seed quality (Dornbos et al., 1989;Smicklas et al., 1992; Heatherly, 1993), but Vieira et al. (1991Vieira et al. ( , 1992 found no High temperature stress during seed filling in controlled environeffect on germination or vigor in field and greenhouse ments reduces soybean [Glycine max (L.) Merrill] seed germination experiments when the stress did not produce shriveled and vigor, but the effect of high temperature in the field has not been and abnormal seeds. Dry conditions at harvest may indetermined. Seeds of two soybean cultivars (Hutcheson, maturity group [MG] V, and DP4690, MG IV) were produced in the field crease physical injury and reduce quality if seeds are in Kentucky, Mississippi, Arkansas, and Texas in 2000 to 2002. Air handled at low moisture levels (TeKrony et al., 1987). temperature during seed filling was monitored and brown (mature) Temperature extremes during seed development also pods were harvested, hand threshed, and all shriveled and abnormal affect soybean seed quality. Freeze injury before physioseeds were removed before determining standard germination and logical maturity caused large reductions in germination vigor (accelerated-aging germination). Mean maximum temperatures and vigor (Judd et al., 1982). High temperatures also during seed filling (growth stage R5 to R7) ranged from 24.0 (Kenreduced seed germination and vigor in growth chamber tucky) to 37.6؇C (Texas). When seed lots infected with Phomopsis and phytotron experiments (Keigley and Mullen, 1986; longicolla (Hobbs) were removed from the analysis, standard germi-Dornbos and Mullen, 1991; Zanakis et al., 1994; Gibson nation and accelerated-aging germination (AA) decreased as mean and Mullen, 1996; Spears et al., 1997; TeKrony et al., maximum temperature during seed filling increased, but the decrease was significant (P ϭ 0.05) only for Hutcheson. Standard germination 2000; Egli et al., 2005). Temperatures of 33/28ЊC (day/ of Hutcheson declined linearly (r 2 ϭ 0.49) from near 100% at 24؇C night) (Keigley and Mullen, 1986), 35ЊC (Dornbos and to 85% at 36؇C, while the decrease in AA was curvilinear (R 2 ϭ 0.86) Mullen, 1991), 35/30ЊC (Gibson and Mullen, 1996), 38/ and germination reached 11% at 36؇C. Seeds of Hutcheson were more 33ЊC (Spears et al., 1997), and 38/27ЊC (TeKrony et sensitive to high temperature than seeds of DP4690 and seed vigor al., 2000;Egli et al., 2005) during seed filling reduced (AA) was much more sensitive to high-temperature stress than was

The development of late‐season K deficiency symptoms in cotton (Gossypium hirsutum L.) fields has become more frequent in the Mid‐South and Far West U.S. production regions. In this study, the objectives were to determine how yield and quality of lint produced by cotton genotypes of varying maturities are affected by different rates of soil applied K and N fertilization. Eight cotton genotypes representing a range of maturities and regional adaptations were studied in Mississippi in 1991 and 1992. All plots received a preplan! application of 112 kg ha−1 N and half the plots also received a 38 kg ha−1 sidedress application of N. Within each N treatment, half the plots received 112 kg ha−1 K surface applied and preplant incorporated, with the remaining plots receiving 0 kg ha−1 K. Averaged across years and N treatments, the K deficiency associated with the 0 K treatment reduced lint yield (9%), boll mass (7%), lint percentage (1%), and seed mass (4%). Varying the N fertilization did not benefit these parameters. The high N treatment reduced lint yield 3% (P = 0.07) and lint percentage 1% (P = 0.06) when coupled with the 0 K treatment. All genotypes suffered yield reductions caused by the K deficiency, except for ‘HS 26’, which was not adapted for production in the Mississippi Delta. Potassium deficiency produced reductions in fiber elongation (3%), 50% span length (1%), uniformity ratio (1%), micronaire (10%), fiber maturity (5%), and perimeter (1%) in all genotypes. Nitrogen application above the 112 kg ha− rate did not increase the lint yield under the growing conditions of our study. The data indicate that genotype was not of importance when dealing with K fertility. The deficiency itself, however, must be dealt with to avoid significant reductions in the yield and quality of fiber produced.

Field experiments were conducted for four years (1981 to 1984) at Lexington, KY, USA (38° N latitude) to determine the causes of reduced yields associated with delayed plantings of soybean (Glycine max [L.] Merrill). Five cultivars were planted in mid‐May and early July in row spacings ranging from 18 to 89 cm and the plots were irrigated to minimize moisture stress. Yields were significantly reduced by delayed planting because of reductions in the number of seed m−2 and mass seed−1. The yield reduction under irrigation indicates that moisture stress is not a major cause of reduced yields from delayed plantings. The reductions in seed number were associated with lower insolation interception during flowering and pod set and with smaller vegetative mass (g m−2) at growth stage R5 (beginning seed fill). The data suggest that a vegetative mass at R5 greater than that required to intercept 90 % or more of incident insolation is required to maximize seed number. The length of the seed filling period was not affected by planting date. Mass seed−1 was positively related to mean air temperature during seed filling, which suggests that the smaller seed from delayed plantings are a result of the lower temperatures associated with the later maturity.

The abortion of reproductive plant parts is an important, but not well understood, part of the yield production process in many crop plants. The objective of this research was to characterize the effects of growth stage and timing on the abortion process of soybean [Glycine max (L.) Merr.] flowers exhibiting variation in abortion. Soybean plants were grown in a greenhouse (cv. McCall) and for 3 yrs in the field (cv. Kent). The soil types were a Donerail silt loam (Typic Arguidofl), Maury silt loam (Typic Paleudalf), and a Lanton silt loam (Cumulic Haplaquolls). Fully‐opened flowers located on nodes from the middle of the main stem were marked during early (R1) and late (R3) flowering and then observed for 2 or 3 weeks to determine the extent and timing of flower abortion. Defoliation and three different depodding treatments were used to alter abortion. Nineteen percent of the early flowers aborted (failed to produce seed bearing pod) in the greenhouse, but 31 to 48% aborted over yrs in the field. Abortion of late flowers varied from 76 to 92% in two greenhouse experiments and from 69 to 90% in the three field experiments. The abortion process occurred more rapidly in the greenhouse where much of the abortion was apparent by 6 days after flower opening (DAFO), compared with the field where less than 50% of the abortion was apparent by 10 DAFO. The three depodding treatments decreased and defoliation increased flower abortion. Removing all pods from a plant except those at one node, reduced late flower abortion at that node from 69 to 54%. However, removing all pods from only one node of the plant reduced abortion at that node from 69 to 36%. The results of these experiments suggest that the processes controlling abortion operate at the individual node level and that the whole plant source‐sink ratio is of secondary importance.