Microgeographic population structure and population life history variability was examined in wild brook charr (Salvelinus fontinalis) from the Indian Bay watershed in northeastern Newfoundland. Assessment of population structure over two years, based on five microsatellite loci, indicated that individual lakes supported reproductively isolated populations. Life history characteristics showed significant variability among three of the genetically distinct populations. Population differences in life history can be explained as optimal solutions to population-specific mortality and growth regimes. These results suggest that wild brook char populations, via selection and/or phenotypic plasticity, may be adapted to their local environment on a microgeographic scale. [Fri PM2]

Haddock have been previously shown to produce sounds during mating. Sounds are produced by the rapid compressing action of drumming muscles located around the swimbladder. In this study, several hypotheses were tested to further investigate the sound patterns generated by courting haddock in captivity. We assessed whether: (i) single males or females generate sounds; (ii) sound production becomes altered with the introduction of an additional male; and (iii) startling fish influences sound production. Analyses were carried out from March 30-June 11, 1999 at the St. Andrews Biological Station. Sounds generated by males in spawning condition were categorized as knocks, hums and an intermediate between these two types. Solitary males and females did not produce sounds. Sounds produced by males were not necessarily elicited by egg release, but occurred even in the absence of the spawning act and in the absence of other spawning males. Sound production began 10.3 +/- 15.9 min. following a disturbance with only a single male and female present and 5.3 +/- 6.0 minutes with two males and a single female present. [Sat AM2]

Loons breeding in lakes of the Maritime Provinces have the highest mercury levels in North America. To understand mercury pathways of these ecosystems, a survey of fish communities was conducted in 19 lakes of Southwestern New Brunswick in 1997. Total mercury of fish (primary, secondary and tertiary consumers) ranged from 0.02 to 1.09 ug/g wet weight, and varied significantly by species and lake. Water chemistry and lake morphology account partially for this variation; to further resolve the influence of trophic structure, seven lakes were surveyed in depth in 1998. Analysis of both total and methylmercury determined proportions of bioavailable mercury at each trophic level, while invertebrates collected established baseline levels for inter-lake comparisons. Trophic positions of fish were assigned based on stomach contents and stable isotope (C and N) analyses. Relationships between assigned trophic position and mercury levels will be presented, to demonstrate the effect of fish community structure on mercury levels within lakes. Results of this study will be incorporated into predictive models of mercury concentrations in N.B. wildlife. [Sat AM2]

The reproductive potential of a fish stock can be estimated by determining the fecundity - body size relationship of a population using linear regression and the general fecundity equation: F = a + X b (where F is fecundity, a is an intercept, X is usually weight or length, and b is an exponent). Variation in fecundity among populations is common for fish species due to stock-specific characteristics affecting fecundity (genetics, temperature, food availability, population density, age/size-structure, and condition). Preliminary data on the Eastern Scotian Shelf haddock show that this stock has a much lower predicted fecundity at body size and greater intrapopulational variability than other historical studies on haddock in the North Atlantic. This could be the result of several interacting factors such as those alluded to above. This is the first study on the fecundity of Eastern Scotian Shelf haddock. Interannual variation is examined over a three-year period and compared to historical data on the neighboring Southwestern Scotian Shelf stock. Contemporary data on the Southwestern Scotian Shelf is also examined, to determine whether there has been a significant change in the reproductive potential since the last decade. Factors affecting variation in fecundity (condition factor, size at age, reproductive effort, mixture/ overlapping of stocks) are examined using multivariate analysis (ANCOVA). This fecundity data may better explain the highly variable recruitment patterns of the haddock stocks (particularly the Eastern stock having undergone drastic decline in abundance and present commercial fishery moratorium) on the Scotian Shelf, and could be a more sensitive approach for recruitment modeling than the traditionally used spawning stock biomass model. [Sat PM2]

It is now widely accepted that whenever necessary, supportive stocking should be made with progeny produced from breeders issued from the same population being stocked. Much less attention as been paid to the effects of the numbers of breeders and progeny being used in such practice. Using an analytical approach, Ryman and Laikre (1991) demonstrated that this could lead to important reduction of effective population size, depending on the relative proportion of captive breeders/progeny relative to the wild population. This approach was however limited to making predictions for a single event of stocking and oversimplified several demographic processes. In this paper, we extend the work of Ryman and Laikre by developing an analytical approach derived from inbreeding theory that we used to make predictions, relative to the effects of various scenarios of supportive breeding on the resulting evolution of inbreeding coefficient in wild populations. [Fri PM2]

Models that predict the effect of changes of environmental conditions on fish communities focus on variables integrated over time and space (total fish yield or production, average phosphorus concentration, mean depth, degree× days). These models reveal general rules on how ecosystems work. However, perturbations do not affect all parts of an ecosystem equally, and all parts of an ecosystem do not equally contribute to maintain fish communities. In the context of the increasing pressure to use natural resources and to modify habitats, procedures to identify areas of key importance for fish communities must be developed. Temporal and spatial scales at which fish and habitats should be studied may be obtained by a better description of fish distribution. Once these scales are identified, ecosystems may be represented as a 3D mosaic of micro-habitats defined using temporal and spatial scales relevant to fish. The ecological value of each micro-habitat may be defined by its potential for reproduction, survival, and growth. Bioenergetic models may be a valid framework to assign values of ‘potential for growth’ to micro-habitats because they can function at various temporal and spatial scales. While objectives are simple to set, making models operational is not.

Our study investigates morphological differences among populations of rock bass (Ambloplites rupestris) captured from lake and stream ecosystems of two central Ontario watersheds. Twenty-six external morphological measures were taken on the rock bass sampled, including five measures of body width, a dimension that is often overlooked in morphological assessments. Discriminant function analysis revealed significant body shape differences in both male and female rock bass. Stream fish were found to be more fusiform (i.e. relatively longer with a reduced body depth), had a longer caudal peduncle and had lateral fins that were more anteriorly placed on their body than lake fish; while lake fish had longer and wider lateral and median fins. We suggest that stream fish use these body form traits to adapt to swimming and foraging in the more hydrodynamically arduous conditions of their environment. Other habitat features (beyond the presence or absence of flowing water) and differences in the fish community structure of the sample sites did not appear to affect rock bass morphology. [Sat PM1]

In order to diversify and expand the product base of aquaculture in Canada, new species are being investigated for their potential. Species such as rainbow trout and Atlantic salmon have a relatively long history of domestication and have attained significant commercial production. Other non-salmonid species, such as the European turbot, sea bass and sea bream have reached impressive production numbers in Europe. In Atlantic Canada, a number of new marine species have been investigated over the last number of years. During the process of domestication, a number of biological realities have to be addressed including broodstock management, juvenile production and ongrowing. In this talk, I will address the area of juvenile production, specifically larval performance. It is during the larval stage that most mortality occurs, and it is here that we need to provide an "ecology" which maximises growth and survival. In achieving this, we can begin to understand some of the ecological constraints which have shaped the behaviour and survival of larvae in the sea. [Sat AM1]

We examined sperm motility and swimming characteristics for shortnose sturgeon (A. brevirostrum) and Atlantic sturgeon (A. oxyrhynchus). Sperm was collected from three shortnose sturgeon males and five Atlantic sturgeon males captured by gill nets during their respective spawning seasons; May-June for shortnose sturgeon and July-August for Atlantic sturgeon. Sperm collected was stored in plastic vials in a cooler with a temperature between 3-4° C. Samples of sperm were placed on a neubauer hemocytometer and videotaped using a camera attached to a B-MAX microscope under 100x total magnification. Percentage of sperm motile after activation by water, was followed through time. We also examined swimming characteristics of the sperm: maximum speed, mean speed, distance traveled and path linearity. Preliminary results indicate that Atlantic sturgeon sperm, on average, remained motile for a longer period of time than did the shortnose sturgeon sperm, however, all sperm decreased in average swimming speed over time. Also, over the period of observation, a greater percentage of Atlantic sturgeon sperm were active. This data will be used to examine the potential for promiscuous mating of sturgeon in the wild. [Poster]

This study focuses on the first year of life of Atlantic salmon in the Western Brook system, Gros Morne National Park (Newfoundland). First, I will look at the egg-to-fry survival by comparing the number of eggs deposited in the stream in autumn with the number of young-of-the-year (0+) that have survived to late summer. The proportion of the 0+ population that have survived but drifted out of the system after emergence (June) will also be assessed. Little is known about the mechanisms regulating this strong tendency of 0+ to drift soon after emergence and that can greatly influence survival within a population. I am hypothesizing that there is a difference between drifting fry and resident fry. I predict that drifting fry are weaker and younger since it is thought that they are forced out of suitable habitats by competition after emergence. If true, the difference should be reflected in their size and age, determined by otolith reading. [Fri PM2]

Habitat quality for salmonids in river is often described 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 (NEG). 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 develop a methodology to evaluate the energetic benefits and costs in the field for juveniles of Atlantic salmon. As an illustration of this methodology, this presentation presents data for one day of sampling. To evaluate the benefits, five fishes were captured, to collect their stomach contents, at each 4 h intervals on a period of 24 h. Daily ration was calculated using the Eggers model. Four times during the 24 h survey and during 1 h each time, we used four underwater strereovideocameras to evaluate fish behaviour. Image processing has been done to measure the time budget and swimming speed for the salmon during all the films. The effective swimming speed is obtained by accounting for the ambient water velocity gradient. To calculate the energetic cost of swimming, we used an empirical relationship that transforms effective swimming speed in oxygen consumption. Other energetic costs have been calculated using a bioenergetic model. Net energy gain for one day was determined as the difference between daily ration and daily costs (swimming and others). This model will be useful in habitat quality models to provide energetic perspective.

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 de développer une méthodologie pour mesurer les gains et les coûts énergétiques sur le terrain, pour des tacons de saumon atlantique. À titre d’illustration, nous présentons les données pour une journée d’échantillonnage. L’évaluation des bénéfices a été effectuée en récoltant le contenu stomacal de 5 tacons capturés à chaque 4 h, durant 24 h. Le modèle de Eggers a été utilisé pour calculer la consommation journalière. À quatre reprises pendant la journée, 1h à chaque fois, nous avons filmé les poissonsaà l’aide de 4 stéréovidéocaméras sous-marines. Le budget emporel et les vitesses de nage des tacons ont été évalués par analyse d’images, pour l’ensemble des films. La vitesse de nage réelle est obtenue en tenant compte de la vitesse du courant à laquelle ont fait face les tacons. Les coûts énergétiques associés à la nage ont été calculés en utilisant une relation empirique entre la vitesse de nage et la consommation d’oxygène. Les autres coûts énergétiques proviennent d’un modèle bioénergétique. Le gain énergétique net pour une journée est la différence entre la consommation journalière et les coûts pour une journée (nage et autres). Ce modèle sera utile pour donner une perspective bioénergétique aux modèles de qualité d’habitat. [Fri PM1]

Atlantic salmon returns in eastern North America have decreased by one-half to two-thirds in the past 20 years. This underlines the need for a new understanding of the ecology of salmon at sea, where the increased mortality responsible for the decline appears to have occurred. Marine-phase Atlantic salmon grow remarkably rapidly, with weight increasing about 75-fold between river exit and return as 1SW adults. These growth rates are much faster than those of other marine fish occupying the northwest Atlantic. Fish growth depends chiefly on temperature and food availability. It is proposed that salmon achieve rapid growth by behavioural thermoregulation, occupying warm surface waters to increase body temperature in a manner akin to basking reptiles, and visiting lower levels as needed to obtain food. There is no cover in the surface layer, and salmon do not form dense schools that reduce predation risk as do other pelagic fishes (capelin, mackerel). Consequently, salmon are subject to high predation mortality. Thus, Atlantic salmon can be viewed as a risk-prone species which trades high annual mortality for rapid growth. Weights of returning Atlantic salmon have not decreased during the period of declining returns. This implies that salmon have a pre-determined target weight at river return, and choose to accept increased mortality risk in order to attain target weights. The decline of Atlantic salmon has co-occurred with increased numbers of endothermic predators (seals and seabirds), decreased numbers of ectothermic predators (groundfish), and episodes of cold water in the Labrador Sea. It is proposed that the decline in salmon returns is due to increased predation in the context of rising endothermic predator numbers and cold water. Predation risk is exacerbated during cold periods by the increased need for basking in the high-risk surface layer in order to maintain growth schedules, and by the advantage that cooler water gives an endothermic predator when pursuing ectothermic prey. [Sat AM1]

Fragmentation and loss of habitat are recognized as the greatest existing threats to biodiversity. Although some degree of fragmentation occurs naturally in aquatic systems, human activities have greatly disrupted habitat connectivity by the construction of weirs, dams, reservoirs and locks. As well as increasing fragmentation, these activities have also impacted aquatic systems via habitat loss and degradation. We are assessing the effects of barrier systems on aquatic biodiversity in a highly fragmented river system (Trent-Severn Waterway) in southern Ontario. By taking a comparative approach at several hierarchical scales, this study will establish explicit links between types of barriers in this cascading reservoir system, as well as the cumulative effects of barriers on fish communities, species, and metapopulations. Preliminary assessment of fish communities using a nested sampling design (river segment, valley segment, site and transect) revealed a diversity gradient with high turnover in species composition. These results will be complemented with genetic analysis of key species, to assess gene flow and metapopulation dynamics among valley and river segments. Together, these findings should provide a quantitative determination of cumulative impacts of barriers and fragmentation on fish communities. [Fri PM2]

Brook charr, Salvelinus fontinalis, exhibit a trophic polymorphism in lakes of the Laurentian Shield (Québec): some individuals feed almost exclusively on benthic invertebrates ("benthic individuals") while others feed almost exclusively on pelagic prey ("pelagic individuals"). This variation in feeding habits has been related to differences in spatial distribution (telemetry), body morphology and coloration pattern. These data on fish diet were obtained from stomach content analyses and are therefore limited indicators of long-term feeding habits. The goal of the present study was to investigate the differential use of benthic and pelagic prey by littoral and pelagic individuals using stable isotopes (d ¹³C and d ¹⁵N) as temporal integrators of diet. Our study was carried out on 9 lakes of the Mastigouche Reserve, Québec, in summer 1999. Preliminary results showed clear differences in the isotopic signature of benthic and zooplanktonic prey. Thus, we predict that the similar differences will be found in brook charr of the two forms if they are faithful to a given type of prey. Results will be discussed at the conference.

L’omble de fontaine, Salvelinus fontinalis, est présent sous deux formes dans les lacs du Bouclier canadien: une forme littorale, qui se nourrit d’invertébrés benthiques et une forme pélagique, qui se nourrit d’organismes zooplanctoniques. Ces différences dans la diète des deux formes ont été obtenues à partir d’analyses de contenus stomacaux. Ces derniers ne donnent cependant qu’une image ponctuelle de l’alimentation des individus. Le but de cette étude vise à confirmer l’utilisation à plus long terme des proies benthiques et zooplanctonique par les formes littorales et pélagiques respectivement, à l’aide des isotopes stables (d ¹³C and d ¹⁵N). Notre étude a porté sur 9 lacs de la Réserve Mastigouche, à l’été 1999. Nos premiers résultats démontrent une différence nette dans la signature isotopiques des proies benthiques et zooplanctoniques. Cette différence devrait se répercuter chez les ombles si ces derniers sont fidèles à l’un des types de proies. Les résultats obtenus, suite aux diverses analyses, seront présentés à la conférence. [Poster]

Acoustic telemetry was used in the autumns of 1996 and 1997, to measure the movements of 29 and 60 Atlantic salmon (Salmo salar L.) presmolt, respectively, in the St. John River system, New Brunswick, Canada. In addition, 60 smolt were followed in the spring of 1998. The objectives were to assess timing of presmolt emigration and determine overwintering positions, rates of downstream fish movement, and losses in the turbines at hydroelectric dams. In the autumn, presmolt moved as far as 120 km downstream from their initial release sites. Some fish overwintered in still water sections of hydroelectric reservoirs located a minimum of 220 km from the ocean. Autumn movements presumably position the fish for a timely exit from the river in spring. However, none of the 60 smolt tracked in spring 1998 passed through the final reservoir and out to sea, and many of them did not make it past the first of the three dams they encountered. The rate of downstream movements of both presmolt and smolt was significantly reduced upon entering two of the reservoirs, perhaps because the loss of currents disoriented them. The acute mortality rate through turbines was 18.2%. [Poster]

We used catch data from index sampling of fish communities of eastern Lake Ontario, to calculate indices to assess annual relative abundance, variability, and a combination of the two to reflect stability, or status. During the past decade, most major species have undergone a dramatic change. Abundance increased to record-high levels in the late 1980s and early 1990s, then decreased abruptly, 2 years earlier in small species than large ones. The period of pivotal change was 1991 to 1993, coinciding with a significant decrease in water temperature from a period of 6 abnormally warm years, 1986 to 1991 (Apr.-Sept.--18.7° C), to 3 abnormally cold years, 1992 to 1994 (17.3° C). Temperature decrease from 1991 to 1992 (2.1° C) was the most extreme in four decades. Intensive rehabilitative stocking of lake trout has helped restructure and mature the cold-water community to a point where large species were at record-high levels in the late 1980s and early 1990s, creating increasing prey demand that required maximum prey production; this was possible in the recent high-temperature regime (late 1980s), but was not sustainable in the low-temperature regime that followed. After 1992, large species lost condition, more dead fish were observed, and abundance decreased. This low-temperature perturbation induced a successional setback (enjuvenation event) in the overall maturity of the ecosystem. [Sat AM1]

Limited dispersal should result in genetic differences among populations proportional to geographic distances of separation. This association between gene flow and distance can be disrupted by; 1) continuing genetic exchange among distant populations, 2) historical changes in gene flow, and 3) physical barriers to dispersal. Accuracy of homing in anadromous brook charr is not clearly known, hampering precise predictions about the spatial scale of genetic structuring. Moreover, direct estimates of gene flow are often difficult to obtain in natural populations and provide only a contemporary picture of migration that may not be accurate in an evolutionary perspective. In this study, we used highly polymorphic molecular tools (six microsatellite loci) for elucidating the evolutionary factors responsible for the observed patterns of genetic variation in brook charr in 36 anadromous populations from eastern Canada (Gaspesia, Bay of Chaleurs, North Shore, Anticosti Island and New-Brunswick). More specifically, we aim to; 1) determine to what extent oceanographic structures act as barriers to gene flow, and 2) estimate the parameters of the migration model that best fits the observed genetic structure. Globally, our results suggest that anadromous brook charr exhibit a strong homing behaviour, even at very limited spatial scales. Moreover, genetic similarity of proximate populations indicate that straying is limited to adjacent rivers. A significant regional component of genetic variation suggests an impact of oceanographic features on the extent of gene flow. One major outcome of this project will be to provide managers the biological basis for the elaboration of a management plan for anadromous brook charr.

