Pollen dispersal and representation in a neotropical rain forest

Results of 3 years of pollen trapping on Barro Colorado Island, Panama, are presented. Minimum pollen dispersal distances are estimated for the most abundant pollen taxa. Dispersal distances for some taxa appear to be as low as 5 m, while for other taxa at least 50% of the trapped pollen had travelled more than 40 m. No consistent pattern of spatial variation was reflected in the pollen rain, however, samples close to large canopy gaps had exceptional abundances of 'disturbance' taxa. A preliminary analysis of the representation of canopy components indicates that as much as 19% of pollen caught in the traps was derived from large tree species.     

Immune response to nylon filaments in two damselfly species that differ in their resistance to ectoparasitic mites

1. Insects commonly resist parasites using melanotic encapsulation. Many studies measuring immune response use the amount of melanin deposited on an artificial object that has been inserted into the animal as a proxy of the amount of resistance that the host is capable of mounting to natural parasites. 2. The relevance of this methodology to immune response in natural insect populations needs further study. Here, we examined two temperate damselfly species to elucidate the relationships among damselfly size, natural resistance to mites, and the immune response mounted by the same damselflies against nylon filaments. 3. The damselfly species that had high rates of melanotic encapsulation of mites in nature deposited more melanin on the nylon inserts than the species with low rates of natural resistance. 4. In females of this species, those that had resisted mites naturally melanised the nylon filament more aggressively than those that did not resist mites. 5. Our results show some support for the use of nylon filaments to assess natural patterns of immune response in these damselflies, but also suggest that caution should be used in interpreting the responses.     

Complex NAPL Site Characterization Using Fluorescence Part 2: Analysis of Soil Matrix Effects on the Excitation/Emission Matrix

Commercially available cone penetrometer (CPT)fluorescence based sensor platforms have been used to detect non-aqueous phase liquids (NAPLs), such as petroleum oils and lubricants, in situf or more than a decade. These approaches have also been used to detect dense non-aqueous phase liquid (DNAPL) source zones by detecting commingled oilsfuels, and naturally occurring organic materials entrained by DNAPLs and carried to depths below the water table. Several neat NAPLs and mixtureswere added to various soil types and analyzedfor specific fluorescence characteristics to determine the optimal excitation source for site characterization efforts. Using excitationlemission matrices (EEMs), we demonstrate that an optimized excitation wavelength can be determinedfor specific fiuowphores within the NAPL mixtures, and that available systems can be ranked based on the specific contaminant and site soil types. An optimal excitation wavelength yields the maximum fluorescence within an EEM spectrum. We ranked commercially available cone penetrometer fluorescence detection systems according to the potential for ease of detection based on maximum fluorescence response. When soils were added tocomplexNAPLmixtures,analytefluorescence emissionwasattenuatedinpreferential portions of the EEM, leading to differences in the optimal excitation source wavelength. Furthermore, impure silica-containing minerals impact the emission signal, potentially leading to incorrect conclusionsf or several commercially available systems. Our find ings suggest that afrequency-agile (e.g., tunable excitation source) probe system would be superior to any other system commercially available, provided the system would be relatively easy to operate and would have rapid in-situ EEM generating capabilitiesfo r optimization in the field.     

Energy expenditure, body-weight and foraging performance of Storm Petrels Hydrobates pelagicus breeding in artificial nesting chambers

Many avian species, such as Storm Petrels Hydrobates pelagicus, are intolerant of disturbance at the nest, which complicates the collection of data relating to metabolic rate and the use of body reserves during incubation. I describe the design of an artificial nest chamber, which is simple and inexpensive to construct and facilitates the collection of such data. Eighty-one nest chambers situated in a large colony of breeding Storm Petrels had high occupancy rates (29/81 in each of 2 years), and the breeding success of birds nesting in boxes was similar to that of pairs nesting in natural crevices. Direct measurement of carbon dioxide production using standard respirometry techniques and estimations of metabolic rate based on the rates of mass loss during incubation indicated close agreement between the two methods of estimating energy consumption. Assuming the metabolic requirements during incubation are furnished entirely from stomach oil, 76% of the daily mass lost represented stomach oil catabolism. The duration of incubation shifts was unrelated to the body mass, and presumably to body reserves, of Storm Petrels on arrival at the nest. Shifts were usually terminated by the return of the foraging partner. The body mass of birds returning from foraging was relatively constant and was unrelated to the amount of time spent foraging at sea, indicating that the decision rule to return from foraging was the acquisition of a threshold level of body mass (about 31 g). There was a negative relationship between the duration of foraging trips and the body mass of Storm Petrels at departure from the nest and a positive relationship between trip duration and the net mass gain at sea. The use of nestboxes based on the design described here would have a wide variety of applications in facilitating data collection for many cavity-or burrow-nesting species which are sensitive to disturbance.     

