Justifying the maintenance of small natural areas requires understanding their contribution to the conservation of specific natural resources. Mount Rushmore National Memorial (MORU) is a small portion of the Black Hills of western South Dakota and eastern Wyoming. Because it has been protected from logging since the late 1930s, it may serve as an important part of the Black Hills forest as a whole. To understand this role, we investigated the extent and degree of logging activities in the memorial and compared the current structure of the MORU forest to that in the rest of the Black Hills today and before Euro-American settlement. Our results suggest that approximately 29% of the park has had no tree harvesting activity, 18% of the park has had only selective cutting of trees, and 66% (344 ha) of the park's area is covered by old-growth forest. Based on current estimates of similar forest in the remainder of the Black Hills, the forest at MORU constitutes the second-largest area of old-growth ponderosa pine forest in the Black Hills. Although the current structure of the forest does not appear to be outside the range of natural variability for this ecosystem, some components of the forest are near the edge of this range. Conservation of this important natural resource will most likely require more active management than has occurred in the past, but this management will require careful consideration because of the rarity of this resource in the region.

Populus tremuloides is one of the most widespread woody species in North America, occurring across Canada, the northern United States, and at higher elevations in the southwestern United States and parts of Mexico. In Texas, it is found only at the highest elevations in a few disjunct populations in the Guadalupe, Davis, and Chisos Mountains. We examined the structure of six isolated P. tremuloides stands in the Davis Mountains. Stand age ranged from 25 to 65 years. The density of tree-sized P. tremuloides was 429 to 15,520 stems/ha, average basal area was 8.2 to 305.9 cm²/plant, and total basal area 5.1 to 22.9 m²/ha. As stand age increased, density of P. tremuloides decreased and average basal area and total basal area increased. The highest density of seedlings and saplings was in the younger stands. Seventeen woody species were found, with four to 10 woody species in any one stand. While these stands were dominated by P. tremuloides, two stands had Quercus gambelii relative basal area of 34 and 49%, and one stand had Pinus ponderosa relative basal area of 40%. Diameter distributions for P. tremuloides in all of the stands was unimodal but positively skewed. There were fewer seedlings and saplings or small tree-sized stems in the older stands. There were seedlings and saplings in each stand but few small tree-sized stems entering the populations, suggesting a recruitment bottleneck. Continued establishment and potential changing composition of other species in these stands makes the future structure difficult to predict.

Management of rare plants living in small populations requires life history and breeding system knowledge for informed conservation decisions, yet often management and recovery planning proceeds without breeding system information. We performed a series of hand pollination treatments to determine the breeding system of Lupinus sulphureus ssp. kincaidii (Kincaid's lupine), a Threatened plant of western Oregon, and followed its most active pollinator, Bombus californica, to estimate selfing rates. We found that Kincaid's lupine is self-compatible and can also produce seeds through facilitated selfing, a condition without pollen transfer between different flowers on the same plant. Bombus spent at least 75% of their total foraging flight time performing geitonogamous flights, foraging between male and female phase flowers on the same plant, but selfing rates were likely underestimated because Kincaid's lupine is capable of vegetative spread. Moreover, the majority of Bombus foraged contrary to conventional flight patterns on protandrous vertical inflorescences, likely increasing selfing rates. Our results indicate that while it is easy to rely on population seed to assess whether or not a population is inbred, and therefore a candidate for augmentation, a molecular study and/or hand pollination experiments are required to assess population inbreeding depression severity, because other plausible explanations for low seed set have been demonstrated.

