CCFFR-99 ABSTRACTS OF PAPER PRESENTATIONS

Aku, P.M.K., and W. M. Tonn. Department of Biological Sciences, University of Alberta, Edmonton, Alberta, Canada, T6G 2E9; (403) 492-4162; (403)492-9234; bill.tonn@ualberta.ca.

INFLUENCE OF WATERSHED DISTURBANCE ON THE GROWTH OF NORTHERN PIKE IN SMALL BOREAL LAKES (O)

In boreal Canada, forest fires are natural disturbances that potentially affect small lakes. Growth in fishes is a vital individual-level variable that is strongly responsive to environmental change. To help understand impacts of fire on lake ecosytems and contrast them to effects of forest harvesting, we used the linear model approach to back-calculation to reconstruct and compare historical patterns of growth from cleithra of northern pike (Esox lucius), the dominant long-lived species, in 17 small lakes in northern Alberta. Pike from multi-species populations grow faster than those from pike-only populations. Growth rate among populations was positively correlated with the availability of forage fish but independent of the disturbance experienced by the watershed.

• Barry, E. and R.A. Curry. New Brunswick Cooperative Fish and Wildlife Research Unit, Biology Department, University of New Brunswick, Fredericton, NB
• MERCURY PATHWAYS AND TROPHIC INTERACTIONS IN NEW BRUNSWICK LAKES (O)
• Understanding the movement of mercury through trophic levels is crucial for developing models predicting its fate in natural ecosystems. We sampled primary, secondary and tertiary consumers of 18 lakes and analysed mercury concentrations and trophic position (stomach contents and stable isotope ratios) in southern New Brunswick in 1997/8. Plankton, benthos and bullfrogs were sampled for methylmercury and stable isotope analyses in seven of these lakes in 1998 to provide a baseline for inter-lake comparisons. Fish mercury levels vary from 0.02 to 1.09 ug/g, large differences in mercury concentrations are found both within and between lakes. Relationships between lake morphology, water chemistry, trophic position and community structure (average of five fish species per lake) are considered when comparing the large range of fish mercury levels. This study is the first in-depth look at mercury concentrations in lakes of this region. Examination of the trophic interactions within these lakes will lead to a greater understanding of mercury pathways within lakes and further the development of predictive models of mercury levels in New Brunswick lake ecosystems.

• Benoit, H. P.1 , P. Pepin2. 1Department of Biology, Ocean Sciences Center, Memorial University of Newfoundland, St.-John's, Newfoundland. email:b62hpb@morgan.ucs.mun.ca. 2Science Branch, Northwest Atlantic Fisheries Center, Department of Fisheries and Oceans, St.-John's,NF.
• METAMORPHIC AGE AND LENGTH IN MARINE FISHES: AN EMPIRICAL REVIEW OF PATTERNS FOR SPECIES, POPULATIONS AND INDIVIDUALS (O)
• We present a comprehensive summary of age and length at metamorphosis, and associated variability, for marine fish. Data from the literature were partitioned into species, intraspecific, and individual levels of resolution in order to examine the factors which affect the timing of settlement. Temperature appears to underlie most facets in the timing of metamorphosis, and is likely due to a strong relationship between temperature and growth rate (mm/day). Length of larvae at metamorphosis was significantly and positively related to temperature, when considered interspecifically, although the pattern did not hold for intraspecific comparisons. Age at metamorphosis showed an exponential decrease with increasing temperature in inter- and intraspecific comparisons, but age did not covary with length. Considered jointly, these results suggest that within species, age at metamorphosis is merely the time that it takes to grow to a somewhat fixed metamorphic length. A strong exponential relationship between mean age and length and their associated variability (SD) was found, with a slope greater than unity in both cases (i.e. variability increases relative to the mean). These results along with data considering the covariance between age and length for individuals suggest that the timing of metamorphosis for individuals may not be as simple as growth to a target length.

• Boss, S.M. and J.S. Richardson. Department of Forest Sciences, University of British Columbia, Vancouver, British Columbia
• SEASONAL RESOURCE LIMITATION OF RESIDENT CUTTHROAT TROUT IN SMALL, COASTAL STREAMS (O)
• Two key resources are generally thought to limit stream-dwelling salmonids, the supply of food and the amount of suitable habitat within the channel. These same two resources can be considerably altered when streamside trees are harvested. To what extent are individual growth and survival of fish limited by food and cover in streams? How does the influence of resources change between seasons, and is over-winter survival strongly influenced by fish size? We performed a summer field experiment to directly test the effect of food and cover on growth and survival of coastal cutthroat trout (Oncorhynchus clarki clarki). Using a 2x2 factorial design, we manipulated food and cover in stream enclosures containing individually marked trout replicated in two streams. Fish receiving food additions experienced growth rates more than 25 times higher than those of unfed fish, indicating severe food limitation during summer. The absence of cover increased mortality by over 50% in one stream, but had no effect in the other. During winter, we combined a mark-recapture study of trout with stream habitat surveys to determine the extent to which habitat complexity limits growth and survival. Finally, to address how summer growth affects over-winter survival, we monitored the released experimental fish and the general population throughout autumn and winter. Preliminary results suggest that over-winter survival was not explained by body size for experimental fish, but was positively related to body size for the general population of one of the two streams.

• Bradford, Mike and Jim Irvine. Fisheries and Oceans Canada, c/o School of Resource and Environmental Management, Simon Fraser University, Burnaby, BC, and Pacific Biological Station, Nanaimo, BC.
• DECLINES IN THE ABUNDANCE OF COHO IN THE INTERIOR OF SOUTHERN BRITISHCOLUMBIA, AND THE 1998 CONSERVATION CRISIS
• In 1998 dramatic restrictions in the British Columbia salmon fishery were instituted to protect endangered coho salmon populations. We focus on the Thompson River population complex, which has declined to less than 10% of recent historical maxima in only 3 generations. In many small streams inspected for adult coho, spawners have not been observed in the last few years. The cause of the declines appears to be primarily a combination of overfishing and declining survival rates for juveniles in the ocean. We use a simple 2-stage life history model to examine the effects of different exploitation and marine survival rates on these populations. The most likely future for these populations are continued declines, unless there is a reversal in the adverse conditions coho are currently encountering in the ocean.

• Bratty, J. M., C. J. Walters, and L. B. Holtby. Fisheries Centre, University of B. C.; Department of Fisheries and Oceans. Vancouver, B. C.
• HABITAT SPECIFIC OVERWINTER GROWTH, MOVEMENT AND SURVIVAL OF JUVENILE COHO IN INTERIOR STREAMS (o)
• The winter ecology of juvenile coho salmon in interior streams is poorly understood. To address this deficiency, we investigated habitat specific overwinter survival, growth and movement of individually-marked juveniles during two years in two tributary streams of the North Thompson River, B. C. Contrary to some previous studies, preliminary results suggest no evidence of size-dependent overwinter survival or growth in either year. However, fish overwintering in pool or side channel habitats generally survived at higher rates than those in runs. Coho densities and spatial distribution throughout the streams differed between years. In the year with much lower densities, coho were found almost exclusively in pools and side-channel habitats. In one creek, one beaver pond accounted for over 50% of the spring pre-smolt population. In the other stream, a large proportion of individuals overwintered in a groundwater sidechannel. These habitat types may be critical for sustaining coho freshwater production in interior systems. However, they may also be limited. Therefore, results may have important ramifications for land-use management decisions, especially when populations are at low levels of abundance. Results may also be used to improve existing stock production models.

• Burton, F. et D. Boisclair. Université de Montréal, Département de sciences biologiques, C.P. 6126 succursale centre-ville, Montréal, Québec.
• DÉTERMINANTS ENVIRONNEMENTAUX DE LA CONSOMMATION JOURNALIÈRE POUR LES TACONS DE SAUMON ATLANTIQUE
• La qualité de l'habitat pour les salmonidés en rivière est souvent décrite en fonction de variables abiotiques simples comme la profondeur, la vitesse et le substrat. La qualité de l'habitat s'exprime aussi par un ratio entre les coûts et les bénéfices d'exploitation d'un habitat, ce ratio étant le gain énergétique net. Une augmentation de la qualité de l'habitat devrait résulter en une augmentation du gain énergétique net pour un poisson utilisant cet habitat. L'objectif de notre travail était d'examiner une composante du ratio coût-bénéfice d'utilisation d'un habitat. Plus spécifiquement nous avons testé la possibilité de prédire la consommation journalière en fonction de variables environnementales simples comme la profondeur, la vitesse, la température et la densité de tacons. Nous avons échantillonné trois sites de la rivière Ste-Marguerite (Saguenay, Qc) durant trois étés. Cet échantillonnage nous a permis de travailler dans une gamme étendue de conditions environnementales. Pendant chacun des échantillonnages, des tacons étaient capturés à tous les 4h sur une période de 24h. Le modèle développé prédit la consommation journalière en fonction de la température et de la profondeur. Les relations entre la consommation et les variables environnementales pourraient être intégrées aux modèles de qualité d'habitat et ainsi augmenter leur précision.
• ENVIRONMENTAL DETERMINANT OF DAILY RATION FOR ATLANTIC SALMON PARR (O).
• Habitat quality for salmonids in river is often describe in function of simple abiotic variables, such as water depth, water velocity and substrate type. Habitat quality can also be defined as a cost and benefit ratio, this ratio being the net energy gain. An increase in habitat quality should result in an increase of net energy gain for a fish in this habitat. The objective of our work was to examine one component of the cost-benefit ratio of habitat utilization. More specifically we tested the possibility to predict daily ration in function of simple environmental variables such as water depth, water velocity, water temperature and density of fish. We sampled three sites of the Ste-Marguerite river ecosystem (Saguenay, Qc) during three summers. This sampling schedule was adopted to estimate ration under a wide range of environmental conditions. For each sampling, fish were captured at every 4h intervals on a period of 24h. The model we developed predicts daily ration in function of water temperature and depth. Relationship between ration and environmental variables, could be integrated to habitat quality models and increase the precision made by these models.

