Momordica charantia L. (Cucurbitaceae), a traditionally used medicinal plant, was the subject of chemical ecology and ethnobotany studies in Togo, West Africa. Momordicins I and II present in leaves were quantified with a novel HPLC method. Environmental factors which are significantly and positively correlated to momordicin content include light intensity and parameters related to drier habitats. These results are consistent with growth-differentiation balance and carbon/nutrient balance theories. Genotypic variation is significant. The Togolese population use M. charantia mainly for stomach aches, chickenpox and measles. Utilization consensus exists between ethnic groups and also between the population and traditional healers. Traditional knowledge of biological activities is confirmed by positive antiviral and anthelminthic activities of leaf ethanolic extracts. Moreover, harvesting criteria used by traditional healers suggest evidence of a certain comprehension of the plant natural chemical ecology.

Forty-seven riffle zones from 21 streams of Eastern Ontario and Western Quebec were sampled in 1998 to describe how characteristics of the benthic invertebrate assemblage (abundance, taxa richness and size distribution) varied as a function of water quality parameters (conductivity, TP, SRP, TSS, NO3+NO2, NH3, TKN, Cl-, SO42-) along a gradient of watershed development. A principal components analysis on water quality parameters revealed that there were two groups of correlated water quality variables that explained the majority of the variability among sites. The first group of variables included chloride, sulfate, nitrate+nitrite and conductivity and represented a gradient of urbanization while the second group represented nutrients and included: soluble reactive phosphorus, total phosphorus, ammonia, total suspended solids and total Kjeldahl nitrogen. Simple and multiple regression models predicting invertebrate assemblage characteristics were fitted using water quality principal components scores as independent variables. Overall, invertebrate assemblage characteristics were related to both groups of water quality variables. Abundances per taxon and size classes generally increased with increased nutrients, and overall abundance and the ratio of abundances of sensitive to tolerant taxa declined with increasing chloride, sulfate, nitrate+nitrite and conductivity. Existing information suggests that the water quality gradient found in these streams is more a reflection of anthropogenic sources than the result of geological differences. Therefore, it appears that human activities affect the distribution and abundance of invertebrates in this region. However our models did not explain a good proportion of the variability. It would seem that stream invertebrates of the Ottawa valley are also affected by other parameters that have yet to be identified.

The benthic macroinvertebrate response to water quality changes was examined in riffle zones of three Eastern Ontario Rivers (Ottawa River, Rideau River and Jock River) during the spring and summer of 1995 and 1996. Water quality gradients were detected within the rivers but these were weakly correlated to invertebrate benthos structure making it impossible to predict one from the other. However, the observed differences in water quality among sites and rivers, although significant, may not provide a large enough water quality gradient to observe a strong signal with the invertebrate data. A principal component analysis on the invertebrate densities revealed that invertebrate variability was greater between seasons than between years. May--June samples tended to have higher variability than July--August samples. These results support a standardized approach to the macroinvertebrate sampling period from year to year and the need to sample during a less variable time of the year such as late summer or early fall. Some of the inherent problems with benthic invertebrate indices are examined and the appropriateness of a coarse taxonomic level are discussed. In general, I found no evidence of strong impacts of water quality on the benthos structure of the three rivers. However, at certain locations where water quality parameters were unusually high particularly compared to the guideline, we observed an increase in invertebrate abundance. This study does provide a reference point for each of our large rivers, from which we can compare to future monitoring work.

Agricultural and urban development incites change in aquatic ecosystems. It is generally hypothesized that change due to agricultural and urban development modifies the aquatic habitat characteristics, which ultimately affect the fish communities. The first objective of this study was to describe the appearance and disappearance of fish species within the Rideau River. Scientific literature, museum archives, newspaper articles, historical atlases, municipal reports, and anecdotal reports were consulted to reconstruct the past 120 years of the fish community. The results indicated that the diversity of fish species increased as a result of fish introductions, through increased boat tragic, fish stocking, transient species, baitfish introductions, live-well dumping and the introduction of exotics. The second objective of this study was to determine whether diversity and abundance of fish in the Rideau River could be attributed to agricultural, urban, or forested land-use type. In order to determine this, the fish community was intensively sampled in each of the three habitat types utilizing various sampling gear (trapnets, seines, backpack electrofisher). Over 9000 fish belonging to 33 species were captured between mid-July through mid-September 1998 and 1999. It was found that fewer species were captured in urban areas as compared to agricultural or forested areas and that the abundance of fish was higher in agricultural areas. Also, land-use was correlated with habitat characteristics and habitat characteristics were correlated with fish community diversity and abundance.

