The sampling on the rubber tree nursery experimentation

Abstract

The present work was carried out with the objective of estimating the number of seedlings for sampling in rubber tree (Hevea spp) nursery experiments, originated from jungle seeds, so as to keep the variance to a minimum. The study was based on data from a density trial with twelve spacings, utilizing only the results of the best six treatments. The ideal sample size is that which estimates fie minimum mean variance. The minimum variance of the variables studied was obtained when four planting rows were taken as a sampling unit, expressing fie least coefficient of variation. The mean variance of fie sample practically stabilized when 'n' > twelve plants were distributed throughout the sampling unit. The coefficients of variation obtained were 3.92% for plant height, and 4.07% for stem diameter, for samples of competitives plants. The confidence interval estimated gives an idea of the experimental accuracy.