Life history tactics of three Central Florida populations of Poecilia latipinna were compared to those predicted by life history selection theory. The three populations were sampled monthly for one year. At each sampling period a series of physicochemical measurements were taken to determine the degree of spatial and temporal environmental heterogeneity within each site. Life history traits determined for each population included size at maturity, size specific fecundity, reproductive effort, and length, weight, and condition of offspring. The population inhabiting the most variable environment was characterized by smaller size at maturity, lower size specific fecundity, lower overall reproductive output, and larger, more robust young. In all cases reproductive effort decreased with increasing female length, although the exact nature of this relationship varied between sites. The observed life history tactics do not fit the hypotheses of bet hedging or r-and K- selection theory, but support the idea that life history traits may vary independently rather than being inherited as coadapted units.

Learning and experience shape mate preferences in many species. My thesis investigates the role of experience on mating behavior of male and female sailfin mollies (Poecilia latipinna). In the first chapter I explore whether adult experience influences male sailfin molly mate preference for their sexual parasite, the Amazon molly (Poecilia formosa), and whether experience could account for reproductive character displacement (RCD) of male mate preference in this species. Sailfin males from sympatric populations show a stronger preference for conspecific females over Amazon mollies than do males from allopatric populations. I exposed males from sympatric and allopatric populations to either a sailfin female or an Amazon prior to a mating trial with an Amazon. For the allopatric population, males with recent experience with an Amazon directed fewer mating behaviors towards an Amazon during mating trials than did males with recent experience with a sailfin. Males from the sympatric population, however, performed the same amount of mating behaviors towards an Amazon regardless of experience. Thus adult experience influences mating preferences and suggests that experience may play a role in RCD in this species. In the second chapter I investigate whether a learned sensory bias could influence female mate preferences. Sensory biases that influence mate preferences can arise through selection on the sensory system in foraging and predator detection domains. I tested whether a learned preference originating outside of the mating domain, specifically a color-based food preference, can be transferred to a color-based preference for a male trait. I trained female sailfin mollies to associate either green or blue with food and then tested their preference for animated male sailfins featuring either a blue or green spot. I found that females did not prefer the male with the same color spot to which they had been conditioned. I discuss the problem of learned preference transfer and suggest directions for future research into the role of learning in sensory bias. / text

Thesis (Ph. D.)--Virginia Polytechnic Institute and State University, 1993. / Vita. Abstract. Includes bibliographical references (leaves 109-117). Also available via the Internet.

Lithium is widely used in the management of bipolar disorder, yet memory impairment is a serious side effect. To assess the effects of lithium on spatial working and reference memories, we have employed a plus maze utilizing spontaneous alternation (SA) and place-learning paradigms in two experiments with the black molly fish. Four treatment groups were gavaged with 20 μl of a 10, 100, or 1000 mM lithium chloride (LiCl) solution or ddH2O vehicle every 12 h for 22 to 24 days. On Day 15, subjects began an 8-day SA task or a 10-day place-learning task. Results indicate that there is a significant difference in SA performance among the treatment groups for Days 1, 2, and 3. Results of the place-learning task indicate that the 1 M dose group needed significantly more trials to reach criterion and made significantly fewer correct first choices than the other dose groups. Capillary ion analysis determinations of plasma and brain lithium levels illustrate linear dose-response relationships to doses administered. Regression analyses indicate that there is a relationship between SA performance and plasma/brain lithium levels during the initial part of testing. Collectively, the results indicate that chronic lithium administration impairs spatial working and reference memories.