• Casselman, J.M., Ontario Ministry of Natural Resources, Aquatic Ecosystems Science Section,Glenora Fisheries Station, R.R. 4, Picton, Ontario K0K 2T0
• TEMPERATURE AND YEAR-CLASS STRENGTH OF WARM- AND COOL-WATER FISHES OF EASTERN LAKE ONTARIO AND THE BAY OF QUINTE (O)
• For four decades, bottom-set gill nets and trawls have been used to index abundance of fishes of the warm- and cool-water fish communities of eastern Lake Ontario and the Bay of Quinte. Year-class strength was indexed by using young-of-year catches from midsummer to early autumn. Relations between year-class strength and mean monthly water temperature of the upper Bay of Quinte for 1972-1997 were positively correlated for the five most abundant species. Average monthly temperatures in spring and summer 1998 were very high (50-yr extreme) and were used to predict abundance and to compare with actual catches. White perch, the most abundant young-of-year in 1998, correlated best with July temperatures (P = 0.017), with predicted abundance 2.87 less than, but not significantly different from, actual. Yellow perch, the second most abundant young-of-year and now the most abundant species in the Bay of Quinte, correlated with March temperatures (P = 0.033), with actual catches 1.67 greater than, but not significantly different from, predicted. Catches of young-of-year freshwater drum correlated best with July-August temperatures (P = 0.002), with actual 3.68 greater than, and significantly different from, predicted. Gizzard shad and alewife predictions were considerably greater than actual catches, correlating with April-May temperatures (P = 0.017) 30.26 and with May temperatures (P = 0.018) 22.96 respectively. Significant differences between actual and temperature-predicted catches for 1998 could be explained by population extremes, zebra mussel abundance, and predator-prey interaction.

• Chiperzak, D.B3., K.L Howland1 and I. Mcleod2. 3Department of Fisheries and Oceans, Yellowknife, NT. 1University of Alberta, Edmonton, Alberta. 2Gwich'in Renewable Resource Board, Inuvik
• RADIO TRACKING RESULTS OF INCONNU (STENODUS LEUCHICHTYS) IN THE LOWER MACKENZIE RIVER AND BEAUFORT SEA COAST. (O)
• Radio tags were mounted on 34 inconnu (Stenodus leuchichtys) in 4 areas of the lower Mackenzie River system and Beaufort Sea coast. Fourteen tags were mounted at Shingle Point (Yukon North Slope), 5 in the outer delta, 5 in the Peel River and 10 in the Arctic Red River. All tags had a battery life of approximately 400 days and were mounted between June and July of 1996. In 1996 radio tracking results indicated three probable spawning locations; one site approximately 100 km up the Peel River, the second at the Ramparts Rapids on the Mackenzie River and the third approximately 100 km up the Arctic Red River. The Arctic Red River spawning location was confirmed by persons on the ground. Inconnu sampled along the coast were not current year spawners. In 1997, seven of the inconnu tagged at Shingle Point were again found along the coast during the summer indicating that these inconnu were likely not alternate year spawners. The results of this tagging study will be discussed with reference to fisheries and habitat management issues as it relates to inconnu.

• Clément, M. and D. L.G. Noakes. Department of Zoology, University of Guelph, Guelph, Ontario, N1G 2W1.
• VALIDATION OF SPAWNING HABITAT MODEL FOR ATLANTIC SALMON IN LAKE ONTARIO TRIBUTARIES (O)
• The objective of this study is to validate the "Sediment Intrusion and Dissolved-Oxygen Transport Model" (SIDO). The SIDO model has been developed to simulate the physical habitat processes, survival and emergence time of chinook salmon (Oncorhynchus tshawytscha) and steelhead trout (O. mykiss) embryos on the North American Pacific coast. This study will determine if the model can accurately simulate the processes that are occurring in Atlantic salmon (Salmo salar) redds (nests) and ultimately predict embryo survival rate and emergence time in Lake Ontario tributaries. This is part of a large-scale project conducted by the Ontario Ministry of Natural Resources to restore a self-sustaining Atlantic salmon population in one or more Lake Ontario tributaries.The validation of the SIDO model first involves determination of the routing of water (water surface profile), sediment (suspended sediment and bedload transport) and dissolved-oxygen concentration at the stream-redd interface. The second validation phase measures intragravel flow, intrusion of fine sediment and dissolved-oxygen concentration in the redds, and quantifies the dissolved-oxygen demand by the organic matter and salmonid embryos during the incubation period. Preliminary results of the validation procedures of the SIDO model will be presented and other ongoing researches related to the Atlantic Salmon Restoration Program will be discussed.

• Cooke, S.J.1, A.F. Weckworth1, R.S. McKinley1 and D.P. Philipp2. 1Waterloo Biotelemetry Institute, Department of Biology, University of Waterloo, Waterloo, ON. 2Center for Aquatic Ecology, Illinois Natural History Survey, Champaign, IL.
• AN ELECTROMYOGRAPHIC AND VIDEOGRAPHIC STUDY OF THE SPAWNING BEHAVIOUR OF LARGEMOUTH BASS (O)
• Although a common species in North America which creates obvious nests, surprisingly little detailed information is known about the spawning behaviour of largemouth bass in non-hatchery conditions. The spawning behaviour of male and female largemouth bass was studies in central Illinois during the spring of 1998. Fish were implanted with electromyographic (EMG) transmitters (16 male, 8 female) several weeks prior to the initiation of spawning. As water temperatures approached 15 °C, fish were monitored continuously using 3 EMG receivers and 4 underwater cameras. Video cameras were positioned near sites where males were observed excavating nests. Several complete spawning acts involving EMG tagged fish were videotaped, in addition to numerous other spawning events involving untagged fish. We were also able to determine when EMG tagged fish spawned based upon patterns of muscular activity. We present detailed descriptions of the spawning behaviour and patterns in axial musculature activity associated with different spawning behaviours. We also discuss the utility of EMG telemetry for monitoring the reproductive activity of largemouth bass in areas of high cover, turbid water, or during low light conditions.

• Cooper J. Andrew, National Marine Fisheries Service, Systematics Laboratory, Museum of Natural History, Washington DC, 20560-0153.
• REVISIONS IN FLATFISH TAXONOMY: SOMETIMES THE NAMES HAVE TO CHANGE (O)
• The genus-level taxonomy in the family Pleuronectidae (right-eyed flounders) has undergone significant revisions in recent years. Changes incorporated into the most recent AFS list of common and scientific names (Robins, et.al., 1991) were based on a phenetic analysis of species interrelationships (Sakamoto, 1984). The revisions proposed by Sakamoto's study affected 26 species in 12 genera. Due to the commercial significance of many of these species, the nomenclature (sensu Sakamoto) has been the subject of much scrutiny and confusion. Taxonomic arrangements and nomenclatorial changes proposed in this phenetic study have been disputed and revised in the light of new and more complete information based on a cladistic analysis for this family (Cooper and Chapleau, 1998). To the non-systematist, the basis for these changes is sometimes obscure and not well understood. However, nomenclatorial changes mandated by a cladistic analysis can be meaningful to both taxonomist and fisheries scientist. Direct comparison of these two recent revisions highlights the utility of the cladistic analysis, especially in studies examining comparative biology, ecology and life history of these commercially important flatfish species.

• Crawford, Stephen S. and Bruce J. Morrison. Axelrod Institute of Ichthyology, University of Guelph, Guelph, Ontario
• ACTIVE ADAPTIVE MANAGEMENT FOR GREAT LAKES FISHERIES (O)
• We discuss the benefits and pitfalls of formally implementing the first 'active adaptive management' program for commercial and recreational fisheries of the Great Lakes. Existing fisheries management regimes in the basin can be classified as 'reactive' or 'passive' in nature; management may result in meaningful feedback and adjustments over the long term, yet problems occur when driving factors clash and the rate of change in management practices exceeds the rate of learning. The Chippewas of Nawash First Nation has developed an 'active adaptive management' program for Lake Huron fisheries that includes: (1) rigorous analysis of stakeholder values and objectives regarding the fisheries, (2) explicit recognition of uncertainty in the decision-making process, (3) pre-treatment modeling of management options, (4) statistically-designed management experiments to test pre-treatment models, (5) direct feedback between management experiments and policy development. The Chippewas of Nawash management program can serve as a pilot study for development and implementation of adaptive management programs for other Great Lakes fisheries.

• Danylchuk, Andy J. and William M. Tonn. Department of Biological Sciences, University of Alberta, Edmonton, Alberta, T6G 2S9
• POPULATION-LEVEL VARIABILITY OF FATHEAD MINNOWS (Pimephales promelas) IN BOREAL ALBERTA LAKES: POTENTIAL CONSEQUENCES OF FOREST HARVESTING (O)
• The fathead minnow is common to small lakes in the western boreal forest. Because it is often the most abundant, or only fish present, the fathead is an important component of aquatic food webs in this region. That little is known about the basic ecology of boreal fathead populations is a concern, since forestry activities are now expanding into this previously unharvested area. We monitored four fathead minnow populations for two years before (1995-1996) and two years after (1997-1998) experimental forest harvesting to: 1) examine inter-population and inter-annual variation in population-level characteristics, and 2) determine the effects of forest harvesting on these populations. Results from mark-recapture surveys indicate a high degree of inter-population and inter-annual variation in fathead abundance. Moreover, there can be large inter-annual differences in the size structure of the population. Preliminary examination of environmental data indicate that regional climatic patterns and lake and watershed characteristics likely contribute to this population-level variation. The high degree of temporal variability observed suggests that the fathead is sensitive, yet resilient to environmental disturbance. However, additional disturbance resulting from forest harvesting could potentially augment population variability and thus increase the probability of temporary extirpation of local populations.

