Because the physical characteristics (such as skin color and hair form) that are used to define race are polygenic, precisely measuring the genetic influence on them hasn’t been possible (although geneticists are getting closer) (Gibbons, 2010). Physical anthropologists and other biologists who study modern human variation have largely abandoned the traditional perspective of describing superficial phenotypic characteristics in favor of examining differences in the frequencies of genes. Beginning in the 1950s, studies of modern human variation focused on the various components of blood as well as other aspects of body chemistry. Such traits as the ABO blood types are phenotypes, but they’re also direct products of the genotype. (Recall that protein-coding genes direct cells to make proteins, and the antigens on blood cells and many constituents of blood serum are partly composed of proteins; Fig.14-2.) During the twentieth century, this perspective met with a great deal of success, as eventually dozens of loci were identified and the frequencies of many specific alleles were obtained from numerous human populations. Even so, in all these cases, it was the phenotype that was observed, and information about the underlying genotype remained largely unobtainable. But beginning in the 1990s, with the advent of genomic studies, new techniques were developed. Now that we can directly sequence DNA, we can actually identify entire genes and even larger DNA segments and make comparisons between individuals and populations. A decade ago, only a small portion of the human genome was accessible to physical anthropologists, but now we have the capacity to obtain DNA profiles for virtually every human population on earth. And we can expect that in the next decade, our understanding and knowledge of human biological variation and adaptation will dramatically increase.