In order to understand particle processing on the labial palps of suspension-feeding bivalves, mucocyte mapping and visual observations were performed on adult Mytilus edulis (Fam. Mytihdae) and Placopecten magellanjcus (Fam. Pectinidae] Mucocyte contents were divided into categories depending on staining affinity for periodic acid-Schiff and alcian blue. The mucocyte distr~bu-tion was determined by counts at representative sites on the palps, done on stained whole mounts or histological sections. Functional correlates for M. edulis were obtained using video endoscopy; this technique was compared to observations of carmine particle transport on specimens from which 1 valve had been removed. Video endoscopy was precluded in P. magellanicus due to the large and extremely sensitive arborescent lips; the behaviour of carmine particles deposited on the palps was therefore observed for this species. Mucus-particle cords are removed from the gill ventral partlcle groove by the cilia of the dorsal fold of the M. edulis palp at right angles to the current flow; highviscosity acid (AMPS) and acid-dominant (ADMPS)-containing mucocytes probably stabilize the cord at this point. The position of the retractable dorsal fold determines whether the cords are rejected or processed for ingestion. When the specimen's ingestlve capacity is overloaded, the arriving cords are transported parallel to and upon the palp crests to the palp ventral margin, and then to the posteriormost tip of the palp for rejection. This rejection pathway is characterized by transport perpendicular to or counter to current flow and high proportions of AMPS mucocytes. When the ingestive capacity of the specimen is not exceeded, mucus-particle cords are directed into the palp particle groove, where they are transformed into a slurry by the mechanical action of the palp crests and dorsal fold, and probably also the addition of low-viscosity neutral mucopolysacchandes present on the palp crests. The resulting slurry is amenable to both ingestion and particle selection; selected particles continue their anteriorward movement in the palp particle groove, while rejected particles are transported ventrally in the AMPS-rich crest troughs to join the rejection pathway along the palp ventral margin. The palp rejection site for P rnagellanicus is also the AMPS-rich ventro-posterior margin, which terminates in a mixed mucopolysaccharide (MMPS)-rich reflected tip, probably related to the valve-clapping voidance of pseudofeces. Both the oral groove and the crest summits contain almost exclusively MMPS, suggesting that in this species the labial palps are mostly Involved in ingestive particle handling, as opposed to M . edulis. Differences in gill structure and functlon may account for thls putative difference in palp function. The combination of mucocyte mapping and video endoscopy (where possible) significantly enhances our understanding of bivalve feeding mechanisms.