Variation in growth within and among families of Douglas-fir through a single season

Abstract

In a study examining the multivariate nature of main stem and branch growth in Pseudotsuga menziesii (Mirb.) Franco, major generalizations from previous studies of ovule development were corroborated. The characteristic increase of individual variable values through time was associated with (i) increases in variation as expressed after both uni- and multi-variate analyses and (ii) changes in variable intercorrelations (= correlation structure). We interpret variation as estimating the number of different states available to a developing system and covariation among traits as a mathematical measure of the organization or integration of the measured plant parts. Together, increases in variation and changes in correlation structure with development reflect increasing complexity in biological systems concomitant with the increase in information expressed in such a system. This change in complexity has been related to the second law of thermodynamics in open (nonequilibrium) systems through recent theoretical developments. Furthermore, this study assesses the effectiveness of known genetic structure in accounting for variation between groups of individuals. Our results attribute the majority of the variation to differences within, rather than among, full-sib families. As well, these families differ in both variable variance and covariance. These results are not consistent with common assumptions in quantitative genetics.