sticks up to nine feet in height supported by a forest sapling and sometimes capped with a hut or tepee-like roof); 2) the avenue type (an elongated mat of sticks with usually two and sometimes four vertical stick walls between which the bird displays); 3) the mat type (a simple platform of ferns and sticks); and 4) the stage type (a cleared area of earth). These structures are decorated in specific ways by the various species. The ornaments, which sometimes number nearly a thousand, take the form of colorful flowers, berries, leaves, snails, insects, and the like. On several trips to New Guinea, I have given much attention to the study of bower-birds. During this work, the detailed results of which will be published elsewhere, I noticed that the species with the most elaborate bowers are often those with the least elaborate plumage. It occurred to me that there might be a correlation between the degree of ornamentation of the bower and that of the bird. This condition proved most pronounced in the "maypole builders." Indeed, the bower in some maypole builders (and perhaps in certain of the avenue builders) may be of such transcendent importance in the behavioral pattern of the species that it nearly or completely replaces the visual morphological signals of the builder and, through a transfer of the forces of sexual selection to inanimate objects, renders morphological ornamentation superfluous. The aptly named Amblyornis inornatus-a tepee builder-is the primary example of this hypothetical phenomenon. The male of this species, which constructs the most complicated and highly ornamented bower known, is crestless and virtually indistinguishable from the female; yet the males of all of its close relatives wear elongated golden-orange crests and differ strongly from the females. A second case in point is A. subalaris, which builds a somewhat less complicated tepee-type bower. This species has the golden-orange crest darker and shorter than any of the other dimorphic species and heavily edged with dark brown as in the crestless female. l•*urthermore, the distribution and extent of brown in the orange crest of A. subalaris are extremely variable, giving the impression of recent release from the controls imposed by sexual selection. On the other hand, A. macgregoriae, which builds the simplest bower (a short cone of sticks around a pole on a mossy saucer with no suggestion of the tepee-like roof) and does not use colorful ornaments or, in fact, any but the simplest, most inconspicuous of paraphernalia (a few bits of charcoal, sometimes a lichen-decorated stick or some animal silk), has by far the longest, most brilliant crest found in the genus (with the exception of A. fiavifrons whose bower is unknown). These facts lead me to believe that the evolution of bower form and ornamentation from simple to complex reflects the degree of transfer of the forces of sexual selection from morphological characters to external objects, namely, the bower. The adoption of inanimate objects as primary sexual releasers apparently has no parallel