The ecology of the propagules of the invasive brown seaweed Sargassum muticum (Yendo) Fensholt was studied both in culture and in the sea. In stationary water the propagules sink at an average rate of >0.5 mm·s −1 and stick on contact. Germlings retained on the surface of the reproductive receptacles beyond the normal incubation time seem unable to liberate themselves. When released forcibly or by the eventual decay of the underlying receptacle, they seem unable to re-attach to the substratum. In the sea the vast majority of the germlings arise within 2–3 m of the parent plants although a few settle up to 30 m away. Seasonal recruitment follows closely the onset and degree of fertility of the adult populations. Successful colonization of the substratum requires not only the close proximity of a fertile plant, but also the availability of cleared rock. The presence of an existing canopy or turf of any one of at least eight species of algae can totally inhibit colonization by interposing a physical barrier between the settling germlings and the substratum. In southern California the barrier can be effectively re-established on denuded surfaces within 1–3 months. Germlings cultured in suspension rapidly lose their ability to attach on settlement but in culture unattached germlings remain healthy and seem able to grow well indefinitely. In nature drifting germlings can be found in the plankton up to at least 1.3 km from the nearest fertile plant. Drifting germlings seem to represent a mechanism for short-range dispersal, yet the species has almost certainly migrated 1100 km down the entire coast of California in a single step. Neither drifting germlings nor abscissed fertile branches could have made this migration, for they are produced too late in the year to take advantage of the inshore currents which are only seasonally favourable for southward dispersal. Floating vegetative branches on the other hand can depart at any time of the year and their southward migration would be greatly accelerated by the southerly winds that augment the currents. The branches could become fertile en route and as the plant is both monoecious and self-fertile, a single migrant could give rise to an extensive population. The possibility that man may have introduced S. muticum into southern California is discussed, but considered unlikely. The facility with which S. muticum not only migrates long distances, but also produces dense populations on arrival, reflects the efficiency of the varied dispersal mechanisms that it possesses.