Non-random assortment of chromosome pairs and meiotic drive in Drosophila melanogaster

Abstract

During meiosis different chromosome pairs are thought to assort quite randomly from each other; thus, an A/a; B/b (A and B refer to different chromosome pairs) chromosome type would be expected to produce four kinds of gemetes, AB, Ab, aB and ab in equal numbers. Several papers have been published reporting that in some cases the above principle was not always true (Novitski and Sandler, 1957; Grell, 1959). In all of these studies, however, the genotypes used were characterized by abnormal chromosome complements or involved chromosomes of more or less abnormal structure such as trans locations or inversions. In the absence of such abnormalities it has been widely assumed that the principle of random assortment is generally true. Recently, however, Hiraizumi and Nakazima (1967) reported that when the second chromosome of £. melanogaster carried segregation distorter but was free of any visible structural abnormality, it tended to assort more frequently with the Y chromosome. They also observed that the amount of non-random assortment was positively correlated with the frequency of recovery of the SD-bearing chromosome. The above phenomenon was well established for the case of the SD system but it is possibly a unique phenomenon characteristic only of this system. If, however, the phenomenon is generally true in nature, then it may be one of the important factors changing gene frequencies in populations. The present investigation established the generality of non-random assortment between the second and sex chromosomes and confirmed the positive correlation between the segregation frequency of the second chromosome and the degree of non-random assortment. Preliminary mapping of the elements of the system was also done. These findings are of interest because in many Drosophila experiments fitnesses of genotypes have been estimated by differential larval stage viabilities based upon the recovery frequencies among progeny flies under the assumption that the distorted segregation was due solely to the differential viabilities of the genotypes involved. with the establishment of the generality of the phenomenon of non-random assortment of chromosome pairs and the changes in segregation frequencies associated with it, many of these conclusions may have to be critically re-examined.