Lorsque la dispersion est limitée, on attend une relation proportionnelle entre distance géographique et différenciation génétique. Cette association entre flux de gènes et distance peut être perturbée par; 1) des échanges continus entre populations distantes, 2) des fluctuations historiques du flux de gènes, et 3) des barrières physiques à la dispersion. La précision de la fidélité à la rivière natale (homing) est peu documentée chez l'omble de fontaine, ce qui rend difficile d'émettre des prédictions quant à l'échelle spatiale de la structure génétique attendue. De plus, des estimés directs des flux de gènes sont difficiles à obtenir en populations naturelles et ne fournissent qu'un image figée, qui n'est pas nécessairement pertinente dans une perspective évolutive. Dans cette étude, nous avons utilisé des outils moléculaires hautement polymorphes (six locus microsatellites) pour déterminer quels facteurs évolutifs sont responsables des patrons de variation génétique chez l'omble de fontaine dans 36 populations anadromes de l'est du Canada (Gaspésie, Baie des Chaleurs, Côte Nord, Ile d'Anticosti, , Nouveau-Brunswick). Plus spécifiquement, nous avons cherché à; 1) déterminer à quel point les structures océanographiques locales constituent des barrières au flux de gènes, et 2) estimer les paramètres du modèle de migration qui correspond le mieux à la structuration observée. Globalement, nos résultats suggèrent que l'omble de fontaine anadrome présente une forte fidélité à sa rivière natale, même à des échelles spatiales très limitées. De plus, la plus grande similarité des populations adjacentes indique que les égarements (straying) sont limités aux rivières adjacentes. Une composante significative de la variation génétique est attribuée au niveau régional, ce qui suggère que les caractéristiques océanographiques ont un impact sur l'intensité du flux de gènes. Une application majeure de ce projet sera de fournir aux gestionnaires les bases biologiques nécessaires en vue de l'élaboration d'un plan de gestion pour l'exploitation durable de l'omble de fontaine anadrome. [Sat PM1]

The habitat variables important to the reproductive success of smallmouth bass were modeled. The model incorporates habitat variables (temperature, fetch, substrate and water quality) that are important to spawning and nesting, as well as the growth and survival of the young. Habitat suitability indices were generated for substrate, wind, fetch and water quality using data available from several North American waterbodies. Warming rates can be predicted as a function of air temperature, fetch and mean depth. Once habitat suitability values have been assigned to different lakes or regions within a lake, population data can be included to determine the number of nests that can be supported in the area being modeled. A bioenergetic model then describes growth and survival in the different habitat areas to yield the number of young of year that will survive to the winter. The model allows one to predict the potential production of young of year smallmouth bass within a lake and to estimate the potential effect of habitat alterations. [Sat AM1]

Increases in atmospheric Greenhouse Gases, may change the hydrology of a number of the country's regions. This will have an impact on aquatic and wetland ecosystems as well as municipal, industrial and power generation uses. It is thus important to get an estimate of the potential changes to the Canadian hydrological cycle, in order to make intelligent decisions concerning mitigation factors which society may be forced to undertake. We divided Canada into eco-climatologically similar regions called "Ecozones". We developed two month-stepped temperature-precipitation-runoff models for the country using an artificial intelligence, neural network (ANN) approach. We modified input temperature and precipitation variables in the ANN models, to match those predicted by the Canadian Climate Centre General Circulation Model II for a doubled CO₂ atmosphere and calculated new monthly equilibrium runoff predictions. Our results predict that much of Canada will experience higher annual runoff than is currently the case. The timing of runoff will change significantly in a number of the ecozones as we show that in many regions, peak runoff will occur approximately one month earlier than is currently the case. The ANN model did not work as well for basins in the Prairie ecozone, as we could not develop a good model with data from regulated rivers. [Fri AM]

Native Atlantic salmon were declared extinct in Lake Ontario by 1898. Mill dam construction throughout the water basin, deforestation, and over-fishing all contributed to their extinction. Habitat suitability for Atlantic salmon has improved recently, due to the decline in the number of dams and the increase of bank stabilization along Lake Ontario tributaries. However, the introduction of several non-native salmonids (chinook salmon Onchorhynchus tshawytscha, coho salmon O. kisutch, pink salmon O. gorbuscha, rainbow trout O. myskiss and brown trout Salmo trutta) in Lake Ontario tributaries is of some concern. The objective of this project is to determine if juvenile Atlantic salmon compete for space with other salmonids present in Ontario streams. Baltimore Creek has sympatric Atlantic salmon (re-introduced), brook trout (natural reproduction: n.r.) rainbow trout (n.r.), and brown trout (n.r.) populations. Wilmot Creek has Atlantic salmon (re-introduced), rainbow trout (n.r.), brown trout (n.r.), chinook and coho salmon (n.r.) populations. Underwater observations were conducted during the summer of 1999, to quantify habitat utilization and agonistic behaviour. Electrofishing was conducted in Baltimore Creek to remove all salmonids other than Atlantic salmon from the site. After a period of 2 weeks, underwater observations were repeated to determine if a shift of Atlantic salmon habitat utilization occurs when the number of trout is reduced. Observations on interspecific competition will be presented, to provide further information on the feasibility of the re-introduction of the native, but extirpated Atlantic salmon into the Lake Ontario water basin. [Sat AM1]

The scientific community and general public have numerous concerns surrounding the environmental impact associated with the accidental release of genetically-modified Atlantic salmon. To dampen fears associated with the escapement of modified salmon, sterilization by triploid induction has been suggested. Because of fundamental differences between triploids and diploids, optimizing conditions for rearing transgenic triploid Atlantic salmon will require a comprehensive evaluation of oxygen consumption and haematology in diploid and triploid transgenic and non-transgenic Atlantic salmon. An experiment has been designed to examine differences in standard oxygen consumption rates and haematology between four populations: diploid transgenic, triploid transgenic, diploid non-transgenic and triploid non-transgenic Atlantic salmon. When fish in each group reach an average body weight of ~10, ~25 and ~50g, oxygen concentrations will be recorded in sealed tank respirometers for three replicates of 45-55 fish from each treatment until the O₂ concentration reaches 6 mg/L. At 6 mg/L, tanks will be reoxygenated. This procedure will be performed repeatedly over a 24 h period. At 50 g, 10 fish from each population will have a 200 ml blood sample drawn for hematological analysis, including: erythrocyte number, erythrocyte size and total blood haemoglobin concentration. [Sat AM2]

The social behaviour of Atlantic salmon smolts was evaluated during their migration period under controlled conditions in an experimental stream tank. Agonistic behaviour, dominance, distance to nearest neighbour, and distance from substrate were examined before and after implanting radio transmitters. Two sections within the tank were used: one wide-medium laminar flow where observations were conducted on four individuals, and one narrow-fast laminar flow where observations were conducted on two individuals. There were differences in mean frequency of aggressive acts (i.e. 'overtly aggressive' and 'threat') before and after tagging in the medium flow velocity section. However, subtle changes in some of the individual's social rank were observed. On two occasions, a dominant tagged fish became subordinate, but only once did both dominant individuals become subordinates after tagging. No significant differences were found (p>0.05) in distance to nearest neighbour for control or tagged individuals before or 1, 12, 24 or 36 h after tagging, indicating that the tagged individuals did not defend smaller/larger territories after tagging. The narrow, faster flowing stream section had similar results to the medium flow velocity. No significant changes in the mean frequency of aggressive acts were observed and only once was there a shift in dominance of a tagged individual. Results indicated that smolt behaviour was not altered due to surgical implantation of transmitters, thereby allowing biologists to make inferences at the population level from data gathered from tagged individuals. [Sat PM2]

We studied the effects of catch-and-release angling on rock bass (Ambloplites rupestris), a small but common centrarchid species in North America. A field study of hooking injury and mortality was conducted in Lake Erie at water temperatures of 16 C. We captured fish using one of four bait types: barbless worm, barbed worm, barbless jig, and barbed jig. No mortality was observed after holding fish for 5 d. Fish captured using worms were hooked more deeply than fish caught on jigs. Fish captured using barbless hooks were removed more easily and rapidly than barbed hooks. Next, we studied the cardiac response of rock bass to 30 sec of simulated angling and either 30 sec or 180 sec of air exposure. These air exposure durations were intended to simulate the conditions faced by fish that were either easy or difficult to remove from the hook, as revealed by our previous field study. Fish experienced a mild bradycardia during the angling and severe bradycardia during air exposure. After being returned to the water, all fish exhibited heightened cardiac output, with the short duration of air exposure recovering more rapidly than the long duration of air exposure. During recovery, elevation in cardiac output was largely due to heightened heart rate despite a decrease in stroke volume. Our results suggest that although hooking mortality did not vary with bait or hook type, the physiological disturbance of rock bass was influenced by air exposure duration, a parameter that would be influenced by bait and hook choice. We discuss the limitations of hooking mortality as an indicator of angling disturbance, and suggest alternative integrated approaches to monitor the real-time disturbance and subsequent recovery faced by fish. [Fri PM2]

The commercial harvesting of Ascophyllum nodosum (rockweed) potentially alters the intertidal architecture influencing fish populations inhabiting the rockweed. The impact on these fish communities cannot be assessed until an inventory of species and a description of their microhabitat use have been compiled. Monthly surveys of unharvested sites near the mouth of the Bay of Fundy over the course of a year were performed. Traps and nets were deployed in the intertidal zone, and fish caught were identified and aged. By means of stomach content analyses, diet choice was described and compared to seasonal prey availability data from the Department of Fisheries and Oceans. I also analyzed underwater video images to quantify the spatial and temporal use of the rockweed beds by fishes. A total of 13 separate species of fish from varying age classes have been noted. Results will provide a critical guideline for assessing the sustainability of rockweed harvesting in the Bay of Fundy. [Sat AM2]

Spatio-temporal migration patterns of anadromous brook charr, as well as many other salmonids, are poorly understood in most rivers of North America. A new non-lethal approach using analyses of scale microchemistry is being developed which is simple, cost-effective and more accurate than other tagging techniques. Scales develop in circulii and the distance between the increments is dependent on water temperature, allowing us to distinguish between periods of slow and fast growth. The scales may incorporate in their matrix different concentrations trace elements present in the water. Since seawater strontium (Sr) concentration is 10-100X higher than freshwater, a higher Sr/Ca ratio in layers corresponding to periods of life in saltwater would be predicted. Using a wavelength dispersive X-ray electron microprobe on a scanning electron microscope (JEOL 733) allows us to assess migration patterns on a finer scale, within the entire ontogeny of individuals. An alternative embedding and cutting technique has been developed to overcome difficulties related to the topographic effect of the scales. Initial trials on brook charr displayed Sr peaks presumed to be associated with residence in seawater. Analyses of known anadromous charr in three rivers and from aquaculture fish will be presented, with a discussion of the enhanced resolution of the anadromous behaviour and the implications for managing this heavily exploited and threatened form of the species. [Fri PM1]

The movements of Atlantic salmon parr (Salmo salar) and resident brown trout (Salmo trutta) between small stream and mainstem where studied in three small tributaries of the Arctic Ponoi River, Russia. The three streams chosen had resident brown trout populations present. Two of the three tributaries were known to be non-spawning sites for adult salmon, but juvenile salmon entered all three from mainstem natal areas. Resident brown trout moved up and downstream in equal proportions throughout the season. Upstream (into tributary) migration of salmon parr was detected for all tributaries, although the number of tributary emigrants generally exceeded that of immigrants. Peak tributary immigration of parr occurred in mid-July when water temperatures reached 10-12° C, but a small degree of upstream movement was observed throughout the summer. The parr may migrate to seek refuge from large mainstem predators, escape warm summer mainstem temperatures, or to seek out more abundant food resources for improved growth opportunities. These results indicate that salmon parr do leave their natal habitat to migrate upstream into small nursery tributaries already inhabited by resident brown trout. [Sat PM1]

Temperature recording data loggers were used to study the winter thermal characteristics of two north temperate lakes in New Brunswick. Thermographs produced at various habitat depths included the autumn temperature decline, lake freeze-up, winter stratification, and spring ice break-up. Lake temperature declined rapidly at all depths during autumn until surface ice formed at 0° C. Depending on immediate local weather conditions, temperatures may reach near 0° C at all depths prior to freeze-up. Under the insulating cover of ice, temperatures begin to increase at depth and classic winter thermal stratification (0 to 4° C) occurs in late winter. Considerable differences in the winter thermal regimes between lakes were noted, despite being in close geographic proximity. These results suggest that in order to survive, northern fish must withstand unavoidable low temperatures (<1° C), and sharp temperature changes during the early winter critical acclimatization period. Potential implications for fish survival in the north and requirements for future research are discussed. [Poster]

The spatial distribution of Atlantic cod (Gadus morhua) in the southern Gulf of St. Lawrence varied substantially between the 1970's and 1980's. It was proposed that cod occupied different temperatures in the southern Gulf depending on food availability as influenced by their abundance. To verify this hypothesis, we measured the final thermal preferendum of Gulf of St-Lawrence cod in relation to food ration. After a 6 to 15 week conditioning to one of three predetermined rations (ad libitum, maintenance and starvation), 72 fish 40 to 50 cm in length were left three at a time to thermoregulate in a horizontal thermal gradient for 45 hours. Except for water content of tissues, the food rations produced significant differences between condition indicators of all three groups. A small but significant difference was found between the final thermal preferenda of fish fed at maintenance and those that were starved (ad libitum 5.94°C; maintenance 6.28°C; starvation 4.56°C). After omitting the first three weeks which showed unwanted signs of interaction, the difference between fish fed ad libitum and starvation also became significant (ad libitum 6.52°C; maintenance 6.36°C; starvation 4.02°C). These findings partly support the hypothesis since fish thermoregulated differently depending on the availability of food. On the other hand, the absence of significant differences between fish fed ad libitum and maintenance rations suggests that cod need to be in poor condition before their temperature preference is affected. The final thermal preferenda found during this experiment are close to temperatures occupied in nature, but are well below the value commonly accepted for this species (13.5°C). This discrepancy can be partly explained by differences in methodology between the two studies. When experimental conditions are well controlled, the final thermal preferendum is a useful tool for the study of behavioral thermoregulation of fishes. [Sat PM2]

The migratory patterns of Atlantic salmon (Salmo salar) smolts were studied in a Newfoundland fiord lake river system during spring and summer. Use of such unique lacustrine habitat by smolts of this species has yet to be investigated. A mark-recapture experiment was used to ascertain smolts movement over time periods and for different size classes as well as to estimate the production of smolts by the system. Special attention was directed toward the use of the fiord lake by migrating smolts. A comparison between smolts caught exiting the lake and smolts caught in the river was done by quantifying differences in age and growth rates for these two groups, as shown by smolt scale analysis. Year 1 results have shown that smolts caught in the rivers were significantly shorter in length than smolts caught exiting the lake (13.8 cm to 15.7 cm respectively). Two separate emigration periods were observed for smolts exiting the lake, the first extending from May 21 to June 28 and the second from July 2 to Aug 26. Smolts exiting the lake in the later period had a significantly greater mean fork length than those smolts exiting in the earlier emigration period (14.6 cm to 16.8 cm, respectively). [Sat AM2]

Field surveys are typically conducted to assess the effects of industrial and municipal effluents on wild fish living in surface waters. Fish populations are sampled in reference areas, unexposed to effluents, and endpoints are compared to fish sampled from effluent-exposed areas. In many aquatic environments, it is difficult to determine the effects of effluent discharges using traditional field survey techniques, because of low abundance of sentinel species, inadequate reference sites for comparison to exposed sites, or a lack of information on effluent exposure due to mobile sentinel species, the presence of other confounding effluent discharges, or complex effluent dispersion patterns. Effluent discharges to coastal environments are good examples of such complexity. In these instances, mesocosms can be used as an alternate monitoring approach to traditional field sampling surveys. Mesocosms are field-based artificial stream systems that can simulate the natural environment (ambient temperature, salinity, photoperiod) while allowing for variable control (sample sizes, sex ratios, effluent exposure), experimental manipulation and statistical replication. The objective of this study was to evaluate if a mesocosm system could be used to assess the effects of pulp and paper effluents on fish in different estuarine and marine receiving environments. Studies were conducted in the summers of 1997, 1998 and 1999 in Saint John and Miramichi, New Brunswick where Fundulus heteroclitus (saltwater killifish; mummichog) were exposed to effluent for 30 to 60 day periods. Results showed that mesocosms could effectively assess effluent effects on both physiological and individual-based endpoints. In addition, the system operated well in highly complex coastal environments and provided information on effluent effects on local fish species that could not have been obtained through conventional field survey techniques. [Fri PM1]

Duplisea, Daniel E., and M. V. Bravington. Lowestoft Laboratory, Pakefield Road, Lowestoft, Suffolk NR33 0HT England.

