Abstract-The relationship of lake water color (mg liter-' Pt) to watershed and lake morphometry, derived from topographical, hydrological, and bathymetric maps, was examined by log-linear multiple regression on a data set obtained from 337 lakes from the northern United States and Canada. Color was positively related to the drainage ratio, and negatively related to watershed slope, mean lake depth, and lake area. Lakes that receive significant indirect drainage from other lakes were less colored than headwater lakes. The best regression model explained 60% of the variance in lake water color. Regression models were calculated separately for four geographic regions (the Laurentians of Quebec, eastem Quebec-Maine, northeastern Wisconsin, and the Experimental Lakes Area of western Ontario). The regression coefficients for the most significant variables, drainage ratio, watershed slope, and lake area were similar over these four areas, but the regional models still had significantly smaller standard errors than the general model based on the whole data set.The color of lake water is a useful index of dissolved humic matter (Gorham et al. 1983). This "gelbstofl" is made up of refractory acidic polymers that dissolve in soil water or in water percolating through peat and sedge bogs (Shapiro 1957; Aiken et al. 198 5). In addition to imparting weak acidity (Shapiro 1957;Gorham et al. 1986) humic matter reduces transparency and thereby the depth of the euphotic zone and the extent of the littoral, thus reducing the abundance and depth range of aquatic macrophytes (Spence 1982). Humic matter can directly reduce phytoplankton productivity (Jackson and Hecky 1980) and phytoplankton AcknowledgmentsWe thank Martin Scott, Lesley Pope, and Serge Ouimet for help with map measurements. Eville Gorham and three anonymous reviewers made suggestions.This work was supported by grants from the Natural Sciences and Engineering Research Council of Canada to J. Rasmussen, the FCAR (Quebec Government) equipe grant to J. Kalff, W. Leggett, R. H. Peters, and J. Rasmussen.A contribution to the Lake Memphremagog Project, McGill Limnology Research Center.

The surface and hyporheic fauna of a riffle area in Salem Creek near Elmira, Ontario, was investigated with a Surber and a corer sampler . A stratified random sampling design estimated the number of invertebrates at 154 561 (109 805-224 155) M-2 in June and at 64 407 (40 905-97 996) M-2 in October with more than 70% of the animals occurring below 5 cm . There was a great spatial variation of the hyporheos density in the three dimensions of the riffle, which seemed to be related mostly to hydrogeological characteristics (superficial groundwater flow and grain size) . Upward water movement in the centre of the stream occurred in the upper two transects, especially from the upper right bank, while the lower transect showed little water transmission . Hyporheos composition is discussed in regard to life history and the input of organic materials . Only two of nine pairs of Surber samples showed a high degree of similarity (0 .68, 0 .81) . The surficial spatial density and diversity were examined in relation to the available surface and to substrate heterogeneity . It is concluded that the great variations between samples caused by many factors present severe problems in the collection of reliable quantitative samples . ResumeLa faune de surface et celle de 1'hyporhelique d'un radier de Salem Creek pres d'Elmira, Ontario, ont ete etudiees a l'aide de l'echantillonneur de Surber et d'un carottier . Un echantillonnage aleatoire stratifie a permis d'estimer le nombre d'invertebres a 154 561 (109 805-224 155) m 2 en juin et a 64 407 (40 905-97 996) M -2 en octobre . Environ 70% de la faune benthique a ete retrouvee entre 5 et 65 cm . La densite de l'hyporheos presente de grandes variations spatiales, et celles-ci semblent etre reliees aux caracteristiques hydrogeologiques (mouvement des eaux souterraines superficielles et taille des particules). U n net mouvement ascendant dirige vers le centre du radier a ete detecte dans les deux transects situes en amont; aucun mouvement n'a ete mesure dans un troisieme transect situe an aval . La composition de l'hyporheos est mise en relation avec les cycles vitaux et avec l'apport des matieres organiques . Des 9 paires d'echantillons recueillis au Surber, deux seulement presentent un certain degre de similarity (0 .68, 0 .81) . Les variations spatiales de la densite et de la diversity de la faune de surface furent mises en correlation avec la surface disponible et l'heterogeneite du substrat . On en conclu que la grande variabilite entre les echantillons daps les deux habitats est due a plusieurs facteurs et qu'elle pose de serieux problemes pour 1'estimation de parametres quantitatifs .

No abstract

No abstract

We document the recent reappearance of anadromous sockeye salmon ( Oncorhynchus nerka ) that were thought to have been extirpated by the construction of hydroelectric dams on the Coquitlam and Alouette rivers in British Columbia, Canada, in 1914 and 1927, respectively. Unexpected downstream migrations of juveniles during experimental water releases into both rivers in 2005 and 2006 preceded upstream return migrations of adults in 2007 and 2008. Genetic (microsatellite and mitochondrial DNA) markers and stable isotope (δ34S and 87Sr/86Sr) patterns in otoliths confirm that both the juvenile downstream migrants and adult upstream migrants were progeny of nonanadromous sockeye salmon (kokanee) that inhabit Coquitlam and Alouette reservoirs. Low genetic diversity and evidence of genetic bottlenecks suggest that the kokanee populations in both reservoirs originated from relatively few anadromous individuals that residualized after downstream migration was largely prevented by the construction of dams. Once given an opportunity for upstream and downstream migration, both populations appear capable of reverting to a successful anadromous form, even after 25 generations.