• Decker, S., and S.G. Hinch. Fisheries Centre, and, Department of Forest Sciences, University of British Columbia, Vancouver, B.C.
• EFFECTS OF PRIMARY PRODUCTION ON SEASONAL CHANGES IN JUVENILE COHO SALMON SIZE AND BIOMASS IN ARTIFICIAL SIDE CHANNELS (O)
• In southwestern British Columbia over 40 artificial side channels have been constructed to rehabilitate and enhance salmon stocks in watersheds damaged by logging, agriculture, or urbanization. To understand how these systems function in terms of juvenile coho production, we selected three pairs of physically similar channels with one member of each pair having high summer production of periphyton and the other member with low periphyton production, and estimated the population sizes of aged 0+ coho fry every four months over one year. We found that average fry size and biomass per unit area at the end of summer were greatest in the high periphyton channels, however, summer density did not differ between channel types. In the subsequent spring just prior to smolting, fry were still larger and biomass greater in the high periphyton channels. In contrast to summer, spring density differed between channel pair members but differences could not be explained by summer periphyton production. Channels with relatively high pre-smolt densities tended to have relatively warm winter temperatures, likely a result of higher groundwater input. We are not certain whether higher densities reflected better over-winter survival or just better over-winter retention.

• Dobson, D.L., S.G. Hinch, S.N. Aitken. Department of Forest Sciences, University of British Columbia, Vancouver, B.C. V6T 1Z2
• HABITAT-SPECIFIC GENETIC AND PHENOTYPIC DIFFERENTIATION IN COHO SALMON: USING MICROSATELLITE MARKERS TO MAKE INFERENCES ABOUT QUANTITATIVE TRAITS IN A COASTAL B.C. POPULATION (O)
• Juvenile coho salmon will spawn and have their offspring reared during the summer in very different habitat types in coastal B.C. streams. For instance, some utilize mainstem river areas while others take advantage of off-channel habitats. However, during winter, juveniles produced in mainstem areas are believed to move into off-channel habitats for reasons of protection. These mainstem juvenile fish tend to be much larger at the start of winter than the off-channel resident juveniles. It is possible that these two size morphs reflect genetic differentiation that would indicate either fine-scale population structure or heritable growth-related differences. On the other hand, the variation may simply reflect phenotypic plasticity. If there is a genetic basis then these results could mean that conserving "stocks" of coho might involve more fine-scale protection of habitats and populations that currently practiced.

• Doucett, R.R.1, Power, M.V.2, Power, G.1, Caron, F.3, and Reist, J.D.4. 1Department of Biology, University of Waterloo, Waterloo, ON, N2L 3G1. 2Department of Agricultural Economics and Farm Management, University ofManitoba, Winnipeg, MB, R3T 2N2. 3Ministere Environnement et Faune, Quebec, QC, G1R 4Y1. 4Fisheries and Oceans Canada, Arctic Fish Ecology and Assessment Research Section, Winnipeg, MB, R3T 2N6.
• PUSHING THE LIMITS: DETECTING ANADROMY IN A SOUTHERN RELICT POPULATION OF ARCTIC CHAR (SALVELINUS ALPINUS) USING STABLE-ISOTOPE ANALYSIS (O)
• We investigated the use of stable-isotope analysis to detect anadromy in a southern relict population of Arctic char (Salvelinus alpinus) in the Riviere de la Trinite system, Quebec. Of 17 char collected while migrating out to sea, 3 had stable C, N, and S isotope ratios (del13C: -22.1 +/- 1.3 per mil, del15N: 12.3 +/- 0.5 per mil, and del34S: 17.7 +/- 0.4 per mil, respectively) representative of marine-feeding fishes. The remaining char had stable-isotope ratios (del13C: -29.2 +/- 1.1 per mil, del15N: 8.9 +/-0.4 per mil, and del34S: 8.0 +/- 0.2 per mil, respectively) similar to those of resident fish (n = 20) sampled upstream at Davidson Lake (del13C: -27.9 +/- 0.7 per mil, del15N: 7.3 +/- 0.7 per mil, and del34S: 7.5 +/- 0.3 per mil, respectively). Sea-surface temperatures, taken at 5 m depth in the lower St. Lawrence in summer 1997, were above the 14 degrees Celsius threshold thought to control anadromous char distributions in northern Norway. We speculate on why anadromy still exists in the Riviere de la Trinite system, and also on what future climate change may do to the anadromous component of char populations at their southern limit. Our results extend the southern range of anadromous Arctic char on continental North America by at least 2 degrees latitude, and also suggest that a sea-surface temperature of 14 degrees Celsius is not the thermal maximum limiting marine distributions of anadromous Arctic char. Furthermore, we show that stable-isotope analysis may provide researchers with a quick and inexpensive technique that will complement more conventional methods in the identification of anadromous fishes.

• Earle, Jennifer and Jim Campbell. R.L. & L. Environmental Services Ltd., Edmonton, Alberta
• SEASONAL MOVEMENTS OF BULL TROUT IN THE BERLAND RIVER, ALBERTA (O).
• In 1996, Trout Unlimited Canada commissioned a study of bull trout movements in the Berland River, a tributary to the Athabasca River, north of Hinton. Fourteen bull trout were implanted with high frequency radio transmitters in order to monitor their seasonal movements, including spawning-related movements. Aerial surveys were conducted from August 1996 until January 1997 and covered the mainstem Berland River from its headwaters to the confluence with the Athabasca River, as well as 17 Berland River tributaries. Five of the fish exhibited typical spawning movements and were located during aerial surveys in the Little Berland River and South Berland River. These fish generally remained in the tributaries from mid-August to early/mid September. Maximum recorded upstream movement for a spawning fish was 45 km. Flights conducted since late September indicated fish primarily overwintered in an 8 km stretch of river, 4 km downstream of the mouth of the Little Berland River. Since oil and gas development and forestry activities are increasing in the drainage, this study provides valuable information on critical habitat for various life stages of bull trout. Additional studies conducted in late summer/fall 1997 included monitoring bull trout spawning movements at fish fences on the Little Berland River and Moon Creek, as well as redd surveys on several streams known to support juvenile bull trout populations.

• Edwards, P.A. and R.A. Cunjak. Department of Biology, UNB, Fredericton, New Brunswick
• FORESTRY AND THE SLIMY SCULPIN (COTTUS COGNATUS)...THE UNTOLD STORY (O)
• Long term data sets of population abundance are invaluable tools by which to quantify fish population change. This study utilizes nine years of electrofishing and habitat data collected from thirty two sites within Catamaran Brook, New Brunswick. Prior to clear-cut logging activity, natural population changes of the slimy sculpin (Cottus cognatus) were quantified both spatially and temporally. Preliminary results suggest spatial variation in sculpin distribution. Bedload movement experiments were carried out to assist in the determination of the impact substrate stability has on sculpin distribution. The density and distribution patterns attributed to natural fluctuation are to be used as templates to quantify sculpin population changes as a result of the timber harvest in the Catamaran Brook basin.

• Fox, M.G.1 and A.J. Crivelli2
• 1Environmental and Resource Studies Program and Department of Biology,Trent University, Peterborough, ON. K9J 7B8. 2Station Biologique de la Tour du Valat, Le Sambuc 13200 FRANCE
• LIFE HISTORY TRAITS OF PUMPKINSEED POPULATIONS IN WARM THERMAL ENVIRONMENTS (O).
• Fish populations frequently show a latitudinal pattern of life history variation; such variation could be due to latitudinal differences in growth rates, or to the direct effect of temperature on age at maturity, size at maturity or reproductive effort. To examine these alternatives, we compared the growth and life history patterns of three populations of pumpkinseed (Lepomis gibbosus) introduced into warm, Mediterranean waterbodies in southern France more than 50 years ago, with native populations in central and eastern Ontario. Growth and life history traits of the French populations were inferred from monthly samples collected from May to September, and from a reproductive allocation experiment conducted on one Canadian and two French populations. Mean age at maturity in the French populations ranged from 1.3 - 2.3 years, and was significantly earlier than that of Canadian populations. Length at maturity was significantly smaller in the French populations, and although there was no significant difference between French and Canadian populations in mean GSI, the results of the reproductive allocation experiment suggest that French populations have a longer breeding season, and shed a greater mass of eggs per unit body mass than Canadian pumpkinseeds. Body length at ages 1 - 4 was as high or higher in Canadian pumpkinseeds, suggesting that the earlier maturity of French pumpkinseeds was a direct effect of the warm thermal environment, rather than an indirect effect of faster growth.

• Gaudreau, N. and D. Boisclair. Universite de Montreal, Departement de sciences biologiques, C.P.6128, Succursale Centre-ville, Montreal, QC. H3A 1B1.
• ACOUSTIC ASSESSMENT OF FISH ABUNDANCE AND DISTRIBUTION PATTERNS DURING DIFFERENT MOON PHASES. (O)
• Estimates of the abundance of fish populations are necessary to establish their economical value and the importance of their role in ecosystems. Sampling for fish abundance requires appropriate strategies that might be developed using recurrent ecological patterns. The different phases of the moon cycle have been shown to affect the daily vertical migration of fish in deep lakes. In shallow lakes, some indications lead to a similar relation between the moon cycle and daily horizontal migration of fish. Our objectives in this work were 1) to confirm the influence of the moon phases on daily horizontal migration and 2) to develop a sampling strategy accounting for that potential influence. We used a hydroacoustic system to insonify a bay of Lake Croche (Laurentian region of Quebec). Sampling took place every 4h during 3 consecutive days every week during 7 weeks, thus covering 2 moon cycles. Our results reveal an 8-fold decrease of the number of fish horizontally migrating during full moon nights compare to new moon nights. We developed 2 sampling strategies that minimise sampling effort and maximise accuracy and precision of estimates.

