Plant Growth Analysis: a Program for the Fitting of Lengthy Series of Data by the Method of B-splines*

A method is presented for fitting curves to lengthy and/or complicatedseries of data on plant growth. A computer program which derivesthe usual plant growth analytical quantities, and their errors,from these fitted curves is also described and offered for circulation.The fitted curves are splined cubic polynomial exponentials.Examples of their application are given, employing both realand artificial data. In any set of data the number of splines,and the position of the knots joining them, may be determinedeither by objective statistical tests or by the experimenterhimself, who thus retains a considerable degree of control overthe process but can call on the assistance of the program ifrequired. The value of this method is considered in relationto other curve-fitting approaches to plant growth analysis andit is concluded that, provided sufficient primary data are available,the method is free from many of the problems which beset earlierwork in this field, and also provides new possibilities of itsown. growth curves, approximating functions, empirical models, regression analysis, growth analysis