The failure of single species approaches to fisheries management has been spectacularly demonstrated in several systems around the world. Managers and scientist realise that multispecies interaction do occur and should be accounted for in better fisheries models, but turning this idealistic goal into a operational management is difficult. A pragamatic approach to fisheries management involving multispecies considerations is with the use of different models of a system which can be used to examine different scenarios. It then becomes the goal to seek management decisions which are robust over the range of models. To this end, two dynamic models of fish communities differing in complexity were subjected to different management scenarios and the resulting output examined for robust conclusions. The first model was a variation on a multispecies virtual population analysis model (MSVPA), which provided detailed size information on every species in a system and the second a biomass size spectrum model which provides information only on overall community size composition. These two models differ considerably in their conception, the former being an extension of single species fisheries population models and the latter firmly based in ecological theory. Size based harvesting strategies were examined for both models and it was found that similar community-wide strategies gave optimal yield for both models. It is concluded, therefore, that this harvesting strategy is more robust than a strategy derived from only one type of model as the conclusion transcends model structure. This approach allows one to consider alternative model structure in providing robust management advice, whether it entails multispecies consideration or not. [Poster]

Just as the physiochemical characteristics of a stream are known to change from its headwaters to its mouth, so too can the resident fish communities. Beginning in 1990, slimy sculpin (Cottus cognatus) densities and distributions were examined over 9 years in Catamaran Brook, New Brunswick. Thirty-six sites within Catamaran Brook were electrofished during the summer and fall of each year. Results indicate that sculpin densities vary between stream reaches. Sculpin distribution does not however change along a continuous gradient from the headwaters to the mouth. Highest densities occur in the Middle reach rather than the Upper reaches of the stream, as suggested by previous studies. Stream temperature and bedload stability were both investigated as potential determinants of observed sculpin distribution patterns. Slimy sculpin are found to be preferentially associated with the coolest reaches of Catamaran Brook. That slimy sculpin preferred reaches with low bedload movement appears to be of secondary importance to temperature. This study indicates that, although slimy sculpin appear to be a hardy species, able to tolerate a range of physical stressors, they do indeed have habitat optima at which they thrive. After further defining the specific abiotic factors affecting sculpin abundance, and examining biotic competition factors, members of the Cotttus genus could be used as environmental barometers. [Sat AM2]

The relationship between standard metabolic rate as a function of the body weight of horse mackerel Trachurus trachurus was investigated over a weight range of 1-400g. The standard metabolic rate was determined by the use of an intermittent flow respirometer, which was characterised by a high temporal resolution of the oxygen consumption rates. The oxygen consumption was measured over short time intervals of 5 min. The fish were tested individually at 13° C. The standard metabolic rate was estimated by a data extraction method. The relation between standard metabolic rate and weight was expressed as a linear regression with the log transformed data. The mean slope of the regression was found to be 0.725. The metabolic costs of spontaneous activity of horse mackerel was investigated using three activity parameter. The swimming behaviour was recorded on video analysed with a PC based image analysis system and related to the observed oxygen consumption rates. Metabolic rates were not increased when fish were swimming with low velocity. However, strong bends of the fish body as well as turns caused an increase of the metabolic rate which declined gradually over 20-30 min. [Sat PM2]

A recent study found significant relationships between historical applications of an insecticide containing 4-nonylphenol (4-NP) and catch data for Atlantic salmon populations, suggesting possible declines in catch related to effects at the smolt stage. To test the hypothesis that 4-NP impairs parr-smolt transformation (PST) we exposed Atlantic salmon smolts to environmentally relevant, pulse doses of water-borne 4-NP. To determine whether 4-NP is operating via its properties as a weak estrogen, smolts were also exposed to sustained doses of estradiol (E2). The smolts capability to withstand sea water and their subsequent growth and survival were evaluated after exposure to nominal concentrations of 20 and 200 µg/L 4-NP, and to 100 and 300 ng/L E2. Osmoregulatory (plasma and tissue ions, gill ATPase), biochemical (glucose/glycogen), and endocrine (thyroid hormones, vitellogenin, cortisol) parameters were measured on smolts throughout the experiment. There were no treatment related increases in mortality during a sea water challenge soon after exposure. Subsequent growth and survival in sea water was impaired in about 25% of fish from the various treatment groups (5% in control). The response was bimodal, with the affected fish showing weight loss from soon after treatments. If the effects exerted by 4-NP are due to its estrogenic potential, then estrogenic activity stemming from other sources (e.g. domestic sewage, agricultural wastes or phytoestrogens from pulp mills) might influence present day salmon populations. [Poster]

Fitzgerald, D. G., A. R. Dale, and P. F. Sale. Dept. of Biological Sciences, University of Windsor, Windsor, ON N9B 3P4

We identify key physical and biological factors that contributed to the development of one of the largest year classes of yellow perch (Perca flavescens Mitchill) documented in Lake St. Clair. This species can be considered as an archetypal temperate species for North America due to its wide natural distribution across rivers and lakes of all sizes. This success is predicated on opportunistic behaviour and life history adaptations that are evident in immature and mature fish. Classically, yellow perch are believed to closely synchronise their spawning to a brief period in the spring when water temperatures first exceed 7C. This results in the hatch of the cohort in association with an abundant spring bloom of plankton. Collections within Lake St. Clair for young-of-the-year (YOY) yellow perch during 1998 were completed from spring to fall and then again the following spring to describe seasonal growth and overwinter survival patterns. We used YOY yellow perch otoliths to back-calculated hatch estimates and revealed that adults initiated their spawning during typical spring water temperatures but unexpectedly protracted their season during the slow warming to typical summer temperatures. Protracted hatch dates ranged across a 10-week period and contributed to the observation of wide seasonal and fall length frequency distributions. End-of-season sizes for YOYs that were hatched early in the spring had a mean size greater than 80 mm, whereas the fish that were hatched later in the spring had a mean size less than 65 mm. The short winter observed during 1999 then minimized overwinter mortality in late-hatched fish since the fall and spring size ranges of YOY fish were similar. The cascade that produced a large year class initially involved spring weather conditions that allowed adult fish the opportunity to protract their spawning season. This resulted in a broad size distribution of YOY in the fall. The subsequent short winter then allowed a majority of the late-hatched YOY fish to survive into their second season. It is this combination of opportunistic behaviour of adult and young yellow perch coupled with favourable weather patterns that produced the large 1998 year-class in Lake St. Clair. [Poster]

Anthropogenic activity such as, timber harvest and road construction in river basins can create significant stress to stream biota. The present study investigates the effect of fine sediment deposition on survival of Atlantic salmon (Salmo salar) eggs in Catamaran Brook - a tributary of the Miramichi River. Known quantities of wild Atlantic salmon eggs were placed in "incubation baskets" in known spawning locations within the brook, at locations representative of different degrees of impact from forestry activities. Survival to the eyed stage (late March - April) and to mergence (late May - June) was determined separately. Survival ranged from 75% to 99%; (eyed stage); and, 0% to 83% for emergence. Finally, to determine if a relation exists between survival and sedimentation, all baskets were analyzed for the size and amount of particles accumulated during the intragravel period. [Sat PM2]

We investigated the effects of temperature on the post-exercise recovery processes in juvenile Atlantic salmon. We measured the metabolic fuels (e.g. Glycogen, Phosphocreatine, ATP) in the white muscle, and the muscle [lactate], plasma [lactate] and plasma osmolarity before, following, and during recovery from exhaustive exercise; (i) in fish fully acclimated to 6 or 18C, or (ii) in fish exercised at 12C and acutely exposed to 6 or 18C during recovery. In general, energy stores and the post-exercise response of fish acclimated to either 6 or 18C did not differ, but fish acclimated to warmer temperatures recovered slightly faster. In contrast, however, the patterns of recovery were markedly influenced in fish exercised at 12C and allowed to recover at either 6 or 18C. For example, muscle phosphocreatine and ATP levels decreased following exhaustive exercise, however, recovery was delayed considerably in fish exposed to 6C during recovery. Muscle lactate levels increased immediately following exhaustive exercise in all fish, but the concentrations remained elevated throughout the 4 h recovery period in fish acutely exposed to 6C. Fish acutely exposed to 18C during the recovery period had cleared all their muscle lactate by 4 h. Recovery of both plasma lactate and osmolality showed similar patterns to that of muscle lactate. Overall, our results suggest that acute temperature shifts can markedly influence the recovery patterns in fish following exhaustive. [Poster]

In nature, organisms are selected to maximise their reproductive success. For anadromous Atlantic salmon (Salmo salar L.), this may involve a larger body size for both males and females. Previous studies showed that female salmon most commonly mate with a single anadromous male, although multiple-male mating is also known to occur. Males are known to generally mate with several partners. In this study, we assessed individual reproductive success in a population of 76 wild anadromous Atlantic salmon that were introduced to a stretch of the Sainte-Marguerite River in Québec, Canada, that was previously unoccupied by salmon. We used microsatellite loci to perform parentage analysis, based on multi-locus genotype information from all spawners and a subsample of the progeny collected several months following emergence. Our results showed that the body lengths of females were neither correlated with the number of offspring produced nor with the number of spawning partners. Moreover, for females, estimated fecundity proved to be a poor indicator of number of offspring produced by an individual. The number of partners seemed to be the major determinant of the number of offspring produced by an individual for both males and females. We also found that the mean and variance in the number of mating partners was as high in females as in males. Under the environmental and ecological conditions of this study, variance in individual reproductive success of both sexes of Atlantic salmon is greater than previously documented. [Fri PM1]

Estimates of fish abundance vary on a daily, monthly, and seasonal basis. Interpretation of data collected at these time scales assume that sampling units, like transects, can be considered as an unbiased estimate of real fish abundance. Furthermore, it is assumed that the information derived from one transect would not vary if replication was performed. Although this assumption is rarely tested, hydroacoustics in small shallow lakes allows the study of abundance variations at short time scales. Our objective was to assess variations of fish abundance and distribution patterns between sampling sessions conducted at high temporal frequency. Sampling was performed in two shallow lakes during the phase of the new moon, because previous work suggests higher numbers of fish are acoustically detectable in these conditions. Sampling consisted of insonifying the complete contour of each lake with a single-beam echosounder using the horizontal scanning approach. A complete circling of a lake could be performed within 30 minutes. In each lake, a total 12 circlings were performed: 6 during daytime and 6 during nighttime. The 6 circlings were conducted successively without pause. Our results support the assumption that successive estimates of fish relative abundance obtained at high frequency do not show large variations. [Sat PM1]

Southern Ontario has both the highest fish species diversity and the greatest degree of anthropogenic disturbance in Canada. Human activities, such as urban and agricultural development, have greatly increased over the last few decades, with potentially major impacts on adjacent stream communities. It is hypothesized that stream fish assemblages in southern Ontario will have undergone considerable changes in the past 25 years as a result of land use practices. Historical data for 34 southern Ontario sites (cool/cold- and warmwater streams) and seven cool/coldwater sites on Lake Superior tributaries were obtained from provincial stream inventory surveys conducted between 1972 and 1977. These same sites were resampled in the summer of 1998, to assess the composition and stability of local fish assemblages. Multivariate analyses of presence/absence data revealed differences between the 1970’s and 1998 datasets. Principal coordinates analysis revealed that coldwater assemblages responded differently in the last 30 years than did warmwater assemblages. At the local scale, coldwater assemblages had predominantly shifted toward assemblages typical of warmwater habitats, whereas warmwater assemblages exhibited less directional change. Regional patterns of temporal stability did not show such patterns. Parametric tests revealed no significant differences in the degree of change between coldwater, warmwater and northwestern Ontario assemblages. Similarly, the magnitude of change within coldwater assemblages from southwestern Ontario were not significantly different from those in southeastern or northwestern Ontario. We conclude that agricultural practices and urbanization have triggered changes in stream fish assemblages in southern Ontario, and to a lesser degree, in northern assemblages. Future work is needed to examine how habitat consequences of specific land use practices drive changes in stream fish assemblages. [Poster]

Shortnose (A. brevirostrum) and Atlantic sturgeon (A. oxyrhynchus) are morphologically similar species and are often sympatric. They are thought to share common nursery grounds as young juveniles. Consequently, a potential for resource competition between juveniles of these species exists. We tested for a difference in flow preference of juvenile shortnose and Atlantic sturgeon to determine if flow regime may be used to partition resources. Trials were completed in an angular diffuser which provided each fish with a wide range of flow velocities. Capture of juveniles in the wild is rare event, therefore, knowledge gained from this study will help us to locate nursery areas in the wild. This study also provides further information on the role of flow rates in resource partitioning between morphologically similar species. [Sat PM1]

The objectives of this study were to determine if trophic dimorphism occurs in populations of pumpkinseed (Lepomis gibbosus) outside of the isolated Adirondack region where it was first reported, and if reproductive isolation of the morphs is a factor in this occurrence. Five central Ontario lakes, Looncall, Wolf, Monch, Four-Mile and Shadow were sampled for morphology. A sample of pelagic (N = 50) and littoral (N = 50) individuals were sacrificed from each lake, and 31 external and eight internal morphometric measurements were obtained on these fish. Preliminary analysis indicates that trophic dimorphism occurs in all but one of the lakes. Internal morphometric analysis showed a significantly larger pharyngeal jaw apparatus, shorter gill rakers and a wider space between gill rakers in littoral fish. Stomach content analysis confirmed that pelagic and littoral individuals were feeding upon prey types consistent with their morphology. Pelagic individuals were feeding on a significantly higher quantity of cladocerans, as opposed to the benthic fauna preyed upon by the littoral individuals. Size and age at maturity were also compared in pelagic and littoral individuals from Looncall and Shadow lakes, with significant differences found in one of the lakes. A mark-recapture study conducted over two field seasons in Looncall Lake (1700 littoral and 600 pelagic pumpkinseeds tagged) showed very little mixing between littoral and pelagic individuals during the breeding season (6% of females and 4% of males found in the opposite zone from where they were tagged), suggesting that reproductive isolation may be contributing to trophic dimorphism in pumpkinseeds. [Sat PM1]

The use of information by fishing fleets, and its influence on fleet dynamics, is difficult to study and therefore poorly understood. In those cases where it has been examined, information sharing within fleets seems to be dynamic rather than fixed in form. A predictive understanding of patterns of information sharing within fishing fleets would allow managers to recognize conditions that could lead to changes in fleet efficiency that do not correspond to physical attributes of the fishing vessels. As a first step, I have developed an individual based model (IBM) of vessels fishing in a spatially heterogeneous environment. Information in the system is modeled as a subjective "map" of the environment based upon the individual experience of each vessel. Information sharing is modeled as a common "map" among those vessels that are fishing together. The relative performance of "sharers" and "loners" is compared through simulated variation in fish overall abundance, spatial distribution and movement. Finally, the impact of regulated season length on the relative performance of the two types of vessels under these circumstance is examined. [Sat PM2]

We compared consumption rates of yellow perch obtained from surveys of digestive tract contents (DTC) with the radioisotope approach using 137Cs as a biological tracer. We also compared the radioisotope method of Forseth et al. (short-term method; 1992) with the radioisotope method of Rowan and Rasmussen (1996). We sampled fish of age 1+, 2+, and 3+ from three lakes for a total of six lake-age classes combinations. The comparison of the two radioisotope methods revealed their equivalence for the consumption rate levels observed in our study and for the time scale covered by our sampling. The relative differences between the DTC approach and the radioisotope approach ranged from - 67 % to 128 % when the sampling interval length was 28-35 days. At the time scale of 70 days, the relative differences between the approaches ranged from - 15 % to 20 %. We propose the ratio of the sampling interval length to the biological half-life of the contaminant as an indicator of the validity of the radioisotope approach. That ratio was estimated at 45 % for the specific conditions encountered in our study. Our work represents a corroboration of the radioisotope approach in systems where the 137Cs is present at trace levels. [Sat PM1]

Past research argues that two species of smelt (Osmerus mordax and O. spectrum) exist in Lake Utopia, based on relatively larger and smaller (dwarf) sizes respectively. At present, the abundance and population structure is unknown. It is believed that their #'s may be lowered by the stocking of Atlantic salmon (Salmo salar) which prey on the smelt. Random smelt samples of both size classes were collected during spawning migrations from four tributary streams of Lake Utopia. Sampling and analysis will determine morphological differences between the two size classes. Larger predator fish (i.e. S. salar, Catostomus commersoni, Perca flavescens) were also collected for stomach analysis and stable isotopes to describe trophic relationships. Results will show whether two distinct species exist in Lake Utopia, as well as their abundance and their role as a food source. This information will provide us with the true status of O. spectrum. [Poster]

We take a small-scale, space-use approach to the study of Atlantic salmon abundance in Catamaran Brook, New Brunswick. Atlantic salmon defend larger territories than most other salmonids of a similar size. Growing salmon parr compensate for a seasonal decrease in food abundance, by increasing the size of their territories and by moving to deeper, faster water. Percent habitat saturation (PHS), the percentage of the stream area occupied by the territories of salmonid fishes, is a better measure of abundance than population density, because the former integrates the effects of (a) several age-classes or species in a stream, and (b) variation in growth rate or sampling date. ‘Effective density' or ‘effective PHS', calculated by weighting crude density (no. m-2) or PHS by the number of organisms in the sampling unit, more accurately reflects density from the organism's point of view than does crude density or PHS. Effective density and PHS of Atlantic salmon in Catamaran Brook, New Brunswick, increased by 0.4 fish per m² and 4%, respectively, for each order of magnitude decrease in the area of the sampling unit. The allometry of territory size was a better predictor of the decline in density of a cohort of Atlantic salmon in Catamaran Brook than the allometry of metabolic requirements. We suggest that information about territory size provides useful insight into both the measurement and prediction of salmonid abundance. [Sat AM2]

We have been developing a strategy for dealing with cumulative effects assessment, which uses the performance of resident fish to help understand the level of stress on a river reach. A large study has been initiated on the freshwater portion of the Saint John River, to develop a strategy for increasing the level of understanding of the system. The major stressors in the upper river are pulp and paper, hydroelectric development and agriculture. We are comparing the health of fish populations in sections of the Little River that drain predominantly agricultural lands with fish inhabiting reaches draining forested lands. Surveys will be conducted on resident fish, habitat suitability and fish and invertebrate communities, to develop a baseline of performance for interpreting other studies. One of the major agricultural sectors in northeastern New Brunswick is related to potato farming. Population surveys of slimy sculpin (Cottus cognatus) populations in agricultural reaches show differences in size distributions and fish condition. Collections will examine age distribution, growth, reproductive development and reproductive physiology in an attempt to understand the relative roles of habitat damage, nutrient enrichment and chemical use in the responses of the fish. [Fri PM1]

The choice of a habitat by juveniles of Atlantic salmon in lotic systems is mainly determined by abiotic factors. Habitat can be described by bio-physical models which define the quantity and quality of habitat at different flow. Construction of such a model requires the combination of a numerical field model which predicts velocity, depth and bed granulometry anywhere in a river section, with a biological model that reflects the propensity of fish for those abiotic factors. The biological model is made with field data of velocity, depth and granulometry used or avoided by fish. The logistic regression (habitat predictive index; HPI) or the preference curves (habitat suitability index; HSI) can be used for the construction of the biological model. The objectives of our study were: 1) to determine the effect of the number of fish sampled on habitat preference; and 2) to see the evolution of HPI and HSI with the increase sample size. We collected a data set for 300 parrs (1+ 2+) of Atlantic salmon in the Sainte-Marguerite River during the summer of 1997. We successively constructed preference curves with 50, 100, 150, 200, 250, 300 fish chosen at random in our data set. A minimum of 150 fish are needed to have a statistically significant HPI. In contrast, 250 fish must be sampled to develop preference curves on which HSI can be based. [Sat PM1]

This study was designed to test the effects of two potential impacts of forestry roads on lake trout (Salvelinus namaycush) lakes in the boreal shield ecozone: 1) loss of reproductive habitat through siltation; and 2) increased access and exploitation. In the habitat removal experiment on Whitepine Lake (67 ha), spawning habitat was either covered with plastic sheeting or infilled with sand to simulate a release of sediment from the catchment area. During and 8 year study (1991-1999), access to seven original spawning sites (total surface area 40 m²) and 213 alternate spawning sites was progressively removed by covering approximately 1000 m² of substrate with plastic. However, no significant effects on recruitment have yet been detected. Mark-recapture estimates of adult and juvenile (<370 mm fork length) abundance remain high in this research sanctuary lake. The alternate spawning sites selected by the displaced fish appear to provide poor quality habitat (shallow water depth, few interstitial spaces), in comparison to the historic sites, but emergent alevins continued to be produced each year. The lack of obvious effects of reproductive habitat loss was in sharp contrast to the rapid and severe effect that fishing pressure exerted on the second experimental lake. Michaud Lake (148 ha) was closed to fishing for 7 years, during which time access to the lake was improved by the construction of a 12 km forestry road. When the fishery resumed, the lake trout population was rapidly decimated. An estimated 3.8 kg.ha^-1, or approximately 72% of the adult lake trout population was harvested in the first few weeks of the open season. These findings suggest that lake trout can tolerate substantial losses to spawning habitat (reproductive habitat was not limiting), but natural populations, particularly in small lakes, must be protected from excessive exploitation that can accompany the construction of forestry roads. [Fri AM]

