Journal of the Southeastern Association of Fish and Wildlife Agencies

Conservation rangers are increasingly coming into contact with violations involving alcohol and drugs. Arrests for hunting or operating a vessel while under the influence of alcohol and/or drugs have become priorities for enforcement efforts. Arrests are also being made for other alcohol and drug violations on Department-controlled properties such as parks, public fishing areas and wildlife management areas as well as other properties. Increased training to detect these activities has produced very good results.

The Tennessee Wildlife Resources Agency (TWRA) is the state agency in Tennessee charged with the enforcement of all wildlife, fishing, and boating laws of the state. These duties include, but are not limited to, the protection of all fish and wildlife (including mussels), and the investigation of boating accidents. As a matter of necessity, the TWRA has for the past several years used the services of Agency employees trained as scuba divers to aid in investigations that involve searching the lakes and rivers of the state for a multitude of reasons related to both law enforcement, and biological investigations. In the past, Agency divers have dove on boating accident sites. They have searched rivers at the request of other law enforcement Agencies looking for stolen property or evidence related to a criminal investigation. They have searched for drowning victims (both accidental and homicide related). And they have searched for lost equipment from both Agency boats and others.

Wild turkey (Meleagris gallopavo) hunting has gained popularity as turkey populations have risen. Consequently, there is a greater chance that illegal turkey hunting will impact turkey populations. Our objectives were to determine the extent of illegal turkey hunting in Mississippi and identify potential violators. We conducted a mail survey of turkey hunters (N = 4,026) who hunted in Mississippi during the 1994 spring gobbler-only season. A response rate of 55% was obtained. Logistic regression analysis was used to predict the probability of a hunter violating a regulation. We entered 20 independent variables into the model and 6 significantly predicted turkey hunting violations.

Wildlife enforcement trainers must be constantly on the lookout for better equipment for enforcement personnel and more effective methods of training individuals in the use of that equipment. For the past 15 years, the Law Enforcement Division of the Texas Parks and Wildlife Department has issued to its officers the patrol rifle, a piece of equipment which has only recently received serious national consideration as a viable alternative to more traditional weapons in the law enforcement defense arsenal. The relative virtues of the compact semi-automatic rifle versus the handgun and/or the police shotgun are discussed here, along with an innovative method of training field personnel of agencies that have to deal with logistical and budgetary considerations. Training issues and methods are explained, and a video used for this block of instruction is reviewed, followed by a discussion of the pro's and con's of this method of administering firearms instruction.

The goal of this study was to assess the law enforcement needs of the Texas Parks and Wildlife Department. Data were obtained from several state agencies and a mail-out survey to Texas game wardens. Game wardens (N = 397) and their captains (N = 33) returned 685 of 690 mailed questionnaires (99% response rate). All 254 counties were represented. A regression model explained 78% of the total variation in the distribution of game wardens and estimated a statewide excess of 16 game wardens. In contrast, the game wardens indicated that 160 more game wardens were needed. Differences between the model estimates and wardens' perceptions were attributed to a lack of data (e.g., number of anglers, boaters, or campers), incomplete data (e.g., number of hunters for dove, pheasant, and waterfowl), and unreliable data (e.g., number of deer, quail and turkey hunters, and game harvests) for each county.

I examined spring and fall electrofishing catch rates of largemouth bass (Micropterus salmoides) in 12 Texas reservoirs from 1986 to 1992 to assess the relative importance of spatial and temporal variation and to evaluate the relation between electrofishing catch rates and future angler catch. East Texas reservoirs were characterized by relatively greater temporal (i.e., annual) variation in electrofishing catch rates, whereas west Texas reservoirs tended to exhibit greater spatial variation. Reservoirs in east Texas shared a common temporal pattern in electrofishing catch rates. Conversely, there were marked differences in catch rates among west Texas reservoirs that remained relatively stable across years. The relation between electrofishing catch rate and angling success was poorer than expected. Spring electrofishing yielded most of the significant correlations with future angler catch, particularly in east Texas reservoirs.

