The main objective of this study was to evaluate how physical stress and herbivores influence the distribution, abundance, size frequency, and mortality of the giant kelp, Macrocystis pyrifera, in the southeast Pacific. These factors were studied for the kelp and the echinoid herbivore Loxechinus albus over a wide latitudinal range (56°—42° S) of South America on RV Hero cruises in 1972 and 1973. There were large kelp forests with many large plants in the southern Isla de los Estados and Tierra del Fuego sites. Sources of kelp mortality in these sites include entanglement with drift plants and heavy encrustation of bivalves. This resulted in a considerable amount of drifting kelp, and the sea urchins appeared to have abundant food and did not attack living kelp plants. Nevertheless, the urchin densities usually were relatively low and the Loxechinus size frequencies skewed to larger size classes, suggesting that their populations may be limited by larval availability in the Westwind Drift. There were also some large kelp forests in the northern Golfo Corcovado and Isla Chiloe areas, but here the kelp forests seemed relatively ephemeral and Loxechinus appeared to be harvested by humans. The intermediate latitudes (54°—46° S) in areas semiexposed to oceanic waves were characterized by situations in which Loxechinus overexploit the kelps and maintain urchin—coralline algae "barren grounds." In these areas kelps occur only where urchin access is restricted by wave exposure, ephemeral clumps of Desmarestia ligulata, and rarely, by a predator, the asteroid Meyenaster gelatinosus. In many areas the shallow distribution of Macrocystis was restricted by competition with the large fucoid Durvillaea antarctica. These relationships were evaluated by removal and addition experiments. Much more important than latitudinal relationships to both kelp and urchin distribution and abundance patterns were exposure gradients, from shores subject to strong oceanic swells to protected fjords. Transect data suggest that Loxechinus disappear from protected waters before Macrocystis. Thus in some semiprotected habitats there are very dense patches of kelps without grazers; kelp removal experiments in these habitats suggest strong density—dependent kelp interactions. In still more protected sites, the Macrocystis growth form changes, the plants appear to have low fertility, and there is very low recruitment. Summarizing, the Loxechinus population seems restricted by larval recruitment in the face of the Westwind Drift in the far south, perhaps by sedimentation or fresh water in protected fjords, and by human fishing in the north. The very low Loxechinus densities in the far south are correlated with large kelp forests in which the main kelp mortality results from drifting Macrocystis plants. The kelp populations appear limited by physical factors in the fjords and by Loxechinus grazing in most other areas.