Fine-Scale Habitat Associations of Red-Backed Voles in Boreal Mixedwood Stands

Abstract: To provide habitat for late-successional wildlife species, new ecosystem-based forest management practices aim to retain elements of complex stand structure, including live residual trees, dead wood legacies, and advanced regeneration, within managed stands. Predicting the effectiveness of these strategies is a challenge for species whose habitat relationships may involve multiple factors and can vary among sites. For 2 years, we live-trapped a common, late-successional microtine rodent, the southern red-backed vole (Myodes [formerly Clethrionomys] gapperi), in 40 1.4-ha boreal mixedwood sites in Ontario, Canada. Using a neighborhood-scale modeling approach, we related red-backed vole capture locations to spatially referenced measures of overstory trees, shrubs and saplings, downed woody debris (DWD), and forest floor substrate. We further assessed how associations with these features varied with availability of the features within a site and as a function of stand management history. In spring, red-backed voles were associated with trap stations that had, within a 26-m radius, a dense shrub layer, abundant late-decay DWD, coniferous understory and litter, and possibly, understory vegetation associated with moist conditions. Positive associations with shrub cover, late-decay DWD, and a moisture-associated understory were most apparent in sites in which these elements were scarce (e.g., <1,500 stems/ha of hardwood saplings and short shrubs; <0.8% projected ground cover of late-decay DWD). The importance of late-decay DWD; shade-tolerant, coniferous understory composition; and substrate varied depending on a site's management history, with each feature having a strong positive effect in 47–64-year-old stands that were harvested using horse skidding and weaker effects in both 31–40-year-old stands that were clearcut with mechanical skidding and >80-year-old fire-origin stands. Our models of fine-scale habitat relationships for red-backed voles may be useful in establishing structural retention guidelines suitable for wildlife species dependent on late-successional habitat structure. In this regard, retaining abundant DWD and 10–30% live trees at harvest may be effective management strategies for providing favorable habitat conditions at localized scales.     

Satellite Telemetry in Avian Research and Management: Sample Size Considerations

Abstract: Satellite tracking is currently used to make inferences to avian populations. Cost of transmitters and logistical challenges of working with some species can limit sample size and strength of inferences. Therefore, careful study design including consideration of sample size is important. We used simulations to examine how sample size, population size, and population variance affected probability of making reliable inferences from a sample and the precision of estimates of population parameters. For populations of >100 individuals, a sample >20 birds was needed to make reliable inferences about questions with simple outcomes (i.e., 2 possible outcomes). Sample size demands increased rapidly for more complex problems. For example, in a problem with 3 outcomes, a sample of >75 individuals will be needed for proper inference to the population. Combining data from satellite telemetry studies with data from surveys or other types of sampling may improve inference strength.     

Higher gregarine parasitism often in sibling species of host damselflies with smaller geographical distributions

1. This study investigated inter‐specific variation in parasitism by gregarines (Eugregarinorida: Actinocephalidae), among sibling species of damselflies (Odonata: Zygoptera), in relation to relative size of geographical ranges of host species. 2. Gregarines are considered generalist parasites, particularly for taxonomically related host species collected at the same sites or area. Prevalence and median intensity of gregarine parasitism was obtained for 1338 adult damselflies, representing 14 species (7 sibling species pairs) across 3 families within the suborder Zygoptera. Damselflies were collected at three local sites in Southeastern Ontario, during the same periods over the season. 3. Five out of seven species pairs had significant differences in parasitism between sibling species. The less widespread host species was the more parasitised for three species pairs with significant differences in gregarine prevalence, and for two species pairs with differences in median intensity. The more widespread host had a higher intensity of infection as expected, in two species pairs. 4. Future studies on ecological determinants of parasitism among related species should examine robust measures of abundance of species and representation of species regionally.