PREFERENTIAL SEGREGATION IN MAIZE 69 



That a structural difference, aside from the extra chromatin of abnormal 

 10, exists between the two kinds of chromosome 10 also is indicated by the 

 pairing relationships in plants trisomic for chromosome 10. in plants with 

 two normal and one abnormal chromosome 10, trivalent associations were 

 observed in 251 (60.2 per cent) among a total of 417 microsporocytes. When 

 a chain of 3 was found at diakinesis, the abnormal 10 occupied a terminal 

 position in 90 per cent of the cells. It was united with a normal chromosome 10 

 by a chiasma in the short arm. A univalent chromosome 10 was found at 

 diakinesis in 39.8 per cent of the pollen mother cells. 



If pairing, as reflected by chiasmata formation, were random among the 

 three chromosomes, the ratio of normal -.abnormal chromosomes 10 in the 

 univalent class should be 2 : 1. Actually the unpaired chromosome was a nor- 

 mal 10 in 28 cells among a total of 166, while in the remaining 138 cells the 

 univalent was an abnormal 10. In individuals again trisomic for chromo- 

 some 10, but possessing one normal and two abnormal chromosomes, the 

 percentage of trivalent associations at diakinesis was 57.9 in a total of 513 

 cells. In the chains of 3, the two abnormal homologues were adjacent mem- 

 bers, joined by a chiasma between their long arms, in 70 per cent of the 

 cases. An unpaired chromosome 10 was found in 42.1 per cent of the micro- 

 sporocytes. 



If pairing were random, two times as many abnormal lO's as normal lO's 

 should be found as univalents; but in a total of 216 cells an abnormal 10 

 was the univalent in 69, while a normal chromosome 10 was the univalent 

 in 147. Chiasma formation among the three chromosomes 10 of trisomic 

 plants clearly is not at random. There is a marked preference for exchanges 

 in the long arm between the two structurally identical homologues. If synap- 

 sis usually begins at the ends and progresses proximally, the non-random as- 

 sociations found in trisomic plants become understandable. Normal recom- 

 bination values for the li-gi and gi-R regions which lie proximal to R (see 

 Table 4.1 for gi-R data) indicate that any suppression of crossing over is 

 confined to the region beyond the R locus in disomic plants heterozygous for 

 the two kinds of chromosome 10. It is no doubt significant that differences 

 in chromomeric structure are not found in regions proximal to the R locus. 



Inasmuch as the R locus is closely linked with the extra chromatin of ab- 

 normal 10, the ratio oi R:r gametes from heterozygous plants gives a good 

 approximation of the frequency with which the abnormal chromosome passes 

 to the basal megaspore. The genetic length of the long arm of chromosome 10 

 is such that at least one chiasma is found in the arm. If one chiasma invari- 

 ably occurs in the long arm of heteromorphic bivalents, each of the two dis- 

 joining dyads of anaphase I will possess one normal chromatid and one ab- 

 normal chromatid. Preferential segregation would be restricted to the sec- 

 ond meiotic division, and occur only if the orientation of the dyad on the 

 spindle of metaphase II were such that the abnormal chromatid passed to 



