Evaluation of a Viability Model for Predicting Soybean Seed Germination during Warehouse Storage
1999
Fabrizius, Eric | TeKrony, Dennis | Egli, D. B. | Rucker, Marcy
A method for predicting soybean [Glycine max (L.) Merr.] seed germination during storage would benefit seed producers and the seed industry. This investigation evaluated the ability of the Ellis and Roberts deterioration model to predict seed germination after warehouse storage and the accuracy of a rapid-aging test to estimate initial seed quality (K₁). Seed lots with low levels of mechanical injury and Phomopsis longicolla (Hobbs) seed infection produced in 1990 and 1991 were stored more than 2 yr in multiwall paper bags in seed warehouses in Kentucky and Indiana. Warehouse temperature was monitored and seed moisture and germination were determined at 3-mo intervals. The rapid-aging test (40°C and 150 g kg⁻¹ seed moisture [fresh weight basis]) was used to estimate K₁ for each seed lot. The model accurately (±10 percentage points) predicted germination most seed lots after one carryover year of storage (1 yr after the first spring planting season, = 430 d), when little deterioration had occurred. After 2 yr of carryover storage (= 800 d), when germination had declined, the model accurately predicted the germination of only six of 16 seed lots. The inaccurate predictions may have been related to variation in K₁ (substantial variation occurred between repeated determinations on the same seed lot) or to variation among seed lots in the rate of deterioration at constant temperature and seed moisture. These results suggest that some of the assumptions underlying the Ellis-Roberts seed deterioration model may be invalid and require further evaluation.
显示更多 [+] 显示较少 [-]