Melhoramento do cafeeiro. XI - Análise da produção de progênies e híbridos de bourbon vermelho
Data
1957-11
Autores
Título da Revista
ISSN da Revista
Título de Volume
Editor
Instituto Agronômico (IAC)
Resumo
Um conjunto de 24 progênies, derivadas de plantas matrizes selecionadas da variedade Bourbon Vermelho, e outro de 19 grupos de híbridos F1 entre essas plantas matrizes, cada um dêles com 20 plantas, tiveram a produção individual controlada durante os seis primeiros anos. Após esse período, cinco melhores progénies e cinco híbridos F1 num total de 200 plantas, continuaram a ser colhidos individualmente por mais de oito anos consecutivos. De outras progênies e híbridos F1, 139 indivíduos também tiveram a produção controlada por 14 anos. Com os dados obtidos realizou-se um estudo a fim de verificar se a eliminação precoce de progénies menos produtivas, logo após o 2.° ano de produção, redundaria em eliminação de progênies que, após seis anos de produção, passassem a pertencer ao grupo mais produtivo. Como resultado verificou-se que hão haveria desvantagem em se realizar essa seleção precoce, pois as progênies que aos dois anos de produção total se mostraram menos produtivas assim se conservaram até aos seis anos, com raras exceções. A seleção das progênies mais produtivas, feita com base em apenas duas primeiras produções, é menos eficiente. A análise, dos dois e dos seis anos de produção total individual dos 856 cafeeiros tomados como uma população única, mostra que há uma acentuada tendência de os indivíduos mais produtivos aos seis anos pertencerem a progênies e híbridos melhor classificados logo após os dois primeiros anos de produção. Também quanto a êste particular não haveria inconveniência em se eliminarem as piores progênies precocemente. Os dados relativos às 339 plantas, selecionadas após seis anos e com produção controlada por 14 anos, foram analisados quanto aos efeitos que teria a eliminação precoce das piores progênies logo depois do primeiro biênio. Verificou-se que tal efeito seria mínimo, pois na verdade, se uma tal seleção precoce tivesse sido realizada, o número de boas plantas perdidas seria bem reduzido. Dentre êstes cafeeiros 200 pertencem a cinco progênies e a cinco híbridos. As médias de suas produções acumuladas por biênios mostram que entre eles se nota tendência de os melhores conjuntos também se revelarem cedo, pois a sua classificação aos dois anos se altera pouco quando comparada com a classificação após 14 anos consecutivos de produção. A comparação entre as produções totais individuais de 14 anos dessas 200 plantas e as comparações com as médias de seis anos apresentadas por êsses conjuntos, mostram também que maior número de melhores plantas aos 14 anos pertence a progénies melhor classificadas aos seis anos. Ainda neste conjunto de plantas comparou-se a distribuição das produções totais individuais aos seis e aos 14 anos, verificando-se que as plantas que aos 14 anos se situam acima da média correspondem a cafeeiros que, em sua maioria, já se encontravam bem classificados aos seis anos. Esta análise indica que é boa a probabilidade de se efetuar seleção individual das melhores plantas após os seis primeiros anos de produções consecutivas. Como medida de segurança aconselha-se, todavia, que a seleção das progênies seja efetuada com base em seis anos de produção, e que a seleção dos cafeeiros individuais seja feita com base em 12 anos de produção, quando o cafeeiro atinge completo desenvolvimento.
During the initial years of a breeding project carried out in Campinas, coffee progenies and hybrids were usually planted in rows of twenty plants each. Progenies were not replicated in the same location but usually they were studied at three to five experiment stations. Although this method did not allow for statistical analysis of individual yields and other plant and seed characteristics, it proved to be sufficiently accurate to select several outstanding progenies and exceptional individual coffee plants. Individual yields of 856 plants belonging to forty-three of such groups of twenty plants each were studied in order to evaluate the effect of an early elimination of progenies which started yielding below the general mean. The 43 groups were derived from 25 Bourbon plants, 24 groups being selfed progenies and 19 F1 hybrids between a few of those trees. In a previous paper it was shown that these progenies and hybrids did not differ in yielding ability, thus allowing treatment as a single population of Bourbon plants. Individual harvests of each plant from this population were recorded during the first six years of yields (1939-44). A preliminary selection was done in 1944. The five progenies and five hybrids which had the highest mean yields were selected as groups of twenty plants and continued under observation for an additional period of eight years. Besides these 200 coffee trees, 139 others were also selected for their high individual yields during the first six years of crop, and their records were also kept for another period of eight years, ending in 1952. The remaining trees were not recorded after the preliminary selection. Average yields per plant were calculated for the 43 groups, accumulating the crops of the first 2, 4 and 6 years, and distributed, for each period, in classes centered at the general mean plus or minus integer multiples of the corresponding standard error. In the distribution for the first two years of crop (1939-40), 10 groups were in class x-1s, two in x-2s and one group in x-3s. After four additional harvests seven groups remained in the lowest classes and four reached the mean class. This relative stability in ranking after successive years of crops suggests that little was gained by keeping records for the whoie group of 43 pro genies and hybrids. Had the lowest group been discarded after the first two crops, only two lines would have been lost for the preliminary selection in the sixth crop. When comparison is made between the distribution of progenies and hybrids according to their mean yields after the first two crops, and the array of six-years total individual yields of the 856 plants, it may be seen that the most productive plants belong, as a rule, to progenies and hybrids which occupy the upper classes in the distribution of the average yields of the 43 groups. This is an indication that early elimination of the lowest yielding groups would not affect selection among the best individual coffee trees. The comparison made between the distribution of the total individual yields of 339 plants recorded to the 14th consecutive crop, and the array of the average yields for the first two crops of the progenies and hybrids to which those plants belong, show that elimination of 13 groups with the lowest yields in the period of 1939-40 would sacrifice 89 out of 339 coffee trees selected in 1944. Only 20 of these 89 plants, however, yielded above the average of the selected population of 339 trees. Considering a more restricted group of plants, namely, the 200 individual trees that belong to the five progenies and five hybrids selected in 1944 for their high average yield in the first six years of crops, it may be observed that among the 40 plants of the progenies with the highest average yields, 22 ranked above the mean of the 200 plants, after 14 years of production. Among the 40 plants of the two groups with the lowest yields, however, only eight could reach a position above the mean. These informations indicate that in trials involving large number of progenies it is possible to restrict observation on the most promising groups, discarding those which start yielding far below the mean. This elimination of low yielding progenies could be made after two or four consecutive yields, even if it considered that definite selection of individual coffee plants must wait until twelve or more years have elapsed, when the trees attain full development. When coffee progeny trials are planted in randomized blocks or in lattices such an elimination would not introduce difficulties in future statistical analysis. It is pointed out that similar procedures could be adopted in selection programs of other perennial crops.
