No abstract

Uniformity trials were conducted with pure stands of three forage species and one mixture of two species. Plot variance decreased with increasing plot size in every case. The relationship was very close and almost completely linear when plot variance and plot size were converted to logarithms. Thus plot size can be used to estimate plot or error variance. In general, increasing replication number reduced error more rapidly than increasing plot size. Using green weight only to estimate dry weight left 10% of the variation in dry weight in error.

No abstract

Synopsis The clones differed in general combining ability for forage yield, seed yield, and leaf spot reactions. For plant height, specific combining ability was significant and general combining effects were not present. In general, there was fair agreement among the rank of clones and the two measures of combining ability, polycross progeny performance and average single cross performance.

Synopsis Tests for randomness of pollination for 13 characters showed only 1 case out of the 13 with significant differences due to replication (i.e. location) in the polycross or parental field.