Em fins de 1949 foi plantado na Estação Experimental de Monte Alegre do Sul um ensaio de variedades comerciais de café, cujos resultados, abrangendo seis anos de produção, são analisados e discutidos neste trabalho. A Estação Experimental de Monte Alegre do Sul situa-se em solo do tipo massapê-salmourão e apresenta a topogrofía acidentado que caracteriza extensa região cafeeira dos Estados de São Paulo e de Minas Gerais. A finalidade do ensaio é a de verificar o comportamento de diversas variedades, em comparação ao café Nacional, que ainda hoje é a variedade predominante nas fazendas que conservaram parte dos seus antigos cafèzais. As seguintes variedades acham-se representadas no experimento: Nacional, Amarelo de Botucatu, Bourbon Vermelho, Bourbon Amarelo, Mundo Novo, Caturra Vermelho, Caturra Amarelo, Laurina, Moragogipe A. D. e Semperflorens. Com exclusão do café Nacional, representado pela descendência de plantas típicos encontrados na própria Estação Experimental, as demais voriedades são formadas pela mistura de progênies de cafeeiros selecionados pela Seção de Genética do Instituto Agronômico, em Campinas. Como delineamento experimental adotou-se o de blocos incompletos com seis repetições. Ao fim de seis anos de colheitas a análise estatística revelou que a variedade Mundo Novo produziu significativamente mais do que as nove restantes. Em segundo plano classificou-se o Bourbon Amarelo, cuja produção, porém, não difere de modo significativo das que se verificaram para o Bourbon Vermelho, Caturra Amarelo e Caturra Vermelho. As quatro variedades com as maiores produções totais, em kg de frutos maduros (Mundo Novo, Bourbon Amarelo, Bourbon Vermelho e Caturra Amarelo), produziram significativamente mais do que as variedades antigas, como o Nacional e Amarelo de Botucatu, e mais tambem do que os variedades de menor interêsse comercial, como Maragagipe A. D., Laurina e Semperflorens. Os resultados são semelhantes quando se analisam os produções transformadas em café beneficiado.
In 1949 on experiment was started in the Northwestern region of S. Paulo to compare the behovior in this area of the best new stra.ns of coffee that hod been selected previously in Campinas. This trial was made with the following commercial varieties: Nacional, Amarelo de Botucatu, Bourbon Vermelho. Bourbon Amarelo, Mundo Novo, Caturra Vermelho, Caturra Amarelo, Laurina, Maragogipe A. D., and Semperflorens. An incomplete block design was used with fifteen blocks of four treatments. A single plot consisted of ten coffee holes, with three seedlings per hole. Protection against erosion was built around each hole. The first commercial yield was obtained in 1952. The data on the total yields, expressed in kilograms of ripe fruit, in the years 1952 to 1956, are presented in table 1. The analysis revealed that the Mundo Novo coffee yielded more than all other varieties. The yields of Bourbon Amarelo, Bourbon Vermelho, and Calurra Vermelho did not differ significantly. The varieties Nacional and Amarelo de Botucatu produced less than the Mundo Novo, Bourbon Amarelo, Bourbon Vermelho and Caturra Amarelo (table 2). The data presented in table I also indicate that the yield of the varieties with yellow fruit is greater than the corresponding ones with red fruit, as follows; the Bourbon Amarelo produced more fruits than the Bourbon Vermelho, Caturra Amarelo more than Caturra Vermelho, and Amarelo de Botucatu more than the Nacional variety. The difference in yield was not significant and may be due to a pleiotropic effect of the xanthocarp allele. An analysis was also undertaken with the yield expressed in kg of clean coffee and the observed results are almost of the same order, when compared to the yield expressed in kg of ripe fruits (table 3). The relation of the weight of ripe fruits to clean coffee was determined every year from samples taken from the total yield of every plot. Only the Laurina variety and Mundo Novo coffee gave poor outturn (table 4). The Mundo Novo strains used in this trial had not yet been selected for good outturn, as the present strains. It was observed that the percentage of flat, normal seeds, was lower in the Caturra Amarelo and Caturra Vermelho coffees, whereas the percentage of peaberry seeds (fruits with one developed seed) was higher in the Amarelo de Botucatu and Caturra Amarelo. The percentage of "shell" seeds (elephant beans) was higher in the Caturra Amarelo and Mundo Novo. It was also observed a high incidence of "shell" seeds for all varieties, in 1954 and, of peaberry seeds, in 1956, which seems not to be associated to the total yield in these years. The Muragogipe A. D. and Mundo Novo varieties produced the largest and the heaviest seeds (table 4), while the Maragogipe A. D. seeds presented the lowest density in comparison to other varieties. The plant height end the diameter of the trees taken at the bottom were determined in 1956 for all plants (table 4). The Mundo Novo revealed to be the tallest variety while the Caturra Vermelho was the shortest one. A correlation was noticed between the height of the tree and its diameter in the varieties; this correlation, however, did not hold true for the Semperflorens and Caturra varieties. The results obtained in this experiment clearly indicate that the Mundo Novo is a high yielding variety in the area and should, therefore, be recommended for the new plantings. The Caturra Amarelo could also be recommended for this mountain area because it gives gocd yields and its low height makes harvesting easier.
