The brown trout, Salmo trutta L., displays one of the most variable and polymorphic life-history strategies of all the salmonids. In some populations, individuals spend their whole life-cycle in the river (freshwater-resident) whereas in others, a varying proportion migrates to sea for variable amounts of time to better feeding conditions before returning to spawn (anadromous). The ‘decision’ if an individual will migrate or not will be determined by the balance of the costs and benefits of following a particular life-history strategy. The balance of these, which do not affect males and females equally, will determine the future success (measured by fitness) of each strategy. This research addresses the influences of parental contribution, mainly maternal effect, of anadromous and freshwater-resident brown trout on offspring performance and subsequent life-history. A partial migratory population of brown trout was studied in the Tadnoll Brook, one of the seven major tributaries on the River Frome. The tributary is classified as a circum-neutral chalk stream, 9.9 km long with a catchment approximately 50 km2. Carbon and nitrogen stable isotope analysis (SIA) was used to quantify maternal reproductive contribution of anadromous and freshwater-resident brown trout to offspring and determine the future success (measured by fitness) in terms of size and time of emergence. A panel of 12 microsatellite loci was used to assign parentage to 0+ parr. Using field data collected over 1.5 years on individual fish, this study tested parental influence on offspring performance in terms of size and growth rate and calculate the reproductive contribution of maternal/paternal anadromous and freshwater-residents. Adult life-history strategy was identified using a combination of results from SIA, PIT tag data and ecological data (body size, temperature). Parr life-history strategy (1+) was inferred using PIT tag detection data. The results of the SIA indicated fry of anadromous females emerged earlier and at a larger size than fry of freshwater-resident females. Parentage assignment of parr was low (28 %), with 8 parr assigned to both parents and 43 assigned to only a single parent. There was no detectable effect of parental life-history on parr size and growth rate, however the raw data may suggest offspring of anadromous parents have an early size advantage but a slower growth compared to offspring of freshwater-resident parents during the first year of the parr stage. Twenty-four percent of the offspring were identified as putative smolts at 2+ and both forms interbred and could produce offspring of each life-history. Estimates of reproductive contribution (SIA and growth) show a higher proportion of anadromous females and males (growth only) contributed to offspring production. The results of this research indicate that the maternal anadromous contribution is higher in the Tadnoll Brook population, affording fitness benefits to their offspring during early ontogeny such as size advantages and emerging at a more profitable time to establish feeding territories. Adult life-history does not appear to influence juvenile (0+ parr) life-history but may have an effect on offspring performance. The presence of both forms in the population suggests the anadromous fitness benefits to offspring may only have an affect during ontogeny and early stages of growth. Then after juveniles reach a size threshold environmental factors influence offspring life-history, resulting in the largest parr with the fastest growth adopting an anadromous life-history.

This thesis describes a relationship between maternal preference and offspring performance in a population of the butterfly Euphydryas editha that used two host plants, Pedicularis semibarbata and Collinsia torreyi from 1979 to 2001, but now no longer uses Collinsia. In the light of the known history of diet change in this butterfly population, it is not surprising that maternal oviposition preference was variable. Although the diet of the butterflies that evolved rapidly in the 1980’s is no longer changing, I still discovered some females with a chemical preference for Collinsia. This seems to be a legacy of recent anthropogenic diet evolution. The evolution of host preference of females in Rabbit Meadow has not finished yet. Variation of offspring weight and survival were measured and showed a complex relationship with adult preference. Although quite a few adults strongly rejected Collinsia, their offspring grew well on this host, and there was no significant trend for the offspring of strongly Pedicularis-preferring butterflies to perform more poorly on Collinsia. / text

Maternal effects are the non-genetic contributions of mothers (or fathers) towards the phenotype of their offspring. Maternal effects are now well recognised as a facilitator for evolutionary change in offspring phenotypes and life history strategies which can have effects on population dynamics, population divergence and even speciation. Furthermore, maternal effects have been shown to have a heritable genetic basis and that they are genetically variable, which suggests that they contribute to maintaining phenotypic variation. Maternal effects may impede or accelerate responses to selection which has implications for adaptive evolution and making predictions about their evolutionary potential. The importance of their contribution to phenotypic variation and life history evolution has made maternal effects an important consideration in fields such as conservation and population biology, evolutionary ecology and evolutionary genetics. The aim of this thesis is to investigate if maternal effects can influence offspring life history traits and fitness parameters through maternal resources via the egg. Main questions that are asked include: can maternal effects help facilitate transition to a novel host-diet (Chapter 2); does maternal diet influence egg composition