Melhoramento do cafeeiro. VI - Estudo e interpretação, para fins de seleção, de produções individuais na variedade bourbon

Abstract

The present study was undertaken to get information on the following questions of primary importance in coffee breeding : a) the number of years during which individual plants must have their yields recorded, in order to select the best mother trees for progeny tests; b) the number of years of individual progeny plant records, necessary for selection of the best progenies on the basis of total progeny yield ; c) the relation between the yields of mother plants and of their progenies. An attempt to answer the above questions was made through the analysis of data furnished by the Coffee Department of this Institute from a plot of 1107 plants (1 hectare), of Coffea arabica, var. bourbon, planted in 1930 for mother tree selection and individual plant recording. These plants have now 19 consecutive years of individual yield records, and some of them have progenies with 12 to 15 years of yields available for the scope of this investigation. From the study of such a group of mother trees and of their F1 progenies, the following conclusions can be drawn, assuming that the best plants are in fact those which have the highest total yields up to the present time : a) Selection of high yielding coffee mother trees is most efficient when they are already fully grown, on the basis of at least 14 years of production. Selection at early ages is not so effective. b) The efficiency is approximately the same whether the selection is based on total yield after 14 years, or when it is based on the latest biennial yield, such as the 13th plus the 14th, or even when only a single high yield of a late year is considered. This conclusion was drawn from the number of plants which, belonging to the 25 and 100 best after 19 years of yield, were found among the 25 and 100 best plants at different periods of time. From the 100 plants with the highest yield after 19 years, 59 were among the 100 best plants after 12 years; 71 were among the 100 best plants after 14 years, and so on (page 183). The conclusion that efficient mother plant selection in possible with basis on a single year of high yield, though valid only for the plot under investigation, has been corroborated by other results obtained in Campinas, where selection of mother trees was made in adult coffee plantations in years of high yield. c) A similar comparison for progenies, when the whole progeny total is considered at different periods of time, indicates that the best progenies reveal themselves much earlier, that is, after the first 6 years of consecutive yields. It must be pointed out, however, that this conclusion is also based on the assumption that the best progenies are those with highest total yield at the time the analysis was made. d) As in the case of mother plants, efficient individual selection within progenies must be based on a higher number of years of yield, when the plants are completely developed. e) It was found to be difficult to establish the relationship between the yields of mother plants and the total yield of their progenies. However, the available data indicate that, as a rule, exceptionally high yielding mother plants give outstanding progenies. In order to have a better insight on the value of a group of selected mother trees the study of their progenies in replicated trials is advisable. When yield data of several consecutive years are available for individual mother trees, the yield increase index, that is, the coefficient of regression of yield on years, can be useful for comparison among these plants. This index has been calculated for mother plants which have progenies at Campinas. Yield data of the next few years will probably contribute to a better understanding of the relationship between the yields of mother plants and their progenies.