• Giannico, G.R. and S.G. Hinch. Institute for Resources and Environment, and Department of Forest Sciences, University of British Columbia, Vancouver, BC
• EFFECTS OF WATER TEMPERATURE AND WOODY DEBRIS ADDITION ON JUVENILE COHO SALMON WINTER MOVEMENT AND SMOLT OUTPUT IN ARTIFICIAL SIDE CHANNELS (O)
• Many side-channels built in southwestern British Columbia in the early 1980s to increase salmonid production are structurally simple, have little woody debris and are largely ground-water fed. Despite their simplicity, they have been reported to support high densities of coho salmon fry and to have higher smolt outputs per unit of area than natural side-channels or streams. Their high fish producing capacity may be attributed to the warm winter conditions created by the input of ground-water. Channel designs have changed and recently built channels are more complex and have many deep pools containing woody debris, but are largely surface-water fed. To study the role that both water temperature and woody debris addition in artificial channels have on juvenile coho salmon winter movement and smolt output, we set up an experiment using two artificial side-channels. One channel is surface-water fed (cold) and the other is ground-water fed (warm). We longitudinally split a 100 m section of each channel into two equal halves, added woody debris to one half of each channel and monitored fish movement over the fall/spring period using two-way counting fences and group marking. In the cold channel, we observed that the side treated with woody debris retained more fish over the winter and produced twice as many smolts in the spring than the side with no-wood. In contrast, in the warm channel, we found no differences in either fish movement patterns or smolt output between channel sides regardless of wood treatment. Although in the fall fish densities were similar in both channels, by early spring fish density and smolt output were twice as high in the warm compared to the cold channel.

• Gillespie, G.J.1 and M.G. Fox2. 1Watershed Ecosystems Graduate Program Trent University, Peterborough, ON.K9J 7B8. 2Environmental and Resource Studies Program and Department of Biology, Trent University, Peterborough, ON. K9J 7B8.
• TROPHIC DIMORPHISM AND REPRODUCTIVE ISOLATION IN TWO POPULATIONS OF PUMPKINSEED (O)
• Trophic dimorphism has been found to occur in populations of pumpkinseed (Lepomis gibbosus) isolated from bluegill (Lepomis macrochirus) in the Adirondack Mountains (NY) but it is unknown whether this phenomenon is unique to this region, and whether reproductive isolation is a factor in the maintenance of this dimorphism. We addressed these questions in a study of allopatric pumpkinseed populations in Looncall and Wolf lakes (central Ontario), where the presence of individuals in the littoral and pelagic zones was confirmed. Thirty external and seven internal morphometric characteristics were measured on littoral and pelagic pumpkinseeds collected from each lake. In Looncall Lake, dimorphism was significant in 12 of the 30 external metrics, pelagic pumpkinseeds were more fusiform than littoral pumpkinseed, and pelagic individuals had a significantly higher number of gill rakers per arch and a greater distance between gill rakers than littoral individuals. External dimorphism was less marked in Wolf Lake, but the gill rakers of pelagic individuals were significantly wider and more dense, and their molars were significantly smaller than those of littoral individuals. Of the 700 pumpkinseeds that were tagged in Looncall Lake during the breeding season, 117 were later recaptured, and only 8 of these were found in a different zone from where they were originally tagged. These data suggest that very little mixing occurs between littoral and pelagic pumpkinseeds during the breeding season, and that reproductive isolation may be an important mechanism in the maintenance of trophic dimorphism.

• Gingras, J and D. Boisclair. Universite de Montreal, Departement des sciences biologiques, C.P.6128, Succursale Centre-Ville, Montreal, Que, H3A-1B1.
• PERCH (Perca flavescens) CONSUMPTION RATES: VALIDATION OF THE Cs-137 METHOD. (O)
• Fish consumption rates has been traditionally measured by the weight of the complete digestive tract content and it evacuation rate, for exemple, the Eggers method. An alternative method, based on the turn-over rate of the Cs-137, was used by Kevern (1966) to estimate fish consumption. However, the Cs-137 method has been validated only for brown trout (Salmo trutta) and arctic charr (Salvelinus alpinus) in a single lake. The objective of this field study is to compare consumption rates obtained from the Cs-137 approach, from the Rowan and Rasmussen model (1996), with the Eggers method. Ten combinaisons of perch (Perca flavescens) age class (I+ to III+) from four lakes of the Laurentiens region (Quebec) were studied. Three samplings were realised during the summer of 1998: june, july and august. Our results reveal consumption rates differing from 0% to 20% between the Cs-137 and the Eggers method. Thus, our data suggests that the Cs-137 method is appropriate to estimate perch consumption rates in the field.

• Guay, J.C.1, Rioux, D.2, Boisclair, D.1, Leclerc, M.2, Lapointe, M.3 1Université de Montréal, Département de sciences biologiques, C.P.6128, Succursale Centre-ville, Montréal, Qc H3C 3J7. 2Institut National de la Recherche Scientifique (INRS-eau) 2800 rue Einstein, suite 105, Québec, Qc G1V 4C7, Canada. 3McGill University, department of geography, 805 Sherbrooke Street West, Montreal, Qc H3A 2K6, Canada.
• EVALUATION OF A BIO-PHYSICAL HABITAT MODEL FOR PARR OF ATLANTIC SALMON (SALMO SALAR). (O)
• The objective of habitat modeling is to predict the impact of natural and anthropogenic disturbances on the quality and quantity of habitat for given species. Numerous studies have suggested that habitat quality for parr of atlantic salmon can be quantified using water depth, current speed and substrate size. Based on these assumptions, bio-physical models have been developed to map expected habitat quality over specified reach of rivers. For any given site of few square meters, a global index of habitat quality (Ig) can be calculated as the product of the quality of that site relative to its depth (Id), current velocity (Iv) and substrate size (Is) such that Ig = Id x Iv x Is. The quality of a site relative to one of these variables is extracted from a preference curve for this variable. Each preference curve represents the % utilization by parr of a given range of environmental conditions divided by the % availability of such range of conditions in the environment. The purpose of our work was to test the basic assumption of bio-physical models that presume that parr of atlantic salmon will distribute themselves preferably in site characterized by high values of Ig. We constructed preference curves of 0+, 1+ and 2+ parr for depth, current velocity, and substrate size in one section of the Sainte-Marguerite river (calibration section; 700m). The prediction for different flows of the bio-physical model was compared to a map of fish distribution obtained by scuba diving the complete area of the test section (800m) for these flows. Our result confirm that parr of atlantic salmon display well defined preferences for specific range of environmental conditions. Furthermore, the preferences of 0+ parr for current velocity and depth are different from that of older parrs (1+ and 2+). Finally, our work supported the existence of a correspondence between maps of Ig (or predicted parr distribution) and observed parr distribution.

• Gunn, .John M. Ontario Ministry of Natural Resources. Cooperative Freshwater Ecology Unit. Laurentian University, Sudbury, Ont. P3E 2C6 email:JGUNN@Nickel.Laurentian.Ca
• THERMAL TOLERANCE OF LAKE TROUT IN WARM SHALLOW SHIELD LAKES DURING AN EL NINO YEAR
• The lake trout (Salvelinus namaycush) is generally considered to be a highly vulnerable species that would be adversely affected by climate warming in the Boreal Shield ecozone. Usually described as a cold stenotherm with a strong preference for cold deep hypolimnetic waters, it has a relatively low incipient lethal temperature limit. Recently it has been shown that the lake trout will occassionally occupy eplimnetic waters with temperatures approaching lethal conditions, however, it is unkwnown whether these observations describe brief feeding forays or represent exposures to warm conditions with significant physiological effects. We conducted a series of experiments to assess seasonal and daily changes in core body temperatures of adult lake trout in warm shallow isothermal lake where bottom temperatures were approximately 20 C. Growth and diet of juvenile lake trout were also determined. The occurrence of the 1998 El Nino event gave us the opportunity to compare effects between two very different years (cold 1997, warm 1998) that may represent effects that can be expected to occur if lakes gradually warm because of climate change.

• Hay, D.E. Pacific Biological Station, Nanaimo, BC V9R 5K6
• CLIMATE CHANGE AND THE DECLINE OF EULACHONS
• The eulachon (Thaleichthys pacificus) is an anadromous smelt that spawns in about 15 rivers in British Columbia. Eulachons range from northern California to the southern Bering Sea but they spawn in few rivers, perhaps no more than 30 river. Their pre-spawning adult life is in open marine waters. They live close to the bottom are taken as incidental catch mainly by bottom trawlers but rarely by mid-water trawlers. Eulachons live for 3-4 years before spawning and dying in the lower reaches of rivers. Their eggs are small (<1mm), demersal and adhesive, attaching to substrate. Eggs incubate for 3-4 weeks before hatching into small pelagic larvae that are rapidly washed in to estuarine and marine waters. Spawning times vary. In most rivers eulachon spawn in March and April. The Columbia River is one of the earliest spawning runs (January-February) and the Fraser one of the latest (April-May). During the last 20 years all runs in the southern part (California to mid-BC) of the range have gradually declined. In 1994 there was a sudden sharp decline in spawning runs in 3 southern rivers: the Fraser, Columbia and Klinaklini (draining into Knight Inlet) and perhaps in other rivers. This 1994 decline led to the closures of the only two commercial eulachon fisheries (Fraser and Columbia Rivers). The scarcity of eulachons in these and other rivers concerned many people, especially some First Nations, for whom eulachons are a very important species. Climate change, as an general explanation for the declines, is consistent with all of the available information about this species, but this explanation is not well received in many areas. In many rivers some eulachon spawning habitat has been destroyed by industrial activities such as logging, shoreline development and dredging. Industrial pollution has negatively impacted spawning fish and habitat. In offshore areas eulachons are killed as bycatch in trawling operations, particularly shrimp trawls. The concern about these issues is valid. Unfortunately, the attention to these issues often serves to preclude climate change as an alternate and credible explanation for much of the eulachon decline. This presents a dillema. On one hand, those who choose to overlook climate change may have unwarranted expectations for the recovery of spawning runs if and when the habitat issues and other problems are corrected. On the other hand, short- or long-term reversal of climate trends could result in partial restoration of some eulachon runs and diminish the effort to correct persisting habitat abuse and other problems.