and, if so, does this have an effect on offspring life-history parameters (Chapter 3); is there a genetic basis to egg composition and is there potential for egg composition to evolve (Chapter 4); and are defensive compounds from the diet transferred into the eggs, if so, are these uni- or biparentally transferred and does this offer protection against predation (Chapter 5)? To address these questions we used a specialist insect herbivore, the large milkweed bug Oncopeltus fasciatus (Hemiptera: Lygaeidae). In the wild, O. fasciatus feed on plants from the genus Asclepias (Apocynacea). However, O. fasciatus can be reared successfully in laboratories on sunflower seeds Helianthus annus. For our experiments we used two populations of O. fasciatus, one population has been maintained on dry seeds of A. syriaca while the other population has been reared and maintained on sunflower seeds. The results of Chapter 2 were suggestive of a maternal host-diet effect on egg mass and hatching success, but we did not find evidence that maternal host-diet was significant in influencing a transition to a novel host. In Chapter 3 we found that there was variation in the free amino acid profiles of the eggs between our treatments suggesting that amino acid profiles may be influenced by maternal diet. The results of our multivariate selection analysis to examine linear and nonlinear (quadratic) relationships between maternal diet and the free amino acid profiles of the eggs suggest that there may be population-specific responses which can influence specific amino acid profiles in relation to hatchling mass. In Chapter 4 we used only the milkweed-adapted population to determine if there was a genetic basis to amino acid profiles in the eggs. We constructed a genetic variance-covariance (G) matrix to determine the strength and direction of the relationships between amino acids and to assess the potential for amino acid profiles to evolve. While we found genetic variation for amino acids, and that there was evidence for positive moderate to strong genetic correlations between many of them, we also found evidence for constraints for the potential for amino acid profiles to evolve as evidenced by the calculation of gmax (which represents the linear combination of components that has the highest genetic variance and which is the most accessible to evolution). In Chapter 5 we found maternal, but not paternal, transmission of cardenolides into the eggs. However, this did not confer protection of all eggs against predation from larvae of the green lacewing Chrysoperla carnea. Overall, results suggest that for our populations of O. fasciatus, maternal effects are significant in influencing early life history traits such as egg mass and hatchling mass. However, we did not find any significant effects on other offspring life history or fitness parameters that we measured. This may be surprising as positive, and negative, effects of non-genetic contributions of females (and males) to their offspring has been widely reported in many taxa. The patterns and implications of maternal resource allocation and their effects on offspring life history evolution are explored and discussed, as are the limitations of our experimental designs. I hope that this research can be used to stimulate further investigations into maternal effects and the relationships between host-plant, maternal allocation strategies and life history evolution.

<p>  </p> <p>Heat stress affects breeder ducks raised in North America and other parts of the world, but the effects of such stress on the progenies is not known. Therefore, the objectives of this study were to investigate: 1) The objectives of this thesis were to first investigate the effect of heat stress or exposure to exogenous glucocorticoid (GC) on fertility, production performance, egg biochemistry, egg quality, and welfare of breeder Pekin ducks. 2) the effects of maternal GC on phenotypic plasticity and behavior of the F1 generation. Three studies were carried out to investigate these objectives.</p> <p>The first experiment was conducted to test the hypothesis that chronic treatment with low levels of either corticosterone or cortisol would alter heterophil to lymphocyte ratio (HLR) and immune organ morphometrics. Further, we wanted to determine if chronic treatment with either GC would elicit an increase in cortisol levels in egg albumen. To test our hypotheses, we implanted silastic capsules subcutaneously under the skin of the neck of adult ducks (n = 5/sex/dose) using propofol anesthesia. Capsules contained corticosterone, cortisol, or empty capsules as controls. Over the course of 2 weeks, blood serum, blood smears, body weights, and egg quality data were collected. After 2 weeks, ducks were euthanized using pentobarbital (FatalPlus, 396 mg/ml/kg) and body weight, weights of spleens, livers, and the number of active follicles were recorded. Blood smears were analyzed for HLR by a lab unaware of the treatment groups. Albumen GC levels were assessed using mass spectrometry. Data were analyzed using a 2- or 3-way ANOVA as appropriate and <em>post hoc </em>with Fishers protected least squares difference (PLSD). There were no treatment effects on egg quality measures or body weight. Corticosterone treatment did elicit an increase in serum corticosterone (p < 0.05), but not cortisol levels, compared to controls in both sexes. Both cortisol and corticosterone treatments increased (p < 0.05) serum levels of cortisol compared to controls. Relative spleen weights were higher (p < 0.05) in hens following corticosterone but not cortisol treatment. No other organs showed any differences among the treatment groups. Both GCs elicited an increase (p < 0.001) in HLR in hens at all time-points over the 2-week treatment period compared to controls. Cortisol, not corticosterone, elicited an increase in HLR for drakes (p < 0.05) compared to controls at day 1 after implants. Chronic treatment with cortisol, but not corticosterone, elicited an increase (p < 0.01) in egg albumen cortisol levels compared to other groups. Corticosterone was not detected in any albumen samples.