Abstract
Experiments are reported which were designed to test the hypothesis that the movement of the translocation stream is driven by the contractile activity of P-protein filaments. The different types of filament found after negative staining of phloem exudates from Ricinus communis and Cucurbita pepo are described. An approximate model is proposed for the quaternary structure of a 20 nm component in the R. communis exudate. None of the filaments showed any ability to bind heavy meromyosin subfragment one. In experiments with cytochalasin B, no evidence of effects on the movement of 14C-assimilates or on the ultrastructure of the sieve elements of Lepidium sativum was found. It is concluded that the available evidence is unfavourable to the view that P-protein resembles known contractile proteins elsewhere.