A worldwide decline (0.3 units) in average trophic level of fisheries landings has been observed in the last 30 years. Elimination of large, long-lived, predatory species from ecosystems has forced many commercial fishing fleets to turn their attention to short-lived, small-sized, pelagic fishes and invertebrates. The extent of this change in species targeted is not well understood for Canadian waters. The average trophic level of commercially exploited marine organisms and their prey in the southern Gulf of St. Lawrence was calculated from published literature (all years available) and supplemented by samples collected between 1991 and 1995. Every attempt was made to account for seasonal patterns in prey consumption in assigning trophic levels, but year-to-year variation had to be ignored due to a dearth of data. The observed changes in mean trophic level with time reflected the domination of catches by Atlantic herring (Clupea harengus) and Atlantic cod (Gadus morhua). When only fishes were examined, the mean trophic index dropped 0.6 units as a result of the elimination of large cod during the mid-1980s, cessation of cod fishing in 1993, and a marked increase in landings of herring since the late 1970s. When landings of all species were considered, the overall change was 0.4 units. The smaller change was primarily due to increased landings of snow crab (Chionectes opilio) and American lobster (Homarus americanus) since the early 1980s. Both of these species have surprisingly high trophic levels (similar to small cod) and act as a buffer between the effects of changes in finfish landings. Nevertheless, the patterns in landings in the fishery in the southern Gulf of St. Lawrence was not different from that observed worldwide - the abundance of the largest predatory species has been greatly reduced and the fishery is now dominated by pelagic fishes and invertebrates. [Sat PM2]

Each spring alewives and blueback herring, collectively termed gaspereau, ascend rivers along the Atlantic Coast to their freshwater spawning grounds. Adults and their young-of-the-year (YOY) migrate to the ocean after several weeks of feeding in freshwater. There is concern that gaspereau may have negative effects on smallmouth bass when they occur in sympatry. Adult gaspereau may prey on YOY bass or YOY of both species may compete for food resources. Our objective was to investigate the trophic interactions between YOY smallmouth bass and gaspereau within the St. John River system, New Brunswick. Preliminary results show that the timing of the gaspereau run affects trophic interactions between YOY. Stomach contents and stable isotope analysis suggest that bass have a low dietary overlap with gaspereau in the first few weeks after hatching in late May, with a peak overlap occurring from mid-July to early August. In a population where an extended gaspereau run produced several cohorts, diet overlap was low and later spawned gaspereau were a potential food source for YOY bass. A population with a contracted spawning run produced larger YOY, greater evidence of diet overlap and less predation by bass on gaspereau. The implications for bass and gaspereau management in the St. John River system are that variation in hatch dates associated with a natural gaspereau run serves to reduce interspecific competition among YOY and promotes piscivory in bass at an early age. [Poster]

We tested the effect of temperature (13, 15, 18, 21° C) on the rate of yolk-sac absorption, size at yolk-sac absorption, development of fright response, size at time of fist fright response, growth, and survival of yolk-sac larvae of shortnose (A. brevirostrum) and Atlantic (A. oxyrhynchus) sturgeon. Yolk-sac of both species was absorbed significantly faster at higher temperatures. However, there was no significant difference in size of the larvae at the point of yolk-sac absorption or size at initial fright response between temperature treatments. Time to 50% survival also occurred earlier at higher temperatures, as was the development of initial fright response. Atlantic sturgeon developed at a faster rate than shortnose sturgeon. These results also suggest that these species are able to grow under a wide range of temperatures, and that they are therefore eurythermal. [Sat PM1]

Hébert*, C. and L. Bernatchez. GIROQ, Departement of Biology, Laval University, Ste-Foy, Québec (Québec)

Despite the abundance of studies of genetic diversity in freshwater fishes, few of them specifically addressed the role of habitat structure in partitioning genetic variance within and among populations. Microsatellite loci have recently been developed for Salvelinus fontinalis (Angers et al. 1995). In this study, we analysed the variability of six microsatellite loci among 24 brook charr populations samples (n=30) in order to correlate hydrographic structure with genetic organisation. These populations originate from three Canadian National parks showing distinct hydrographic structure. Considering the general characteristics of these habitats, we formulated specific predictions about genetic structure principally based on gene flow and population size potential conferred by each habitat. The hierarchical analysis of molecular variance and the genetic distances computed among populations revealed that habitat structure analysis constitute an important but insufficient predictor of genetic structure. We discuss the importance of habitat complexity on genetic structure in the context of management and conservation. . [Poster]

Vingt-quatre populations d'Omble de fontaine (Salvelinus fontinalis) provenant de trois parcs nationaux présentant chacun une structure hydrographique particulière ont été analysées au niveau de six loci microsatellites. Cela nous a permis d'évaluer le niveau de corrélation entre la structure génétique et la structure hydrographique de l'habitat. Plusieurs prédictions concernant la répartition de la structure génétique ont été formulées sur la base des caractéristiques de chaque habitat afin de vérifier l'hypothèse voulant que cette structure génétique soit prédictible. L'analyse hiérarchique de la distribution de la variance génétique, ainsi que l'évaluation de groupements basés sur les distances génétiques, ont montré que certaines de ces prédictions sont fausses. Ceci nous porte à croire que, dans l'état actuel des connaissances, la prédiction de la structure génétique est difficilement envisageable et cela nous amène à conclure que la structure hydrographique de l'habitat constitue un critère nécessaire mais non suffisant de prédiction de la structure génétique. [Poster]

The Atlantic halibut is a sexually dimorphic species in which females grow larger and mature later than males, making monosex (all-female) culture economically advantageous. Monosex populations can be produced by using hormones (sex steroids) to direct sexual differentiation towards the desired sex (male or female). We have determined histologically that sexual differentiation in Atlantic halibut has occurred by 37 mm fork length (FL), which coincides with the weaned stage. Therefore, steroids can be administered by using steroid-treated pelleted food, as is done with salmonids. However, fish directly treated with steroids are not deemed suitable for human consumption, so indirect feminization must be used. In such a procedure, an androgen is used to masculinize genetic females, which are then mated with untreated females to produce all-female offspring due to the total maternal gene source. This assumes that females are homogametic (XX), which has yet to be confirmed for halibut. The estrogen and androgen of choice for feminization and masculinization are, respectively, 17b-estradiol (E2) and 17a-methydihydrotestosterone (MDHT). An experiment, presently underway, using 30 d treatments at specific doses (10 mg/kg of E2; 1 and 5 mg/kg of MDHT) to sex reverse halibut, will be described. The results of these experiments are directly relevant to the halibut aquaculture industry in Canada. [Sat AM1]

Radiotelemetry was used to determine winter movement, habitat preference and cover use of 10 juvenile Atlantic salmon (Salmo salar; mean weight (ñ SE), 21.65 ñ 1.84 g and mean fork length (ñ SE), 129.67 ñ 4.65 mm) in a small Newfoundland river. Transmitters were surgically implanted and fish were tracked from 20 January 1999 to 10 February 1999. During this 23 day period, a total of 133 daytime and 69 nighttime observations were recorded. Data collected included: 1) position, recorded visually and with a GPS; 2) habitat variables (i.e. water depth, bottom and mean column water velocities, substrate and cover); 3) size of cover stone; 4) visibility of fish or antenna; and 5) fish activity. Thermographs, placed in four locations along the river, recorded water temperatures hourly throughout the study. Due to mild winter temperatures during this study conditions were ice-free. Individuals preferred habitats with a bottom water velocity <5 cm/s, mean column water velocity <40 cm/s and a depth >30 cm. Non-active individuals preferred cobble/boulder substrate while active individuals, only observed at night, preferred gravel/cobble substrate with greater water velocities. This switch in microhabitat preference may allow individuals to forage in areas with higher prey densities and an increased visual field to detect prey and predators. The number of recorded movements (>6 m) ranged from 1 to 13 and the total distance moved ranged from 22 m to 722 m. These movement results were unexpected since environmental conditions were ice-free and relatively stable. Individuals in this study may have moved due to high interspecific competition for space or in search of more suitable habitats. Site fidelity is this study was related to habitat suitability. Individuals in more suitable winter habitats (i.e. deeper with lower water velocities) had higher site fidelity than those in less suitable habitats. [Sat PM1]

The goal of this project is to determine the need and scope of monitoring for the effects of the Bruce Nuclear Power Development (BNPD) on whitefish species in Lake Huron. Lake whitefish (Coregonus clupeaformis) have historically been a major component of the commercial fishery in Lake Huron. Commercial harvests of lake whitefish have increased consistently since 1975, and now exceed values comparable to the best historical harvests during the early part of this century. About 90% of the annual whitefish harvest is derived from the Main Basin, where the BNPD is located. Ontario Power Generation (formerly part of Ontario Hydro) has supported environmental studies at the BNPD site since 1973. These studies have focussed on round whitefish (Prosopium cylindraceum), because this is the only commonly occurring fall spawning species in the vicinity of the BNPD site. Round whitefish ecology and habitat use may be similar to those of lake whitefish, but the amount of spatial and temporal overlap between these species is unknown. Assuming that round whitefish are an appropriate surrogate indicator of lake whitefish, the likelihood of important direct or indirect short-term interactions between lake whitefish or round whitefish and the BNPD was judged to be low. However, the data on which this judgement is based (monthly sampling during the open-water season, 1979-1989) were not of sufficient duration (10 years) to reliably detect longer-term reproductive effects on whitefish populations in Lake Huron resulting from the operation of the BNPD. Furthermore, the Lake Huron ecosystem has undergone changes since monitoring was discontinued in 1989 as a result of local (e.g. reduced P loadings), regional (invasions of exotic species such as zebra mussels, Dreissena polymorpha), and biospheric processes (e.g. global warming, increased UV-B levels). Some of these changes may be important to whitefish ecology and population dynamics in Lake Huron. [Fri PM1]

Despite vast advances in fisheries science, understanding and predicting the behaviour of the ultimate predators, i.e. the anglers, remains illusive. To demonstrate the utility of collecting information from anglers, we present the results of a survey conducted with walleye anglers in Timmins, Ontario. We found that angler’s reported fishing effort on area lakes closely mimicked creel derived data on walleye abundance. Also, through a discrete choice experiment (DCE), we discovered the affinity that local anglers have for surrounding water bodies. The trade-off capabilities of the DCE demonstrated that pushing anglers from their favourite lakes is a difficult task. [Fri PM1]

A 52 month study of individually marked brook trout in Freshwater River, Cape Race, Newfoundland, has revealed evidence of sexbiassed dispersal and, perhaps paradoxically, extraordinarily limited movement over time. After initial marking by PIT tags in June 1995 and 1996, subsequent biannual recaptures (every June and October) made throughout the entire 2 km length of the river indicated that mature males dispersed significantly further from their locations of initial capture than mature females. Notwithstanding these sexspecific differences in movement, which can be interpreted as an adaptive consequence of a polygynous mating system, dispersal on the scale of hundreds of metres was rare. These observations have implications for brook trout population ecology and conservation biology. [Fri PM1]

Hypotheses (A and B), to explain growth variability among larval hake cohorts and among individuals within cohorts, are developed using data collected from Western Bank, Scotian Shelf in Autumn 1997. A) Length-at-age does not vary among cohorts of larvae and is independent of: 1) temperature growing degree days (317 to 436 degrees celsius); 2) zooplankton biomass (5 to 24 g/100m^3); and 3) larval concentration (23 to 493/100m^3). B) Growth rate variation within cohorts is a function of spatial variation in water mass structure (i.e. depth and salinity) and the dynamic processes that produce the spatial distribution of larvae. These two hypotheses are tested using independent data collected from the same region in Autumn 1998. The general hypothesis (A) is rejected as length-at-age is significantly different among cohorts in which individuals in one cohort exhibited a size advantage of ~1.5 mm at age 25-d over larvae collected later in the Autumn. Hypotheses A1 and A2 are also rejected as size at age differences are most easily explained by temperature (9 to 14 degrees celsius) and zooplankton biomass (11 to 17 g/100m^3). Hypothesis (B) concerning growth rates within cohorts is partially rejected as spatial variation in water mass structure (depth and salinity) could not explain the growth variations. Different water mass characteristics (temperature and density), but not zooplankton biomass or larval concentration, could explain a significant proportion of the variation within cohorts. We use these results to challenge the utility of the all too prevalent pattern in the literature of forming conclusions based on unique observations (e.g. single-year of data). We also explore the alternative hypothesis that growth variability is determined by the interaction between parental influences (heritable and non-heritable) and oceanographic processes (biological and physical). [Sat PM2]

Egg size is the primary factor determining size at first feeding and can exert a strong influence on early life survival of fishes. The size of egg produced may be influenced both by maternal and environmental factors. Egg size (dry mass) was determined in 34 walleye populations from across their native range to assess variability in the egg size vs female age/size relationship, and test the hypotheses that egg size declines with increasing latitude (a trend seen in other species) and declines with increasing productivity of the juvenile food source (Lack's hypothesis). Among populations, the slope of the egg size vs female size relationship was generally positive, but varied considerably and was significantly different from 0 for only a third of the populations. Mean population egg size (for a standard female age/size) declined with increasing latitude as predicted. This effect was assumed to be temperature driven, as egg size showed an even stronger positive relationship with mean annual growing degree days. After adjustment for the temperature effect, walleye egg size showed a negative relationship with mean epilimnetic total P concentration, an index of lake productivity and crustacean zooplankton (food of walleye larvae) abundance. The ecological and management implications of egg size variation within and among walleye populations will be discussed. [Sat AM1]

The amount of genetic variation present in a population is assumed to be positively associated with its long-term sustainability. A previous study found significantly lower heterozygosity levels (a metric of genetic variation) in lake than stream populations of brook trout (Salvelinus fontinalis). This finding was unexpected and suggested lower effective population sizes for the lake populations as compared to those in streams. Further genetic analysis on an additional seven lakes and associated streams found this trend to not be universal. Abiotic factors were examined to determine those associated with low heterozygosity levels. Mark-recapture population size estimates for the original lakes indicated that these populations are large but can experience severe declines, possibly explaining the low heterozygosity levels observed. This study demonstrates the potential utility of molecular information as an indication of possible demographic stochasticity. [Poster]

In 1991, diamonds were discovered in the remote Barrenlands region of the Northwest Territories, Canada. In preparation for mineral extraction, two lakes and their tributary streams were drained. As part of a habitat compensation agreement, the mining company initiated the design and construction of a diversion channel and associated fish habitat structures. It is anticipated that the channel will restore watershed connectivity, allowing fish migration, while enhancement structures will provide spawning and nursery habitats, improving the productive capacity for fish. We are examining the effectiveness of the diversion channel and habitat structures in providing productive fish habitat, with particular focus on Arctic grayling (Thymallus arcticus). Comparisons are being made among natural, undisturbed streams and the diversion channel. Preliminary data suggest that the diversion channel and reference stream contain similar fish densities, however, the reference stream supports a significantly greater fish biomass. Reduced periphyton and sediment organic matter in the diversion channel suggest that it's lower productivity results in reduced growth of YOY grayling. [Poster]

This study investigated the metabolic changes associated with forced activity in juvenile shortnose sturgeon. Five minutes of activity resulted in an elevation of oxygen consumption and ammonia excretion rates. In addition, both muscle and plasma lactate as well as muscle pyruvate levels increased substantially following activity. Plasma osmolality, however, did not significantly increase following forced activity. Of the physiological endpoint measured, only muscle lactate required a considerable period to return to resting levels (about 6 h). All the other variables required less than 1-2 h for full recovery. Compared with other fish species, juvenile sturgeon exhibit a reduced anaerobic capacity and recover from forced activity quickly. We suggest that these ancient fish rely on aerobic, rather than anaerobic processes to support exhaustive exercise. [Poster]

When investigating genetic diversity in the wild, it is of special interest to study genes involved in the adaptive differences between populations. For their implication in parasite resistance and their high level of polymorphism, MHC genes have allowed us to examine how natural selection may adapt the populations to their own environment. Here we describe the genetic variation at a MHC Class II B gene exon in four populations of landlocked and three populations of anadromous Atlantic salmon using PCR, single strand conformation polymorphism (SSCP) and DNA sequencing. Samples size varies between 40 and 80 adult fishes caught in 1994, and 1999. Allelic diversity found was higher in the anadromous population of the Ste-Marguerite River than of the landlocked population of the Metabetchouane River with respectively 19 and 5 alleles found. These populations have been previously characterized with neutral markers as mitochondrial and microsatellite loci. Originally, in this study we compared the distribution of the genetic variability at genes submitted to selection pressure. [Sat AM2]

This investigation studied the relationship between interannual growth variations in seven populations of lake sturgeon from across North America and neighbouring tree stands. Past annual growth variation was documented through the development of growth chronologies from annuli contained in calcified or xylem tissues in fish and trees, respectively. Growth chronologies developed from lake sturgeon that demonstrated significant interseries correlation were among member of a population displayed significant, negative correlation with tree growth chronologies. This negative relationship was related to differing responses of fish and trees to annual measures of extrinsic factors. [Sat AM2]

By tagging and resampling individual Atlantic salmon parr in a small stream (1 km study section of West Brook, Whately, MA, USA), we are developing a record of the growth, movement and maturity of individual fish. From 5/14/97 to 12/11/99, we tagged and resampled approximately 3500 age-0 through age-2 parr during 24 sampling sessions. We used electroshocking to sample the study stream for one-half of the samples and used seines at night during the other samples. By the seventh sample, 96.7 percent of the parr were recaptures. Individual fish grew rapidly during the early summer (instantaneous growth = 0.28 d-1) but most fish lost mass (-0.03 d-1) during the summer and early fall. The first mature fish were observed on 8/6/97 and by 9/29/97 50.8 percent of the fish in the stream were mature (expressing milt). In spring, fish that eventually matured were 13 percent heavier (25.6 g vs. 22.6 g) than fish that did not mature, but by fall, maturing fish grew slower and were 5% lighter than non-maturing fish. Spring growth of all fish was rapid. Previously mature smolts grew exceptionally fast (5% BW/d) during the month before migrating. Non-migrating, previously mature residents also grew rapidly in the Spring and were the same size as the smolts 2-3 weeks following migration. Approximately one-half of the smolts caught in the smolt trap were previously mature and all of the age-2 fish remaining in the stream had been previously mature. Movement of individual fish was minimal over the course of the summer; but more fish, especially those that had been mature, moved during the fall. [Sat PM2]

Several studies have demonstrated the existence of two whitefish races known as Atlantic and Acadian in northeastern North America. However, to date, the distribution of these races has not been clearly defined. Whitefish of the two races were found in sympatry in several lakes of northern Maine (USA) where they are associated with distinct ecotypes (dwarf and normal). These findings suggested that secondary contact between two races that evolved allopatrically in the Atlantic and Acadian refugia during the last glaciation events best explained the origin of sympatric pairs in that area. The objective of this study was to determine the distribution of the two races by performing mtDNA and nDNA phylogenetic analysis. A total of 760 whitefish, representing 18 sympatric and 16 allopatric populations were sampled from eastern Quebec and northern Maine. MtDNA variation was analyzed by restriction fragment length polymorphism (RFLP), whereas nDNA variation was quantified by six microsatellite loci. Results from mtDNA analysis demonstrated, (1) the contact zone did not involve two, but rather three races, (2) sympatric pairs did not always originate from the same scenario of allopatric divergence and secondary contact, and (3) the different races are distributed moisaically, reflecting the dominant influence of historical hydrology of the area in explaining the contemporary patterns of genetic diversity. Results derived from microsatellites were not congruent with mtDNA results. Two glacial races (Atlantic and Acadian), regarded as ancestries of different ecotypes inferred from mtDNA haplotypes, were not clearly identified by allelic distribution in any microsatellite loci. These incongruent results suggest microsatellite and mtDNA data do not reflect the same evolutionary architecture in whitefish. [Poster]