• Herunter H, E. MacIsaac, and S. Macdonald. Fisheries & Oceans Canada, Co-operative Resource Management Institute, Simon Fraser University, Burnaby, BC V5A1S6.
• EFFECTS OF FOREST HARVESTING ON THE CHEMISTRY, SEDIMENT DEPOSITION, AND WATER TEMPERATURES OF SMALL STREAMS IN CENTRAL INTERIOR BRITISH COLUMBIA.
• Studies of the effects of forestry activities on the characteristics of small upland and tributary streams in the central interior of British Columbia have been conducted for several years now as part of the Stuart-Takla Fish-Forestry Interaction Study. Clearcutting in some of the study areas began in January 1997 and this paper reports on the preliminary findings from the first 2 seasons of post-harvest research. The study streams incorporated both aggressive and conservative riparian treatments for individual streams as well as additional road crossing study sites. The objective of the research is to evaluate the importance of small upland and tributary streams to downstream fish habitat as well as the ability of current riparian management provisions within the Forest Practices Code to protect small streams and their downstream habitat. Changes in water chemistry, nutrient concentrations, and stream and groundwater temperatures in response to forest harvest treatments have all been recorded and their significance to downstream fish habitat will be discussed.

• Higgins, Paul, B.C. Hydro, Power Supply Operations, Burnaby, British Columbia
• DESIGN OF A LARGE SCALE HABITAT MANAGEMENT EXPERIMENT TO ESTIMATE THE FUNCTIONAL RELATIONSHIP BETWEEN STEELHEAD SMOLT PRODUCTION AND INSTREAM FLOW
• Large scale ecosystem experiments are increasingly being recognized as the most scientifically defensible method to manage and rectify the effects of flow regulation on fish populations. I evaluated options for the design of a flow release experiment to empirically estimate the functional relationship between steelhead smolt production and instream flow in Bridge River, a small regulated tributary of the Fraser River, B.C. Simple ecosystem state and management response models were developed and embedded into a Monte Carlo policy accounting framework to 1) articulate and simulate competing conceptual hypotheses about the response of steelhead to flow and how managers make decsions about optimum flows; 2) simulate future learning of managers to assess whether the long term benefits of improved knowledge about the functional relationship between fish productivity outweigh the costs of flow experimentation and, 3) to determine the optimal experimental design (number, magnitude and duration of flow treatments) and monitoring efforts that maximize both fish production and hydroelectric power generation. My simulations suggested more flow treatments of longer duration increased statistical precision of optimum flow predictions, however, the time discounted costs of foregone energy production needed to produce the flow treatments lead to considerable reduction in net long term management performance. The trade-off between these factors produced an optimum experimental duration of between 8-12 y. Optimum treatment effort allocation was estimated to be 4 treatments (0.5%, 1% , 2%, 5% MAD) each lasting 3 consecutive years, and this design was robust to shifts between regimes of low and high variability in marine survival rate. My suggest that the long term benefits of obtain higher certainty about the form of the relationship between steelhead smolt production and instream flow out weighs the short term costs of experimentation.

• Howland,K.L., W.M. Tonn and R.F. Tallman
• LIFE HISTORY DIFFERENTIATION OF INCONNU IN THE MACKENZIE RIVERSYSTEM: GENETIC OR ENVIRONMENTAL BASIS? (0)
• Our previous work has shown that freshwater and anadromous inconnu of the Mackenzie River system vary substantially with respect to growth rate, age at maturity, and fecundity. In this study we used a "common garden experiment" approach to examine the relative influences of genetic versus environmental factors on observed life history differences of inconnu. Eggs of both life-history types were hatched and offspring were reared in a common environment to see if differences in traits such as development, growth, survival, and salinity tolerance persisted. Based on preliminary results, egg development and growth rate differ among life history types, indicating that both traits are under a significant degree of genetic control. We suggest these genetic differences have evolved due to different temperature regimes within spawning rivers and differences in the growing seasons experienced by the two populations examined in this study.

• Imre, I. and D.L.G. Noakes. Department of Zoology, College of Biological Sciences, University of Guelph, Guelph, Ontario
• A STUDY OF THE VARIATION IN BODY SHAPE OF YOUNG-OF-THE-YEAR BROOK CHARR (SALVELINUS FONTINALIS) INHABITING MICROHABITATS WITH DIFFERENT WATER VELOCITY REGIMES (O)
• Salmonines inhabit variable stream environments for at least part of their life cycle. Many species show behavioural and morphological variation among species , among populations of the same species, and among individuals within populations. This study addresses the basis for adaptive morphological variation among individuals within local populations in stream environments. I studied the influence of hydraulic conditions on the body morphology of young-of-the-year (YOY) brook charr, Salvelinus fontinalis. Individuals captured from focal points with higher water velocities have a more streamlined body shape than individuals captured from focal points with lower water velocities. I tested the hypothesis that the difference in the body shape of YOY brook charr swimming at different water velocities is a result of the species' phenotypic plasticity rather than genetically determined polymorphism. I predicted that fish reared at higher water velocity will have a more fusiform body shape than fish reared at lower water velocity. I reared YOY brook charr from a common stock at 2 different water velocities under controlled laboratory conditions for 4 months. Fish were preserved at regular intervals and measured for a number of morphometric characters. The results of this study contribute to our understanding of the relative importance of phenotypic plasticity and genetic polymorphism in determining fish body shape. This study will also identify factors influencing body shape under controlled rearing conditions. This could be useful for manipulation of body shape of fish reared under intensive culture conditions for stocking in streams and lakes, where fish with different body shape would have different survival potential.

• Jeffrey, J. and C. Taggart. Department of Oceanography, Dalhousie University, Halifax, Nova Scotia
• OCEANOGRAPHIC VARIABILITY AND OTOLITH DAILY GROWTH INCREMENTS IN SILVER HAKE (MERLUCCIUS BILINEARIS) LARVAE FROM THE SCOTIAN SHELF (O)
• It has been argued that growth variation in larval fish is functionally related to variation in the oceanographic environment. Studies of these relations (using temperature, wind mixing, prey availability etc.) provide mixed results, at a variety of scales, that may stem from averaging temporal and individual growth variation. If so, relations between growth and the environment among individuals, cohorts and populations may be masked. Silver hake (Merluccius bilinearis) larvae in the Western Bank region of the Scotian Shelf represent an ideal model species and system for quantitatively determining the nature and scale of natural variations in growth and in the environment, and the relationship between these because: 1) otoliths of silver hake larvae show well-defined daily increments that are ideal for estimating growth variation among individuals at various scales (days to months); 2) multiple cohorts of larvae persist (days to months) in the Western Bank region; and 3) oceanographic conditions in the Western Bank region are continuously varying (days to months). Multiple cohorts of silver hake larvae of various ages (8 to 60 d) were collected in September and November 1997 on Western Bank where an oceanographic instrument array recorded temperature, salinity and current velocity at various depths, and where meteorological data from nearby (approximately 100 km) Sable Island were recorded. Growth variation is determined by comparing the otolith increment radii among individuals of common age within and among cohorts and water masses using repeated measures multivariate analysis of variance. Comparisons of length-at-age relationships are used to examine cohort-averaged growth variation among cohorts within a season. Variations among individuals within and among cohorts are compared to variations in the oceanographic and meteorological conditions. Inter-annual variation in all of the above is examined using more extensive data collected in October 1998. Funded by NSERC and GLOBEC-Canada.

• Lester, N. P.1, B. J. Shuter1, and R. S. Kushneriuk2. 1Aquatic Ecosystems Science Section, Ontario Ministry of Natural Resources, 300 Water St., Peterborough, Ontario, K9J 8M5 2Centre for Northern Forest Ecosystem Research, OMNR, Lakehead University, Thunder Bay, Ontario, P7B 5E1
• LIFE HISTORY VARIATION IN ONTARIO WALLEYE POPULATIONS: THE ROLE OF CLIMATE AND SECCHI DEPTH (ORAL)
• Life history characteristics of walleye in 66 Ontario lakes varied with growing degree days (range: 1100 - 2100 degrees Celsius) and secchi depth (range: 0.5 - 6 m). Populations in warmer, clearer lakes experienced higher growth rates early in life. Asymptotic length was higher in females than males. Female asymptotic length (range: 55 - 90 cm) was positively correlated with early growth rate and lake area. Length at first maturity in females (n = 22 lakes, range = 25 - 50 cm) increased with early growth rate, asymptotic length, and survival. This relationship was consistent with life history theory that predicts how the timing of reproduction should vary in response to environmentally induced changes in growth and survival. These relations provide a basis for developing regional guidelines for managing walleye exploitation in Ontario. We use them here to assess the potential impacts of climate change on walleye populations.

