Seed production in Caucasian clover (Trifolium ambiguum Bieb.) cv. Monaro : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in the Institute of Natural Resources of Massey University, Palmerston North, New Zealand
A number of factors which may affect seed production in Caucasian clover (Trifolium ambiguum Bieb.) cv. Monaro were investigated viz. root system development and its relationship with seed yield; morphological and reproductive characteristics of different genotypes from within the cultivar; the effects of plant density on seed yield; and, the effects of insect pests on seed yield and quality. Caucasian clover is a rhizomatous legume pasture plant with a large root system. A sand bed was used to investigate root system development and the relationship between the root system and seed yield. During vegetative growth (September to November) root dry matter (DM) accumulated more rapidly than shoot DM, and the root to Shoot ratio was 2.45 by mid November. However as reproductive growth accelerated the rate of root DM accumulation decreased and by early February the root to shoot ratio was 0.65. Plants developed primary, secondary and tertiary crowns in this first growing season, with both primary and secondary crowns producing reproductive shoots. The protracted flowering period (four months) was a result of an initial production of reproductive shoots from primary crowns, and an extended production of reproductive shoots from secondary crowns when primary crown shoot production had ceased. There were therefore two flowering peaks. A single primary crown produced an average seed yield of 3.8 g compared with 0.81 g seed from an average secondary crown. However as there was only one primary crown but 26 secondary crowns per plant, the total seed yield from secondary crowns reached 21.1 g per plant. Root DM was highly correlated (R2 = 0.80-0.97) with the number of leaves, reproductive shoots and inflorescences, as well as the number of secondary crowns. Seed yield per plant was almost entirely dominated by the number and size of the secondary crowns, which in turn depended on the size of the root system prior to reproductive development. The seed production potential of three genotypes selected from within cv. Monaro was investigated. The genotypes differed significantly in their morphological and reproductive characteristics, including the number of reproductive stems and inflorescences produced. Genotypes 2 and 12 had seed yields of 3.65 g and 2.99 g per plant respectively, which genotype 9 had a seed yield of only 0.18 g per plant, primarily because it produced very few inflorescences. Poor inflorescence production by some genotypes within a cultivar will limit seed yield and may alter the genotypic composition of the cultivar as seed multiplication progresses. A radial trial was used to determine the optimal density for seed production of the cultivar. Plant density significantly affected leaf number, reproductive stems, inflorescences number and seed yield per plant in that plants at the highest density (38.2 m-2) produced about 7 times fewer reproductive stems and over 15 times fewer inflorescences per plant compared with those in the lowest density (3.1 m-2). Inflorescence number increased as plant density decreased (R2 = 0.82-0.89). The lowest density plants had the greatest reproductive growth per plant but were not capable of creating the highest reproductive production per unit area. Cultivar Monaro showed a parabolic relationship for reproductive production per unit area with plant density, both for inflorescence number and seed yield m-2. Plants grown at the density of 10.8 plants m-2 produced the highest inflorescence number and seed yield per unit area. The number of inflorescences per plant was the most important seed yield component determining final seed yield in cv. Monaro (R2 = 0.95). Plant density affected seed yield only through reducing inflorescence number; it had no effect on thousand seed weight, or germination and hard seed percentage. It is probable that plant density affects reproductive development through its effects on root system development in Caucasian clover. A semi-radial trial was also used to examine the effects of plant density on seed production of twelve genotypes from within cv. Monaro. These genotypes demonstrated significant differences in reproductive capability. Genotypes 1, 2, 4 and 12 represented a high or normal reproductive capability group compared with others and these genotypes produced inflorescences across all five densities. Genotypes 3, 5, 8, 9, 13 represented another group which either failed to produce or produced only a small number of inflorescences even at the lowest density. Genotypes 6, 7 and 10 were intermediate. Seed yield per plant at the 10.8 plants m-2 density ranged from 0 g for genotypes 5, 8, 9 and 13 to 0.89 g for genotype 1. Reproductive response to plant density did not differ from that when seeds of the cultivar were sown in the full radial trial. Insect pests, particularly thrips (Thripidae) may reduce seed yield and quality. The insecticide taufluvalinate (Mavrik) was applied at 150 ml ha-1 either once (just prior to peak flowering) or every 14 days until harvest, beginning at first inflorescence appearance. Both adults and larvae of onion thrips (Thrips tabaci) and red clover thrips (Haplothrips niger) were detected in inflorescences during peak flowering and seed development. The multiple insecticide applications significantly increased seed number per inflorescence and thousand seed weight, therefore increasing seed yield. However the single application had no effect on seed yield or quality.