Habitat use and movements associated with changes in dissolved oxygen (DO) levels and hydrology for 20 radio-tagged largemouth bass {Micropterus salmoides) were studied in the Kissimmee River, Florida, in 1992 and 1993. River channel littoral vegetation, smartweed (Polygonum hydropiperoides) and spatterdock {Nuphar luteum), were habitat types bass most commonly used. Bass maintained home areas in remnant channels and the C-38 canal that had DO >2 ppm throughout the year. During summer when DO dropped slowly over several weeks, bass were frequently located in waters with DO between 1 and 2 ppm. Daily movements, which averaged <100 m in more than half of the location samples, were greatest when previous location DOs were <1 ppm during summer and <2 ppm during fall. Daily movements >1 km from June through October were coincidental to acute decreases in DO <2 ppm at previous location sites.

We developed a modified index of biotic integrity (IBI) for the Tallapoosa River system based on small-bodied fishes. The modified IBI comprised 9 metrics in 4 categories: (1) species richness and composition, (2) indicator species, (3) trophic function, and (4) abundance. We used distribution records and collection data from 1990-1995 to derive expected values for metrics. The IBI was most sensitive to changes in percentage of insectivorous cyprinids, percentage of intolerant species, fish abundance, and number of darter species, and least sensitive to total species richness. IBI scores generally were lower at sites experiencing more severe flow fluctuations as a result of hydropeaking dam operation. We recommend that the IBI be further tested, refined, and used as part of long-term monitoring programs in regulated southeastern river systems.

A fish community index of biotic integrity (IBI) was used to compare 6 streams with minor anthropogenic impacts to 3 streams receiving source and non-point source effluents from gold mine operations in the Piedmont of South Carolina. IBI ratings for streams receiving gold mine effluent ranged from very poor to marginally fair, whereas those for the remaining 6 streams ranged from fair to good. The stream selected as a “candidate” reference stream rated consistently good, establishing a potential baseline for attainable conditions for the ecoregion. The IBI supplied biological data that were useful in identifying perturbations based on alterations occurring in fish communities. The results of this study suggest the IBI could be an excellent tool for monitoring the biological components of streams in the Piedmont ecosystem.

We located 22 potential striped bass (Morone saxatilis) thermal refuges created by groundwater inflows along the Flint River in southwestern Georgia. Line transect counts made by SCUBA divers June-October 1990 were used to develop striped bass abundance indices for 8 of the springs. Abundance differed among springs and was positively correlated with potassium concentration and distance upriver and negatively correlated with pH and dissolved oxygen concentration. Groundwater feeding the springs flows through conduits created by the dissolution of underlying limestone. Locations of these conduits, revealed by linear arrangements of surface depressions, can provide information for managers making decisions concerning conservation of habitats that serve as striped bass thermal refuges.

Fifteen radio-tagged striped bass (Morone saxatilis) were tracked to determine seasonal, inter-river distribution patterns, and habitat use in the vicinity of the Ohio, Tennessee, Cumberland, and Mississippi rivers from November 1992 to November 1993. All 4 river systems were used by study fish during 1 or more seasons. Two fish inhabited only 1 river during the entire survey, while 13 fish used up to 3 river systems. Inter-river movement occurred primarily during November-December (36%) and April-May (58%), which coincided with highest movement periods. Striped bass distribution was skewed toward tailwaters and study segments immediately downstream. General habitats annually included channel borders (47%), tailwaters (37%), and heated discharges (7%); however, specific habitats included open bottom substrates (29%), tailwater boils (28%), and woody debris and trees (24%).

Seven reaches of the lower Pearl River system were sampled with bottom set gill nets from 1992 through 1995. One hundred fifty-eight Gulf sturgeon (Acipenser oxyrinchus desotoi) were captured in 3 reaches. Mean fork length (FL) was 734 mm, and weight (W, g) to length (FL, mm) relation was: W = 1.786 x lO^CFL)3204. Mean weight loss during summer was 1.9 g/day. Ninety-two percent of Gulf sturgeon were captured in the West Middle River reach which is comprised of deep holes and little current. Radio telemetry indicated that the West Middle River was an important summer habitat for juveniles and subadults. Estimated mean summer population size in the West Middle River, determined from mark and recapture, ranged from 67 to 124 individuals, but low intra-year tag return indicated this may be part of a larger population. Annuli on pectoral fin rays were formed during May-July and 94% of fish aged were <6 years old.

