The Tunicate Locus in Maize Dissected and Reconstituted.

Tunicate maize, commonly known as pod corn, is a peculiar type in which the kernels are enclosed, as they are in wild grasses, in floral bracts called "glumes." Pod corn has been regarded as the ancestral form of cultivated corn by a number of nineteenth century students of maize and in this century by Mangelsdorf and Reeves, 2 whohave reviewed the extensive literature on the subject. A genetically reconstructed ancestral form of maize was developed by Mangelsdorfs by crossing pod corn with popcorn, a type also regarded as primitive. The principal characteristics of tunicate maize are controlled by a locus represented by the symbol, Tu, on the long arm of the fourth longest chromosome. In the course of our experiments this locus has several times mutated to a weak form of pod corn which we call "half tunicate" because its effects when homozygous in the genotype, tuhtuh, are about equal to those of the tunicate locus when heterozygous in the genotype, Tutu. This has led us to suspect that the tunicate locus may be a compound one and that the "mutations" which occur at this locus are, in fact, the product of crossing over which separates the components. To test this possibility we crossed a uniform inbred strain, A158, into which the Tu locus had been incorporated by nine generations of crossing and backcrossing, with another inbred strain carrying two recessive marker genes on chromosome 4: su on one side and gl3 on the other side of the Tu locus. The F1 plants heterozygous for the three loci represented by the genotype, SuTuGl3/sutugls, were then backcrossed to a second inbred strain homozygous for the three recessive loci. A total population of 10,248 plants of this backcross was grown over a period of three seasons, 1958-1960. Approximately half of these plants were expected to be heterozygous tunicate; 5,273 were found. There were, in addition, four heterozygous half tunicate plants, all of which proved by progeny tests to be crossovers, two being of the genotype Sugl3/sugl3, and two of the genotype suG13/sugl3. This represents a "mutation" or crossover rate of one in 1,319. Since the average percentage of crossing over between the loci Su and G(3 is 34 per cent,4 the chances of a mutation at the Tu locus being accompanied by a crossing over are approximately 1/3 if the two events are independent. The chances of four mutations being accompanied by crossing over are therefore 1/8,. If this were the first case on record of a locus being separated by crossing over, we should not