1. The evidence so far obtained indicates that mosaics in maize are due to losses or rearrangements of chromosome fragments rather than to somatic crossing over as Stern finds for Drosephila. Paired mosaics involving different chromosomes have been found for nearly all of the easily identified endosperm characters. In seeds heterozygous for C and Pr the following results have been obtained:

These figures indicate a more or less random exchange between the 60 chromosome arms in this triploid tissue.

The secondary paired mosaics (twin spots within the dark part of primary twin spots) can not be accounted for by somatic crossing over but are understandable on the basis of translocation followed by further breaking at other places. Variegated waxy tissue in areas that have previously lost the C gene show an unstable condition that would not be expected from somatic crossing over. Similar variegation has been found involving C, Pr, and Su.

In seeds resulting from the pollination of C wx by c Wx light and dark aleurone twin spots were found indicating a shift of one C allele. If this resulted from an exchange of homologous segments the endosperm underlying the dark part of the twin spot should be waxy. In many such twin spots examined no waxy areas were found.

D. F. Jones