| The courtship displays of male manakins (Pipridae) involve an array of acrobatic and postural elements. Previously it was found that species with elaborate displays exhibit specializations in motor planning and coordination areas of the brain. Several studies have suggested a relationship between cerebellum (Cb) morphology and distinct motor-related functions for the anterior Cb (somatosensory, flying, hopping/walking), posterior Cb (vision, audition, flying, hopping/walking), and vestibular Cb (flying, hopping/walking, vestibular). The anterior, posterior and vestibular Cb as well as basic morphological features of the Cb were measured and tested for a relationship with courtship display complexity in manakins. I compared Cb morphology of four species of manakins: Manacus vitellinus, Pipra mentalis, Chiroxiphia lanceolata, and Lepidothrix coronata as well as one species of flycatcher, Mionectes oleagineus, representing a range of display complexities. I scored each species's display for overall complexity, taking into account acrobatic elements, sound production by the wings and level of coordination between two displaying males. The following features of Cb morphology were measured: Cb volume, white matter volume, molecular layer volume, granular layer volume, volume of the lateral cerebellar nucleus (CbL), volume of the medial cerebellar nucleus (CbM), CbL cell density, CbM cell density, Purkinje cell (PC) size and density, and the sizes of the anterior Cb, posterior Cb and vestibular Cb cortices. Morphology variables were corrected for allometry, if necessary. Parallel analyses were performed on data corrected for phylogenetic relatedness using independent contrasts and on non-corrected data. Data reduction was accomplished by performing individual linear regressions of each Cb morphology variable vs. display complexity were performed, and variables for which p<0.1 (CbM volume, vestibular Cb size, white matter volume and PC size for both phylogenetically corrected data and non-corrected data) were then tested in stepwise multiple regressions. For non-corrected data, both vestibular Cb size (negative relationship) and CbM volume (positive relationship) best predict display complexity. For phylogenetically corrected data, only white matter volume predicted display complexity (positive relationship). This study is the first to provide evidence that specific morphological features of the Cb may evolve in conjunction with a sexually selected behavior. |
