Effects of plantation density and spacing on competitive interactions among half-sib families of maritime pine

Abstract

The effects of competition among eight half-sib families of the Landes geographical race of maritime pine (Pinuspinaster Ait.) were studied in a field trial involving four spacings with two replications, in southwestern France. Families were picked from the open-pollinated progeny in a clonal seed orchard. Interfamily competition was controlled in 30-tree plots, with two families planted alternately in a checkerboard pattern. Each family was represented in one pure and two mixed plots, according to a balanced, incomplete, diallel design. Static competition effects from 4 to 16 years were assessed using (i) a diallel ANOVA model and (ii) a factorial ANOVA model. In the first model, genetic variance was split into a between-plot (kind of mixture) and a within-plot component. In the second model, components due to the direct effect (measured family) and the associate effect (competitor family) were separated. Genetic variance was much more significant at high density (2500 trees/ha) than at low density (1250 trees/ha). At high density, self-thinning occurred at age 16. With square spacing, a significant part of genetic variance, both between and within plots, was attributable to principal effects of families, direct as well as associate. With rectangular (1:4 ratio) spacing, genetic variance was chiefly expressed as family interaction at the plot level and reached a maximum at a more advanced age. Mixed plots averaged significantly better than pure plots for both height and girth in the rectangular spacing modality. Favourable root interaction between families is suggested as the reason for this effect.