A new method to predict seed yield of moist-soil plants

Multiple linear regression can be used to predict seed yield of moist-soil plants; however, measurement of multiple predictor variables is tedious, su bject to variation, and these models can exhibit multicollinearity. Thus, w e tested if simple linear regression models could predict seed yield of 5 s pecies of moist-soil plants as precisely as multiple linear regression mode ls. The single predictor variable was number of dots on a grid covered by s eed. Simple regression models explained as much variation in seed mass (R-a dj(2) = 0.92-0.97) and predicted (R-pred(2) = 0.91-0.96) as well as or bett er than multiple regression models. Precision of models was attributed to t he strong positive linear relation between the dependent variable and the p redictor, accurate dot counting, and lack of multicollinearity. Dot countin g also was easier and more efficient than measuring multiple phytomorpholog ical variables. This new method is useful for researchers and managers esti mating seed yield of moist-soil plants; however, additional models should b e developed for other plant species, and the method should be tested in oth er regions.