To assess the potential influence of coastal development on the quality of estuarine habitat for nekton, we characterized land use and the intensity of land development surrounding small tidal tributaries of Tampa Bay. Based on this characterization, we classified tributaries as undeveloped, industrial, urban or man-made (i.e., mosquito-control ditches). Over one-third (37%) of tributaries were determined to be heavily developed, while fewer than one-third (28%) remain relatively undeveloped. We then examined the nekton community from eleven tributaries in watersheds representing the defined land-use classes. Whereas mean nekton density and species richness were both independent of land use, nekton-community structure differed between non-urban (i.e., undeveloped, industrial, ditches) and urban tributaries. In urban tributaries, the community was skewed towards high densities of poeciliid fishes while typically dominant estuarine taxa were in low abundance or nearly absent. Densities of economically important taxa in urban creeks were also only half that observed in most non-urban creeks, but were similar to those observed in mosquito ditches. Furthermore, six of nine common taxa were found to be in relatively poor condition (6-22% smaller in mass), or were rarely collected, in urban creeks. Reproductive output was reduced for both sailfin mollies (i.e., fecundity) and grass shrimp (i.e., very low densities and few ovigerous females) in urban tributaries. Canonical correspondence analysis differentiated non-urban and urban tributaries based on greater impervious surface, less natural mangrove shoreline, higher frequency of hypoxia and lower, more variable salinities in urban tributaries. These characteristics explained 48% of the variation in nekton data, including the high densities of poeciliid fishes, greater energy reserves in sailfin mollies and low densities of several common nekton and economically important taxa from urban creeks. Our results suggest that urban development in coastal areas has the potential to alter the quality of habitat for nekton in small tidal tributaries as reflected by variation in nekton metrics between urban and non-urban tributaries. To further evaluate the link between coastal development fish-habitat quality, we examined the relationship between landscape development intensity (LDI) and the body condition of juvenile sailfin mollies (Poecilia latipinna), a dominant forage fish in tidal tributaries. Morphometric condition, measured as least-square mean dry weight, did not differ statistically among tributaries (P = 0.85). In contrast, biochemical condition, measured as the concentration of triacylglycerol (TAG), the predominant storage lipid, was significantly different among tributaries (P < 0.0001). LDI explained less of the observed variation in TAG content (R2 = 0.18, P = 0.11) than long-term mean salinity (R2 = 0.81, P < 0.0001), which also tended to be lower in more intensively developed watersheds. We hypothesized that urban land use, characterized by considerably greater impervious surface than undeveloped lands, contributed to altered watershed hydrology, high freshwater runoff and low salinities in urbanized creeks. Together these factors appear to foster conditions conducive to lower energetic cost of osmoregulation in urban creeks, and development of a benthic microalgal community of greater nutritional value than the food resources available in non-urban tributaries. To our knowledge, this is one of the first studies to relate urbanization to the condition of resident fishes. While not directly related to coastal urbanization, the level of predation risk within a habitat is a direct measure of habitat quality that could be reflected by the reproductive strategy of potential prey. To evaluate the use of reproductive metrics of fish-habitat quality, we examined reproduction in P. latipinna from eleven tidal tributaries. Our results revealed a gradient along which females produced many, small offspring at one extreme (mean=42 offspring, 17 mg each) and fewer, larger offspring at the other (24 offspring, 29 mg each). Reproductive allotment ranged from 14.9 - 21.5% maternal biomass. Based on our observation of divergent reproductive strategies, we experimentally tested the null hypothesis of no difference in predation risk among tributaries using a novel quantitative approach to estimate predation. We predicted greater risk in tributaries where mollies produced many, small offspring. Tethering confirmed increasing risk from 16.2 ± 5.3% SE to 54.7 ± 3.6% fish lost h-1 across sites in agreement with observed variation in reproduction. Predation was unexpectedly higher than predicted at one of the four sites suggesting that additional factors (e.g., food) had influenced reproduction there. Our results provide insight into the well-studied concept of predator-mediated variation in prey reproduction by quantitatively demonstrating differential risk for mollies exhibiting divergent reproductive strategies. While the observed range of variation in reproductive traits was consistent with previous studies reporting strong predator effects, higher than expected predation in one case may suggest that the prey response does not follow a continuous trajectory of incremental change with increasing predation risk, but may be better defined as a threshold beyond which a significant shift in reproductive strategy occurs.

The utility of hypervariable DNA polymorphisms as a general population genetics method was studied in three fish species by the use of multilocus DNA fingerprinting. Laboratory lines and field caught specimens from Belize and Florida of the clonal species <i>Rivulus marmoratus</i> were examined to determine the relative contributions of mutation and migration to genetic variation in the species. Specimens of <i>Poecilia latipinna</i>, the sailfin molly, from Florida and Georgia were used to explore the properties of hypervariable markers in the context of an outbred and abundant species that exhibits typical levels of genetic variability at nuclear loci. The results were compared to those of a previous allozyme survey of the same populations. Samples of <i>Morone saxatilis</i> from the Chesapeake Bay system were used to investigate the utility of hypervariable markers in the description of genetic variation of an outbred species depauperate in other measures of genetic variation. The results of this study indicate that variation observed among among <i>R. marmoratus</i> clones characterized by hypervariable loci may be the result of natural selection; based on the analyses of mutation rates and population structure. Results from the work with <i>P.latipinna</i> showed that hypervariable loci could have general utility as a method for studying population structure. This utility was demonstrated in the examination of Chesapeake Bay populations of <i>M. saxatilis</i>. Large degrees of interindividual variation at hypervariable loci permitted the characterization of population structure within Chesapeake Bay populations of this species. / Ph. D.