Seed Number and an Asymmetry Index of 'McIntosh' Apples

In ‘McIntosh’ apples (Malus domestica Borkh.), the number of well-formed seeds and the number of fertilized ovules (well-formed plus aborted seeds) were correlated to yield an index of asymmetry for individual fruit. In spite of different flowering and pollination conditions in 1990 and 1991, the index of asymmetry decreased curvilinearly with increases in the number of well-formed seeds and of fertilized ovules in both years. Québec apples are classified into two categories: Extra-fancy and Fancy. Assignment of apples to these categories affects their commercial value. Apples exhibiting pronounced shape defects are not classified and most often destined for the juice market. Apple growers thus receive only a quarter of the price they would get for Fancy fruit. In this context, an assessment of certain factors influencing the shape and, consequently, the commercial quality of apples is relevant. Apple flowers are hermaphroditic, but in most cultivars they are strongly self-incompatible (Free, 1960). Good harvests depend on cross-pollination by insects. Flowers of most apple cultivars contain 10 ovules. Thus, 10 pollen grains are necessary for the formation of 10 seeds. In ‘McIntosh’, pollination with highly compatible cultivars results in 2.6% to 14.6% of the fruit being lopsided, although there were always more than two seeds per apple (Latimer, 1931, 1933, 1937). Roberts (1946) found that irregularly shaped ‘Wealthy’ and ‘McIntosh’ apples were most frequent when apples contained fewer than two and three seeds, respectively. This irregularity has been attributed to inadequate pollination (Way, 1978). Latimer (1931) and Brittain (1933) noted that the distribution of seeds among the carpels also influenced the shape of apples and that the percentage of lopsided apples decreased with an increase in the number of seed-containing carpels. Heinicke (1917) reported that tissue development was reduced near empty carpels. The fact that fertilized ovules provide an important source of growth hormones, such as auxins (Luckwill, 1948, 1953, 1957), suggests a link between the distribution and the number of seeds among carpels, and the shape of the apple. 44 Received for publication 2 May 1994. Accepted for publication 12 Oct. 1994. We express our appreciation to the apple growers, Réjean and Céline Guertin, for access to their orchard; to Michèle Roy of Ministry of Agriculture and Fisheries of Québec for her collaboration; and to Michel Junger and David Hilbert for revision of this article. The cost of publishing this paper was defrayed in part by the payment of page charges. Under postal regulations, this paper therefore must be hereby marked advertisement solely to indicate this fact. The objective of this study was to determine the influence of seed set on the radial symmetry of ‘McIntosh’ apples. Materials and Methods The apples used in this 2-year study (1990 and 1991) were sampled from trees of ‘McIntosh’/M.26 planted in 1980 at St-Paul d’Abbotsford, Mont Yamaska, Québec. Ten beehives were introduced in the 10-ha orchard in the first year of the study, and 12 in the second. In 1990, flowering began 13 May and ended 26 May, while in 1991, flowering lasted from 13 to 18 May. Low temperature and high precipitation did not favor pollination in the first year. The percentage of flowers injured by low temperature during Winter 1990 was determined during bloom, by counting the number of open and killed flowers on a 42-cm length of branch on each of 35 randomly chosen trees. The number of apples per tree was counted at harvest (15 Sept. 1990 and 4 Sept. 1991) on the same 12 trees. Random sampling as well as reasoned-choice sampling were used to collect apples in 1990. ReasonedH Fig. 1. Cross section of an apple showing the carpels, choice sampling allowed systematic selection of apples with either very low or very high asymmetry to ensure that both extremes of the distribution were represented. Only random sampling was carried out in the second year of the study. Sample size in 1990 was 146 apples, 96 from random sampling and 50 from reasoned-choice sampling, while 60 apples were used in 1991. Each apple was sliced transversely, and the fertilized ovules were classified into two seed categories: well-formed and aborted. Full, fleshy seeds were defined as well-formed seeds. Seeds were aborted when the endosperm was resorbed, whether or not the tegument was shriveled. For each apple, the radius passing through each of the five vascular bundles between carpels was measured from the corresponding seed attachment site to the cuticle (Fig. 1). An index of asymmetry (IA) of shape of each apple was calculated using the equation IA = [(r max – r min)/r max] × 100, where r max corresponds to the longest and r min to the shortest radius measured for each apple. With the exception of the number of apples per tree, all variables were tested for differences between the two years by analysis of variance. Because residuals were not normally distributed, the Mann–Whitney U test was applied to test for differences between the number of apples per tree in 1990 and 1991 (Scherrer, 1984). For both years of the study, regression analysis was performed to determine the relation between the IA and the number of well-formed seeds and the number of fertilized ovules.