It is generally acknowledged that wild fish stocks cannot support any increase in exploitation by commercial fishers. Reports that extol the virtues of aquaculture indicate that the increased demand for fish flesh will be met by this industry. However, there is a tremendous controversy about aquaculture as it is currently practiced. Possible negative aspects of aquaculture include: the potential for exotic introductions, genetic pollution, disease transmission, escape of transgenics, pollution of coastal ecosystems, and whether it is sustainable. Hence, one of the most important questions facing fisheries and aquaculture scientists is what should we really be doing to meet the ever increasing demand for fish flesh? Clearly, we need to better manage wild stocks but we also have to work to understand both the potential positive and negative impacts of aquaculture. Here, I review some of the current objections to the aquaculture industry and resent a modest proposal of how, if at all, we should proceed, with its development. [Sat AM2]

As with terrestrial systems, dependence on aquaculture of one or two species increases the socio-economic significance of production failure/disaster. Diversification of the number of species being cultured not only buffers such losses, but broadens market options and seasons. A recent ACOA-NB Dept. of Fisheries and Aquaculture program was developed in recognition of this need, and is currently supporting several research initiatives investigating "Alternative Species for Aquaculture". Since historic experience has shown that adapting most organisms from a wild existence to intense or semi-intense culture, inevitably raises health problems, the Alternate Species Shellfish Aquaculture component of the Program includes a project aimed at determining the health profiles of three clams species being considered for adaptation to culture in Atlantic Canada - namely, Mya arenaria (soft-shell), Mercenaria mercenaria, M. mercenaria var. notata (quahaugs) and Spisula solidissima (bar or surf clams). In addition to historic data collected by DFO, this Alternate Species project has now complied a comprehensive reference base, similar to those established for other molluscan species. This can now be used to identify problems as, and if, they emerge, and is the first pro-active molluscan health program of its kind. Health problems arising in other species developed for aquaculture have, traditionally, suffered from the lack of a reference base to determine the source of the health problem (e.g. increased population densities enhancing host-to-host transfer of infectious agents; opportunistic agents taking advantage of stressed or handling-damaged hosts or exposure to "new" infectious agents as stocks are transferred from one area to another for use as broodstock or juvenile stocking.). Infamous examples of such cases for as broodstock or juvenile stocking.). Infamous examples of such cases for molluscs include Bonamiasis of European oysters (Ostrea edulis), which evaded detection and was transferred to most European oyster growing countries over a period of 20 years, before the agent was finally identified as the intrahaemocytic protistan Bonamia ostreae. Since 1996, the clam parasite, pest and disease project has highlighted at least three conditions and infections which merit close attention, as we adapt these clams species to culture conditions. Gonadal neoplasia of soft-shell clams has the potential to affect reproduction and recruitment success in affected stocks. A haemic neoplasia condition has been linked to high mortality and losses during processing, and a labyrinthulid parasite of quahaugs, known as "QPX", has been linked to broodstock mortalities of quahaugs under hatchery conditions, as well as mass mortalities of open-water populations in Massachusetts. Research findings, which are being applied to the culture of these species, will be presented. [Sat AM1]

Water temperature is the main factor limiting the geographic distribution of brook trout. Land clearing, climate changes, and impoundments have reduced the amount of cool water in summer in many rivers. My study was conducted on the Midgell River, a warm water impounded stream, and the Valleyfield River, a cool water impounded stream. The study was undertaken in 1994 and 1995, during two relatively warm summers. Maximum water temperatures reached 30° C in 1994, and 29.6° C in 1995 in the Midgell River, while maximum temperatures remained relatively cool (i.e. <23.4° C) in the Valleyfield River. When waters become warm, most Midgell River brook trout move to cold water refuges, which represent less than 5% of the watercourse length. Snorkelling surveys were conducted in springs in the Midgell River system. Brook trout were observed in springs at intermediate to high numbers when mean daily water temperatures surpassed 19° C for about 55 days from mid-June to late-August. Atlantic salmon were observed in the open river during warm periods. The incidence of scars and abrasions (i.e. predator marks) on brook trout increased with the time spent in springs. Brook trout were not observed to use cool water refugia in the Valleyfield River and the incidence of predator marked trout was 2 to 10 times lower compared to the incidence of predator marked trout in the Midgell River. Salmonids that are not thermally restricted to small regions of watercourses, therefore, may be much less susceptible to predation by kingfishers, herons, mink, and older year class salmonids. The Midgell River population was estimated as 3,505 (95%c.i. 2,174 - 5,967) brook trout (i.e. 408 km^-1 of stream), and the Valleyfield River population was estimated as 41,237 (95%c.i. 34,094 - 49,884) brook trout (i.e. 3,586 km^-1 of stream length). The results of my study suggest that warm water temperatures do indeed serve to limit brook trout production in some Prince Edward Island streams. Watercourse alterations, which warm summer temperatures to levels stressful to salmonids, may have negative implications to the recreational fishery. [Fri PM2]

The goal of this study was to evaluate the response of white sucker (Catostomus commersoni), brook charr (Salvelinus fontinalis), and other fish species to the mass removal of white sucker in five Québec (Canada) lakes. The population structures before, the year of, and three years after white sucker mass removal were compared. White sucker removal ranged between 14.2 and 31.3 kg/ha, which corresponded to reductions in catch per unit of effort (CPUE) between 0 and 72% of their initial abundance, and reductions in biomass per unit of effort (BPUE) between 0 and 64% of their initial biomass. In four of the study lakes, white sucker age frequency distributions showed an increase in the proportion of 2+ to 4+ individuals following mass removal. Mean CPUE and BPUE of 1+ brook charr increased significantly after three years of mass removal in the three lakes where white sucker removal was highest. All white sucker populations experienced growth increases after mass removal, while improved brook charr growth was observed in lakes where the most intensive mass removal occurred. Better growth led to an increase in mean adjusted fecundity in white sucker and brook charr, both in lakes where mass removal was most intense. Growth increases also led to higher mean length at maturity in white sucker females, as well as to decreases in mean age at maturity in white sucker males and in brook charr males and females. The present study suggests that white sucker and brook charr exhibit compensatory recruitment, growth, fecundity, and length and age at maturity responses following a reduction of intra- or interspecific competition. The intensity of mass removal and differences in lake characteristics are, thus, factors that may influence the amplitude and type of life history response to a modification in competitive intensity.

The development of a national freshwater fish database is a critical component of identifying and protecting Canada’s freshwater fish biodiversity. It is a necessary precursor to analyses identifying patterns of biodiversity and the historical, environmental and anthropogenic processes that influence these patterns. It is also critical for the development of strategies by both federal and provincial governments to evaluate, conserve and protect freshwater fish biodiversity. The first phase of this project identified existing sources of fish distribution data. Over 30 provincial and territorial agencies, and North American museums were contacted regarding the availability of Canadian fish distribution data. These organizations made over 400,000 digital distribution records available for this project. The second phase of this project is currently underway and involves developing a single standardized database using the available distribution records. Each record in this standardized database will be georeferenced and assigned a tertiary watershed code with the aid of a GIS. A watershed-level distribution map will then be plotted for each species to determine the accuracy and completeness of the database. Outliers will be examined for accuracy, and gaps may be filled using additional data or spatial interpolation techniques. The completed database will be initially used to generate species by watershed lists for use in the DFO project, "Defensible Methods of Assessing Fish", and in biogeographic analyses examining the processes that structure species richness patterns and fish faunal regions at a national scale. [Poster]

Brook charr, Salvelinus fontinalis, shift their diet through summer season, from benthic organisms of the littoral zone to zooplankton of the pelagic zone in lakes of the Canadian Shield, due to a seasonality in the abundance of benthic invertebrates. This diet shift is more pronounced when charr live in sympatry with white sucker, Catostomus commersoni, which use more efficiently benthic organisms. A bioenergetic-based approach was used to test the hypothesis that charr shift their diet according to energy return in these lakes. The specific objective of the present study was to evaluate the energetic costs associated to feeding on zoobenthos and zooplankton by juvenile brook charr in the field. We use a portable system of underwater cameras in field enclosures to estimate the costs of activity in both habitats, with the stereocinematographic approach. The costs of activity, as estimated from activity rates (fish second Ÿ m^-3 Ÿ h^-1) and mean swimming speed (cm Ÿ s^-1) with an empirical model, were 0.237 ± 0.194 Kj Ÿ 12h^-1 in the pelagic zone and 0.175 ± 0.104 Kj Ÿ 12h^-1 in the littoral zone. When integrating these values in a cost/benefit model, the energy left for growth and metabolism (Kj Ÿ 100 g wet^-1 of fish Ÿ d^-1) was significantly higher in the littoral (19.01 ± 10.0) than in the pelagic (9.03 ± 4.8) zone. No clear effect of temperature was observed, probably because our experiments were done at a low range of temperatures. Our results support the hypothesis that brook charr shift their diet to maximize their energetic return in lakes containing white sucker. [Sat AM1]

With the exception of large, economically important fisheries, it is unfeasible to consider management of Ontario’s 3500 walleye lakes and 2000 lake trout lakes on an individual lake basis. Instead, rapid assessment techniques must be applied to provide a series of ‘snapshots’ of fisheries resource status, measured as angling effort and population abundance, which would be used to track the success of management prescriptions. Three quaternary watersheds in northwestern Ontario were selected to assess this approach. Approximately 850 lakes occurred within this 500,000 ha area, including 13 that were known to support lake trout and 147 with walleye. Angling effort was estimated through aerial surveys of all lakes, identifying the number and location of fishermen in both winter and summer. Current levels of effort were then compared with that considered sustainable, as identified by empirical population models. Population abundance of walleye was estimated through standard gill net surveys on a subsample of 32 lakes, representing different lake productivity and effort levels. Density was then compared with that expected in a lightly exploited situation, based on known relationships between walleye density and lake attributes. Status-of-resource maps were produced to illustrate at landscape scale the overall distribution of ‘healthy’ (acceptable levels of exploitation and high walleye density) versus ‘unhealthy’ lakes (those over-exploited or in a low-density, population recovery phase). Management objectives would be to minimize the number of lakes in the latter group. [Fri PM1]

This project examines the effect of community structure within and among trophic levels on smallmouth bass growth in four New Brunswick lakes. We hypothesize that food web complexity or critical forage species abundance controls growth of smallmouth bass. Stable isotopes carbon (del. 13C) and nitrogen (del. 15N) will be used as indicators of feeding activity (i.e. benthic/pelagic) and trophic level in order to evaluate food web dynamics. Isotope samples were taken from forage base to top predators, and baseline signatures were established for each of the four lakes. Isotope signatures, compared with species abundance data, will be used in regression based analysis to determine how inter and intraspecific interactions may affect recreational fish production in these lakes. We will report on population and community structure data from lake assessments in 1999, as well as initial results from stable isotope analysis as it relates to trophic interactions of smallmouth bass in these lakes. [Poster]

The lake herring (Coregonus artedii) is the primary prey of lake trout (Salvelinus namaycush) in Lake Opeongo, and acts as an important link between the primary predator and the zooplankton community. Lake herring aggregate into schools under moderate to high light conditions (approx. >-1 log Lum/sqm) but disperse under low light conditions (approx.< -1 log Lum/sqm). We used suspended gill nets to capture lake herring aggregations during the daytime period within the epi- and hypolimnion, as well as at the depth associated with the preferred temperature of the species. The method was repeated during the nighttime period to capture dispersed individuals. We measured the nearest neighbour distance between individuals captured within the net, to determine the relative packing density of catches from nets set at different depths and under various light intensities. We will discuss the spatial structure of lake herring aggregations as influenced by the sex and size of individuals within. We will also discuss foraging differences between schooling and non-schooling lake herring. [Sat AM2]

Ontario’s northeast region contains 1,454 known walleye (Stizostedion vitreum) lakes scattered over an area of 441,122 square kilometres. Lakes vary is size from under 2 hectares to over 90,000 hectares. Historically data was collected on very few of these lakes using a variety of methods and throughout the year making it difficult to compare between lakes or over time. In 1993, Ontario began testing a fall walleye index netting (FWIN) method that was developed in Quebec. After a slight modification, the FWIN method is now the recommended provincial index netting standard for sampling Ontario and Quebec walleye populations. To understand the status of walleye populations in the northeast, a representative sample of natural walleye lakes were assessed using the FWIN methodology from 1994 to 1999. Lakes were classified according to surface area, climate, and transparency with a minimum of 2 lakes sampled for each possible combination. To evaluate the impact of past management actions, lakes that had been stocked or had more restrictive fishing regulations were also sampled. Results from this extensive survey program will be used to identify management approaches which are robust to the observed variation in walleye life history parameters seen in northeastern Ontario. [Fri PM1]

We studied the biology of a resident population of cod in Gilbert Bay (52° 35'N, 55° 52'W), Labrador. Gilbert Bay has an area of ~15 km² with limited connection to the coast, and is generally less than 30 M deep. Water temperatures are subzero in winter but surface water warms rapidly after ice leaves the bay, usually in mid May. Spawning occurred from May to early June in 1999, while time to hatch took approximately 24 days. Larvae and pelagic juveniles collected in early August 1998, compared to those collected in 1999, indicated that larval growth is highly variable. Fish, ranging in size from 15 to 100 cm (2 - 15+ years), were captured in shallow water (<5m) throughout the summer. Sexual maturity is attained at 32 cm tl, and 5 years of age. Recapture information also indicated that fish returned to, or remained in, the same location in the bay, from one year to the next. Life history characteristics of this population are compared with other cod within NAFO divisions 2J3KL [Sat PM2]

Once abundant in the waters of Lake Superior a century ago, "coaster" brook trout, so named because they utilize both lake and stream habitat throughout their lifecycle, have been reduced to a few remnant stocks scattered mainly along Superior's north shore. Construction of dams, logging, mining, commercial and sport fishing, resulted in habitat loss and exploitation of the once abundant coaster brook trout. The purpose of this research is to further understanding of the coaster brook trout, specifically their habitat. Radio transmitters were surgically implanted into 20 coaster brook trout to identify the type of lake habitat utilized and at what time in their lifecycle they use it, as well as the type of tributary stream habitat required for spawning and rearing of young. Implantation of the transmitters was completed in May 1999, with telemetry tracking commencing in June. Telemetry locations throughout the months of June and July found coaster brook trout scattered along points and shoals with a shallow basin slope, with primary substrate being cobble and coarse gravel. The vast majority of locations were in water depths of less than 4 m and within 150 m of shoreline. Brook trout tended to remain in this specific habitat type for extended time periods to then move up to several kilometers to another area of similar habitat. In August, fish began staging at tributary stream mouths. In September, flows increased and 12 tagged brook trout ascended the streams with some migrating back out to the lake to ascend a different tributary. Brook trout continued upstream, at times over 8 km, passing through areas previously thought impassable to brook trout, to reach spawning areas of moderate gradient with a small cobble and coarse gravel substrate. Preliminary results suggest brook trout have fairly specific habitat requirements, which will be investigated further, along with the relationship of this habitat to landscape scale variables. [Sat PM2]

Protecting fish habitats and fish populations requires that we do a more effective job of assessing the potential impacts associated with industrial development. The Moose River study encompassed a series of collections between 1991 and 1998 focused on developing an approach for cumulative effects assessment for the Moose River basin in Northern Ontario. There are three main approaches to undertaking an assessment of the potential cumulative effects of development: stressor-based, effects-based and values-based. The dominant approach currently in practice is stressor-based, consisting of the documentation of existing and potential stressors, valued ecosystem components, predicted impacts and possible mitigative strategies. The approach developed for this project is an effects-based approach, emphasizing the use of fish performance measurements to drive the assessment process. The studies examined fish at 20 different sites, through a series of 50 different fish collections. The analyses addressed baseline data, annual and seasonal variability at reference sites, and provided a background of data at 10 undeveloped sites in the Basin. The report outlines the benefits of using whole organism characteristics in an iterative, effects-based assessment program. The collections highlighted differences between rivers, and differences in undeveloped and developed sites. These differences were related to the possible ecological mechanisms limiting fish performance. A study design rationale, methodology and interpretation framework was developed, along with a strategy for decision-making that is amenable to an effects-based approach to cumulative effects assessment. Such an approach requires a level of commitment to baseline monitoring, adaptive management and post-operational monitoring that is currently lacking in most situations. [Fri AM]

Initial studies of the adfluvial bull trout (Salvelinus confluentus) population in Lower Kananaskis Lake, a 646 ha reservoir, 1.5 h west of Calgary, Alberta, showed that adult densities had declined to less than 60 spawning adults in 1992. New regulations were imposed in the spring of 1992 and included a zero bag limit, bait ban, and closure of Smith-Dorrien Creek, the primary spawning stream, to angling. The study continued through 1993 and from 1995 to present. Since the implementation of the regulations, a steady increase in the number of adult spawners has been observed, with a total of 1278 adults in the fall of 1998. This study has provided a unique opportunity to observe the dynamics involved in the recovery of a growth overfished population. Of particular interest are the affects of increasing density on the growth, survival and spawning behaviour of adult bull trout. Current trends show a decrease in the body size of female adults recruiting into the spawning population, and growth rates within the entire population as densities increase. Further investigation will help determine if the rate of alternate year spawning is increasing and if survival rates within a cohort are decreasing. [Fri PM2]

Low head barrier dams are an alternative to chemical control of parasitic sea lamprey (Petromyzon marinus) in streams tributary to the Laurentian Great Lakes. The dams are intended to block the access of spawning adult lamprey to upstream spawning areas. We investigated the impacts of these dams on non-target fish species by a combination of four procedures. We compiled and analyzed historical data from fish sampling in tributary streams over the past century. We conducted an extensive field sampling survey of matched pairs of barrier and reference streams around all five lakes. We carried out intensive studies on selected pairs of barrier and reference streams to estimate movements of fish, and age and growth of fish species. We also compared streams with natural barriers (i. e. waterfalls) to those with low head barrier dams. We suggest the most likely mechanisms for the impacts of barrier dams on fishes, and modifications of dams to minimize those impacts. [Poster]

Peat moss harvesting is a lucrative industry in eastern New Brunswick and was valued at 64 million dollars in 1996. Current methods used in excavating peat have raised concerns regarding its possible effects on the aquatic ecosystem and subsequently on coastal shellfish culture. To better understand the potential impact of peat on the aquatic fauna, a study site was chosen at Mill Creek, a tributary of the Richibucto River, where peat, originating from a commercial operation, has accumulated over several years. A preliminary survey of species from this area revealed the presence of the sand shrimp, Crangon septemspinosa. This crustacean has been used in previous studies as an indicator of environmental stress. The current work consists of two components: first, to establish if there is a relation between peat concentration and the distribution of sand shrimp in its natural habitat; and second, to determine whether peat is present in the digestive system, and if it interferes with the digestive process. Our findings showed that there is an inverse relationship between peat concentration and the distribution of shrimp in Mill Creek. Furthermore, when given a choice between sand and peat substrate in the laboratory, C. septemspinosa displayed a clear preference for sand. Initial histological observations of shrimp exposed to peat, confirmed its presence inside the digestive tract. The ingestion of peat particles may translate into physiological changes affecting growth and development, since shrimp found on substrates displaying high peat concentration were had lower indices of condition then those over sand. [Poster]