• Mackereth, R.W. and D.J. McCormick. Centre for Northern Forest Ecosystem Research, OMNR, Lakehead University, Thunder Bay, ON.
• COMPARATIVE AND EXPERIMENTAL APPROACHES TO ASSESSING THE IMPACT OFTIMBER HARVEST ON COLDWATER STREAM FISH HABITAT IN ONTARIO'S BOREALFOREST (O)
• The purpose of our research program is to quantify timber harvest impacts on stream fish habitat in Ontario's boreal forest, determine the spatial and temporal scale of these impacts and develop harvest guidelines to minimize habitat loss. We are using a comparative study to understand how the extent and timing of disturbance (i.e. timber harvest and fire) influences the physical and biological characteristics of streams. A GIS is used to select study sites which differ in catchment area and to summarize catchment attributes (including timber harvest area) and compare these to in-stream variables. A complimentary experimental study is examining specific processes, such as sedimentation, by which timber harvest impacts fish habitat. To date our results suggest that the natural variability in fish communities and fish habitat, related to factors such as spatial scale (i.e. catchment area) and landscape variables (e.g. surficial geology, valley slope), is much greater than changes attributable to timber harvest disturbance. This observation is supported by our experimental results which show very small changes in upland sediment movement and no detectable change in stream habitat adjacent to the harvest area. Further examination of temporal and cumulative impacts are required to fully understand the ecological impacts of timber harvest on stream fish habitat.

• MacNeill, W. Scott and R. Allen Curry. New Brunswick Cooperative Fish and Wildlife Research Unit, Biology Department, Bag Service #45111 University of New Brunswick, Fredericton, NB. E3B 6E1.
• JUVENILE BROOK TROUT (SALVELINUS FONTINALIS)ABUNDANCE AND HABITAT USAGE IN RELATION TO SEDIMENT LOADS IN PRINCE EDWARD ISLAND STREAMS. (O)
• Prince Edward Island (PEI) recreational fisheries resources have been significantly degraded by the construction of dams, pollution from agricultural runoff, and soil erosion from land cultivation. Erosion is considered the most serious environmental problem facing PEI streams. We measured juvenile brook trout (Salvelinus fontinalis) abundance in watersheds with varying amounts of agricultural land use. We observed a lower density of juveniles in watersheds with a greater proportion of land in agriculture (80% vs. 60%). Annual suspended sediment concentrations increased with the proportion of agricultural land use, to a maximum >20,000 mg/l in a productive brook trout stream. Juvenile brook trout were found to use different habitat types in streams with higher amounts of suspended sediment. Land use effects on juvenile brook trout abundance and behavior will be discussed.

• McPherson, A.1, D. Cook2, and C. Taggart1,2. 1Department of Oceanography, Dalhousie University, Halifax NS. 2Marine Gene Probe Laboratory, Dalhousie University, Halifax NS
• GENETIC STRUCTURE OF SCOTIA-FUNDY ATLANTIC HERRING (CLUPEA HARENGUS) BASED ON MICROSATELLITE VARIABILITY (O)
• Genetic studies, based on allozyme and mtDNA methods reveal limited discrimination among groups of northwest Atlantic Herring (Clupea harengus). Recent data based on microsatellite loci originally developed for Pacific herring provide evidence of genetically distinguishable herring populations in the northeast Atlantic. In the study reported here, the genetic structure of adult Atlantic herring from Scotia-Fundy spawning locations is quantified using 5 highly polymorphic Pacific herring microsatellite loci (Cha17, 20, 63, 113, and 123) that include one not previously used with Atlantic herring. Pacific and northeast Atlantic herring (courtesy of P.Shaw, Univ. Hull) are compared to Scotia-Fundy herring using microsatellite allele frequencies as well as Fst and Rst measurements to estimate the degree of genetic differentiation among geographically isolated groups. As contemporary management schemes in the Scotia-Fundy region are partly based on geographic location, and because simple geographic isolation and more complex oceanographic processes (currents, bathymetry) can both serve as mechanisms for population structuring, genetic structure in herring is assessed within and among the management units to evaluate the appropriate scale for population-based management. Patterns of genetic differentiation are compared to geographic distance (Mantel test) in order to assess the significance of simple geographic separation in determining population structure. The results and patterns identified provide insights to genetic structuring processes in Atlantic herring that are testable using the recent reappearance of larvae from herring spawning offshore on Western Bank of the Scotian Shelf and cohorts of larvae from different sources that are advected along the Nova Scotia coastal current. Funded by NSERC and The Pelagics Research Council.

• Mellina, E. and, S. G. Hinch. Forest Sciences Department, University of British Columbia, Vancouver, B.C.
• PHYSIOLOGICAL STRESS RESPONSES OF STREAM DWELLING RAINBOW TROUT TO CLEARCUT LOGGING IN THE CENTRAL INTERIOR OF B.C. (O)
• In British Columbia coastal watersheds, clear-cut logging to streambanks often causes dramatic changes to fish habitat (and to fish populations). However, little is known about impacts of clear-cut logging to fish and habitat in the streams of the province's central interior, although we speculate that impacts to habitat may be relatively subtle. To detect effects of logging on rainbow trout (Oncorhynchus mykiss), we focussed attention on physiological stress indicators. Stress affects a variety of functions including reproduction, growth and resistance to disease. In fall 1996, we sampled trout and assessed their habitat in 15 streams: 5 unlogged - controls; 5 newly logged - clear-cut to streambank within last five years; 5 older logged - clear-cut to streambank 15-20 years prior. Acute stress indicators (levels of hematocrit, plasma cortisol, glucose, and chloride) and impaired plasma cortisol response (a chronic stress indicator) showed no affect of logging. However, fish from newly logged streams had large interrenal nuclear diameters relative to fish in control or older logged streams, suggesting that fish in newly logged streams were exposed to some type of chronic stressor. We suspect that poorer quality of habitat within newly logged streams may have affected fish feeding and refuging, thus leading to relatively higher stress levels.

• Mercer, J., A. Hunter, J. Hunter, J. Leonard and D-P. Achneepineskum
• LONG-TERM ECOLOGICAL PROJECTS IN THE RAINY RIVER WATERSHED (O)
• The Rainy River Watershed is a distinct ecoregion which lies between Lake of the Woods and Rainy Lake in northwestern Ontario and Minnesota. Rainy River First Nations have a spiritual responsibility to the watershed and are maintaining our historic guardianship by establishing a number of long-term Ecological Projects in the Rainy River watershed, on both sides of the border: 1) Compiling an inventory on all available information on the watershed; 2) Supplementing inventory with field checks, such as aerial surveys for bird nests.; 3) Implemented self-imposed moratorium on commercial fishing and logging; 4) Founded a successful Lake Sturgeon Hatchery; 5) Initiating a 5 year genetic study of Lake Sturgeon from the Rainy River and from the hatchery: 6) Continuing to establish a baseline data set on three of Rainy River's using the Stream Assessment Protocol for Ontario; 7) Continuing the rehabilitation of spawning habitat in the Pinewood and Sturgeon Rivers; 8) Monitoring Rainy River's tributaries every week in an effort to identify leaky sewage disposal systems; 9) Designed an environmental effects monitoring of the pulp and paper mills on the Rainy River; 10) Implemented a recycling and composting program, weekly waste audits, river clean-up, and erosion control at a community level. We feel there is opportunity to establish partnerships with individuals and organizations outside of our watershed and we would like to foster these relationships.

• Morrison, Bruce J. and Stephen S. Crawford. Axelrod Institute of Ichthyology, University of Guelph, Guelph, Ontario.
• APPLICATION OF TIME SERIES ANALYSIS TO LAKE WHITEFISH COMMERCIAL CATCH DATA.(O)
• Commercial catch data for a lake whitefish fishery in Lake Huron is analyzed as a time series. The time series has irregularly-spaced gaps, spanning from 1984 to 1997. The data for 1997 are also separated into two groups of vessels; punts and tugs. Auto-regressive (AR), moving average (MA) and auto-regressive integrated with moving average (ARIMA) models, are fitted to the harvested effort and weight data from daily catch records. For 1997 data, time series analysis is first applied to the data pooled among vessel types, and then separately by vessel types. Time series analysis of commercial data provides a useful method for analyzing commercial catch data because it not only accounts for the lack of independence among fishing events, but also allows for prediction of future lake whitefish harvests, at least for the short term (1-5 years).

• Mushens, C. J. and J. R. Post. Department of Biological Sciences, University of Calgary, Calgary, Alberta.
• INVESTIGATION OF THE INFLUENCE OF THE PRESENCE OF ADULT BULL TROUT SPAWNERS ON THE BEHAVIOUR AND MIGRATION OF JUVENILE BULL TROUT IN SMITH-DORRIEN CREEK. (O)
• Lower Kananaskis Lake, located in Peter Lougheed Provincial Park, Kananaskis Country, contains a native adfluvial population of bull trout. Smith-Dorrien Creek, which flows in from the northwest, is the only suitable spawning habitat for this particular population. Juvenile adfluvial bull trout generally reside within their natal streams for the first two to three years before migrating downstream to the lake. It has been noted that the majority of the downstream migration of age 2 and 3 juveniles begins around mid-August once adults have migrated upstream to spawn. It is hypothesized that this migration of adults upstream has a strong influence on the initiation of juvenile migration due to the adults displacing the juveniles out of their habitat and also the increased risk of predation. In order to test this, ten telemetry tags were implanted in juvenile bull trout (size range 160-190mm); five on August 6, 1998 and five on September 8, 1998. Individuals were tracked on a daily or bi-daily basis to monitor movement, diurnal habitat use and proximity and behaviour to adults. Results form the initial five tags showed that juveniles maintained small home ranges within the creek before adults moved upstream. Upon the incidence of adults within the creek, two juveniles proceeded to migrate downstream and three remained within the creek, of which one was consumed by an adult bull trout. As the adult population continues to increase in number, the presence of adults in the creek during the fall spawning run may limit juvenile densities.