Lake Norman is a 13,516-ha reservoir frequently devoid in late summer of the cool, well-oxygenated water preferred by mature striped bass {Morone saxatilis). Large summer striped bass kills, however, are rare. The objective of this study was to determine the summer temperature and dissolved oxygen (DO) levels in habitat used by striped bass in Lake Norman. We tagged 48 striped bass (2.2-7.8 kg) with temperaturesensing radio tags (40-50 MHz) during spring 1992 and 1993. Tag frequency, tag temperature, location, and a water column temperature/DO profile were recorded for each tag encounter. Mean tag temperatures (25.6 and 26.1 C) and associated mean DO concentrations (4.0 and 3.1 mg/liter) were similar for large (>4.5 kg) and small (<4.5 kg) striped bass in the warmest weeks of summer 1992. Habitat <28.0 C and containing >2.0 mg/liter DO was always present in most of Lake Norman both years.

A 3-year study was initiated in 1992 to characterize the St. Johns River freshwater commercial striped mullet (Mugil cephalus) gill net fishery and estimate bycatch. Each year in fresh water, an estimated 137 fishers using 84 boats fished 1,554 days and harvested 233,109 kg of striped mullet. Approximately 86% of the fishers used the strike technique exclusively, while 14% used both strike and set-net techniques. Striped mullet comprised 99% of the total catch by number. The average annual freshwater game fish bycatch ranged from 86 to 274 fish (< 1 % by number). The extremely low bycatch would obviously have no negative impact on game fish populations in a system as large as the St. Johns River.

Human chorionic gonadotropin (HCG) is often used to induce spawning of fish. However, the U.S. Food and Drug Administration (FDA) has not approved HCG for use in fish because they claim additional efficacy data are required. Efficacy of HCG for spawning walleye (Stizostedion vitreum), was evaluated at 3 hatcheries for 3 years. During the 1993-1995 spawning seasons, 17.5% (54 of 308) of female walleyes spawned after a single injection of HCG. Of 253 females receiving a second injection, 92.1 % spawned for an overall spawning success of 93.2%. For females, a HCG dose as low as 330 IU/kg in each injection was as satisfactory as higher doses. Fish from ponds had a significantly lower spawning rate (8.7%) than wild fish (30.6%) after a single injection of HCG, but there was no significant difference in spawning success for these groups after 2 injections.

During 1984-1992, adult shortnose sturgeon (Acipenser brevirostrum) from the Savannah River were spawned and their offspring reared to various sizes prior to release in this river as part of a stock enhancement program. A total of 79,270 untagged and 18,213 tagged fish were released at various sites from river kilometer (rkm) 54 to 273. Mid and upriver stocking areas appeared to provide higher survival than downriver areas and stocking during November and December appeared more beneficial than during other months. However, regardless of stocking time or site, all stocked juveniles that were captured came from the same downriver nursery area (at the fresh-brackish water interface) used by wild fish. Stocked fish comprised at least 35.4% of juveniles captured and the low total number of juveniles captured suggests that the Savannah River population of shortnose sturgeon may be experiencing poor recruitment.

Production of smallmouth bass (Micropterus dolomieu) in 11 -ha Beaver Nursery Pond was estimated from 1990 to 1994 to determine numbers of fish stocked in Beaver Reservoir. In 1990, 2 rapid population sampling methods, seining a known area and SCUBA transects, were tested against a Petersen mark-and-recapture estimate. Both sampling methods provided population estimates that did not differ significantly from mark-and-recapture values, but both generated greater confidence intervals. Annual production in the nursery pond by late June varied between 57,000 and 164,733 fish/year, with a mean of about 97,500 ± 59,750 (confidence interval, CI); mean total length of the smallmouth bass produced was 50 mm. Smallmouth bass were usually released into the reservoir in late June to reduce cannibalism, but during 1993 and 1994 fish were retained in order to assess growth and population changes.

Genetically-marked Florida largemouth bass (Micropterus salmoides floridanus) fingerlings with rare or unique genotypes were stocked into nursery ponds at Lake Conroe (8,484 ha) and Lake Gladewater (323 ha). Following grow-out, advancedsize individuals were stocked into respective reservoirs. We measured the contribution of stocked individuals to corresponding year classes in receiving reservoirs. Largemouth bass survival in nursery ponds at the end of the grow-out period was 15.6% at Lake Conroe and 20.9% at Lake Gladewater. A total of 2.8 fish/ha (mean total length = 138 mm) were stocked in Lake Conroe and 8.8 fish/ha (mean total length = 121 mm) were stocked in Lake Gladewater. Following reservoir stockings, largemouth bass populations were sampled with electrofishing gear in fall and spring and electrophoretic analysis was conducted to evaluate the contribution of stocked fish to corresponding year classes.

