104 BRE 10997 Brain Research, 343 (19S5) 104-, 1i2 Elsevier Neural Correlates of Female Song in Tropical Duetting Birds ELIOT A. BRENOWITZ j . ARTHUR P. ARNOLD 1and RACHEL N. LEVIN2,3 1Department of Psychology and Laboratory of Neuroendocrinology, Brain Research Institute, University of California, Los Angeles, CA 90024, 2Section of Neurobiology and Behavior, Division of Biological Sciences, Cornell University, Ithaca. N Y 14853 (U.S.A. ) and 3Smithsonian Tropical Research Insatute, Balboa (Panamal (Accepted December llth. 19841 Key words: bird song - - duetting bird - - female song - - vocal control - - sexual dimorphism Duetting involves production of song by female and male birds in close temporal coordination. We studied the neural network con- trolling song in 3 tropical duetting species. The volumes of song control regions (SCRs) in the brain, neuronal density in nucleus robus- tus of the archistriatum (RA) which is one of these SCRs, total number of neurons in RA, and somal size of neurons in RA were meas- ured and compared to values published for zebra finches and canaries in which only males sing. The extent of sexual dimorphism in SCRvolumes, RA neuronal density, and total neuronal number in RA varied in a graded fashion across species and was correlated with extent of sexual dimorphism in song repertoire size in any one species. Somal size of RA neurons was identical in mates and fe- males of each duetting species, regardless of relative repertoire size. Of all SCRs, the caudal nucleus of the ventral hyperstriatum ap- peared to have the greatest relative size in the song system of duetting birds compared to non-duetting species. INTRODUCTION Investigation of the neural basis of song produc- tion in oscine birds has provided much insight into the causal relationship between structure and function in the nervous system. The song control system consists of a network of discrete connected brain nuclei: le- sioning one of several of these nuclei may lead to se- vere deficits in song behavior 3,24,26. Previous re- search has concentrated on two species in which only the male sings, the zebra finch (Poephila guttata) and canary (Serinus canarius), in which there is pro- nounced male-biased sexual dimorphism of neural song control region (SCR) attributes. However. there are several avian specms in which females also sing. In this report we examine the neural correlates of female song in species where both males and fe- males engage in complex vocal duets with each other. In zebra finches, only the adult male is normally capable of song. Song in canaries is generally restrict- ed to adult males, although some females will sing if they are in breeding condition and isolated from males. Such female song contains far fewer syllable types than does male song 2s. These sex differences in vocal behavior correlate well with sex differences in the neural system controlling song in both species. The gross volumes of the caudal nucleus of the ven- tral hyperstriatum (HVc), the robust nucleus of the archistriatum (RA), Area X of the parolfactory lobe. and the hypoglossal motor nucleus (nX!I) are significantly larger in malesZ3. At a finer level of anal- ysis. celt body size and/or dendritic field spread of neurons is generally smaller in females in HVc. RA. nXII, the magnocellular nucleus of the anterior neo- striatum (MAN), and the dorsal medial portion of the intercollicular nucleus (DM of ICo) 8A3 In RA of female zebra finches, the total number of neurons and size of neuronal soma are estimated to be less than half that in mate RA, and spacing between RA neurons is only about 20% that in malesl3. The occurrence of female song in a number of spe- cies ratses fascinating evolutionary questions. Do singing females have male-like neural song control systems, or have they evolved novel female-typical vocal control systems? No previous reports have ad- dressed the neural basis of naturally-occurring re- Correspondence: E. A. Brenowitz. Department of Psychology, 1283 Franz Hail, University of California. Los Angeles. CA 90024. U.S.A.