ABSTRACrFlowering can be modified by gibberellins (GAs) in Pharbitis nil Chois. in a complex fashion depending on GA type, dosage, and the timing of treatment relative to a single inductive dark period. Promotion of flowering occurs when GAs are applied 11 to 17 hours before a single inductive dark period. When applied 24 hours later the same GA dosage is inhibitory. Thus, depending on their activity and the tmin of application there is an optimum dose for promotion of flowering by any GA, with an excessive dose resulting in inhibition. Those GAs highly promotory for flowering at low doses are also most effective for stem elongation (2,2-dimethyl GA4>> GA32 > GA3 > GAs > GA7 > GA4). However, the effect of GAs on stem elongation contrasts markedly with that on flowering. A 10-to 50-fold greater dose is required for maximum promotion of stem elongation, and the response is not influenced by time of application relative to the inductive dark period. These differing responses of flowering and stem elongation raise questions about the use of relatively stable, highly bioactive GAs such as GA3 to probe the flowering response. It is proposed that the 'ideal' GAs for promoting flowering may be highly bioactive but with only a short lifetime in the plant and, hence, will have little or no effect on stem elongation.There are many reports of both promotive and inhibitory effects of applied GA3 on flowering of herbaceous higher plants and for woody species (20, 21, 30 and references therein). Even for a single woody species, apple, both inhibition and promotion of flowering have now been reported (9 and references therein). Depending on the species, the site(s) and mode(s) of GA action could differ, and this may account for the contradictory responses observed between and within a species. The complexity of response has been much discussed (7,26). As a further illustration, Warm (24) concluded that GA3 acted at two sites, in either or both the leaf and the shoot apex, in promoting the flowering of Hyoscyamus niger.Although some of the contrasts noted in the responses to GAs may reflect differences in the site and timing of treatment, the GA used and its uptake and metabolism in the plant could also be important. To examine the importance ofthese various factors '