In nurseries, it is important to determine the number of plants to be evaluated so that inferences about seedling growth are reliable. The aim of this research was to estimate the sample size (number of plants) necessary to evaluate plant height, collar diameter and relation plant height/collar diameter of canjerana seedlings (Cabralea canjerana (Vell.) Mart.) and to verify its variability among tubes, substrates and ages evaluation. An experiment was conducted in the nursery, in which nine treatments, organized in randomizes blocks design with four replication, were formed by the combination of three sizes of tubes (100, 180 and 280cm³) and three substrates (100% peat, 80% peat + 20% rice hulls and 60% peat + 40% rice hulls). It was evaluated the following characters, in ten ages: plant height, collar diameter and relation plant height/collar diameter, in ten plants sampled at random in each experimental unit. Analyses of variance, means comparison tests and verification of normality were carried and the sample size was calculated. The sample size to evaluate the plant height, collar diameter and relation plant height/collar diameter of canjerana seedlings is dependent on age evaluation. For the tubes, substrates and ages evaluation studied, 18 plants per experimental unit (pot) are enough to predict the these characters average, with an estimation error equal to 10% of estimated average, with a degree confidence of 95%.
Canjerana; experimental planning; sampling; experimental precision
