Seed Production in Radiata Pine Clones on Four Different Sites

Seed production was studied in cones from 18 clones of radiata pine (Pinus radiata D. Don) replicated within and between four contrasting sites. At one site which was phosphate-deficient, number of full seeds per cone, along with cone volume #nd mean seed weight, was about one-third less than elsewhere. Empty seeds, although never completely absent, were prevalent only in occasional cones. Such cones were common at the phosphate-deficient site, but were not a consistent feature of particular clones. All clones showed inherently high fertility within cones. There was indirect evidence of a high incidence of ovule abortion: firstly, number of seeds per cone showed much more non-genetic variability than number of scales, and secondly, within-clone correlations between these two characters were not high. Such ovule abortion is believed to resemble conelet drop, which was prevalent within the experiment, in being often not caused by inadequate pollination. Almost without exception the cones contained totals of 40 or more seeds, suggesting that satisfactory cone development is related to a threshold frequency of either ovule abortion or ovule survival. INTRODUCTION It is becoming clear that in radiata pine (Pinus radiata D. Don) much wastage of seed production occurs between conelet receptivity and seed filling. The wastage first becomes evident in the abortion of conelets, customarily known as "conelet drop", which mostly occurs soon after pollination, and may involve over half the conelet crop (Sweet and Thulin, 1969). In cones that do mature many of the ovule sites may not contain seeds, and a varying proportion of the seeds that do develop may be empty. However, there is very little concrete information concerning the incidence of missing and empty seeds in radiata pine. This paper reports a study of the numbers of full and empty seeds together with some other related characters, which was made with a collection of cones from a trial with clones replicated within and between sites. One objective was to obtain circumstantial evidence as to the causes of empty and missing seeds in radiata pine. Another paper (Burdon and Low, 1973) covers cone characters and mean seed weight in the same experiment. N.Z. JI For . Sci. 3 (2): 211-219 212 New Zealand Journal of Forestry Science Vol. 3 MATERIAL AND METHODS