This study assessed the effectiveness of a newly constructed 4.2 hectare wetland in Eastern Ontario as a sink for nutrients within an urban fringe watershed. During the initial two years after construction, the wetland was a sink for total phosphorus (TP) during the spring, summer and early fall. However, it was a large source of TP during late fall. During the spring, summer and early fall the wetland was a minor sink for total nitrogen (TN), but it was an effective mechanism for reduction of TN in late fall. An analysis of the nutrient retention capabilities of plant litter indicated that sediments of lower organic content may be more effective at P retention. The very different reactions of P and N to late fall anoxia in 1997 imply the need to focus design and management of constructed wetlands on the nutrient of major concern within the watershed. (Abstract shortened by UMI.)

Four herbaceous plant communities were described along a natural soil depth gradient in an old field. Biomass and canopy height increased with soil depth. Species richness showed a quadratic relationship, whereas the proportion of native species increased with soil depth. The limiting resource was found to be nutrients, not water, even though the year was abnormally dry. I hypothesized that competition was the main factor structuring these communities. The relationship between competition intensity and soil depth was determined through experimentation using two phytometer species, Liatris spicata and Lythrum salicaria. Competition intensity decreased with soil depth across the four communities, which contradicts the Competitor/Stress-Tolerator/Ruderal Hypothesis (CSR) of Grime and Tilman's Resource Ratio hypothesis. Competition in this old field was primarily below ground. Herbivory from large herbivores did not change along the gradient, whereas stress was found to be greatest at both ends of the gradient.

The mechanisms affecting metal retention by surface-flow wetlands were investigated by relating variations in metal retention to variables such as hydraulic residence time, temperature (i.e. seasonal effects), phytoplankton biomass, photosynthesis and the presence of emergent vegetation. The importance of hydraulic residence time was investigated at the Monahan surface-flow wetland in Kanata, Ontario. This wetland retained metals best during summer and fall whereas during winter, the metal retention was significantly lower. The first-order removal model predicted Fe and Mn retention in the spring and Zn retention from spring to fall in both years of the study. Hydraulic residence times, greater than 7 days, provided the maximum retention of Fe, Mn, and Zn. However, first-order removal models failed to fit summer, fall and winter data for almost every metal under investigation (Fe, Mn, Cu, As) suggesting that hydraulic residence time (<1--25 days) do not regulate metal retention during these seasons. The Monahan wetland also affected the partitioning of metals between particulate and dissolved phases thus potentially affecting the bioavailability of metals to downstream systems. On a yearly basis, the wetland showed significant retention of the dissolved phase, but the retention of total Fe and Mn was poor. The wetland transformed dissolved into particulate metals from spring to fall whereas during the winter, dissolved metals were released. Changes in pH, alkalinity and temperature could explain 11% and 40% of the variation in the ratio of dissolved to total Fe and Mn respectively. Furthermore, from spring to late summer, planktonic algal biomass was negatively related to the ratio of dissolved to total Fe and Mn, which suggests the importance of phytoplankton in affecting the partitioning of metal in young plankton-dominated wetlands. In the Monahan wetland and in an acid mine drainage wetland (Falconbridge) near Sudbury, Ontario, diel changes in metal concentrations followed diel changes in pH and oxygen induced by photosynthesis. During the day, metal concentrations in the water column were low because high pH and oxygen favored the precipitation of Fe and Mn oxides at the sediment-water interface. At night, Fe and Mn oxides were reduced and released to the water column because intense biological respiration decreased oxygen and pH. Diel changes in metal concentrations have to be considered when evaluating the retention performance by surface-flow wetlands. Finally, the effect of emergent vegetation on the concentration and partitioning of metals in surface sediments of four wetlands was investigated. (Abstract shortened by UMI.)

Mercury accumulation in fish is a substantial problem in northern temperate regions. Highly variable levels of mercury have been measured in fish from the St. Lawrence River. Here, I used stable isotopes of carbon (delta13C) as tracers of energy source to distinguish food webs of Lake St. Francis (a section of the St. Lawrence River near Cornwall, Ontario) from those of surrounding tributaries and coastal wetlands. I investigated the role of stable isotopes of nitrogen (delta15N) as estimates of trophic position, and I compared mercury accumulation in yellow perch ( Perca flavescens) from Lake St. Francis and the tributaries and coastal wetlands nearby. Since wetlands and tributaries were thought to be the major sites of methylmercury production, I predicted that fish feeding in tributaries would be more contaminated with mercury than fish of the same size or trophic level that are deriving energy from Lake St. Francis. Several components of the food webs of Lake St. Francis and its tributaries and coastal wetlands were sampled for mercury and/or stable isotope analyses (dissolved inorganic carbon (DIC), particulate organic matter (POM), zooplankton, benthic invertebrates, and yellow perch). (Abstract shortened by UMI.)