Prescribed fire and thinning are commonly employed management practices in mixed-oak forests of the central Appalachians. The effects of these practices on the soil seed bank are important to consider in order to evaluate the full impact of these treatments on plant community dynamics in the understory. Species composition of the soil seed bank was examined under three treatments: thin, burn, and thin followed by burning, and an untreated control. Thinning was conducted in Fall 2000, and burns were conducted in Spring 2001. Soil samples were collected in March 2004, three years following treatment. Species composition was assessed by seedling emergence. Seventy total species were found in the seed bank. Ruderal species such as Erechtites hieraciifolia, Carex spp., and Rubus spp. were the most commonly occurring species across all treatments. No significant difference in species composition or proportion of functional groups was found between any of the treatment groups. However, as found in previous studies, species composition of the seed bank and aboveground vegetation were significantly different in all treatments. Species composition and canopy cover were each significantly spatially autocorrelated in the thinned and burned treatment. While thinning and burning did not produce detectable changes in the species composition of the soil seed bank at these sites, they may influence the spatial heterogeneity of the soil seed bank.

The Northwest Florida coastal landscape consists of a mosaic of sand pine scrub, longleaf pine, and ecotonal communities. Fire suppression has altered the structure and composition of these communities. The effects of environmental variation and the recent reintroduction of fire on plant distribution and abundance across this landscape were investigated. Relative cover of herbaceous and woody ground cover species and density of sapling and canopy trees along with elevation, percent bare ground, percent canopy cover, and litter depth were determined in burned and long-unburned treatments for all community types. Canonical correspondence analysis (CCA) identified elevation and time since fire as the most influential environmental variables for species distribution and abundance. Species most closely associated with lower sites included Gaylussacia mosieri Small, Ilex glabra (L.) Gray, Kalmia hirsuta Walt., Magnolia virginiana L., Pinus palustrus P. Mill., Vaccinium myrsinites Lam., Aristida L. species, and Calamovilfa curtissii (Vasey) Schribn.. Species associated most closely with higher sites included Ilex vomitoria Ait., Quercus chapmanii Sarg., Quercus geminata Small, Quercus myrtifolia Willd., Vaccinium stamineum L., Baptisia lanceolata (Walt.) Ell., Cladonia Hill ex P. Browne species, Eriogonum tomentosum Michx., Rhynchospora megalocarpa Gray, and Yucca filamentosa L. Eight months post fire, burned sites had lower density of woody overstory and saplings compared to long-unburned sites. Herbaceous species richness did not differ between burned and long-unburned sites. Repeated prescribed fires during the growing season and direct seeding may be needed to promote flowering and establishment of additional herbaceous species.

The amounts of old-growth forest present under current and pre-harvesting era disturbance regimes in the dominant higher elevation (900–2300 m) forests in eastern British Columbia (B.C.) – wet cold Engelmann spruce (Picea engelmannii Parry ex Engelm.) - subalpine fir (Abies lasiocarpa (Hook.) Nutt.) forests (ESSFwc), and one important group of ESSFwc forests, the Northern Monashee (ESSFwc2) biogeoclimatic variant – were quantified using a GIS forest age database, with the assumption that oldgrowth forests were forests >140 years old. This was done in order to inform natural disturbance based management in this part of British Columbia. Database constraints restricted the analysis to the post 1800 period only and resulted in estimation of a range of old growth for any time period. The oldest trees in old-growth forests do not necessarily indicate when the forests were last disturbed by fire, as 14C dating of charcoal indicated that, for two of five 210- to 320-year old stands sampled, the most recent fire event probably occurred over 1000 years ago. The amount of old growth in both ESSFwc and ESSFwc2 forests, since 1800, decreased to a minimum in the mid to late 1800s, then increased. In the case of the ESSFwc2, this increase occurred until the 1960s before the amount of old growth decreased again. Amounts of old growth in 2003 (58–59% of the forested area) were within the pre-harvesting era range of 30–60% in ESSFwc2 forests, but may be above the range of 20–50% in ESSFwc forests. Old-growth forests have dominated most subalpine landscapes in eastern B.C. for at least the last several decades. If management of ESSFwc forests is to emulate historical disturbance regimes, greater protection of old-growth ESSFwc forests than at present will be necessary.

