A uniformity test using ‘Prophet’ crambe(Crambea abyssinica Hochst.) in a “solid seeding” of 1,296 basic units (one basic unit equaled 1.47 m2) was conducted to determine an appropriate plot shape and size for use in crambe yield testing.In analyses using Smith's regression procedure, the soil heterogeneity index, b, was calculated as 0.5361 and the optimum plot size was estimated to be 6.70 m2.Results of analyses using the modified maximum curvature technique developed by Lessman and Atkins (1963) indicated a plot size of 5.35 m2 as optimum.Plot shape had an effect on plot‐to‐plot variability. Long, narrow plots arranged with their greatest length in the direction of greatest variation resulted in a lower estimate of among plot variance.

No abstract

Reciprocal crosses between 3‐dwarf and 2‐dwarf lines of grain sorghum differing only at the Dw2 locus were grown in the summer of 1963. The effect of plant height on yield was studied to elucidate mechanisms or causes of superior yields of tall genotypes and to determine if higher yields might be achieved in combine height sorghums. It was concluded from light availability data that the taller 2‐dwarf parent and F1's were perhaps more efficient in utilization of available light due to spatial arrangement of leaves. The data suggested that the yield level of the taller 2‐dwarf line may not be recovered rapidly in the shorter 3‐dwarf line by manipulation of alleles at the Dw2 locus conditioning height.

One hundred sixty‐two lines were used to detect genotypic diversity for 7 characters inCrambe abyssinica Hochst during 1970 and 1972. Lines used were from 2000 randomly selected out of PI247310 and PI279345, and reselected on the basis of their progeny performance in head row nurseries. The cultivars, ‘Prophet,’ ‘Meyer,’ and ‘Indy,’ were used as checks. Characters studied were seed yield, test wt, plant ht, % oil (measured gravimetrically and by nuclear magnetic resonance), days to bloom, and % glucosinolates. Statistical differences were detected among lines for all characters except % glucosinolates, based on individual year and combined analyses of variance. Broad sense heritability estimates ranged from .22–.88. Gains from selection, as a % of the mean for each character, were estimated at 1.13–14.11. Intercharacter relationships were analyzed by correlation. The ranges in variation for traits are considered limited for good progress from selection within the germplasm represented, but advances can be expected.

Eleven crambe introductions grown in 14‐ and 38‐inch rows in southern and northcentral Indiana in 1964 were evaluated for number of branches and seed pods in different regions of the plant, plant height, weight of seed pods and seeds per volume of seed pods, and yield. A highly significant difference between the two locations was found for most characters. Different row spacings significantly affected most of the characters measured. However, variation among introductions was significant only for the number of branches attached to the lowest 4 inches of the main stem and the number of primary branches on the whole plant. Plant‐to‐plant variation within introductions in wide rows was heterogeneous for number of primary branches and seed pods in the lower regions of the plant. Desirable plant habit for combine harvest is possible with the material available by using appropriate cultural practices and selecting among and/or within introductions. For yield, however, to insure an effective improvement program in crambe, additional introductions of new diverse material or the possible development of more variable populations by hybridization seems necessary.