It is generally accepted that a large fraction of genomic sequence variations within and between species are neutral or nearly so. Whether the same is true for phenotypic variations is a central question in biology. On the one hand, numerous phenotypic adaptations have been documented and even Kimura, the champion of the neutral theory of molecular evolution, believed in widespread adaptive phenotypic evolution. On the other hand, phenotypic studies are strongly biased toward traits that are likely to be adaptive, contrasting genomic studies that tend to be unbiased. It is thus desirable to test the neutral hypothesis of phenotypic evolution using traits irrespective of their potential involvement in adaptation. Here we present such a test for 210 morphological traits measured in multiple strains of the yeast Saccharomyces cerevisiae and two related species. Our test is based on the premise that, under neutrality, the rate of phenotypic evolution declines as the trait becomes more important to fitness, analogous to the neutral paradigm that functional genes evolve more slowly than functionless pseudogenes. Neutrality is rejected in favor of adaptation if important traits evolve faster than less important ones, parallel to the demonstration of molecular adaptation when a functional gene evolves faster than pseudogenes. After controlling the mutational size, we find faster evolution of more important morphological traits within and between species. By contrast, an analysis of 3466 yeast gene expression traits fails to reject neutrality. Further, intraspecific and interspecific variations in yeast gene expression conform to the phylogenetic relations of the strains rather than their ecological environments. Thus, yeast morphological evolution is largely adaptive, but the same does not apply to the transcriptome, suggesting that phenotypic variations at different levels are shaped by different evolutionary forces.