</p> <p>The goal of our second experiment was to test the hypothesis that heat stress (HS) would alter welfare, egg quality, and morphometrics of breeder ducks. Furthermore, we wanted to test if HS would increase cortisol levels in egg albumen due to recent exciting findings that cortisol, not corticosterone, is isolated in egg albumen. To test our hypothesis, adult Pekin ducks were randomly assigned to two different rooms at 85% lay with 60 hens and 20 drakes per room. Baseline data including body weight, body condition scores (BCS) (such as footpad quality, eyes, nostrils, feather cleanliness, and feather quality scores), and egg production/quality were collected the week preceding heat treatment. Ducks were subjected to cyclic HS of 350C for 10h/day and to 29.50C for the remaining 14h/day for 3 weeks while the control room was maintained at 220C. Eggs were collected daily, and body weights were taken on days 0 and 21 relative to the onset of heat treatment. BCS were collected weekly. Eggs were collected weekly for quality assessment and albumen glucocorticoid (GCs) levels assessment using mass spectrometry. One week before the exposure to HS, 10 hens and 5 drakes were euthanized and the same number again after 3 weeks of HS or control exposures using pentobarbital and birds necropsied. Body weight, weights of the liver, spleen, and the number of maturing follicles were recorded. Data analyses were done by 2- or 3-way ANOVA as appropriate with a Tukey-Kramer post hoc test. BCS were analyzed using a chi-squared test. A p value ≤ 0.05 was considered significant. Circulating levels of corticosterone were significantly (p < 0.01) elevated at week 1 only in the HS hens while there was no significant difference in the circulating levels of corticosterone in drakes compared to the controls. The circulating levels of cortisol increased significantly at week 1 (p < 0.05), week 2 (p < 0.05), and week 3 (p < 0.01) in the hens and at week 2 and 3 only (p < 0.05) in the drakes compared to the controls. Feather quality scores (p < 0.01), feather cleanliness scores (p < 0.001) and footpad quality scores (p < 0.05) increased significantly in the HS group compared to controls, higher BCS indicate a decline in welfare. HS elicited a significant (p < 0.001) decrease in egg production at weeks 1 and 3 and a descriptive decrease in the number of fertile eggs upon candling at 10 days of incubation, numeric decrease hatchability and increase in the number of dead embryos in the HS group after the incubation period. Hens in the HS group showed a significantly decreased BW (p < 0.001), and number of ovarian follicles (p < 0.05) compared to controls. Shell weight decreased significantly at week 1 (p < 0.05) compared to controls. Yolk weight decreased significantly at week 3 (p < 0.01) compared to controls. HS elicited a significant increase in albumen cortisol levels at week 1 (p < 0.05) and week 3 (p < 0.05).</p> <p>The third experiment was conducted to determine if parental exposure to heat stress would impair performance, hypothalamic pituitary adrenal (HPA) axis response, welfare, or behavior of their offspring. To achieve these goals, we treated adult drakes and hens at peak lay to heat stress or control temperature for 3 weeks and incubated eggs collected from the last 3 days of the experiment. A total of 76 ducklings were placed into pens from each parental treatment group: control (CON-F1) and heat stress (HS-F1) and raised as grow-out ducks. Weekly data for body weights, body condition scores (BCS), and novel object test (NOT) were collected weekly. At 3 weeks of age, ducks (n = 6 per treatment group) were subjected to adrenocorticotropic hormone (ACTH) (ACTH/cosyntropin, 0.0625 mg/kg) challenge or vehicle as control. Blood samples were collected from the metatarsal vein into serum-separator tubes at 0, 1, 2, 3, and 4 hours relative to treatment for the determination of serum glucocorticoids. Blood smears were also produced from these same samples to determine heterophil to lymphocyte ratios (HLR). All injected birds were euthanized with pentobarbital on the second day relative to ACTH administration, spleen and bursa were removed and weighed immediately. Duck level analyses were completed using 1-, or 2 -way ANOVA as appropriate. BCS were analyzed using a chi-squared test. We observed that HS-F1 had a lower hatch weight (p < 0.05) compared to CON-F1. However, growth rates during the 5-week grow-out period were not significantly different between the two flocks. NOT (N = 4) analyses showed that the HS-F1 had a greater fear response (P< 0.001) compared to CON-F1. Similarly, an ACTH stimulation test showed that the HS-F1 ducks had significantly heightened corticosterone and HLR responses compared to CON-F1 ducks (p < 0.05). The HS-F1 showed altered baseline and ACTH-stimulated levels of cortisol compared to controls.</p> <p>In conclusion, GC elicit differential effects and although corticosterone has been stated to be the predominant GC in avian species, cortisol may provide critical information to further understand and improve welfare. HS decreased performance, fertility, and productivity of breeder ducks. In addition, HS and exogenous GC elicited a selective deposition of cortisol, not corticosterone, in the egg albumen. The maternal cortisol deposited in eggs alter the hypothalamic-pituitary adrenal (HPA) axis and behavioral responses of the F1 generation. This suggests that maternal hormones can alter the phenotypic plasticity of the offspring and can be used to produce offspring that have better adaptation to the rising temperatures as a result of climate change. Finally, the measure of cortisol in egg albumen can be used as a non-invasive marker of stress.</p>