"Intersectional" hybrids (Corn Belt lines crossed with local strains) of the following types have been tested: single crosses, top crosses, multiple top crosses, three-way crosses, and double crosses. The average of all "intersectional" hybrids was 20.8 percent in 1939 and 18.2 percent in 1940, higher than the average grain yield of the local varieties. These hybrids were made up entirely at random except for morphological observations of the parent lines. Besides grain yield the "intersectional" hybrids approach or equal the local varieties in pest resistance and grain quality. When compared with Corn Belt double crosses, the "intersectional" hybrids are much superior in general adaption to North Carolina conditions.
In six locations across the state 5 × 5 lattice square designs on 1/140 acre plots were utilized in 1940. The lattice square design showed an average gain in precision of at least the equivalent of an extra replication in a complete randomized block design. Complete randomized blocks of more than 30 entries have proved very unsatisfactory in our studies. Since 5 × 5 lattice squares have been of doubtful value in the Corn Belt, it seems worth while to mention our results on the heterogeneous soils of the Southeast.
Paul H. Harvey