Although test weight is used as a grading criterion and an indication of quality in wheat (Triticum aestivum L.), its relationship to specific quality parameters in soft wheat is not well documented. Seed characteristics, flour yield, and baking quality were studied in 24 soft red winter wheat (SRWW) genotypes grown in six environments to determine their relationship to test weight. Flour yield, flour protein, alkaline water retention capacity (AWRC), and softness equivalent (SEQ) were evaluated by the micro-procedures of the USDA-ARS soft wheat carly-generation milling and baking quality evaluation program. Despite removal of shriveled kernels prior to evaluation, environmental effects had a significant impact on quality parameters, ranging from 68% of total variability (SEQ) to 5% (AWRC). Test weight was correlated with flour yield, but was significantly correlated with flour protein content (r = 0.54, P < 0.05) as was kernel density (r = 0.49). Thousand-kernel weight, diversity of seed size, proportion of large seed, and average kernel length and width were not correlated with flour yield or other quality parameters. Test weight did not predict flour yield in SRWW when shriveling was absent, but it was related to flour protein content, which is associated with baking quality. Kernel size or size distribution did not affect end-use quality. The best predictive model based on the characters above explained little of the total variation in flour yield (R 2 = 0.22). S OFT WHEAT BREEDERS must consider milling and baking quality in cultivar improvement. Criteria of good baking quality of soft wheat flour may differ according to intended end-product use (Bass, 1990). Milling quality describes traits important in the milling process such as kernel texture and flour yield. Flour yield is expressed as the percent of flour obtained from a given amount of

Although genotype affects test weight in wheat (Triticum aestivum L.), little is known about how physical kernel characteristics affect this quality parameter. The influence of seed and spike characteristics on test weight was investigated in 24 soft red winter wheat (SRWW) cultivars grown in six environments. Seed characters and test weight were measured using gram samples from which shriveled kernels had been removed. Removal of the shriveled kernels increased test weight an average of 36 kg m−3 and reduced the total variation in test weight due to environment and environment by cultivar interaction. Of the seed and spike characters measured, only flour protein content and the average number of seed within a spike were significantly correlated to test weight (r = 0.56 and r = −0.41, respectively) at P = 0.05. Average kernel length and width and the average number of seed per spikelet showed weak negative correlations with test weight (r = −0.38, r = −0.35 and rs = −0.35, respectively) at P = 0.10. Increasing kernel density was weakly associated with increasing test weight (r = 0.39, P = 0.10). Thousand‐kernel weight, diversity of seed size, kernel length to width ratio, spike length, number of spikelets per spike, and spike density were not directly related to test weight. Results of this study indicate that among the characters measured, the amount of shriveling and the protein content of the grain were the most important determinants of test weight in SRWW cultivars.

No abstract