70 M. M. RHOADES 



the lower pole of the spindle. Normal segregation would occur in those mega- 

 sporocytes which had homomorphic dyads. 



If the terminal segment of abnormal 10 determines preferential segrega- 

 tion, it follows that loci near the end of the long arm will be preferentially 

 segregated more frequently than loci further removed from the end of the 

 chromosome. From the data in Tables 4.1 and 4.2 it is evident that the dis- 

 tortion from a 1 : 1 ratio is greater for the R locus than for the more proximal- 

 ly situated gi locus. The li locus which is proximal to gi was less affected 

 than ^1. 



Longley (1945) reported non-random segregation at megasporogenesis for 

 chromosome pairs other than chromosome 10 when one of the two homologues 

 had a prominent knob and the other was knobless. Segregation was random 

 for these heteromorphic bivalents in plants homozygous for the normal chro- 

 mosome 10, and non-random if abnormal 10 was heterozygous. He studied 

 preferential segregation of chromosomes 9 and 6. The data for chromosome 9 

 are the most instructive. Some strains of maize have a chromosome 9 with a 

 knob at the end of the short arm, others have a knobless chromosome 9. The 

 C, Sh, and Wx loci lie in the short arm of this chromosome, with Wx nearer 

 to the centromere. C and Sh are in the distal one-third of the short arm. Ap- 

 proximately 44 per cent recombination occurs between Wx and the terminal 

 knob — they approach independence — while C and Sh are 23 and 26 recombi- 

 nation units distant from the knob. * 



When plants of knob-C/knobless-c constitution, which were also heterozy- 

 gous for abnormal 10, were pollinated by recessive c, 64 per cent of the func- 

 tioning megaspores possessed the C allele. The Sh locus, close to C, showed a 

 similar degree of preferential segregation in comparable tests, but the Wx 

 locus was little affected. Such a progressive decrease in effect is expected if 

 the terminal knob on the short arm is instrumental in producing preferential 

 segregation. The part played by the knob of chromosome 9 was wholly un- 

 expected. Obviously this heterochromatic structure can no longer be con- 

 sidered as genetically inert. The data on various loci in chromosomes 9 and 

 10 prove that the degree of preferential segregation of a locus is a function 

 of its linkage with heterochromatic regions which, in some way, are con- 

 cerned with non-random segregation. 



The data presented above show that alternative alleles are not present in 

 equal numbers among the female gametes when abnormal 10 is heterozygous. 

 We have here an exception to Mendel's first law. Are deviations from Men- 

 del's second law, the independent assortment of factor pairs on non-homolo- 

 gous chromosomes, also occurring? This question is answered by Longley's 

 data where the C and R loci are both segregating preferentially. In separate 

 experiments he found the C locus was included in 64 per cent and the R locus 

 in 69 per cent of the functioning megaspores. Assuming that these percent- 

 ages hold in plants where both are simultaneously segregating, the observed 



