Abstract:
The rate of cross-pollination in Coffea arabica has been demonstrated to be variable according to the marker used. The following mutants have been tested in Campinas: cera (ce) 7.0 - 9.0% purpurascens (pr) 2.6 - 12.3%, cera-purpurascens (ce, pr) 8-9 - 20.0%, polyorthotropica (po) 15.0% and xanthocarpa (xc) 12.4%. The trial here reported was accomplished in order to study the outcrossing rate among coffee trees planted in the same hole, as coffee is commercially grown in São Paulo. Five treatments were included in a randomized block design. The treatments differed in the proportion of individuals producing yellow or green endosperm per hole: A - four individual cece (yellow endosperm); B - three cece and one cece (green endosperm); C - two cece and two CeCe; D - one cece and three CeCe. A fifth treatment with a single cece plant per hole was included in order to compare the results with those previously obtained. The rate of outcrossing was found to increase linearly with the increase of Cece individual in the four plant hole. The treatment D showed an average of 16.0% of cross pollination. The reaults for holes with a single cece plant confirmed those obmed by other authors (9.1%). Differences were found in the proportions of the various seed types (peaberry, elephant and flat beans) in the green seeds resulting from crossfertilization in comparison with the same seed types of the cera type. The results obtained indicate that a coffee population derived from a plantation with four plant per hole should have a higher genetic variability than previously thought. This information would explain, in part, the success obtained in the breeding program by the pedigree method.