Provincial and federal fisheries managers identify that conserving recreational fish stocks is of primary importance. However, in the face of reduced budgets, managers may become increasingly reliant on angler reports to assess the status of fish stocks. This practice has the potential for disastrous results. In this presentation, we illustrate the point for a mixed-stock fishery from Alberta consisting of native cutthroat (Oncorhynchus clarki) and bull trout (Salvelinus confluentus) and non-native brook trout (S. fontinalis). Since 1998, anglers have been involved in an organized program to selectively harvest brook trout from Quirk Creek in an effort to enhance native stocks. Percent composition of the total angler's catch has shown a steady increase in the two native species, which in turn, has been used as a measure for the efficacy of the project. In contrast, population estimates conducted by electrofishing suggest selective angler harvest on brook trout has had little impact on their population. Using population models, we show how size-dependent vulnerability of brook trout to angling coupled with fast growth and early maturity can lead to the erroneous conclusion that their populations are in decline and native stocks are recovering. This has several implications to managers looking at conserving native fish stocks. [Fri PM2]

The traditional difficulties of age interpretation are in achieving precision and accuracy. These problems become critical when aging the whole operculum of older slower growing fish (10-40 years). In these fish, the presence of opaque dense bone tends to obscure early annuli leading to a possible misinterpretation of age. To overcome this, we developed a second technique for sample preparation. A transverse section was cut from the frontal plane of the operculum allowing the interpreter to view beneath the dense bone. We used an index of precision to determine the relative precision of the two techniques. Using both methods to age all sampled fish, we compared mean age to determine if a bias existed. We used the results of the two aging techniques to create age, year class and growth rates for two populations of longnose sucker sampled in western Labrador. [Poster]

Winter flounder (Pleuronectes americanus), which can tolerate a wide range of temperature and salinity conditions, is being examined as a potential species for cold water aquaculture. Our objective was to determine the best diet to maintain healthy fish in captivity over long periods of time. Different feeding treatments were tested: 5 mm trout pellets, fresh fish (capelin, herring, or mackerel), a mixed fish/pellet diet, amphipods, and a mixed amphipod/fish diet. Flounder were captured in spring and fed daily until late autumn with a ration of 1.5% of their wet weight. Fish were not fed during the winter, when the sea water temperature dropped below 0oC, and were fed again in spring for about one month. In May, we measured condition indices, liver and muscle water content, and energy content in both liver and muscle tissues. Total proteins, total lipids, and carbohydrates were also measured in liver. Flounder fed on amphipods had the lowest energy content as well as the highest water content in both liver and muscle. The highest energy content was observed in fish fed a mixed fish/pellet diet. The results indicate that muscle water content is a quick and advantageous measurement for evaluating the overall condition and energy reserves in winter flounder. [Poster]

Recreational fisheries have an aggregate economic value of about $7 Billion annually, yet we are observing collapses at an increasing rate which is eroding economic and social value. Yet these collapses receive little attention, because they tend to be small stocks with only localized importance, and we don't have the manpower to detect and monitor them. In addition, our management strategies tend to mask the collapse of native recreational fisheries stocks. We have identified four processes involving a dynamic combination of fish and human behaviour that act in a depensatory way to hasten the decline of exploited recreational fisheries. These processes will be identified empirically, and their impacts at the population level explored with simple age-structured models. [Fri PM1]

Assessment of the relative contribution of different populations to a mixed-fishery is an important tool of fisheries management. However, stability of these estimations over spatial and temporal scales is rarely assessed. Lake St-Jean system (Québec, Canada) harbours four genetically distinct populations of Atlantic salmon (Salmo salar L.) that reproduce in distinct tributaries. The genetic structure observed has previously been shown to be stable over many generations. The present study aims to determine the origin of individual land-locked Atlantic salmon coming from recreational fishing in Lake St-Jean. For this purpose, analysis of microsatellite DNA have been done on archived scales. Samples of three fishing periods (years 70', 80' and 90') coming from four different exploited areas of the lake have been analysed. Classification of individuals in their population of origin was done by maximum likelihood procedure. Overall classification success of fish in their population of origin reached 88%, as shown by simulation procedure. Our first results show a differential distribution of fish from different populations in the lake. Global exploitation of the four populations is unbalanced, the majority of the fish coming from a single tributary, whereas another population is almost not contributing to the mixed sport fishery. This pattern seems to be stable over many years. [Sat PM2]

L'évaluation de la contribution relative de différentes populations dans une pêche mixte est un outil important dans la gestion des pêches commerciales qui peut également être utilisé pour la pêche sportive. Cependant, la stabilité spatio-temporelle de telles estimations est rarement vérifiée. Le Lac St-Jean (Québec, Canada) est un système où se retrouvent quatre populations génétiquement distinctes de saumons atlantiques résidents (Salmo salar L.) se reproduisant dans quatre tributaires différents. Il a été démontré précédemment que la structure génétique observée au niveau de ces populations est stable dans le temps. Le but de la présente étude est de déterminer la rivière d'origine de saumons récoltés lors de la pêche sportive au Lac St-Jean. Pour ce faire, nous avons effectué l'analyse de l'ADN microsatellite à partir d'écailles séchées. Des échantillons provenant de trois époques distinctes (années 70, 80 et 90) et représentant quatre secteurs de pêche du lac ont été analysés. Le classement de chacun des poissons dans leur population d'origine a été effectué par analyse de maximum de vraisemblance. Des simulations sur ordinateur ont permis d'évaluer à 88% le succès global de reclassement des individus dans leur rivière d'origine. Les premiers résultats obtenus montrent une distribution différentielle des saumons venant des différentes populations dans les quatre secteurs du lac, une des populations étant surreprésentée alors qu'une autre contribue très peu à la pêche en lac. Cette répartition semble être stable sur plusieurs années et permet donc d'évaluer le niveau d'exploitation des différentes populations. [Sat PM2]

Of 761 walleye tagged in Rice Lake in early June of 1989 and 1990, 66 tags were returned within the current fishing season. We assumed tagging mortality, tag-shedding, and non-returning of tags at several levels, retrieved from the literature. Using models with F = 0.30, .35, and .50 (µ = 0.22, .26 and .35), survival ranged from 50-30% for 1989, and 69-46% in 1990. Largemouth bass showed better survival rates. Of 2238 largemouth bass tagged, 206 returns were made over a two-year period. This corresponded to rates of survival of as high as 74.1% in the first year, to 100% for those bass which had overwintered into the second year. Survival of largemouth bass tagged in June and July was higher than for those fish tagged and released in August. As with walleye, survival was higher at the F = 0.30 -0.35 level (as opposed to F = 0.50), a range which we consider to be representative of the fishery in Rice Lake. Assumptions of the model are discussed. [Fri PM2]

YOY walleye habitat and species association patterns were investigated in Big Clear Lake, a mid-sized mesotrophic eastern Ontario waterbody, during the summer of 1999. Fish communities were sampled using a rapid visual assessment technique, and habitat variables were categorized for logistic regression analysis. Preliminary results suggest a temporal shift in prey association and habitat use. In early summer, there were no strong habitat associations, and YOY walleye were closely associated with the presence of YOY yellow perch, one of their primary prey. During mid-summer, YOY walleye were found at all depths in heavy submergent cover areas (>30%) while, by late summer, they exhibited a preference for shallow water (<2 m) and low submergent cover (<15%). Previous data from an Ontario riverine population suggested that YOY walleye have no strong habitat preferences, and that the observed habitat patterns could be explained by the location and density of their preferred prey (Leis and Fox 1996). Our results suggest that, in lake environments, walleye habitat selection is based on both physical and biological parameters, but one or the other is important at different stages of their first year of life. [Sat AM1]

Stock identification is important for fisheries managers when attempting to implement effective regulations and strategies. Stocks may differ in their fundamental characteristics such as growth rate, survival, recruitment, and reproductive rates. Each requires consideration by a manager when developing a plan for a fisheries resource. Many methods have been used for stock identification studies with varying degrees of success. One method is the use of otolith shape. Otolith shape has genetic and environmental influences but appears to reflect the growth rate, which can be significantly influenced by the environment. This study will address the degree to which brook trout populations can be distinguished based on otolith shape on a lake-specific basis within a watershed in Newfoundland. It is also assessing if the segregation reflects differential growth rates in the fish sampled. The otolith shapes were analyzed using Fourier or harmonic shape analysis. We will report if growth rate is the mechanism responsible for differences in otolith shape and the role of environment. [Poster]

Huntsman Marine Science Centre (HMSC) is involved in flounder culture since late 1980's and the results were encouraging. The current program was initiated in February 1998, aiming to establish a pilot hatchery, and to develop protocols and hatchery techniques to produce commercial quantities of juveniles for grow-out facilities. It is a joint venture of HMSC, IRAP, ACOA, DFA and Grand Manan Inc. Major target species selected for the first year are yellowtail flounder, winter flounder and witch flounder. As the program progress, other species such as American plaice, windowpane and smooth flounder will be added to the list. Results from the first six months are very encouraging. [Poster]

Interest in yellowtail flounder (Limanda ferruginus) aquaculture is greatest in Atlantic Canada, due to the relatively high market value and a moratorium on commercial fishing. Research on yellowtail flounder culture at the Ocean Sciences Center has focused on broodstock management, egg incubation, larval rearing and on-growing. Experiments on larval rearing, initiated in 1994, and focused on the effect of prey concentrations, prey type, light intensity and photoperiod. Results of these experiments will be discussed in terms of behaviour, growth and survival. [Sat AM1]

Fish production rate is a key measure of the quantitative performance of salmonid fish populations in any area of interest, as production is linked to habitat carrying capacity and potential fish yield. Estimating salmon production requires the collection of information on the growth, maturation, density and survivorship of fish on a real basis, which is a time consuming and demanding process. For this reason, P/B ratios (ratio of production to standing biomass) are sometimes used as a shortcut method of estimating production, if biomass is known. P/B ratios vary among fish species and populations in different areas, depending on their size and age structure. The hypothesis that annual P/B ratios for salmonids can be predicted from the equation: P/B = a W –0.35, was investigated in this study. The allometric coefficient (-0.35) for this equation was obtained from the literature, and W was the average size of fish in the population. For validation, the calculated P/B ratios were compared to published values of four species: Salmo salar, Salvelinus fontinalis, Salmo trutta and Oncorhynchus mykiss. Species, population and habitat differences in P/B ratios (intercept a of above equation) were examined. As an application, the efficacy of P/B ratios with biomass to measure the productive capacity of salmonid habitat is proposed and examined. [Sat AM1]

The resurgence of an offshore, Scotian Shelf population of Atlantic herring, is described from the spatial distribution of larvae from samples across the entire central Scotian shelf in November 1997 and 1998. The sampling grid was designed using information from circulation models, bathymetry, and the spatial correlation of the seasonal temperature and salinity fields to scale inter-station distance appropriate to the dynamical features of interest. The spatial distribution of herring larvae was not significantly different from that predicted from combination of spawning location, and subsequent advection in the surface layer circulation inferred using the dynamic height method during the survey periods. Atlantic herring larvae were associated with two distinct geographical areas of the Scotian Shelf, specifically, the coastal zone and Western-Sable Bank. The lack of small larvae on the other marine banks (Emerald, Middle, and Banquereau) indicates that Western-Sable Bank is the only important offshore spawning habitat on the Scotian Shelf. The spatial distribution of small herring larvae along the coast extended seaward to the axis of the Nova Scotia coastal current, consistent with rapid cross-shelf transport within the surface mixed layer and broad dynamical retention within the coastal current. The presence of large larvae in a band nearly contiguous with the main axis of the Nova Scotia coastal current also indicates that larvae were advected within the Nova Scotia current from upstream sources. Offshore, herring larvae were most abundant on the flanks of Western Bank, and not found on the crest, where a null circulation is apparent. We hypothesize that the offshore resurgence of herring represents a unique population that spawns in a highly advective environment unlike the typical well-mixed coastal-spawning locations in Southwestern Nova Scotia, and the coastal Gulf of Maine. [Sat PM2]

A multi-stream comparative study conducted in southwestern British Columbia revealed that juvenile coastal cutthroat trout were found at highest abundance in narrow streams (<3 m) with fine substrate (sand and gravel). To assess how cutthroat trout grow and behave within different physical habitats, we selected three streams for intensive examination from the physical structure gradient that we derived in the multi-stream study: (A) narrow width, fine substrate; (B) narrow width, course substrate; (C) wide width, course substrate. We used radio telemetry and bi-monthly recapturing of PIT tagged fish from set trap locations to examine individual growth and movement among the stream types over the fall, winter and spring of 1998/99. PIT tagged individuals tended to be heavier and grew faster during the spring than the winter or fall in all streams. Growth did not seem to be dependent on the physical habitat of the stream. Large fluctuations in water depth during the fall in stream type (C) displaced several radio tagged fish (>100 m) and caused a general downstream movement in PIT tagged individuals (0.78 m/day). Almost no movement was detected during the winter in the three streams for PIT tagged fish (<0.09 m/day), however during the spring movement was in the upstream direction in stream types (A) and (C) (0.11 m/day, 0.33 m/day respectively), with no movement in the stream type (B). Past and current provincial logging practices usually result in removal of riparian vegetation on small streams. This can alter their physical habitat, usually by widening and aggrading the stream bed. Therefore, in streams where riparian logging has occurred, displacement of cutthroat trout during floods may be a concern. [Sat AM2]

Variance in age interpretation, in terms of both precision and accuracy, is a real and important phenomenon of age and growth studies, and consequently of population assessment. However, few studies have examined the subsequent effects of variance on population year-class strengths and age-structure. Previously, we demonstrated that in the absence of systematic bias, random errors affecting accuracy and random errors affecting precision can be considered together using a standard measure of precision. Four series of repeated interpretations of 297 muskellunge (Esox masquinongy) cleithra showed a precision level (mean C.V.) of 5.35% produced no statistically significant differences between the resulting proportional year-class strengths. We now use Monte Carlo simulations to determine what level of variance does result in significant differences in year-class strength, and to examine the effects of varying levels of variance on theoretical year-class strengths representing different recruitment strategies. We manipulated an age-specific error distribution derived from the muskellunge replicate series, to effect different levels of variance and applied the error distributions, using Monte Carlo techniques to age distributions, producing year-class distributions which were then compared to the original year-class strengths. Muskellunge year-class strengths were found to be statistically robust with respect to age interpretation variance. We discuss means of assessing biologically significant differences and the effect of variance on commonly calculated population parameters such as mortality. [Sat PM1]

Brook charr are an integral component to the biodiversity and recreational angling economy of the Miramichi River, New Brunswick. They invaded the 14,000 km² watershed approx. 10,000yb and historical data indicates they frequently migration or movements among habitats in the system. Previous microsatellite studies of brook charr have primarily assessed lake environments. Our objectives were to determine if a metapopulation structure existed in brook charr inhabiting a riverine environment using microsatellites. A total of 12 tributaries encompassing all major drainages were sampled (N = 440 fish). Six primer pairs known a priori to be polymorphic in brook charr were used. Gene diversity was significantly high, HE = 0.94 in some cases, while variance among populations as established by AMOVA was very low (0.94%). HWE tests revealed significant heterozygote deficits for three of six loci. This corresponded to a deviation in HWE for a majority of the populations. Analysis of migration as established by Nm values indicated significantly high amounts of gene flow between populations (18%). Isolation by distance also revealed significant gene flow among populations. It was found that significant pairwise differences existed between populations, but a definitive population structure by drainage groupings could not be elucidated. This may be due to the high extent of gene flow among charr within the Miramichi or the recent evolutionary history inferring limited divergence. Our results suggest subdivisions among populations within this river system, but do not support a definitive phylogenetic structure. [Sat AM2]

While relatively common for terrestrial and freshwater organisms, evidence for extensive introgressive hybridization is scarce in marine organisms. In this paper, we provide evidence for such a phenomenon between two economically important fishes from the Northwest Atlantic; S. fasciatus and S. mentella are, respectively, found in allopatry south and north of a zone of sympatry that extends in the Gulf of St. Lawrence. Using eight microsatellites loci, we documented the extent of genetic divergence among 17 samples; six of S. fasciatus, four "allopatric" samples S. mentella and seven "sympatric" samples S. mentella from the zone of sympatry. Fst estimates were congruent in showing little divergence among populations within S. fasciatus (Fst= 0.0101) and S. mentella (Fst= 0.0008) found in allopatry. In contrast, a pronounced differentiation was observed between these two groups (Fst=0.113). However, samples of S. mentella within the zone of sympatry were less genetically differentiated from S. fasciatus (Fst=0.088) than were allopatric samples of S. mentella (Fst=0.113). This pattern was also confirmed by a Principal Component Analysis (PCA) based on the allelic frequencies of the 17 samples, and a Factorial Correspondence Analysis(FCA) based on individual multilocus information. A detailed analysis of allelic composition also showed an admixture of both S. fasciatus and S. mentella specific alleles in S. mentella found in sympatry, with their relative contribution being 20% and 80%. Altogether, those results provided evidence for pronounced introgressive hybridization between the two species when found in sympatry, and suggested that introgression is more towards S. mentella. . [Poster]

Tandis que l'hybridation introgressive est un phénomène relativement commun chez les organismes terrestres et d'eaux douces, il est beaucoup plus rare dans le milieu marin. Dans cette étude, ce phénomène a été mis en évidence entre deux espèces vivipares de sébaste, économiquement importantes dans l'Atlantique du Nord-Ouest. S. fasciatus and S. mentella sont retrouvés en allopatrie au nord et au sud d'une zone de sympatrie qui s'étend jusqu'au Golfe du Saint-Laurent. Par l'analyse de 8 loci microsatellites, nous avons documenté l'étendue de la divergence génétique entre 17 échantillons de sébaste; 6 échantillons de S. fasciatus, 4 échantillons de S. mentella en allopatrie et 7 échantillons de S. mentella dans la zone de sympatrie. Les estimés Fst n'ont montré qu'une très faible structure des populations des échantillons de S. fasciatus (Fst=0.0101) et S. mentella (Fst=0.0008) en allopatrie. Par contre, une nette différence était observée entre ces deux groupes (Fst=0.113). Pourtant, les échantillons de S. mentella de la zone de sympatrie étaient moins différenciés génétiquement de S. fasciatus Fst=0.088) que ne l'étaient les échantillons de S. mentella en allopatrie fst=0.13). La divergence intermédiaire de S. mentella de la zone de sympatrie a également été confirmée par deux analyses multivariées: une "Analyse en Composantes Principales" (ACP) basée sur les fréquences allèliques des 17 échantillons, et une "Analyse Factorielle de Correspondance"(AFC) basée sur l'information génétique multilocus des 780 individus. Une analyse détaillée de la composition allèlique a montré que les chantillons de S. mentella dans la zone de sympatrie comprenaient un mélange d'allèles spécifiques de S. fasciatus et de S. mentella allopatriques, avec une contribution respective de 20% et 80%. Ainsi, cette étude a mis en évidence une hybridation introgressive prononcée entre les deux espèces lorsqu'elles sont en sympatrie, et suggère que l'introgression est surtout dirigée de S. mentella vers S. fasciatus. [Poster]

