Breeding Behaviour of Cenchrus Ciliaris in Kenya

Abstract

In 1955, one tuft of each variety was split, and the splits were planted in adjacent rows to facilitate cross-fertilization if this should occur naturally. To avoid incompatibility due to differences in the time of flowering, seeds were collected only from those panicles that had flowered simultaneously with the other varieties, and this precaution was observed for each generation. Seed from each row was planted in boxes in 1956, and 70 seedlings of each variety were transplanted to adjacent plots at a uniform spacing. This same process was repeated in 1957 and 1958, when 105 plants of each variety were grown. A total of 245 plants of each variety was thus examined during three generations. In all three generations, the plants of each of the six varieties were invariably uniform and true to type. This shows that our varieties do not mix even if grown closely together. They breed in the same way as those investigated by Snyder et al. and should be considered as apomicts.