The primary objective of this study was to determine if soybean grain yields could be predicted by the P and K content of the growing plant. A supporting objective was to find what plant parts should be taken, and at what stage of growth, to give the best relationship. Multiple curvilinear regression analysis was used to determine this relationship from data collected in four field experiments.
Yields, chemical composition of soybean plant parts and other data were available from four similar fertilizer experiments conducted at various locations in Iowa in 1958. A randomized block 9 x 9 central composite design, containing various combinations of P and K fertilizer rates, was used in all experiments. Soybean plant samples from each plot, taken in three different growth stages, were separated into various plant parts and chemically analyzed for total P and K contents. (Growth stages used were: Stage 5-Nine to 10 trifoliate leaves unrolled with stem branching evident; full bloom with withered flowers in lower leaf axils. Stage 7-Pods plainly evident in plant tops; lower pods nearly full length with beans developing; flowering ceased. Stage 9-Bottom leaves beginning to yellow; top pods almost fully developed with beans nearing "green bean" stage.)
Miller, R. J.; Pesek, J. T.; Hanway, J. J.; and Dumenil, L. C.
"Soybean yields and plant composition as affected by phosphorus and potassium fertilizers,"
Research Bulletin (Iowa Agriculture and Home Economics Experiment Station): Vol. 35
, Article 1.
Available at: https://lib.dr.iastate.edu/researchbulletin/vol35/iss524/1