Roussel, J-M.¹, A. Haro², and R. A. Cunjak¹. ¹Department of Biology, University of New Brunswick, Fredericton, NB; ²S.O. Conte Anadromous Fish Research Center, National Biological Service, Turner's Falls, MA

Studying movement and habitat use by individual fish in streams is often required to address the functional mechanisms in fish-habitat relationships. Whereas radio-tracking is used for large fish, the technology is unsuitable for small species because of the size of the transmitter. We have investigated a new method for tracking fish using the Passive Integrated Transponder (PIT) technology. The portable device consists of a chest-mounted reading unit (equipped with a palm-top computer), a 12 V backpack battery, and a 50 cm diameter antenna mounted on a 4 m long pole. The method was field-tested with wild Atlantic salmon parr (fork length (FL) range = 84-126 mm), using 23 mm long and 4 mm diameter PIT-tags surgically implanted in the body cavity of the fish. Laboratory experiments indicated no post-tagging mortality for fish >85 mm FL, and no tag loss when sutures were employed. In the field, tag detection distance was 70 cm. While holding the antenna above the stream surface, the operator could locate a fish's position to within a square metre. Enclosure experiments indicated that 80% of tagged parr, on average, were detected by the reader. The technology can be applied to address numerous questions in the fields of animal behaviour, habitat use and population dynamics. [Fri PM1]

Organotins are potent immunotoxins for animals yet are used widely in industrial and agricultural applications and are routinely found in aquatic and marine environments in potentially harmful concentrations. We investigated the effects of one organotin of particular environmental concern - tributyltin (TBT) and its dealkylated metabolite dibutyltin (DBT) - on the immune function of a common, coastal, marine, bivalve mollusc, the blue mussel (Mytilus edulis). Groups of 20 mussels were exposed continuously to water-borne TBT or DBT (0, 5, 10, 20, 40,or 80 ng Sn/l) under flow-through conditions and were monitored for membrane injury (MI), phagocytotic activity (PA), lysosome retention (LR), and haemocyte count (HC) on each of days 0, 1, 4, 11, 18, 25 and 32. All immunological measures were significantly affected by both TBT and DBT. MI increased with time of exposure to both butyltins to a maximum of 40-50% of haemocytes showing damage. TBT produced greater and more rapid MI than did equivalent doses of DBT. Rapidity of MI was dose-related, evident as early as day 1 for >10 ng/l TBT and >20 ng/l DBT, but even the lowest dose tested produced MI in 4d (DBT) to 11d (TBT) and by day 32 all doses of TBT and all doses >5 ng/l of DBT had produced a similar, significant magnitude of MI. Similarly, PA was reduced by all doses of TBT and DBT in a time- and dose-responsive manner. TBT reduced PA more rapidly and to a greater extent than equivalent doses of DBT. PA depression was evident by day 1 for >10 ng/l TBT and by all doses of both chemicals from day 18-32. LR was elevated by 80 ng/l DBT between days 1 and 11 but not thereafter. High doses of TBT produced a similar, though non-significant, pattern of elevation and then fall in LR. Significant elevations and depressions in HC were observed in mussels exposed to both butyltins though effects were highly variable and not clearly related to either dose or time of exposure. Additionally, groups of 10 mussels exposed for 32 d to the same concentrations of TBT or DBT described above were challenged with a low concentration of the common bacterial pathogen Listonella anguillarum (=Vibrio). For both butyltins, a significant dose-related impairment in mounting a successful defense against Vibrio was observed. Although no accumulation of either butyltin could be detected in the mussels exposed to doses lower than 40 ng/l, this study established links between exposure to environmentally relevant, water-borne concentrations of TBT and DBT, immunocompetency and susceptibility to disease in the blue mussel. [Fri PM1]

La pêche sportive à la ouananiche (Salmo salar) du lac Saint-Jean est une activité d’importance conomique et culturelle majeure pour la région. Cependant, on observe des fluctuations interannuelles d’abondance et de croissance très importante de cette ressource, ce qui influence directement les retombées de cette activité. Une des hypothèses retenues pour expliquer ces fluctuations est celle d’une variation dans la production de la proie préférentielle de la ouananiche, l’éperlan arc-en-ciel (Osmerus mordax). La compréhension de la structure populationnelle des éperlans dans le lac pourrait donc s’avérer très importante afin de mieux prédire la production de ouananiche. Nous avons récemment observé deux mode de taille très distincts (longueur à la fourche = 93 mm et 187 mm ) de géniteur d’éperlan se retrouvant au même aires de fraie et au même moment, et ce dans deux tributaires du lac. Afin de vérifier l’hypothèse nulle voulant que les éperlans représentant ces modes de tailles appartiennent à un même pool génique, nous avons premièrement réalisé une étude morphologique basée sur l’analyse de caractères morphométriques et méristiques. Nous avons de plus développé et appliqué les premiers locus microsatellites spécifiques au genre Osmerus. Ces locus se sont avérés très polymorphe avec un nombre d’allèle par locus variant entre 9 et 17, se traduisant en un estimé de diversité génétique (He) variant entre 75% et 80 %. Nous présenterons les résultats des analyses et discuterons de l’utilité des marqueurs génétiques développés pour l’assignation populationnelle à tout les stades de vie, de même que pour l’analyse de contenus stomacaux de ouananiche. Genetic and morphological differentiation between dwarf and normal forms of smelt in Lac Saint-Jean. [Fri PM2]

Landlocked Atlantic salmon (Salmo salar) sport fishery is of primary economical and cultural importance for the region of Lac Saint-Jean. However, strong interannual fluctuations in growth and abundance directly influence the potential benefits of this activity. One hypothesis considered to explain such fluctuations is that they are related to that of smelt (Osmerus mordax) production. Consequently, understanding smelt population structure in the lake may be of importance to better predict salmon production. We recently observe a strong bimodal distribution (fork length = 93 and 187 mm) in the size of adult smelt found on the same spawning grounds at the same time in each of two tributaries. In order to test the null hypothesis of a single gene pool for smelt, we have undertaken a morphological study implying both morphometric and meristic characters. We also developed and applied the first microsatellites specific to Osmerus. These loci proved to be highly variable, with the number of alleles per locus varying between 9 and 17, and translating into gene diversity estimates varying between 75% and 80%. We will present the results of our analyses and discuss the potential usefulness of those markers for population assignment at all life history stages, as well as for salmon diet. [Fri PM2]

Identification of eggs and larvae of Pacific mackerels (Scomber japonicus and S. australasicus) will make it easier for us to predict 0 group abundance of these fishes. Species identification using DNA markers is especially helpful for this purpose. As a preliminary study, we sequenced and compared ribosomal (r) internal transcribed spacers (ITSs) from several Pacific mackerels. Sequencing of plasmid clones containing PCR-amplified regions revealed 3' terminus of 18S rDNA, ITS1, 5.8S rDNA, ITS2 and 5' terminus of 28S rDNA. While sequences of coding regions are stable among samples, two ITSs showed greater differences between species including insertion/deletion of a stretch of several nucleotides. Sequence differences are few among individuals within species. Sequencing several PCR-clones revealed some variabilities within individual, which is hardly likely to be the PCR error. An S. australasicus individual has a few S. japonicus-like ITS2 copies as well as the typical S. australasicus copies. Coexistence of a number of paralogous copies gives an insight into the evolutionary history of the Pacific mackerels. We designed species-specific PCR primers in the sequenced region. An allelic PCR using these primers successfully distinguished these two species. [Fri PM2]

We cloned and sequenced the complete mitochondrial genome of Japanese flounder Paralichthys olivaceus. Sequencing by serial nested sets of deletion and primer walking revealed a circular 17,090bp genome containing 37 structural genes as in other vertebrates so far reported. The organization including gene order is quite similar to that of other teleostean fishes as well as placental mammals. The putative control region of the flounder mitochondrial genome contains a length variable region of about a 74bp tandem repeat cluster. As a preliminary study, we adopted the maximum likelihood and neighbor-joining inference methods to examine phylogenetic relationships among teleostean and related fishes. Comparisons of amino-acid sequences of protein-coding genes and nucleotide sequences of tRNA genes indicated that phylogenetic resolving performance is different from gene to gene. While a large portion of mitochondrial genomic sequence is suitable for phylogenetic inferences among teleostean fish, some individual coding genes gave non-canonical trees. Unless entire mitochondrial sequence is available, we recommend to use sequences of such genes as NADH dehydrogenase subunits 1, 4, 5, cytochrome oxidase subunit 2 as a compromise between sequencing economy and phylogenetic resolution for the highest biodiversity of teleostean fish among vertebrates. [Poster]

Electrofishing is one of the most commonly used techniques for capturing stream fish, although recently, its utility has been questioned due to concerns for animal welfare. While numerous studies have examined its effects on mortality, injury, growth, and behavior, little has been done to assess direct physiological responses. Cardiac output (CO), heart rate (HR), and stroke volume (SV) were monitored in rainbow trout, Oncorhynchus mykiss, electroshocked (pulsed DC) with various voltage, frequency, pulse width, and shock duration settings. Additionally, behavioural recovery times and internal hemorrhaging were also examined. During electroshock, heart function became erratic and typically ceased for much of the event. Following electroshock, CO increased almost entirely due to an increase in SV. Cardiac function generally returned to resting levels within 2-3 h. Behaviourally, recovery was much more rapid, typically taking only a few minutes for fish to regain equilibrium and begin swimming normally, and only rarely exceeded 1 h. Internal injury ranged from 0 to 7 square cm of hemorrhaging along the spine and surrounding musculature, although only 4% of the fish had corresponding damage to the vertebrae. Comparisons across the various electrofisher settings indicate that response and injury are highly variable, but generally increased with more intense settings. Studies examining the physiological and behavioural response of fish and subsequent recovery will be necessary for the development of electrofishing guidelines which minimize the disturbance to fish. [Fri PM2]

The associations between spiny dogfish (Squalus acanthias) and hydrographic variables (temperature, salinity, and depth) were examined in the Bay of Fundy and Scotian Shelf from 1970 to 1998. Data were obtained from standard groundfish bottom trawl surveys. Males were found to occupy bottom-water of significantly higher salinities and depths than that occupied by females. Length also significantly affected habitat associations where occupied salinity increased and depth decreased as the total length of dogfish decreased. Overall, the occupied temperatures, salinities and depths were significantly different from those which were available. Dogfish occupied warmer temperatures along a narrow range (6.62 - 9.19°C) compared to those which were available (1.57 - 9.35°C). Occupied salinity (32.70 - 34.43 ppt) and occupied depth (88.62 - 184.66 m) were also distributed along a narrower range than available salinity (32.16 - 34.79 ppt) and available depth (55.00 - 218.10 m). Sex specific, length specific and overall environmental preference by dogfish may bias traditional ‘offshore’ groundfish surveys while large scale changes in hydrographic parameters may alter dogfish distribution and their interactions with other marine fauna. [Poster]

Triploid female fish are unlike normal diploids in that they show impaired oocyte growth and are unable to produce viable offspring. Compared to diploids, triploid brook trout (Salvelinus fontinalis) have significantly lower plasma estradiol and testosterone levels during their first and second reproductive years, and can produce only a fraction of the normal number of oocytes. As well, triploids in their third reproductive year show a great variability in oocyte size, ranging from pre-vitellogenic stage (<1.0 mm diameter) to mature stage (4.0-4.4 mm diameter) oocytes. Thus, unlike in diploid brook trout which can produce mature oocytes as 2 year-olds, triploid females cannot produce mature oocytes until 4 years of age. The purpose of this study was to compare in vitro steroid production by diploid and triploid ovarian follicles, in order to determine whether delayed oocyte growth in triploids is the result of impaired ovarian steroid production. For a given experiment, intact oocytes (i.e. including follicular tissue) were manually separated from surrounding connective tissue of age 3+ triploids, pooled and then sorted according to small (< 2 mm diameter), medium (2.2-3.0 mm diameter) and large (4.0-4.4 mm diameter) size. Pre-maturational intact oocytes (2.5-3.0 mm diameter) from a single, age 3+ diploid fish were similarly collected and used for comparison. Five intact oocytes were placed into individual wells of 24-well tissue culture plates with 1 mL medium. Incubations were for 2-16 h at 15° C in the presence or absence of gonadotropin stimulus (SG-G100) and steroid precursors (i.e. testosterone). Following incubation, medium was removed and subjected to radioimmunoassay for quantitation of estradiol and testosterone. The results, as well as their importance to triploid reproductive physiology, will be discussed. [Sat AM2]

In this case history, we describe how the Aurora trout (Salvelinus fontinalis timagamiensis), a unique race of brook trout, endemic to only two small lakes located in northeastern Ontario, was saved from extinction and restored to its native habitat. The Aurora trout was endangered by high acid deposition and extirpated from its native habitat in the Temagami area of Ontario during the 1960's. A captive breeding program that began in 1958 preserved the gene pool until self-sustaining populations could be reestablished during the 1990's. The recovery program included whole-lake liming to improve water quality in the native habitat, reintroduction of captive brood stock, the establishment of reproducing populations in two non-native lakes, and perhaps, surprisingly, the creation of limited angling opportunities to generate public support for conservation. [Sat AM1]

The mobility of individuals has important implications for the population regulation and management of stream-dwelling salmonids. In particular, young fish with limited energetic reserves often face high levels of competition, frequently resulting in slow growth and high rates of mortality. Unfortunately, because of technical difficulties involved with tagging and monitoring small fish, data on salmonid movements are extremely biased towards larger fish. In our study, a total of 187 young-of-the-year Atlantic salmon (3.3-5.1 cm) were individually tagged, between 6-27 July 1999, in Catamaran Brook, New Brunswick, and their location recorded in five consecutive snorkeling surveys conducted over the summer. The study area (total length 120 m) consisted of eight 15 m sites which included five tag-recapture sites and three recapture-only sites. In each of the five tagging sites, all young-of-the-year salmon were caught via snorkeling using two aquarium dip nets, tagged with fluorescent elastomer dye, and released at the exact location of capture. In general, young-of-the-year Atlantic salmon in Catamaran Brook showed remarkable site fidelity over their first growing season. One month after tagging, 87.2% of the 187 fish remained within the study area and most of the tagged fish moved less than 2-3 m from their original location. Similarly, in late September, 54.5% of the 187 fish remained within the study area. These results will be discussed in the context of the restricted movement paradigm. [Sat PM1]

The Ontario Ministry of Natural Resources uses index-netting programs to provide abundance estimates of fish populations. Catch per unit effort (CUE) supplies an index of abundance and it is assumed that CUE is proportional to fish density. The Algonquin Fisheries Assessment Unit is testing this assumption. Two lakes in Algonquin Park with similar surface areas (Bena Lake = 10.54 ha and Cecil Lake = 11.37 ha) and no record of brook trout were chosen. The lakes were stocked with equal numbers of yearling and three-year-old brook trout and were sampled using a standard netting procedure 4 days after stocking. Bathymetric and limnological data were also collected. The catch rate was 2.5 times greater in Bena Lake. Catch rates were compared to the limnological data to determine which factor(s) accounted for the difference in catch. Total volume and light penetration may be related to the spatial distribution of fish and to a fish's ability to avoid capture. Results to date emphasize a need for further research into the effect of environmental variables on the spatial distribution of fish. Further study is planned to investigate temporal variability of the relationship and to determine whether the same factors are important in larger lakes. [Poster]

Fisheries biologists in Alberta are faced with the dilemma managing low productivity, boreal fisheries under the pressure of a large and increasing human population. Commercial fisheries for the most desirable species collapsed in the 1940s, and many of the remaining sport fisheries for walleye and pike have recently collapsed. Success rates for anglers are typically below 15%, catch rates are under 0.1 fish/h, and the sizes of pike and walleye are a fraction of the historical fish size. Existing management strategies (requiring extensive biological data) were ineffective in preventing these collapses, because of there being very few biologists responsible for managing the fisheries in a large geographical area. As a result, meaningful regulations were seldom instituted or were implemented too late for providing effective protection. In addition to affecting the sport fisheries, these sport fishery collapses may have caused large changes to ecosystems. Comparisons of seine net catches shows that small prey fishes have increased greatly in abundance at collapsed fisheries. This may result in depensatory predation on walleye fry. Piscivourous birds have also increased, creating the conditions for a widespread trophic cascade. [Fri PM1]

L'exploitation de la tourbe au Nouveau-Brunswick crée de nombreux emplois et génère un important revenu pour la province. Cependant, l'exploitation des tourbières a des impacts sur l'environnement. Le travail étudie la relation entre le mercure et un dépôt de tourbe causé par l'exploitation d'une tourbière près de Richibouctou, N.-B.. L'eau générée par un drainage extensif du terrain est acheminée vers des bassins de sédimentation avant d'être drainée dans le système aquatique naturel, dans ce cas au ruisseau Mill. Au cours des 15 dernières années, un dépôt de tourbe s'y est formé. Le mercure, qui se concentre dans les tourbières au cours des années par le fait de la déposition atmosphérique, est alors remis en circulation dans l'environnement aquatique. Les objectifs étaient (1) de comparer les caractéristiques de l'eau de drainage à des sites-contrôle dans le bassin versant, (2) de mesurer si l'eau et les sédiments retrouvés près de la tourbière contenaient plus de mercure qu'aux sites-contrôle, et (3) d'évaluer si les espèces indicatrices choisies (Crangon septemspinosa, Fundulus heteroclitus et Mytilus edulis) affichaient des concentrations de mercure plus élevées que la moyenne dans les sites-contrôle. Les résultats obtenus sont les suivants: Les analyses d'eau près de la tourbière démontrent une plus grande acidité et un apport en matières organiques, en nutriments et en mercure. Les concentrations de mercure dans l'eau du ruisseau Mill ne sont cependant pas très élevées en raison de la dilution des marées et de la salinité. Dans les sédiments, le niveau de mercure dans le dépôt de tourbe est presque dix fois plus élevées que dans les autres sédiments du ruisseau Mill. Toutefois, on ne réussit pas à démontrer une plus grande bioaccumulation de mercure dans les sites avec impact de la tourbe. Cela s'explique par la non-méthylation du mercure due à la présence d'ions chlorure qui affecte le pH de l'eau et à la variation des conditions physico-chimiques de l'eau. Aussi, la dilution et le lavage résultant des marées réduisent la quantité de mercure disponible aux espèces. [Fri PM1]

Relation between a peat deposit and mercury bioaccumulation in three estuarine species, Mill Creek, Richibucto, New Brunswick

Peat exploitation in New Brunwick has an important economic role. However, peat exploitation has negative impacts on its environment. This research studies the relation between mercury and a peat deposit caused by the exploitation of a peatbog near Richibucto, N.B.. Exploitation needs an extensive drainage which generates high volumes of water that end up in the natural aquatic system, in this case Mill Creek. Within the last 15 years, a peat deposit has formed. Mercury, found in peat bogs due to atmospheric deposition, is recirculate in the aquatic habitat. The objectives were (1) to compare the drainage water's characteristics with control sites in the watershed, (2) to determine whether or not the water and the sediments found close to the peat bog contained more mercury than those at the control sites, and (3) to find out if the chosen indicator species (Crangon septemspinosa, Fundulus heteroclitus and Mytilus edulis) showed higher mercury concentration than the average recorded at the control sites. The results are as follows : the characteristics of the water sampled in proximity to the peat bog demonstrate greater acidity and an inflow of organic mater and of mercury. However, the mercury concentrations in the Mill Creek water are not very high due to the washing of the tides and the water's salinity. As for the sediments, the mercury concentrations are nearly ten times higher in the peat deposit than in the other Mill Creek sediments. Still, we do not find higher mercury bioaccumulation in the sites impacted by the peat than in the other reference sites. This can be explained by pH and by the fluctuating physico-chemical conditions, factors that can inhibit the methylation process of mercury. Furthermore, the dilution and the washing caused by the tides reduce the amount of the mercury accessible to the species. [Fri PM1]

