A Contribution to an Understanding of Crossing over within Chromosome Rings of Oenothera.

There are two theories to account for linkage phenomena as observed in Oenothera. (1) Linked genes are always found within the same chromosome, that is, linkage is of the conventional or Drosophila type. The number of linkage groups is determined by the number of chromosomes in the genom, and bears no relation to chromosome configuration (SHULL 1928). (2) In addition to linkage of the ordinary type, genes may be distributed among several chromosomes and yet be linked, such linkage being induced by the cohesion of these particular chromosomes to form the circles so characteristic of Oenothera. The number of linkage groups is consequently determined by the number of chromosome groups, and varies from form to form (CLELAND 1923, 1924, 1926). Neither of these theories is capable, unaided, of explaining all the facts. But while the difficulties in the way of “theory 1” would seem to be insuperable, an hypothesis supplementing “theory 2” has been advanced which seems to account satisfactorily for the few cases of exceptional behavior which constitute the only difficulty in the way of this theory (DARLINGTON 1931, EMERSON 1931). The present paper is devoted to a discussion of this supplementary hypothesis. Leaving out of account the relation of the segmental interchange hypothesis to the problem of linkage (which hypothesis would definitely exclude “theory 1,” if proved [CLELAND 1933]), “ theory 1” meetsits Waterloo in the phenomenon of linkage alteration described by RENNER (1925). It is impossible on the basis of “theory 1” to explain how two or more genes can be linked in one complex-combination and be independent in another. The painstaking and brilliant researches of RENNER leave no reason to doubt, however, that such alterations in linkage do occur. Furthermore, the investigations of CLELAND and OEHLKERS (1929, 1930), RUDLOFF (1929, 1930), GERHARD (1929) and EMERSON (1930, 1931) indicate clearly that these alterations are related to and accompany alterations in chromosome linkage. Genes which are linked when the chromosomes cohere into a large circle become independent when they come to lie in a hybrid in which the chromosomes are no longer organized into a large circle, but form several independent pairs. The alteration in linkage relations, therefore, and the correlation between linkage relations and chromosome con-