Caroline Pellew and Eva Richardson Sansome 33 



the theory may be required to cover the whole field (see Mrs Sansome's 

 report, p. 46). 



In applying this theory to the connection between the factors for 

 round wrinkled and semi-steriUty in the Thibetan hybrids, we may 

 represent Th. 7 as AB {I{.).AB (R), CD. CD, and Duke of Albany as 

 AD {r).AD (r), BC.BG. The half-steriles in F^ will be Rr; in F^, etc., 

 they wiU also be Rr, except when crossing-over has occurred, the r 

 factor being transferred to AB, or the R factor to AD. Thus a critical 

 test of the theory may be carried out by intercrossing the various ex- 

 ceptional classes. Such crosses have been made and the hybrids will be 

 grown in the coming season. In a very similar case in maize (Burnham, 

 1930) the critical tests have been made, and the results are in agree- 

 ment with the theoretical expectation. Moreover, in this chromosome 

 ring in maize ("semi-sterUe 1") the morphology of the segmental inter- 

 change has been demonstrated very clearly by McChntock (1930). 

 McClintock was able to observe in spores with the normal number of 

 chromosomes, the presence of two homologous chromosomes belonging 

 to the set concerned with ring formation, and the absence of a member 

 of the second pair. To explain the 50 per cent. steriUty, she assumes that 

 "in half of the sporocytes any two adjacent chromosomes in the ring go 

 to the same pole, forming sterile combinations, and in the other half of 

 the sporocytes the adjacent members go to opposite poles, forming fertile 

 combinations." 



In Pisum, two other examples of ring formation associated with semi- 

 sterility are known, both discovered by Hammarlund and Hakansson. 

 In the case first discovered, Hammarlund (1923) found that the two pairs 

 of factors for colour-white and green-yellow pods were strongly linked, 

 although they are usually independent, and that the heterozygotes were 

 in general half-sterile. In the half-steriles Hakansson found ring forma- 

 tion (1931). Hakansson is inclined to think that the two factors in 

 question are, in the semi-sterile plants, in different chromosomes, and 

 that the appearance of linkage is brought about by the fertility of 

 parental chromosome combinations and the sterility of recombinations. 

 He considers however that on the genetical evidence the two factors 

 may equally well be in one chromosome with one of the factors in a 

 translocated segment. Such an arrangement in the ring may be repre- 

 sented thus (AB, CD are used as above, (A a) and (Gp gp) represent the 

 two pairs of factors) : 



A {a) B B (A) (Gp) C C (gp) D D A 



Joum. of Genetics xxv 



