JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org. This content downloaded from 169.230.243.252 on Tue, Abstract. We studied bank vegetation between the spring high-water level and the summer low-water level along two rivers in northern Sweden. The hypothesis that natural and ruderal species would show different downstream patterns of species richness was tested by sampling species composition and environmental variables along 200 m long stretches of riverbank, 10 km apart. Natural species richness was highest in the midreaches of both rivers, whereas ruderal species showed a significant, monotonic increase downstream. There are no obvious mechanisms producing the quadratic pattern of natural species richness. The downstream increase in ruderal species suggests a founder effect depending on larger artificial disturbances near the coast, but alternative explanations are also given. Total species richness did not exhibit any interpretable downstream patterns. The only factors significantly correlated with total species richness along both rivers were substrate heterogeneity and substrate fineness. Total species richness increased with substrate heterogeneity and was at maximum at intermediate levels of substrate fineness. This suggests that co-occurrence of natural and ruderal species is most likely where substrate types are numerous and dominant substrate particle size intermediate.

NILSSON, C., GARDFJELL, M., and GRELSSON, G. 1991. Importance of hydrochory in structuring plant communities along rivers. Can. J. Bot. 69: 2631-2633. Using wooden cubes as experimental seed mimics we tested whether differences in seed deposition among riverbank sites predictably affect the structure of riparian vegetation. Although cube deposition was not interpretable in terms of environmental variables like current velocity, riverbank width, and substrate, it was related to species composition of the riparian vegetation. We found an excess of species with short-floating seeds among those that were most frequent in sites capturing few cubes and an excess of species with long-floating seeds among those that were most frequent in sites capturing many cubes. This result provides the first evidence that water dispersal interacting with floating time may play a role in smallscale structuring of riparian plant communities. hydrochory in structuring plant communities along rivers. Can. J. Bot. 69 : 2631-2633. En utilisant de petits cubes de bois comme modele experimental pour imiter les graines, les auteurs ont vCrifiC si les differences dans la dCposition des graines de long des rives de la rivikre permettrait de prCdire la structure de la vegetation riparienne. Bien que la deposition des cubes ne soit pas interprktable en terme de variable environnementale, telle que la vtlocitC du courant, la largeur des rives et le substrat, elle est cependant relike a la composition des especes de la vCgCtation riparienne. Les auteurs ont trouvC une plus grande quantitC d'espkces ayant des graines de courte flotabiliti, parmi celles qui se retrouvent le plus souvent dans les sites oh Cchouent les cubes et un surplus d'especes avec des graines a longue floatabilite parmi celles qui Ctaient des plus frCquentes dans les sites oh on retrouve un grand nombre de cubes. Ce rCsultat donne la premikre preuve que la dispersion par I'eau, en interaction avec la ptriode de flotabilitk, pourrait jouer un r61e dans la structuration a petite Cchelle des communautCs de plantes ripariennes.

No abstract

The hypotheses that litter deposition on riverbanks would depend on riverbank features and that litter mass would affect riverbank vegetation were tested by sampling vegetation, litter, and soil along a seventh-order stretch of the unregulated Vindel River in northern Sweden. Sampling was carried out within a horizontal litter gradient at the top of the riverbank where most of the litter deposition occurs. The amount of leaf litter left on the riverbank after flooding was significantly correlated (P < 0.05) with soil organic matter. In contrast, wood litter mass and total litter mass were not correlated (P > 0.05) with riverbank features such as soil organic matter, riverbank slope, or stem density. Species richness, species diversity, and the proportions of stemmed and nonstemmed species showed quadratic relationships (P < 0.05) with leaf litter mass, whereas ground vegetation cover decreased monotonically (P < 0.05) with increasing leaf litter mass. Tall, stemmed species were most frequent at eroded sites with low amounts of leaf litter and at sites where leaf litter had accumulated. In contrast, the highest proportion of low-growing species was found at sites with intermediate leaf litter masses. Community characteristics such as species richness, species diversity, and the proportions of stemmed and nonstemmed species (but not ground vegetation cover) were less strongly correlated with wood litter mass and total litter mass. Linear equations explained about half of the variation in ground vegetation cover along gradients of wood and total litter masses. Density and richness of seeds increased with increasing litter mass. We discuss the litter effects in terms of productivity and disturbance and conclude that flooding in pristine rivers may exert a considerable influence on riverbank vegetation as a result of litter displacement.

No abstract