According to the River Continuum Concept, the structure of invertebrate communities changes from small headwater tributaries to larger mainstem streams. This leads to the question: do fish distribution patterns reflect food (invertebrate) availability? The objective of this study is to quantify the biotic diversity found in streams of different orders in Fundy National Park, New Brunswick, Canada. Predictions are twofold: 1. invertebrate abundance decreases with increasing stream order; and 2. biomass of brook trout (most common fish species) is directly related to invertebrate abundance. Four sites on each of two rivers were chosen, representing 1^st, 2^nd, 3^rd and 4^th order streams. Aquatic invertebrate abundance and brook trout density were determined at each site over three seasons. Preliminary results show that invertebrate abundance, trout biomass, and trout density were higher in small order streams. These results support the inverse relationship between biotic abundance and stream size. The markedly low trout biomass in larger streams in Fundy Park may be due to warm water temperatures (>20 C) and/or biological constraints (e.g. food availability). [Sat PM2]

Largemouth bass (Micropterus salmoides) and smallmouth bass (M. dolomieu) are the top predators in many aquatic systems, and form an important recreational fishery in many parts of North America. Catch-and-release angling for bass during the brood-guarding stage can influence a male's decision to continue parental care or abandon his brood, thereby directly impacting his reproductive success. Recurring decreases in nesting success within a waterbody can alter year class strength and/or recruitment. Largemouth bass in a small Ontario Lake have been closely monitored over the last four years, in an effort to understand the characteristics of the nesting population. Observations revealed that large males spawn earlier than small males, that mating success is proportional to male size, and that the percentage of males that successfully raise broods is positively correlated with brood size. In addition, large nesting males were more susceptible to angling than small nesting males. Large males, therefore, have greater fitness potential than small males, but are more likely to be affected by angling during the nest-guarding stage. It is not known, however, what aspect(s) of angling during nest-guarding induce abandonment. Behavioral experiments involving male removal through angling and brood size manipulations were used to assess parental care decisions in smallmouth bass. The results of this experiment showed that: 1) nest abandonment in smallmouth bass was influenced more by brood loss to predators than by the physical stress of short duration angling; and 2) the defense of offspring declined following brood reductions and angling. These studies are being used to develop and assess alternative conservation strategies for smallmouth bass and largemouth bass in Ontario. [Fri PM2]

The commercial feasibility of haddock (Melanogrammus aeglefinus) aquaculture is being explored in Atlantic Canada. A challenge in the development of a haddock aquaculture industry is the solving of problems associated with the species sensitive culture phases or bottlenecks. Bottlenecks are life history phases or aspects thereof which pose a serious impediment to profitable production. Bottlenecks, once located, can be partitioned into their components, research conducted to solve or alleviate them, followed by appropriate revision of the culture system. This approach has been applied to haddock culture through the development of a pilot-scale hatchery at the St. Andrews Biological Station. Under this approach, each year a target number of juveniles to be produced is established and, based on expected life stage-specific survival rates, a starting quantity of eggs is used to begin production. Mortalities and growth rates are recorded throughout the life history. In parallel with annual hatchery production operations, experiments are designed and undertaken to solve the observed production impediments. New information resulting from these experiments is then incorporated into the pilot-scale hatchery for the next year of production. After several years of evaluation, technology appropriate for profitable production is then potentially transferred to industry. The "systems approach" when applied can (i) reduce the time to market, (ii) be cost cutting, (iii) introduce predictability in production and (iv) generate a higher quality product. [Sat AM1]

M. Trudel*, A. Tremblay, R. Schetagne, and J.B. Rasmussen. Department of Biology, McGill University, Montreal (Quebec), and Environment Department, Hydro-Quebec, Montreal (Quebec).

Stunted or dwarf fish are usually characterized by a much lower growth, reach maturity earlier, and tend to have a shorter lifespan than normal fish. Dwarf fish typically occur in dense populations, presumably because of low food supply, suggesting that stunting may be due to exploitative competition. In this study, we tested the hypothesis that dwarf fish consume less food than normal fish. The energy budgets of sympatric populations of dwarf and normal lake whitefish (Coregonus clupeaformis) were estimated using feeding rates derived from a mercury mass balance model that we recently developed. Our analysis showed that lake dwarf whitefish consumed more food than normal lake whitefish. Thus, our analysis did not support the hypothesis that the low growth of stunted fish is not limited by an unusually low feeding rate. Instead, our analysis indicates that the growth efficiency of dwarf lake whitefish is lower than normal lake whitefish, and that dwarf lake whitefish allocate a larger fraction of their energy budget into metabolism than normal lake whitefish. Our analysis also suggests that the earlier maturation and shorter lifespan of dwarf fish may be due to their higher metabolic rates. [Poster]

Variation in larval input is thought to be a major factor regulating the population abundance of marine populations. On South Caicos, fishing for spiny lobster (Panulirus sp.), conch (Strombus sp.) and various reef fishes accounts for almost 90% of the local economy, yet little is known of the processes regulating larval recruitment to this area. The objectives of this study were to quantify seasonal, lunar and tidal periodicity in larval and early juvenile input through tidal channels connecting the nearshore waters of South Caicos to the deep waters of the Turks Islands Passage. This was accomplished by deploying various collection devices, including channel nets, Witham collectors and small artificial reefs (mini-casitas). Channel net samples indicated that larval lobsters recruited mainly around the new moon, while peak input of reef fishes occurred between full moon and last quarter. Abundance of fish in the channel net increased with tidal flow speed and volume filtered, whereas flow speed and volume filtered had no effect on density of larvae per unit volume filtered. This suggests that the density of lobster and fish larvae in the surface water is not affected by lunar phase or tidal flow, but the increased flow around full and new moon brings more larvae through tidal passes. Pulses of larvae in the channel net were generally followed by catches of newly settled juveniles of the same species in the Witham collectors and mini-casitas several days later. This indicates that channel nets are a useful tool in predicting the timing of spiny lobster and reef fish settlement. [Poster]

When traditional taxonomy based on morphology is poorly supported or resolved, molecular phylogeography can bring insights on the evolutionary legacy that should be preserved by conservation efforts. The North American ciscoes (Coregonus spp.) represent an important segment of the fish fauna heritage of Canada, as they are the most phenotypically variable freshwater fish of North American postglacial lakes. The NA ciscoes are also a group at risk, as seven of the 67 fish species listed in Canada as extinct, endangered, or threatened are ciscoes. However, the taxonomy of this group remains confuse and the phylogenetic relationships among recognized taxa are unresolved. This study documents the genetic variation at mitochondrial and nuclear (microsatellites) markers in samples from 34 locations, including all recognized extant ciscoes taxa and most reported ecomorphotypes of the 'C. artedi complex'. MtDNA and microsatellite polymorphisms did not validate current taxonomic assignations. However, microsatellites could detect low but significant genetic differentiation among sympatric species/morphs, and revealed that differentiation was more pronounced among lakes than among taxa, thus indicating that recent, repeated and intralacustrine divergence is the most plausible scenario for the origin of similar polymorphism in ciscoes. These results are discussed in the context of conservation approaches focusing on species, evolutionary significant units and/or habitats to preserve biodiversity. [Poster]

Lorsque la taxonomie traditionnelle d'un groupe d'organismes est mal définie, la phylogéographie moléculaire peut aider à identifier le patrimoine évolutif qui mérite de faire objet d'efforts de conservation. Les ciscos d'Amérique du Nord représentent un segment important du patrimoine faunique du Canada et constituent le groupe de poissons affichant les polymorphismes éco-morphologiques les plus variés des lacs post-glaciaires. Ce patrimoine est par ailleurs menacé, car sept des 67 espèces de poissons éteintes, en péril ou menacées au Canada sont des ciscos. Par ailleurs, la taxonomie de ce groupe demeure confuse car les relations phylogénétiques entre les taxons de ciscos actuellement reconnus sont mal connues. Cette étude documente les polymorphismes génétiques mitochondriaux et nucléaires (microsatellites) d'échantillons de ciscos de 34 sites incluant tous les taxons reconnus et la majorité des écomorphes du complexe C. artedi. Ces marqueurs révèlent l'absence de concordance entre la taxonomie actuelle et le polymorphisme génétique. Par ailleurs, les microsatellites détectent une différentiation génétique faible mais significative entre les espèces/morphes sympatriques, et indiquent que la différentiation est plus prononcée entre les lacs qu'entre les espèces/morphes. Ces résultats suggèrent l'apparition récente et répétée de polymorphismes phénotypiques similaires chez les ciscos et seront discutés dans le cadre d'efforts de conservation ayant pour cible l'espèce, l'unité évolutive significative et /ou l'habitat. [Poster]

Gynogenetic research, has among other things, been applied to determining sex-controlling mechanisms in teleosts and in production of monosex fish populations. The first step towards these goals is the production of gynogenetic haploids. A gynogenetic individual does not possess any paternal genetic material. The male genome is excluded during fertilization by exposing the milt to UV light. The goal of this study was to find a combination of UV intensity and sperm dilution, which yielded the highest number of gynogenetic haploids. The haploids rarely survived until hatch, which made the use of microsatellite an effective tool to detect any traces of the paternal genome prior to hatch. Larvae that were treated and survived to hatch showed a so-called "haploid syndrome". [Poster]

Anadromy in brook trout is fairly common in rivers with marine access through much of their northern range, but little is known of their movements and habitat utilization in estuarine and marine environments. With the use of new acoustic telemetry technology the movement of 20 anadromous brook trout >30cm were monitored from May through August 1999 in Baie Laval, Forestville, Quebec. Four brook trout were captured in the Laval River during out migration and the remaining 16 were captured in the estuary of Baie Laval. An array of ten VR1 remote receivers was used to monitor movements in near shore estuarine and marine environments. Remote receiver data was supplemented with extensive active-tracking, including measurements of temperature, depth, and salinity upon location of all fish. All of the study fish appeared to remain near shore (<500m). There was a strong tidal periodicity to the movement, with the fish moving up into the bays at high tide and then dropping back to the mouth of the bays at low tide. In August the fish began pre-spawning movements back into the Laval River. Details of movement and habitat use will be discussed in relation to anadromy as a life history strategy and the implications for conservation of wild anadromous brook trout. [Fri PM2]

Experiments were designed to examine the effect of temperature on oxygen consumption and ammonia excretion rates in juvenile Atlantic salmon. Salmon acclimated to 15 C were acutely exposed to either 20 C or 25 C for a period of three hours. Oxygen consumption and ammonia excretion rates were monitored before, during and after the temperature shift. Both temperature challenges significantly increased oxygen consumption and ammonia excretion rates. To investigate a more ecologically relevant temperature challenge, a third group of fish was exposed to a gradual increase in temperature from 15 to 20 C over a period of three hours (approx. 1.7 C/hr). Temperature was maintained at 20 C for three hours after which, temperature was returned to 15 C at the same rate as the previous increase. This last experiment showed, again, a trend of higher oxygen consumption and ammonia excretion rates at higher temperatures although no significant results were obtained. [Poster]

Previous researchers have used the outlines of scales and otoliths to discriminate between distinct stocks in mixed stock fisheries. Fourier analysis has most often been used to create a mathematical description of a silhouette that forms the basis for discrimination. However, theoretical problems exist when this methodology is applied to irregular shapes with non-periodic signals. Fourier analysis is able to obtain information from an outline, but many unique outlines can have nearly identical Fourier decompositions. Wavelets, a relatively new signal processing methodology, is suited for the detection of local features of aperiodic data that may be misrepresented by Fourier analysis. Using walleye (Stizostedion vitreum) scales collected from Manitoba Lakes, we performed a study to address the use of Fourier and Wavelet transforms for quantifying scale outline and discriminating walleye stocks. Comparisons were made between walleye collected from northern and southern basins on Lake Winnipeg and between Lake Winnipeg and Cedar Lake. [Sat PM1]

The Magaguadavic River serves as the principal Canadian indicator site to document interactions among wild and escaped aquaculture salmon. Work here has shown that a time when wild salmon returns to North America have generally been declining, populations located within the Bay of Fundy region are particularly hard hit, and are now on the verge of biological extinction. More than 90% of Canada’s East Coast salmon production occurs in the Bay of Fundy, and in this region, wild runs are struggling to cope with an overload of escaped, domesticated fish. Ecological impacts are, in some cases, similar to those observed in Europe and, in other cases, differ. In the Magaguadavic River, escaped adults have bred with wild fish, possibly introducing domesticated germ lines to the wild. In addition, wild smolt runs are also being swamped by escaped smolts from hatcheries within the system. By contrast, and contrary to the European experience, sea-cage escapees may be homing to the Magaguadavic River, and many of the escapees entering the river are sexually immature. Disease screening has found that escapees are generally free of contagious viral or bacterial infections, although in 1999, a first escapee has tested positive for ISA. Urgent measures are needed to conserve these wild stocks. [Fri PM2]

The geographic and genetic structure of freshwater fishes are shaped by historical and ecological influences. To evaluate these influences on piscine phylogeography in Ontario we investigated the mitochondrial DNA (mtDNA) sequence diversity in the control region of three ecologically dissimilar percids (Teleostei: Percidae) using single-stranded conformation polymorphism (SSCP) and DNA sequencing. We rejected the null hypothesis of shared phylogeographic structure of Johnny darter (Etheostoma nigrum), logperch (Percina caprodes), and yellow perch (Perca flavescens). Although measures of molecular diversity varied according to predictions based on species' ecology, phylogeographic patterns did not. Northern and southern Ontario lineages of Johnny darters were identified, with a secondary-contact zone across central and eastern Ontario. In contrast, Ontario logperch were divided into eastern and western lineages with only minimal secondary contact. Analysis of yellow perch populations is ongoing. Hierarchical population structure of each species was better explained by historical watershed boundaries (hypothesized from patterns of Pleistocene deglaciation) than by contemporary ones. Phylogeographic incongruence among these species appears to be unrelated to ecological parameters. Hierarchical population structure further suggests that incongruent patterns are the result of the contingent use of aquatic dispersal corridors. This work demonstrates that the identification of distinct phylogenetic units of Canadian fishes for conservation should be done on a case-by-case basis. [Sat PM1]

The genetic population structure of European and American eels has been based purely on allozyme and mtDNA. Nearly all previous studies suggest that these catadromous fishes are composed by two unique panmictic populations. Nevertheless, past results have been based upon moderately variable genetic markers and limited sampling efforts. Arguably, a genetic structure may not have been resolved in these methods. A possible genetic structure may exist, based on larval retention, cohort fidelity or homing. Using eight microsatellite loci, we documented the extent of genetic divergence among 22 populations, 12 of A. anguilla (N = 600) and 10 of A. rostrata (N = 500). Estimates of inter- and intraspecific genetic differentiation were calculated from allelic frequencies as well as Fst and Rst estimates. Marine currents are the primary factors causing passive larval dispersal. We, therefore, tested the effects of geographic distance against population structure using a mantel test. An AMOVA approach assessed the partition of variance among several marine units (i.e. Baltic Sea, Atlantic Ocean and Mediterranean Sea in A. anguilla). These results will be discussed with reference to speciation processes, population expansion and microsatellite mutations models. [Fri PM2]

The parasite Diphyllobothrium dendriticum can significantly reduce fish growth and increase mortality in salmonids, especially Arctic charr (Salvelinus alpinus). However, there appears to be some confusion about the copepod intermediate hosts that transmit this parasite to salmonids. According to the literature, Cyclops scutifer should be the primary natural host, yet when I compared host specificity for the European species C. scutifer and Eudiaptomus graciloides, I found E. graciloides was a much better host. Trials indicated that by 14 days post infection no parasites had developed to an infective stage in C. scutifer but by 21 days 5% were infective. In contrast, in E. graciloides 34% of parasites were developed on day 14 and 81% were completely developed by day 21. I found the Canadian copepods Diaptomus leptopus and D. minutus to be excellent hosts for this parasite and routinely use them to maintain the D. dendriticum life cycle in Arctic charr. These findings have important implications for understanding the seasonal pattern of parasite acquisition in northern lakes and why some salmonid populations become more heavily infected than others. [Poster]

Identifying and preserving unique species variants is essential for conservation efforts aimed at the maintenance of biodiversity. A unique colour variant of lake trout (Salvelinus namaycush) identified in southern Algonquin Park, ON, is known locally as "salmon trout", is silver without markings and is restricted to Kingscote Lake. Our objective was to characterize, through analysis of mitochondrial and microsatellite DNA, the genetic identity of the silver lake trout. Through comparative genetic analyses of both the silver and normal varieties of lake trout found within the lake, as well as several outgroup populations, we hope to determine the unique characteristics that define the silver morph with respect to normal lake trout. [Poster]

I used the well studied system of juvenile coho salmon and steelhead trout to experimentally address three questions relevant to theoretical predictions about how interacting species coevolve under dynamic habitat partitioning in variable environments. In natural environments, competitively dominant coho predominate in energetically superior pool environments, while competitively inferior steelhead occupy riffles and undergo niche expansion into pools in allopatry. Steelhead has evolved a broader fundamental niche than coho: a "generalist" fitness set (sensu Levins). The fitness sets of the two species are related to foraging efficiencies in the two environments. There is a tradeoff in competitive ability and foraging efficiency under natural size differences based on emergence time. Coho are superior competitors in pool habitats, yet inferior foragers in riffle environments, while the opposite holds for steelhead. The competitive advantage of coho breaks down in the absence of naturally observed size asymmetries. These experimental results help explain regional patterns of coabundance and suggest testable hypotheses about how patterns of coevolution may vary throughout the species' sympatric range. I discuss the implications of these results to the conservation of freshwater fish communities in the Pacific Northwest of North America. [Sat PM1]

Zhu, Xinhua ¹, Dong Liu¹, and M. K. Litvak². ¹Department of Marine Ecology, Institute of Oceanology, The Chinese Academy of Sciences, Qingdao 266071, P. R. China, ²Department of Biology and Center for Coastal Study and Aquaculture, The University of New Brunswick, Box 5050, Saint John, NB E2L 4L5

The effects of temperature and substrate on survival and settlement of bastard halibut, Paralichthys olivaceus (T. et S.), during the early development stages, were studied under laboratory-reared condition. Average survival of larvae (Y: individual) changed with water temperature (T: ^oC) as polynomial function as Y=-4.5163T²+161.81T-1419.20 (r²=0.5845, n=8, p<0.05), which yielded an optimum temperature of 17.9^oC. We provided larvae with three size grades of sand (small, medium and large) for settlement. Larvae in the tanks with the lager grain size experienced the highest survival. These results suggest that the both of these factors should be used as indicators for monitoring and assessment of recruitment dynamics. [Sat PM2]