The time of initiation of segregational pairing between nonhomologues in Drosophila melanogaster: a reexamination of wm4.

HE distributive pairing model of meiosis for the female of Drosophila melanoTgastcr (R. F. GRELL 1962a) postulates a chronological series of events which includes exchange pairing, exchange, distributive pairing, and segregation, in the order given. The model was dictated by genetic data, derived from studies of nonhomologous pairing, which showed that the frequency with which nonhomo!oguos paired, as m2asured by their segregation, did not influence the frequency with which homologues paired, as measured by exchange, even though the same chromosome was involved in both processes. Furthermore, nonhomologous pairing, although selectively involving noncrossover chromosomes, did not produce the noncrossovers that it utilized, since it did not increase their frequency. Thus, two independent pairings were indicated and a temporal order w’ils conferred upon them such that the homologous type, as well as the process of exchange itself, necessarily preceded the nonhomologous type. The distributive model represents the single sequence consonant with this genetic information. Subsequent studies, including those of secondary nondisjunction (R. F. GRELL 196213) , of heterozygously inverted heterologues (ROBERTS 1962) , of compound chromosomes (E. H. GRELL 1963,1970), of free fourth chromosomes (R. F. GRELL 1964b) , of heterozygous translocations ( CHANDLEY 1965; R. F. GRELL, unpublished), and particularly free X duplications (R. F. GRELL 1967) have extended and clarified certain aspects of the model; but, more importantly, they have in every case provided additional evidence as to its validity. In 1967, MERRIAM reported certain data which he interpreted to mean that nonhomologous associations, hearing segregational consequences, were established prior to exchange. His conclusion was based upon a genetic test of a particular inverted X chromosome known as white-mottled 4 . The test was intended to distinguish between the distributive model and an “alternative” model proposed by NOVITSKI (1964). In reality, the model which MERRIAM tested turned out to be not the NOVITSKI model hut la third model which will be designated the MERRIAM model. A comparison of the three models can be found under RESULTS