4.3. Red color and the Bekko pattern
We have found that the red and Bekko phenotypes to be inherited independently suggesting that the two colors are controlled by different and unlinked loci. For instance, the proportion of the Kohaku offspring (red pattern without Bekko pattern) was similar to the proportion of the Sanke offspring (red pattern with Bekko pattern) as previously reported by Gomelsky et al. (1996, 1998). However, both red and Bekko patterns segregated in offspring of transparent parents. It is possible that the “Tp” locus may be epistatic to both genes that separately control these two colors. Under this hypothesis, Kohaku koi are probably recessive homozygotes to the bl allele and hence the absence of Bekko pattern in progenies of Kohaku. Crossing F1 transparent offspring of Kohaku parents could verify that. Under the same hypothesis, the Shiro-Bekko phenotype does not have red color because of the genotype in the red color loci (R" and Rb) and not due to the dominant Tp allele. Another possibility is that there are two separate “inhibitor” loci, one epistatic to the red color genes (Ra and Rb) and the other to the black color gene (Bl). Two separate “inhibition loci” would be supported if red-colored offspring were found in crosses of Shiro-Bekko parents or if Bekko offspring were found in crosses of Kohaku parents. We did not cross Shiro-Bekko parents, but crosses of Kohaku parents never yielded offspring with Bekko pattern.