Electrical barriers were successful in blocking movements of gizzard shad (Dorosoma cepedianum), golden shiners (Notemigonus crysoleucas), rainbow trout (Oncorhynchus mykiss), brown trout (Salmo trutta), and largemouth bass (Micropertus salmoides) stocked in a 24-m long canal during simulated modes of pumped hydropower operation. Blockage rates were highest during nongeneration (95%-97%) and generation (94%-97%), and lowest during pumping (83%-84%). These results indicate an electrical barrier may be useful in blocking fish from migrating into areas around hydropower projects where they are vulnerable to being entrained.

Poor water quality upstream of the Carolina Power and Light Company (CP&L) Walters Hydroelectric Project (WHP) resulted in establishment of a mitigation fund in lieu of immediate water releases as a condition of a new Federal Energy Regulatory Commission (FERC) license. The negotiated agreement between state management agencies and CP&L includes provisions for future instream flow releases, but until that time, CP&L will make annual payments to the Pigeon River Fund (PRF). The PRF will grant money to nonprofit groups and government agencies for projects related to water quality, fish and wildlife habitat improvement, fishery management, and angler access outside the WHP boundaries. We suggest that agencies avoid, minimize, rectify, reduce, or eliminate projects over time before considering monetary compensation as mitigation.

We conducted a screener telephone survey of 5,209 anglers in Mississippi to recruit participants for a statewide, 5-year, longitudinal angler survey that will administer a total of 10 surveys (waves), 2 per year. Of these 5,209 anglers, 4,393 agreed to participate in future surveys. Of these 4,393 anglers, 52% responded to a mail survey and 76% participated in a telephone survey. Twenty-four demographic and fishing preference variables were selected from the initial screener survey to compare data from respondents and nonrespondents. A greater percentage of respondents to both surveys were generally Caucasian, married, in higher income brackets, and tended to be employed full-time. A greater percentage of respondents to both surveys fished more days and preferred to fish for black bass (Micropterus spp.) than did the original screener population.

Anglers who fished Lake Fork Reservoir June 1994-May 1995 were intercepted during creel surveys conducted by the Texas Parks and Wildlife Department (TPWD) and asked to participate in a follow-up, self-administered mail survey. Anglers were asked questions about their fishing activity at Lake Fork, and their motivations, satisfaction, and expenditures on the trip they were intercepted. Lake Fork anglers could be characterized as high specialization anglers, indicating they had been fishing in fresh water for an average of 28 years, and fishing an average of 62 days in the previous year. Additionally, 34% of anglers belonged to a fishing club, and 57% indicated they participated in fishing tournaments. Lake Fork anglers had an average of $13,174 invested in fishing equipment, twice as much as the average black bass (Micropterus spp.) angler in Texas. About 89% of anglers targeted largemouth bass (M. salmoides) on their trip which lasted an average of 1.74 days.

This study examined factors which can affect the quality of fluorescent oxytetracycline (OTC) marks on walleye (Stizostedion vitreum) otoliths. A 1 -time exposure to a strong ultraviolet (UV) light source, such as is done when viewing under a UV microscope, significantly decreased mean OTC mark qualities from an initial value of 2.8 (3.0 is maximum and 0.0 is no mark) to <1.0 3 months following the initial examination. Mark intensity continued to decrease over time (<0.5 after 6 months, 0.3 after 12 months). Otoliths that were stored in an unlit environment consistently had higher intensity marks over time than those stored in a lighted environment. Dimethyl sulfoxide (DMSO) was examined as a possible potentiator to improve OTC mark intensity. Otoliths that were marked by immersion in 200 ppm OTC and 0.81 % DMSO did not exhibit higher quality marks than those which were immersed only in 200 ppm OTC.

Diets, food selection, and factors affecting feeding success of small (5-9 mm total length, TL), medium (10-19 mm TL), and large (20-29 mm TL) larval and early juvenile gizzard shad (Dorosoma cepedianum) were examined in 2 large Missouri reservoirs. Diets and food selection shifted from mainly copepod nauplii and cyclopoids to rotifers and phytoplankton as fish grew. Mean length of ingested zooplankton rose slightly as gizzard shad grew from the small to medium length groups and then decreased for the large length group. Feeding success (feeding incidence and mean gut fullness) increased with fish size. Feeding incidences of small and medium gizzard shad were significantly higher at an upstream site than at a downstream site in 1 reservoir, but not in the other. Feeding incidence and, to a lesser degree, mean gut fullness of small and medium gizzard shad usually increased with food abundance, although correlations were weak.

