Self-incompatibility in Eucalyptus

Eucalypts are generally preferentially outcrossing (Pryor 1976; Griffin et al 1987), with relatively high outcrossing rates (e.g. 0.69-0.84 Moran and Bell 1983) maintained by varying degrees of self-fertility (Pryor 1957, 1976; Hodgson 1976b; Griffin et al 1987), aided by protandry (Pryor 1976) and reinforced by selection against the products of self-fertilization in later stages of the life cycle (see Potts et al 1987). Most species exhibit a marked reduction in seed yield following self-pollination compared to outcrossing, although within species there is considerable variation in the level of self-fertility (Pryor 1957, 1976; Table 1). In most of the species examined in detail to date, the majority of individuals are partially self-fertile, but individuals range from fully self-incompatible to fully self-fertile (Table 1). Postmating barriers to self-fertilization are thus rarely complete, and Eldridge (1976) notes of the genus that "persistent attempts at artificial self-pollination have been successful to some degree on almost every tree tested". Griffin et al (1987) later noted that with only one possible exception, all the E. regnans trees they had investigated were self-fertile to some degree, and Pryor (1957) states that of some dozens of individuals examined of nine species only one tree of E. bicostata was found to be fully self-incompatible. Early reports of self-sterile species (Krung and Alves 1949) have not been substantiated (e.g. E. grandis - Hodgson 1976a,b), although Eldridge (1976) did note the failure of self-pollination of E. deglupta while outcrossing was successful.

Considerable variability in the success of replicate self-pollinations has been reported (Tibbits 1988), and clearly adequate controls and several independent replications of a cross are required before a plant can be reliably classified as fully self-incompatible.