• Norton, G. Matthew B., Daniel Boisclair, Pierre Legendre and Bernadette Pinel-Alloul Département de Sciences Biologiques, Université de Montréal, C.P. 6128, succursale Centre-ville, Montréal, Québec H3C 3J7
• THE SPATIAL DISTRIBUTION OF ZOOPLANKTON IN THREE LAURENTIAN LAKES: THE EFFECTS OF MACROPHYTES AND FISH (O)
• Investigations into the spatial distribution of zooplankton have considered the role of either macrophyte or fish distribution, but rarely of both factors simultaneously. In order to determine the effect of each and the possible interaction between them, three Laurentian lakes were studied in the summer of 1998. The lakes varied by their degree of residential development and by their macrophyte and fish communities. A roughly 20000 m2 section of each lake was sampled intensively using a systematic-cluster sampling design in order to maximize spatial extent (200 m) while at once capturing small scale dynamics (10 m). The sampling grid was positioned to cover both the littoral and pelagic zones. Each lake was sampled during both day and night in one 24 hour period, using a submersible centrifugal pump. The zooplankton density, biomass and size spectrum at each site were measured using an Optical Plankton Counter. Once during the summer, the macrophyte distribution within the sampling area was determined by direct observation. Throughout the summer, day and night fish distributions were obtained using a horizontal scanning single beam hydroacoustic system. Zooplankton density was found to display significant spatial autocorrelation at 10 and 50 m scales, and significant differences between the littoral and pelagic zones were observed. The results of this study can help to elucidate the predator-prey relationship between fish and zooplankton, and the relative importance of macrophyte beds as refuges.

• Paradis, A.R., and P. Pepin. Ocean Sciences Centre at Memorial University of Newfoundland, St.Johns, Newfoundland, Canada, A1C 5S7, E-mail: aparadis@morgan.ucs.mun.ca; and Department of Fisheries and Ocean, PO Box 5567, St John's, NF, A1C 5X1, E-mail: pepin@athena.nwafc.nf.ca.
• SIZE-SELECTIVE VULNERABILITY TO PREDATION OF MARINE FISH LARVAE: CONFRONTING PREDICTIONS OF AN INDIVIDUAL-BASED SIMULATION MODEL TO FIELD SAMPLES OF AN ICHTHYOPLANKTON COMMUNITY OF CONCEPTION BAY, NEWFOUNDLAND.(O)
• We investigate the effects of encounter and susceptibility in shaping the size distribution of fish larvae. We demonstrate that the two components of vulnerability (encounter and susceptibility) counteract each other which leads to a lack of size-selective survival for fish larvae in our simulation. The implications of our findings suggest that it may be difficult to detect the impact of size-dependent processes in the field. A better understanding of factors influencing encounter rates represents a critical element in extrapolating laboratory studies of predation to the field. Can we find evidence of size-dependent predation in field populations of fish larvae? To answer this question we contrast changes in the length frequency distribution of cohorts of larval fish, obtained from field samples, with predictions based on an individual-based model. We use observations of the predator community to assess whether losses due to predation could be the primary cause of possible discrepancies between observed and predicted length frequency distributions. By contrasting model predictions with observed size frequency distributions, we found that the stochasticity of field samples are too important to allow possible detection of size-selective mortality. We elaborate on the conditions that would allow the detection of size-selective mortality and on the power of such analyses.

• Paul, A.J. and J.R. Post. Department of Biological Sciences, University of Calgary, Calgary, Alberta
• STOCK-RECRUITMENT RELATIONS IN POPULATIONS WITH INTERACTING COHORTS (O)
• The importance of stock-recruitment (SR) relations is well recognized in fisheries biology. Often, simple single-curve relations are assumed with the best known examples being Ricker or Beverton-Holt SR relations. However, empirical data on stock and recruitment rarely show such simple relations. While others have shown from stochastic grounds why empirical SR relations maybe difficult to determine, we present an entirely deterministic model that results in SR relations that are not single-curved relations. Our model is no different than the classic model which gives rise to the Beverton-Holt SR relation except that we have allowed the juvenile cohorts to interact. Through both numerical and analytical exploration, we show that a range of SR relations can be obtained depending on only a few parameters. These findings have important implications at the population level. Processes that always give rise to dynamically stable populations for non-overlapping cohorts can lead to population instability when cohorts overlap. A key component to the instability of these populations is thetime period in which cohorts interact. The longer the time cohorts overlap the more likely the dynamics of a populations will show instability. We conclude that fish species with long development times and where juvenile life stages are forced to interact may exhibit population instability (e.g., damped oscillations, quasi-cycles or limit cycles). Finally, we present examples from freshwater salmonids where such conditions may exist.

• Plate, E. and C.W. Hawryshyn. Department of Biology, University of Victoria, British Columbia
• HORMONAL INFLUENCE ON IMPRINTING IN SOCKEYE SALMON (ONCORHYNCHUS NERKA)
• It is commonly accepted that the accuracy of homing in anadromous salmonids is based upon the functionality of their highly sensitive olfactory organ. It has been shown that the internalization of stimuli (imprinting), i.e.chemical properties of the prevailing environment can only be accomplished at certain time periods or windows of opportunity. Hormonal fluctuations are known to open and close these windows in0 many animals. In this project, we attempt to find evidence for a direct link between thyroid hormones and imprinting performance. Sockeye juveniles were therefore exposed to odours at different stages before and during smolting. Simultaneously, thyroxin levels were artificially increased in selected groups. The fish were subsequently reared until maturation and tested for odour recognition in two behavioural setups. Thyroxin levels were monitored at all stages of the project. The behavioural tests provided strong evidence for the modulation of imprinting performance by thyroid hormones. The hormones might have an effect on memory storage, the sensitization of the olfactory system or both. Neuroethological (heart-rate-conditioning and compound action potential recordings) experiments are currently being employed to establish the neural basis for olfactory imprinting. A greater understanding of the imprinting process could make it possible to use artificial additives in enhancement facilities like spawning channels or hatcheries to imprint sockeye fry at emergence. Then, enhanced adults returning to these enhancement facilities could perhaps be attracted to and concentrated at particular fishing locations apart from the wild stocks within a bigger watershed using the same chemical, that was added at the enhancement facility. The non physical separation of enhanced stocks from wild stocks could not only lead to a more species selective fishing but could also establish a stock selective harvest within a given species like sockeye. This research was supported by The Git'ksan and Wet'Suet'En Watershed Authorities, Skeena Green Plan, The Science Council of British Columbia and the Department of Fisheries and Oceans, Canada.

• Post, J.R. and E.A. Parkinson. Biological Sciences, University of Calgary, Calgary, Alberta and B.C. Ministry of Environment, UBC, Vancouver, British Columbia.
• ENERGY ALLOCATION STRATEGY IN YOUNG FISH: THE ALLOMETRY OF LIPID STORAGE (O)
• Energy allocation strategy of young fishes can be inferred from the allometry of lipids and other body constituents. In age-0 rainbow trout, smaller and slower growing individuals allocated proportionately less energy to storage in the form of lipids. At body sizes greater than approximately 10 cm this size differential in energy allocation was significantly reduced but was still hyperallometric. There was a similar allometric pattern for non-lipid structural components (i.e. proteins, bones and scales) as measured as non-lipid dry mass but this allometry, and the size based inflection in allometry, was weaker than in lipids. Our data suggests that the net growing season size-dependent pressure imposed by piscivory on the smallest age-0 rainbow trout is approximately 2.4 times higher than that on the largest individuals and that the selective pressure to accumulate lipid reserves in small fish was approximately 6 fold greater in small fish relative to large ones.

• Reinhardt, U.G., Yamamoto, T. and Nakano, S. Experimental Forests, Hokkaido University, Japan
• INTERACTIONS BETWEEN DOMESTICATED AND WILD MASU SALMON JUVENILES UNDER FIELD CONDITIONS (O)
• We observed survival and growth of age-0 Masu salmon (O. masou ) from a a fully domesticated and a wild population in large stream enclosures. The mixed groups of juvenile salmon were subjected to the manipulation of food availability through exclusion of invertebrate drop and to the presence/absence of predators. Under predation pressure by adult Hucho (Hucho perryi), the wild fish suffered a slightly higher mortality than the domesticated fish. Despite lower final densities, the growth rate of both types of juvenile salmon was depressed under predation compared with the control treatment. Irrespective of predation pressure, the domesticated fish outgrew their wild counterparts. Domesticated Masu juveniles maintained high growth rates even under the low-food treatment which caused a sharp decline in growth of the wild fish. This suggests that the domesticated fish excluded the wild fish from access to food. In accompanying laboratory experiments, domesticated Masu juveniles dominated over wild fish in pairwise encounters and showed a greater willingness to feed in presence of predator cues. These experiments provide an example of sublethal effects of piscivorous predators on their prey under natural conditions. Furthermore, our observations show that the the intentionalor accidental release of artificially enhanced salmonids into the wild can have detrimental effects on wild salmon stocks.

