No abstract

No abstract

Resumo -O objetivo deste trabalho foi determinar o número de repetições, de plantas por parcela e de medições necessário para avaliação de caracteres de crescimento em testes clonais de eucalipto (Eucalyptus spp.). Foram avaliados dados de altura, diâmetro à altura do peito e volume de madeira, em três testes clonais, realizados em delineamento de blocos ao acaso, com 30 (testes 1 e 2) e 10 clones (teste 3), com seis repetições de seis (teste 1) e de 10 (testes 2 e 3) plantas. As avaliações foram realizadas dos 14 aos 74 meses de idade. Foram estimados os coeficientes de repetibilidade e de determinação, bem como os efeitos do acréscimo na frequência das medições sobre a eficiência seletiva. O volume de madeira é a característica com maior variabilidade na parcela e menor repetibilidade; portanto, exige maior número de plantas para que se obtenha elevada exatidão na estimativa do valor real do indivíduo. Em testes clonais de híbridos de Eucalyptus grandis x E. urophylla, recomenda-se o uso de três repetições, com seis plantas por parcela, e de três avaliações, em idades superiores a 2 anos, para que se alcance acurácia seletiva em torno de 90%.Termos para indexação: Eucalyptus, correlação intraclasse, eficiência seletiva, híbrido Urograndis, modelos mistos, repetibilidade. Number of replicates, of plants per plot and of evaluations for eucalyptus clonal testsAbstract -The objective of this work was to determine the necessary number of replicates, of plants per plot, and of measurements for evaluating growth traits in eucalyptus (Eucalyptus spp.) clonal tests. Data on height, diameter at breast height, and wood volume were evaluated in three clonal tests, performed in a randomized complete block design with 30 (tests 1 and 2) and 10 clones (test 3), with six replicates of six (test 1) and 10 (tests 2 and 3) plants. The evaluations were performed from 14 to 74 months of age. The coefficients of repeatability and determination were estimated, as well as the effects of increasing measurement frequency on selective efficiency. Wood volume is the trait with the greatest variability in the plot and the lowest repeatability; therefore, it requires more plants to obtain high accuracy in the estimate of the individual's true value. In Eucalyptus grandis x E. urophylla hybrid clonal tests, the use of three replicates of six plants per plot and of three evaluations, at ages above 2 years, is recommended in order to achieve a selective accuracy around 90%.

No abstract

The survival, initial productivity and the genetic parameters of <i>C. citriodora</i> subsp. <i>citriodora</i> (CCC), <i>C. citriodora</i> subsp. <i>variegata</i> (CCV) and <i>C. torelliana</i> (CT) were used to develop a breeding strategy for the <i>Corymbia</i> species. Survival, height and diameter at breast height (DBH) data were assessed 24 and 36 months after planting and the mean annual volume increment estimated in three trials. Longitudinal DBH data analysis was applied to each trial individually, to identify the best and the poorest families at both ages. The mortality ranged from 5% in CT to 27% in CCC and the mean annual increment varied from 17.8 to 20 m³ ha-¹ year^-1 at 36 months after planting. The 36-month narrow-sense heritability (h_a²) was high for CCV (0.69±0.17), moderate for CCC (0.41±0.11) and low for CT (0.21±0.09). The genetic parameters indicated the need for different breeding strategies, for each species. Selecting the best families while roguing the poor families allowed forward selection of CCC and CCV. It was possible to select good CT trees for hybrid breeding, however, improving the species population requires focusing on increasing the effective size and expanding the genetic variability in the CT population.