The perennial grass Phalaris arundinacea (reed canarygrass) is a widespread invader of North American wetlands. We asked if the use of herbicide with burning, clipping, or seeding could reduce the cover of P. arundinacea and increase cover of native species in a wet prairie that receives stormwater runoff. Two glyphosate treatments decreased P. arundinacea cover, and spring seeding of 33 native species doubled species richness and floristic quality compared to no seeding. Despite an initial decrease in abundance, P. arundinacea cover was no different than control plots two years after seeding and overall native species richness and cover decreased from the previous year. Application of the grass-specific herbicide sethoxydim in the third year of the study reduced P. arundinacea cover and height while allowing native forbs and graminoids to persist. Most of the 75 species identified during the study were perennial, native forbs, and most of the sown species that commonly established in the field had high germination rates in the laboratory. Continued management of P. arundinacea is needed to maintain desirable native wetland flora.

To identify and prioritize natural areas for consideration as research natural areas (RNAs) on the Superior National Forest (National Forest hereafter NF) in northeastern Minnesota, we developed a six-step approach that could be applied to natural area selection anywhere. Our goal was to select a limited number of sites to represent the NF's landscape and local ecosystem diversity. We also sought to minimize management conflicts. The steps are: (1) review existing information to create a list of potential sites; (2) develop a scoring system and score sites based on attributes that assess “natural area” quality of sites from maps, data, and other office information; (3) list ecosystems represented by each site and review sites based on their quality scores and the ecosystems they represent; (4) conduct field evaluations to confirm ecosystem quality for as many high scoring sites as feasible; (5) integrate the information from steps 1–4 with regional and local guidelines for conservation goals, ecosystem representation, and standards for site quality; and (6) review sites to minimize management conflicts and to maximize ecosystem representation and document the findings. Even with limited funds, using the six steps, we pared down the list from 114 to 41 sites for consideration in forest planning. Managers later used results to select 11 of the 41 sites for RNA establishment.

The Darlingtonia fens, found on serpentine soils in southern Oregon, are distinct communities that demonstrate a balance among high water tables, shallow soils, the presence of heavy metals, and limited nutrients. While conservative efforts have been made to preserve them, soil patterning or the specific processes that appear to underlie soil variability are poorly understood. I investigated the relationship of fire-free period on soil gradients and succession in three Darlingtonia fens, each with a different amount of time since last fire. Changes in vegetation and belowground variables were distinguished across fen, savanna, and shrub communities and across the three sites. I suggest that belowground gradients may be more important than time since fire at preserving these uniquely adapted systems, since vegetation undergoes post fire succession from fen to shrub to savanna after approximately 100 years since fire. Although restoration of ecosystem structure and processes was not the primary focus of this study, my data suggest that time since fire may drive ecosystem processes in a trajectory away from the normal succession cycle and that soil moisture decreases and nitrogen levels increase with an increase in fire-free period.

Humans are increasingly altering forested habitats, in part by forestry practices, but also in more subtle ways such as creation of recreational trailways through forested areas, especially in nature preserves and parks. The effect of these trails on wildlife has only recently begun to be addressed. In this paper, I hypothesize that trails alter the distribution of terrestrial salamanders in forested habitats. The routine clearing of fallen trees from trailways may act to increase microhabitat availability for salamanders around trails, and thus move salamanders closer to the trails, or salamanders may simply stay away from trails in general to avoid disturbance associated with humans. To address this issue, I conducted a 3-month study in a nature preserve in northeast Georgia, using standardized, paired surveys of cover objects (logs and stones) in 78 m² plots along trails and in similar plots 25 m away from trails. I found significantly more logs in along-trail plots versus off, but there was no difference in stone abundance. There were more salamanders under logs in along-trail plots versus off-trail, but on a per-log basis, there was no significant difference between along-versus off-trails, suggesting that trails result in more microhabitats for salamanders around them, not that salamanders move closer to trails per se.