Conservation of biological diversity has become a priority in the natural resource management planning process, yet information on the effects of management practices beyond the scale of single stands is lacking. The objective of this study was to determine whether timber management was compatible with the goal of maintaining forest bird species diversity in the northern hardwoods forest type. Twelve 64-ha study areas, 6 managed for quality sawlogs and 6 not managed for timber, were selected in the White Mountain National Forest, New Hampshire. Managed areas included approximately 60% sawtimber, 20% pole, and 20% sapling even-aged northern hardword stands while unmanaged areas were 100% sawtimber stands $>$80 years old. Young stands were $<$20 ha in size; older stands ranged from about 10 to 140 ha. Bird species abundances were determined in each area from 8-minute counts in June 1989-90 at 16 points spaced 200 m apart. Bird species diversity, as measured by Shannon-Weaver, Simpson, log series alpha, and species richness indices, was greater (P $<$ 0.005) on managed areas than on unmanaged areas. Managed areas included a total of 53 species compared to 33 on unmanaged areas. All 33 species present on unmanaged areas were also present on managed areas. Of 19 species with mean abundances $>$1 individual/area and coefficients of variation $<$50%, the ovenbird was the only species more abundant (P $<$ 0.10) on unmanaged areas, yet ovenbirds were the third most abundant species on managed areas. American redstart, rose-breasted grosbeak, and veery were all more abundant (P $<$ 0.05) on managed areas. The above results suggest small scale timber harvesting can increase forest bird species diversity without loss of any species. These results apply to extensively forested areas of northern hardwoods in the northeastern United States and may not apply in other areas or in areas where harvesting practices are more intensive. Analysis of minute-by-minute bird counts and evaluation of bird surveys for unmanaged areas suggested point counts should be 8-minutes in duration at points $\ge$400 m apart. Travel time between points had a greater effect on efficiency (individuals detected/hour) than count duration. Each point should be surveyed $\ge$3 times, and for a uniform habitat such as mature northern hardwood forest, a minimum of 24 points should be established.

In the dry subtropical forest of Tucuman Province, Argentina, I assessed effects of habitat fragmentation on (1) the insect pollinator fauna by monitoring visits to flowers of two dominant leguminous trees and by sampling bees with yellow pan traps, and on (2) pollination and seed production levels by determining stigmatic pollen load sizes, number of pollen tubes at the base of the style, and number of fruits per flower (fruit set) and seeds per fruit (seed set) in 16 plant species representing a variety of growth forms, and pollination and breeding systems. The experimental design involved five study sites each with continuous forest, a large ($>$2 ha) forest fragment, and a small ($<$1 ha) fragment. Comparisons in pollinator abundance and diversity between fragments and continuous forest were replicated across 4 sites. Comparisons in pollination levels and fruit and seed set were replicated across 1-4 sites depending on the plant species. The exotic honeybee (Apis mellifera) dominated the flower-visiting fauna associated with the two leguminous species. Frequency and taxon richness of native flower-visitors to both plant species declined with decreasing fragment size, but frequency of honeybee visits increased, so neither species experienced significant effects of fragmentation on total visitor frequency. Similarly, in pan trap samples the relative frequency of honeybees increased with decreasing fragment size, whereas native bees increased in number and species diversity with increasing size of forest patch and as spring progressed. Fragmentation-related declines in pollination and seed output (fruit set x seed set) prevailed in the subset of the Chaco flora I studied. Among species, fragmented plant populations recorded an $\sim$20% median decrease in pollination and seed output levels. However, mechanisms behind these declines differed across species. In three self-incompatible species, decreases in the number of pollen tubes with fragmentation could be related to declines in pollination quality rather than in the number of pollen grains deposited on the stigmas. Correlative evidence showed that fragmentation-related declines in pollination could limit seed output in some plant species, but an increase in seed abortion due to inbreeding depression could also affect fragmented plant populations. Comparisons among self-compatible and self-incompatible species did not show any differential effect of fragmentation with breeding system. Overall, the results of this study show that fragmentation may exert chronic, negative effects on pollinators, pollination, and seed output with potential effects on plant regeneration, forest dynamics, and total species diversity. These results also indicate that fragmentation may alter community and population processes through subtle and complex mechanisms.

One hundred and two virulent (V) strains of Cryphonectria parasitica were isolated from chestnut blight cankers on American chestnut trees (Castanea dentata) in western Massachusetts, and vegetative compatibility was investigated. The 102 strains fell into 54 vegetative compatibility groups (VCGs): 10 groups contained more than 2 strains in each, 6 groups had 2 strains per VCG, and 38 VCGs contained only one strain each. A wide diversity of VCGs may be due to the presence of C. parasitica in Massachusetts for at least 80 years. A representative strain from each of the 10 VCGs with more than two strains, was paired with 22 hypovirulent (H) strains not native to Massachusetts (4 strains with French double-stranded RNA (dsRNA), 17 strains with Italian dsRNA, and 1 strain with American dsRNA). Some of the V strain could be converted to hypovirulent by the French and/or Italian H strains, but none were converted by the American H strain. Ten V strains from different VCGs were compared with 25 H strains, and with 29 strains converted to hypovirulence in cultural characteristics, pathogenicity, and phenol oxidase activity. Fungal growth, colony color, and pycnidial development were compared on various media. Fungal growth could be accentuated by the addition of ground bark or wood tissue to the media. Cultural characteristics were consistent among the V strains, but much variability was noted with the H strains. A rapid method (bark/wood test) for testing the virulence was developed, and compared with living tree inoculations and excised stem sections. The bark/wood test gave the fastest and most consistent results. Phenol oxidase activity of V strains was much stronger than those of H strains on the modified Bavendamm's medium. Several H strains were cured by the treatments of glucose and cyclic nucleotides (cAMP, cGMP). When the H strains were cured, they took on the cultural characteristics of normal V strains and virulence was restored. DsRNA bands were not detectable by gel electrophoresis from the cured strains, while uncured strains retained dsRNAs. It is assumed that RNA-dependant RNA polymerase gene, which regulates RNA replications, was depressed by cyclic nucleotides. Subsequently, RNA replication was inhibited and virulence returns as all genes can be expressed.

Thesis (M.S.)--West Virginia University, 2010. / Title from document title page. Document formatted into pages; contains vi, 26 p. : ill. (some col.), col. maps. Includes abstract. Includes bibliographical references (p. 25-26).

Thesis (M.S.)--Oregon State University, 1986. / Typescript (photocopy). Includes bibliographical references (leaves 77-81). Also available on the World Wide Web.

No description available.

Thesis (Ph. D.)--Oregon State University, 2010. / Printout. Includes bibliographical references (leaves 125-134). Also available on the World Wide Web.

Thesis (doctoral)--Radboud Universiteit, Nijmegen, 2008. / Description based on print version record. Includes bibliographical references (p. [183]-188).

Thesis (M.S.)--Virginia Polytechnic Institute and State University, 2003. / Vita. Includes bibliographical references (leaves 69-71).

Thesis (doctoral)--Radboud Universiteit, Nijmegen, 2008. / Includes bibliographical references (p. [183]-188).