SYNOPSISWith 7 corn hybrids at various locations in Virginia, hybrids were more important than growing conditions as a source of variation in oil content; the reverse was true for the effect on protein content. With 3 levels of N, P, and K, and at 10,000 and 16,000 plants per acre, the location, planting rate, and N level all significantly affected protein content. Protein content was significantly higher under drought conditions than under good growing conditions. Highest protein yields occurred with the higher plant stand and heaviest N application. None of the factors had any appreciable effect on oil contenL ALTHOUGH the corn kernel has long been and prob-

Two populations of maize (Zea mays L.), Virginia Corn Belt‐Southern Synthetic (VCBS) which was developed from a broad genetic base, and Virginia Long Ear Synthetic (VLE) which was developed from intercrosses of seven Corn Belt inbred lines, were advanced two cycles of recurrent selection based on S1 and testcross progeny yield. The 10 populations and their diallel crosses were grown at 3 locations with 6 replications at each location. Significant improvement in population yield and combining ability was achieved only with VCBS.S1 selection in VCBS was more effective than testcross selection in increasing both population yield and combining ability. Testcross selection was effective in increasing the frequency of genes that contribute to yield in crosses but not in populations per se whereas S1 selection increased both combining ability and population yield. The yield of the second‐cycle S1 selection population yielded 9.0 q/ha (11%) more than the second‐cycle testcross population and its mean population cross yield was 1.5 q/ha higher. The two populations derived from VCBS by S1 selection produced significantly higher yields when intercrossed or crossed with VCBS than either population derived by testcross selection. S1 selection in VCBS was as effective as testcross selection in increasing the frequencies of genes that contribute to yield, whether in inbred (derived) or hybrid populations.

Two recurrent selection series, originating from the same 153 S1 lines, were carried through two complete cycles. Selection was based on testcross yield in one series and on S1 progeny yield in the other. The mean S1 yield increased 31.4% with 2 cycles of selection for S1 yield and 17.9% with 2 cycles of selection for testcross yield. The more productive S1 lines tended to produce the more productive crosses, but the correlation between S1 and testcross yields decreased with each cycle of selection. Four S1 lines from the second cycle synthetic of the series selected for progeny yield were more productive than the original cross‐pollinated population. In a second experiment, lines that were derived from the more productive S1 progenies were most frequently maintained under visual selection. Nine of the 10 S1 lines that were still represented after selection in S3 progenies were above average in yield. Seven of the 31 S1 lines that were still represented after selection in S2 progenies produced testcrosses which ranged from 105 to 110% of the trial mean and 6 of these 7 were above average in S1 yield.

No abstract

Soybeans (Glycine max L.) for grain and sorghum (Sorghum vulgare L.) and maize (Zea mays L.) for both grain and silage were planted in late June and in July following winter barley (Hordeum vulgate L.) to evaluate these crops in a double cropping system in eastern Virginia. Soybeans have been widely grown after barley in that area. Production and harvest costs were estimated and net returns were calculated from these estimates and the values of the crops produced. Grain yields were obtained over a 2‐year period. Both mid‐season and late maturity soybeans produced crops of satisfactory quality each year. Early or very early maturity sorghum and maize hybrids were planted and appeared to be satisfactory for grain when planted in June but one July planting was severely damaged by an early frost. When the June plantings were harvested for grain, the net return from maize was significantly greater than from either sorghum or soybeans. Soybeans, however, appeared to be more dependable than maize or sorghum for grain if planted in July. Maize and sorghum silage yields were obtained the second year of the experiment. Net returns from both crops for silage were consistently greater at each date of planting than from any grain crop. All but one of the differences were significant. Net returns were greater from sorghum silage than from maize at each date of planting and the difference was significant for the July planting. Better adapted strains of all three crops are needed for use in a double cropping system.