Diet and telemetry studies were conducted in La Plata Reservoir, Puerto Rico, to investigate potential interactions between largemouth bass (Micropterus salmoides) and butterfly peacocks {Cichla ocellaris). Partitioning of food resources between the 2 species was evident and consistent over several size classes. Interspecific differences were observed in home range size, diel movement, and habitat preference during telemetry studies conducted on adults for up to 250 days. Maximum activity for both species corresponded with periods of elevated food consumption. Although the current studies were restricted to 1 reservoir, low overlap in diet and habitat preference indicates ecological compatibility in tropical reservoir systems.

Stomachs of 5,818 largemouth bass (Micropterus salmoides) were examined from 10 north-central Florida lakes to determine differences in food habits of largemouth bass in lakes with abundant aquatic macrophytes (vegetated) and lakes nearly devoid of aquatic macrophytes (unvegetated). We found significant differences (P < 0.05) between stomach contents of 6 length groups of largemouth bass (range: 60-640 mm TL) from vegetated and unvegetated lakes. The 152- to 254-mm length group exhibited the greatest number of diet differences. Atherinids, decapods, and odonates were consumed more frequently (P < 0.05) by largemouth bass in vegetated lakes, while cichlids, clupeids, and dipterans were more common in the diet of largemouth bass from unvegetated lakes. Largemouth bass from unvegetated lakes were piscivorous by 60 mm TL; those from vegetated lakes did not become piscivorous until they were >120 mm TL.

Otoliths from 822 trophy (>4.5 kg) largemouth bass (Micropterus salmoides) caught by anglers in Florida were obtained from taxidermists during 1987-1993. The fish were harvested from 211 different water bodies. Longevity and mean growth exceeded those reported previously in Florida. Mean age of trophy largemouth bass statewide was 9.7 years and ranged from 4.0 to 16.5 years. Mean growth averaged 0.54 kg/year with a range of 0.28-1.27 kg/year. Growth rate of trophy largemouth bass increased from north to south Florida. Accelerated growth, longevity and strong year classes were important for trophy largemouth bass production in Florida. Fifty-six percent of the trophy fish were harvested by anglers during January-March. Peak months for trophy harvest were progressively later in the year from south to north Florida and coincided with the largemouth bass spawning season.

Surveys of shorebirds from a 13-year period were compiled from the T. E. Maxson wastewater treatment plant in Memphis, Tennessee, a site bordering the Mississippi River. The data represent one of the few long-term shorebird surveys from the Mississippi River valley. Data were used to produce a migrational chronology for species both in spring and fall based on mean abundance. Least sandpipers (Calidris minutilla), pectoral sandpipers (C. melanotos), killdeer (Charadrius vociferus), and lesser yellowlegs (Tringaflavipes) were the most abundant migratory species recorded. Fall migration occurred over a much longer period than did spring migration, and most species were more abundant during this time; however, shorebirds used the facility throughout the survey period of March to November. This information is useful in providing wetland management guidelines on timing and duration of flooding for migratory shorebird conservation.

Much of the original hardwood bottomland in the Mississippi Alluvial Valley is converted to crop lands. Land management agencies began restoring hardwood bottomland because of its importance to wildlife. To provide an initial evaluation of progress toward restoration, we used point counts to compare bird communities among plantations of 0 to 4, 7 to 15, and 21 to 27 years in age with natural sawtimber stands (>50 years in age) in the southern Delta region of Mississippi in 1994 and 1995. Mean number of species per point (species richness) increased with stand age (P < 0.05). Mean total abundance did not differ among age classes. Relative to bird communities of natural sawtimber forest at Yazoo National Wildlife Refuge and Delta National Forest, respectively, Morisita's index of similarity was 85.4% and 74.3% for 21- to 27-year-old plantations, 41.9% and 35.0% for 7- to 15-year-old plantations and 4.6%. and 2.6% for 0- to 4-year-old plantations.