• Reist, J.D.1, M. Power2, J.B. Dempson3, and G. Power4. 1Dept. of Fisheries and Oceans, 501 University Crescent, Winnipeg, MB, R3T 2N6. 2Dept. of Agricultural Economics, University of Manitoba, Winnipeg, MB, R3T 2N2. 3Dept. of Fisheries and Oceans, St. John's, NFLD, A1C 5X1. 4Dept. of Biology, University of Waterloo, Waterloo, ON, N2L 3G1.
• THE INFLUENCE OF LOCAL ENVIRONMENTAL VARIATES ON BIOLOGICAL VARIABILITY OF ANADROMOUS ARCTIC CHAR (SALVELINUS ALPINUS) FROM NORTHERN LABRADOR.
• We investigated the significance of environmental variates as explanations for biological variability from 1976 to 1997 in an anadromous stock complex from northern Labrador. Exploitation and/or environmental forcing factors not examined here may mediate change in the biological parameters, therefore, where necessary first order temporal trends in the data were standardized by regressing the fisheries data series against time and subtracting this result from the raw data. The raw (age) or de-trended (length, weight, condition) data were then regressed against a suite of environmental variates to assess the influence of environment on the biological variables. Mean age was negatively associated with cumulative summer precipitation lagged by four years, mean summer sea surface temperature, and January precipitiation all of which enhance early survival thus aiding earlier recruitment to the fishery. Age was positively associated with summer salinity and air temperature both of which likely favour larger, thus older, fish. Mean length was negatively related to spring air temperature perhaps by causing earlier migration to the sea which increases osmotic stresses thus decreasing growth. Length was positively related to summer air temperature, July precipitation, average winter air temperature, and average summer precipitation lagged by seven years which likely enhance growth directly or increase local productivity thus indirectly enhance growth. Mean weight was positively associated with length of the fish (as expected) but also with August sea surface temperature, average summer precipitation and air temperature both lagged by 4 years, all of which optimize growth especially in the critical first year at sea (age 4). We conclude that significant effects of short-term climate history are observable in relevant biological parameters in this species. Although subject to much variation due to local biological responses, this work can lead to better understanding of the range of possible responses to climate change scenarios for this species and northern fishes generally.

• Rosenfeld, J.S. B.C. Ministry of Fisheries, Research Branch, 2204 Mail Mall, Vancouver, B.C. V6T 1Z4
• SENSITIVITY OF ANADROMOUS CUTTHROAT TROUT TO THE EFFECTS OF LOGGING (O)
• Juveniles of sea-run coastal cutthroat trout rear in small coastal streams for two to three years before smolting. Associations of juveniles with different habitats (e.g. pools vs. riffles) and the effect of LWD on channel structure were evaluated in order to assess the long-term impact of different riparian management practices on cutthroat populations. Cutthroat densities were highest in smaller streams (less than 4 m bankfull width), and smaller parr (0+ and 1+) were typically associated with shallower hydraulic units, while parr larger than 100m were usually found in pools. High LWD loading increased pool frequency and maximum depth, and LWD was responsible for more than 50% of pool formation in most streams. This indicates that long-term depletion of LWD by removal of riparian forest may seriously reduce the capacity of smaller streams to support juvenile cutthroat.

• Shaw, Al and John Richardson Department of Forest Sciences, University of British Columbia, Vancouver, British Columbia
• LIKE SAND THROUGH THE HOURGLASS; THESE ARE THE EFFECTS OF SEDIMENT PULSE DURATION ON RAINBOW TROUT GROWTH AND MORTALITY AND INVERTEBRATE ASSEMBLAGES IN AN INTERIOR B.C. STREAM (O)
• Sediment is a common feature to all stream ecosystems, but the introduction of excess sediment through riparian land-use activities (forest harvesting, livestock grazing) has the potential to negatively impact stream biota. Models predicting the response of invertebrates and fish to sediment events have largely focussed on sediment concentration, ignoring pulse duration. The purpose of this study was to address the effects of sediment pulse duration exclusively and through the use of flow-through stream troughs (n=14) concurrently assess the impact of introduced sediment on invertebrate assemblages and rainbow trout fry. Treatments consisted of sediment pulses ranging in duration from control (no addition) to six hours, applied every second day for a total of 21 days. Trout growth (length and weight) showed a negative relationship (length R2=0.69, p<0.0005 weight R2=0.70, p<0.005) with sediment pulse duration while mortality exhibited no relationship with treatments (p>0.78). Both length and weight were significantly impacted by the three hour treatment. Trout condition (ANCOVA, weight as covariate) did not change with treatment as both length and weight were negatively affected. Preliminary analysis suggests invertebrate abundance and richness decrease as treatment duration increased. These data provide evidence that sediment events should be monitored for concentration as well as pulse duration, to more accurately predict the biological response of stream biota. Also, this design allows testing a series of causal mechanisms describing direct (visual impairment, behavioural modification) or indirect (reduction of invertebrate prey) effects of suspended sediment on rainbow trout.

• Surette, C.1, G. Brun 2 and V. Mallet 3, 1Faculté des études supérieures et de la recherche, Université de Moncton, Moncton, NB; 2Ecosystem Science Division, Environmental Conservation Branch/ CWS, Environment Canada, Moncton, NB; 3Faculté des sciences, Université de Moncton, Moncton, NB.
• MERCURY UPTAKE IN AQUATIC SPECIES FROM THE RICHIBUCTO ESTUARY (O)
• In 1996 the Richibucto Environment and Resource Enhancement Project (REREP) was officially launched. REREP encourages partnerships and community involvement. The main objectives of REREP are to prevent environmental damage, to preserve and to rehabilitate ecosystems within the Richibucto River watershed. A study on mercury uptake in estuarian biota is being conducted as part of REREP. This study consists of two elements. The initial part of the study consisted of analyzing mercury content in four aquatic species. The species studied are: sand shrimp (Crangon septemspinosa), mummichog (Fundulus heteroclitus), Atlantic tomcod (Microgadus tomcod) and American eel (Anguilla rostrata). Three sites, representative of the Richibucto Estuary, were chosen. Mercury concentrations ranged from 0.005 mg/kg to 0.086 mg/kg. Observations suggest that spatial variations exist within the estuary and in comparison with other estuaries in the province. Also, results indicate that mercury concentration increases with trophic levels. The second part of the mercury study, currently in progress, consists in studying the effects of a peat spill, in Mill Creek, on the bioaccumulation of mercury in three aquatic species. Sand shrimp and mummichog, which are abundant and sedentary in Mill Creek, were chosen. In addition, blue mussels (Mytilus edulis), which are universal indicators and absorb contaminants very fast, were introduced in cages. Mercury concentration will also be determined in water and sediments. The methods used are cold vapor atomic absorption spectrometry, for fish samples, and cold vapor atomic fluorescence spectrometry, for water and sediment samples.

• Trivedi, Tanya. , Henderson, B. and N.C. Collins. Department of Biology, University of Toronto at Mississauga, Mississauga, Ontario
• DO ENERGETIC CONSTRAINTS LIMIT FREQUENCY OF REPRODUCTION OF YELLOW PERCH (PERCA FLAVESCENS)? (O)
• Yellow perch (Perca flavescens) abundance in Lake Erie has been on the decline since the late 1980's and annual recruitment has also varied considerably during this time. A spawning event has great energy requirements and therefore the ability of an individual fish to recover from such an event could affect it's ability to spawn yearly and produce intermittent spawning. Recruitment then would depend upon the proportion of mature females that could develop gonads in a given year. A substantial proportion of mature yellow perch females from Lake Erie have resting gonads prior, during, and after the spawning season. These resting gonads are substantially lower in lipid and energy density and are much smaller in size than normal gonads. We explore an energetics mechanism to explain intermittent spawning by measuring energy and lipid density in somatic and gonadal tissue monthly for a year, to determine when and how energy is allocated to reproduction.

• Tucker, S.1, J.B. Rasmussen1, Z. Weng2, N. Mookerji2, and A. Mazumder2. 1McGill University, Dept. of Biology, 2Université de Montréal, Dept. of Biology
• FOOD EXPLOITATION RATES OF SALMONIDS IN STREAMS-A REAPPRAISAL OF THE ALLEN PARADOX
• Salmonid feeding rates were coupled with density estimates to derive total fish exploitation rates for two streams with different hydrological regimes. Feeding rates were measured for juvenile Atlantic salmon (Salmo salar) and brook trout (Salvelinius fontinalis)in the field with the 137Cs mass balance method. Independent estimates of benthic invertebrate standing stocks showed that fish were consuming between 32% and 47% of invertebrate production or half of the invertebrate P/B. The application of age- and site- specific feeding rates solve a long standing paradox in stream ecology, the Allen paradox, as all previously inferred salmonid exploitation rates have been in excess of prey turnover. The discrepancy in salmonid densities, population structures and exploitation rates between the two streams, despite similar food resources, points to physical habitat criteria and the link to food availability as determinant in production. The potential relation between habitat, food !availability, and intrinsic energy demands of fish of varying life-history strategies was subsequently evaluated.

• M. Jake Vander Zanden1, John M. Casselman2 and Joseph. B. Rasmussen1, 1Dept. of Biology, McGill University, 1205 Ave. Dr. Penfield, Montreal, PQ, H3A 1B1 Canada. 2Ontario Ministry of Natural Resources, Fisheries Research, Glenora Fisheries Station, RR 4, Picton, ON, K0K 2T0, Canada
• FOOD WEB SHIFTS FOLLOWING BASS INVASION OF LAKES AS MEASURED USING STABLE ISOTOPES
• Two species of littoral piscivores, smallmouth bass (Micropterus dolomieui) and rock bass (Ambloplites rupestris) have recently invaded a number of lakes across central Ontario. We found that invaded lakes had substantially reduced prey fish diversity and abundance. Stable isotope ratios of nitrogen and carbon indicated that lake trout (Salvelinus namaycush) from invaded lakes exhibited reduced piscivory, resulting in a shorter food chain leading to lake trout. Furthermore, stable carbon isotope ratios indicated that invasion was accompanied by a trophic habitat shift from littoral to pelagic prey. Littoral predator-induced food web alterations diminished the flux of energy and materials between littoral and pelagic habitats, which may have serious implications for native populations of lake trout.