(R.T.); and Electron Microscopy Services and Consultants, Phoenix, Arizona 85022 (R.M.)The incidence of plasmodesmata in the minor vein phloem of leaves varies widely between species. On this basis, two pathways of phloem loading have been proposed: symplastic where frequencies are high, and apoplastic where they are low. However, putative symplastic-loading species fall into at least two categories. In one, the plants translocate raffinose-family oligosaccharides (RFOs). In the other, the primary sugar in the phloem sap is sucrose (Suc). While a thermodynamically feasible mechanism of symplastic loading has been postulated for species that transport RFOs, no such mechanism is known for Suc transporters. We used p-chloromercuribenzenesulfonic acid inhibition of apoplastic loading to distinguish between the two pathways in three species that have abundant minor vein plasmodesmata and are therefore putative symplastic loaders. Clethra barbinervis and Liquidambar styraciflua transport Suc, while Catalpa speciosa transports RFOs. The results indicate that, contrary to the hypothesis that all species with abundant minor vein plasmodesmata load symplastically, C. barbinervis and L. styraciflua load from the apoplast. C. speciosa, being an RFO transporter, loads from the symplast, as expected. Data from these three species, and from the literature, also indicate that plants with abundant plasmodesmata in the minor vein phloem have abundant plasmodesmata between mesophyll cells. Thus, plasmodesmatal frequencies in the minor veins may be a reflection of overall frequencies in the lamina and may have limited relevance to phloem loading. We suggest that symplastic loading is restricted to plants that translocate oligosaccharides larger than Suc, such as RFOs, and that other plants, no matter how many plasmodesmata they have in the minor vein phloem, load via the apoplast.Phloem loading is an energy-requiring process that elevates the solute content of sieve elements (SEs) and companion cells (CCs) to levels far above those of surrounding cells (Geiger et al., 1973;van Bel, 1993;Grusak et al., 1996;Turgeon, 1996Turgeon, , 2000Beebe and Russin, 1999;Hellmann et al., 2000;Komor, 2000;Lalonde et al., 2003). The pressure gradient between source and sink organs drives long-distance transport. It has been suggested that there are two loading strategies-apoplastic and symplastic. This hypothesis derived originally from the observation that there is great variation in the number of plasmodesmata in the phloem of minor veins (Turgeon et al., 1975;Gamalei, 1989Gamalei, , 1990Gamalei, , 1991Gamalei, , 2000van Bel, 1993). Gamalei has classified over 1,000 species on this basis and finds that it is a relatively consistent feature at the family level. He terms those with the highest number of minor vein plasmodesmata as type 1.The physical basis for phloem loading via the apoplast is reasonably well understood; Suc enters the cell wall space and is driven across the plasma membranes of CCs and/or SEs by cotransport with protons (Lalonde ...