Clearcutting has been reduced in frequency in national forests of Arkansas. This will affect populations of birds that require early successional forest habitat. I studied avian distribution, abundance, and reproductive success in young (i.e., 7 to 10 years post-harvest) and old (i.e., 17 to 19 years post-harvest) forest clearcuts from 1993 to 1995. The number of species was greater in young than old clearcuts, with abundance of 5 species, white-eyed vireo (Vireo griseus), common yellowthroat (Geothlypis trichas), prairie warbler (Dendroica discolor), yellow-breasted chat (Icteria virens), and indigo bunting (Passerina cyanea) higher in young clearcuts. No difference in fledging success was detected between young and old clearcuts for any species; however, the ratio of hatching-year to after hatching-year birds captured in mist-nets was lower in older clearcuts for every species studied.

Three taxa of cave bats in Arkansas are listed as endangered: the gray bat {Myotis grisescens), the Indiana bat (M. sodalis) and the Ozark big-eared bat (Corynorhinus townsendii ingens). All occur primarily in the Ozark Plateau region of the northwestern and northcentral portion of the state. Population monitoring and ecological studies of endangered bats inhabiting important hibernation, summer, and transient caves were conducted annually since 1978. Through the efforts of several federal, state, and private agencies and organizations, as well as numerous private landowners, 14 caves important to bats were gated or fenced to protect colonies of bats from disturbance by humans. Several additional caves were afforded protection by intrusion alarm systems, control of access roads and cooperative management agreements.

A survey of red-cockaded woodpeckers (Picoides borealis) was conducted on Goethe State Forest, Levy County, Florida, in 1994 and 1995. Four hundred fifty-one living cavity trees, active and inactive, were located. The majority of living cavity trees (97%) were longleaf pines (Pinus palustris). Mean age of cavity trees was 123.6 years (N = 108). One hundred ninety-seven living cavity trees occurred in 26 clusters, and 25 of these clusters were active in 1995. Nestling production was confirmed in 73% and 65% of the active clusters monitored in 1994 and 1995, respectively. To maintain the red-cockaded woodpecker population at its current status, management activities should focus on improving habitat quality in active clusters, establishing replacement stands to provide future nesting and roosting habitat, and providing adequate foraging habitat.

The alligator snapping turtle (Macroclemys temminckii) occurs throughout most of the southeastern United States, but concern exists regarding possible decline in populations throughout the range of the species. Little is known about the alligator snapping turtle in Arkansas. The species was previously documented from only 20 counties in the state. We sampled 1,905 net nights with baited hoop nets and captured 445 alligator snapping turtles in 56 counties. Captures in 41 counties were new records. We failed to capture an alligator snapping turtle in 5 counties where the species was previously reported. The alligator snapping turtles that we captured averaged 338 mm in carapace length. Catch rates were higher in counties that were previously open to commercial turtling than in counties not previously open to commercial harvest. There was no difference in average size of turtles between counties with and without a history of commercial harvest.

A survey of northern pine snake occurrences (Pituophis m. melanoleucus) was completed in North Carolina in 1990 to determine the distribution and habitat of this state-listed species of special concern. A total of 196 snakes at 163 locations was mapped and visited. Most of the snakes were killed by vehicular traffic. Sixteen pine snakes were recorded at 12 sites in 3 mountain counties, a single snake from each of 2 sites in 1 foothill county, and 26 snakes at 24 sites in 2 coastal counties. However, the majority of the snakes (78% of total) were recorded at 119 sites in 7 contiguous counties in the sandhills of central North Carolina. Most snakes were recorded in the vicinity of sandy soils, but several (N = 12) were observed in bottomlands. Data were recorded on several habitat variables at snake locations. The present study confirmed that northern pine snakes still have 3 population centers in North Carolina.

We compared rat, mouse, and museum special snap traps to determine if differences existed in capture efficiency of small mammals and whether type of trap affected indices of richness, evenness, and diversity. Small mammals were trapped in 57 streamside study areas in 1990 to 1995 in the Ouachita Mountains, Arkansas. Efficiency of mouse traps was equal to or greater than that of museum special traps in capturing all small mammal species. Rat traps were most efficient for capturing the 2 largest small mammal species recorded, the eastern wood rat (Neotoma floridana) and the cotton rat (Sigmodon hispidus). We found no difference among the types of traps in their susceptibility to being inadvertently sprung by extrinsic factors such as rain. Mean species richness, evenness, and diversity did not differ among trap types. A combination of mouse and rat traps increased species richness and diversity over mouse traps alone.