Isotope ratios of carbon (13C/12C) and nitrogen (15N/14N), on a parts per thousand (‰) basis relative to a standard, were used to trace carbon flows (delta 13C) and trophic positions of organisms (delta15N). Values of delta13C and delta15N were obtained for sediments, suspended particulate matter (SPM), plants, and animals found in the Miramichi River Estuary (New Brunswick, Canada) and used to show the relationship between food web structure and patterns of mercury biomagnification. On a delta13C and delta15N basis, three different sources of energy were identified: terrestrial carbon, in situ estuarine primary production, and marine phytoplankton. Isotopically depleted delta13C and delta15N values verified terrestrial carbon was a major source of energy to estuarine sediments and SPM. A cluster analysis of mean delta13C and delta 15N values for 47 abiotic and biotic groups in this study helped to structure the community of plants and animals. Eight distinct clusters were formed: (1) estuarine sediment, and estuarine and freshwater SPM; (2) estuarine and marine primary producers; (3) freshwater fish and submerged terrestrial leaf litter, and estuarine oysters; (4) estuarine filter-feeding invertebrates; (5) estuarine deposit-feeding invertebrates; (6) estuarine planktivorous and benthivorous fish, some benthic invertebrates, and two filter-feeding marine fish; (7) estuarine carnivorous and two benthivorous fish; and (8) double crested cormorant eggs (whites and yolks). The results of this analysis show that for this coastal ecosystem, no distinction can be made in delta13C and delta15N values between estuarine and marine primary producers and filter-feeding fish, whereas freshwater fish and submerged terrestrial leaf litter were characterised by their isotopically light delta13C values. Based on measurements of delta13C and delta 15N differences between a predator and its prey, an average delta 13C- and delta15N-trophic enrichment factor (TEF) of 1.87 +/- 0.16‰ and 2.94 +/- 0.14‰, respectively, were calculated for the Miramichi River Estuary food web. Consistent with observations in other aquatic studies, a total of 4.7 delta15N-defined trophic levels were identified for the Miramichi River Estuary food web. (Abstract shortened by UMI.)

The effects of non-polar environmental contaminants on components of the hypothalamo-pituitary adrenal (HPA) axis and the hypothalamo-pituitary-ovary (HPO) axis were examined in herring gull (Larus argentatus) embryos and cultured avian hepatocytes. In the HPA axis studies, regression analysis of herring gull embryo yolk sac organochlorine residues against basal plasma corticosterone concentrations indicated statistically significant inverse relationships for polychlorinated dibenzodioxins/polychlorinated dibenzofurans (PCDDs/PCDFs), total polychlorinated biphenyls (PCBs), non-ortho PCBs, and 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) equivalents (TEQs). The activities of two intermediary metabolic enzymes were inversely correlated with PCDDs/PCDFs. In a second study, high incidences of early onset of hatching were observed for herring gull eggs collected from some Great Lakes sites. Embryos from Scotch Bonnet Island were observed to have higher than average renal PEPCK activities and lower than average corticosterone stress responses. Regression analysis of embryo yolk sac organochlorine residues from three Great Lakes sites indicated a statistically significant positive relationship for basal plasma corticosterone concentrations and chlorinated hydrocarbons (CHCs), including organochlorine pesticides. HPO axis studies involved development of a reverse-transcription-polymerase chain reaction (RT-PCR) bioassay to semi-quantitatively measure mRNA for the estrogen-inducible egg yolk precursor protein, vitellogenin (VTG), in avian embryo hepatocyte cultures. Short regions of VTG, beta-actin and albumin cDNAs were cloned and sequenced for several species of birds. Hepatocyte cultures were prepared from both chicken and herring gull embryos and treated with the estradiol analogue, moxestrol, or the organochlorine insecticide, o,p'-DDT. Herring gull embryo hepatocyte cultures responded with VTG mRNA induction at 1 nM moxestrol, compared with 10 nM for chicken embryo hepatocyte cultures. Both herring gull and chicken embryo hepatocyte cultures responded with significant VTG mRNA induction when treated with 10000 nM o,p'-DDT. In chicken embryo hepatocyte cultures, 4-tert-octylphenol (OP) was determined to be very weakly estrogenic, whereas TCDD, benzo[k]fluoranthene (B[k]F), three halogenated dimethyl bipyrroles (HDBPs) and an extract prepared from a common tern ( Sterna hirundo) egg were determined to be anti-estrogenic. These results indicate that the semi-quantitative avian RT-PCR VTG mRNA bioassay may be useful for predicting whether wild birds may be sensitive (or exposed) to estrogenic or anti-estrogenic environmental contaminants.