Plant density-dependent variation in density, frequency, and size of watermelon fruits

Quantitative analyses of the continuous response of components of the bioma ss of fruits of watermelon [Citrullus lanatus (Thunb,) Matsum & Nakai] to v ariation in plant density could provide insight into the mechanisms underly ing effects of plant density on marketable geld. Per unit area, the linear response of fruit biomass to plant density recently has been shown to expla in the linear response of marketable yield. In the current study we quantif y plant density-dependent variation in the size, density (no, per unit area ), and frequency (no. per plant) of watermelon fruits. In single-row plots, at least 3.7 m apart, plant density varied from 0.4 to 4.1 plants m(2) (10 00-9000 plants ha(-1)). In each experiment, the linear effect of plant dens ity explained more than 80% of the variation in fruit density. Fruit densit y increased at linear rates of 0.6 to 1.1 thousand fruits ha(-1) per thousa nd plants ha(-1). The plant density-dependent response of the size of fruit s varied considerably among experiments but the frequency of fruits respond ed consistently. In four experiments, there was no evidence of an effect of plant density on fruit size but in three experiments, fruit size decreased at a curvilinear rate of approximately 2.0 (kg(-1) fruit(-1) per thousand plants(-1) ha(-1))(plants ha(-1))(2). Frequency of fruits decreased with pl ant density at curvilinear rates of 0.8 to 2.8 (fruits plant(-1) per thousa nd plants ha(-1))(plants(-1) ha(-1))(2). The response of size of fruits and frequency of fruits, respectively, probably measured an environment-depend ent and an environment independent effect of intraspecific competition.