Transfer of florigenic stimuli between tillers in photoperiodically split plants of Dactylis glomerata and Bromus inermis

The flowering response was studied in single plants of Dactylis glomerata cv. Apelsvoll and Bromus inermis cv. Løfar, with one half exposed to inductive short days (SD = 8 h) and the other half exposed to non‐inductive long days (LD = 24 h) during primary induction in a daylength tunnel. Although unsplit control plants of D. glomerata showed an absolute SD requirement for primary induction, 38% of the LD‐treated half‐plants headed if the opposite half‐plant had been exposed to SD. In B. inermis SD‐exposure of one side resulted in a 5‐fold increase in panicle production on the opposite LD side. These data suggest that flowering stimuli can be translocated from one tiller to another in these short–long day grasses, but it remains to be established whether this transport is specific or merely a result of carbohydrate flow to the LD sink of the photoperiodically split plants. Since LD exposure of the opposite half‐plant had relatively little influence on percent heading or panicle number on the SD side, the translocation of flowering inhibitors appeared to be more limited than that of flowering stimuli. A preliminary experiment using radiolabelled gibberellin confirmed that intertiller translocation of this primary induction inhibitor was virtually absent in D. glomerata and only amounted to 10–15% of the applied radioactivity in B. inermis. Although the specific roles of flowering stimuli and flowering inhibitors, including various gibberellins, remain to be elucidated, these data help to explain why it is often observed that late‐emerging tillers, which have not been directly exposed to inductive conditions of sufficient duration, become reproductive in these grasses.