During the initial years of a breeding project carried out in Campinas, coffee progenies and hybrids were usually planted in rows of twenty plants each. Progenies were not replicated in the same location but usually they were studied at three to five experiment stations. Although this method did not allow for statistical analysis of individual yields and other plant and seed characteristics, it proved to be sufficiently accurate to select several outstanding progenies and exceptional individual coffee plants. Individual yields of 856 plants belonging to forty-three of such groups of twenty plants each were studied in order to evaluate the effect of an early elimination of progenies which started yielding below the general mean. The 43 groups were derived from 25 Bourbon plants, 24 groups being selfed progenies and 19 F1 hybrids between a few of those trees. In a previous paper it was shown that these progenies and hybrids did not differ in yielding ability, thus allowing treatment as a single population of Bourbon plants. Individual harvests of each plant from this population were recorded during the first six years of yields (1939-44). A preliminary selection was done in 1944. The five progenies and five hybrids which had the highest mean yields were selected as groups of twenty plants and continued under observation for an additional period of eight years. Besides these 200 coffee trees, 139 others were also selected for their high individual yields during the first six years of crop, and their records were also kept for another period of eight years, ending in 1952. The remaining trees were not recorded after the preliminary selection. Average yields per plant were calculated for the 43 groups, accumulating the crops of the first 2, 4 and 6 years, and distributed, for each period, in classes centered at the general mean plus or minus integer multiples of the corresponding standard error. In the distribution for the first two years of crop (1939-40), 10 groups were in class x-1s, two in x-2s and one group in x-3s. After four additional harvests seven groups remained in the lowest classes and four reached the mean class. This relative stability in ranking after successive years of crops suggests that little was gained by keeping records for the whoie group of 43 pro genies and hybrids. Had the lowest group been discarded after the first two crops, only two lines would have been lost for the preliminary selection in the sixth crop. When comparison is made between the distribution of progenies and hybrids according to their mean yields after the first two crops, and the array of six-years total individual yields of the 856 plants, it may be seen that the most productive plants belong, as a rule, to progenies and hybrids which occupy the upper classes in the distribution of the average yields of the 43 groups. This is an indication that early elimination of the lowest yielding groups would not affect selection among the best individual coffee trees. The comparison made between the distribution of the total individual yields of 339 plants recorded to the 14th consecutive crop, and the array of the average yields for the first two crops of the progenies and hybrids to which those plants belong, show that elimination of 13 groups with the lowest yields in the period of 1939-40 would sacrifice 89 out of 339 coffee trees selected in 1944. Only 20 of these 89 plants, however, yielded above the average of the selected population of 339 trees. Considering a more restricted group of plants, namely, the 200 individual trees that belong to the five progenies and five hybrids selected in 1944 for their high average yield in the first six years of crops, it may be observed that among the 40 plants of the progenies with the highest average yields, 22 ranked above the mean of the 200 plants, after 14 years of production. Among the 40 plants of the two groups with the lowest yields, however, only eight could reach a position above the mean. These informations indicate that in trials involving large number of progenies it is possible to restrict observation on the most promising groups, discarding those which start yielding far below the mean. This elimination of low yielding progenies could be made after two or four consecutive yields, even if it considered that definite selection of individual coffee plants must wait until twelve or more years have elapsed, when the trees attain full development. When coffee progeny trials are planted in randomized blocks or in lattices such an elimination would not introduce difficulties in future statistical analysis. It is pointed out that similar procedures could be adopted in selection programs of other perennial crops.
Descrição
Palavras-chave
Citação
ANTUNES FILHO, H.; CARVALHO, A. Melhoramento do cafeeiro. XI - Análise da produção de progênies e híbridos de bourbon vermelho. Bragantia, Campinas, v. 16, n. 13, p. 175-195,1957.