This paper summarizes the issuance of federal depredation permits to commercial aquaculture facilities in the Southeast Region by the U.S. Fish and Wildlife Service from 1989 to 1996. Data were evaluated from 1,103 depredation permits issued to individual aquaculture facilities allowing lethal take of fish-eating birds. The take of 21 species of birds comprised of waders, fliers, and swimmers was permitted. In the 9 states where permits were authorized, the number of facilities receiving permits increased from 54 in 1989 to 228 in 1996. The number of birds allowed in the take increased from 7,401 in 1989 to 28,991 in 1996. Sixty percent of the birds permitted for take at aquaculture facilities were reported taken. The double-crested cormorant (Phalacrocorax auritus) was the species most often requested for taking (N = 70,888). Most depredation permits were issued to aquaculture sites in Mississippi (N=459) and Arkansas (TV = 448).

We developed a Monte Carlo simulation approach to examine statistical power in analysis of population trend data. Our stepwise approach was to perform a regression analysis to test the null hypothesis that the slope of the time series regression line was equal to 0 (i.e., Ho:b = 0 for population count data collected over i years), to use Monte Carlo simulations to calculate the statistical power of the test of H0:b = 0 when Ho was not rejected, and to estimate sample size requirements within and across years to detect a population trend at a specified power, Type I error, and coefficient of variation. To demonstrate this approach and illustrate important considerations when conducting power analysis, we analyzed 5 sets of shorebird count data collected by a single observer in the International Shorebird Survey, Marco River, Florida, in 1975 and 1980 to 1987.

I tested the effectiveness of squirrel excluder devices (SQEDs) in deterring southern flying squirrels (Glaucomys volans) from using artificial red-cockaded woodpecker {Picoides borealis) cavities by placing them on approximately one-half of the cavities in 14 inactive recruitment clusters on the Savannah River Site, South Carolina. SQEDs consisted of 2 pieces of 35.5-cm wide aluminum flashing placed 7.6 cm above and below the cavity entrance. Cavities with (/V = 37) and without (7V = 35) SQEDs were checked once per month from February 1995 to January 1996; all flying squirrels found in cavities were removed and destroyed. Cavities with and without SQEDs did not differ in cavity height (P = 0.70), distance to first branch >1 m in length (P = 0.09), distance to the nearest tree (P = 0.29), number of trees within 8 m (P = 0.82), or previous use by flying squirrels (P = 0.67).

We compared the response of herpetofaunal communities in 16 hardwood stands treated with a high-leave harvest (7 m2/ha residual basal area), low-leave harvest (3.5 m2/ha residual basal area), clearcut harvest or no-harvest prescription in the Daniel Boone National Forest, Kentucky, from 1992 to 1996. Animals were captured with straight-line drift fences and pitfall traps. We sampled 800 trap nights and captured 24 species of amphibians (N = 1,363) and 12 species of reptiles (N = 163). Diversity of amphibians was lower in low-leave harvest stands after removal of timber than in noharvest stands (P < 0.05). Numerical abundance and species richness of reptiles were higher after timber removal in high-leave, low-leave, and clearcut harvest stands than in no-harvest stands (P < 0.05), and diversity of reptiles was higher after removal of timber in low-leave harvest stands than in no-harvest stands (P < 0.05).

Management of the flatwoods salamander (Ambystoma cingulatum), a species in decline throughout its range, is hindered by a lack of information on the habitat requirements of the species. Because Ambystomatids are generally philopatric, preservation of quality natal habitat is important for long-term population health and stability. Conservation of breeding sites in managed landscapes is impeded by insufficient knowledge of the habitat components required for successful larval development. We sampled 10 ponds known to be breeding sites of flatwoods salamanders in the winters of 1992 and 1993 to gather baseline information on the habitat and ecology of the larval amphibian assemblages present. Ornate chorus frog (Pseudacris ornata) larvae were the dominant amphibian in both years and exhibited biomass peaks in late winter. Leopard frog (Rana utricularia) larvae gained dominance in mid-spring as the ponds approached dry-down.

A 3-year study was conducted to evaluate the availability and use of cavity trees and snags in hardwood stands regenerated using the group selection method. A survey of cavity trees and snags was completed before and after group selection harvest in 16 2.9-ha plots in the Ozark National Forest, Arkansas. Cavity trees and snags were identified to species, and the height, dbh, state of decay, and number of visible cavities recorded. We marked and measured 66 cavity trees and 126 snags with cavities, and observed the use of these cavities by wildlife. Use of cavity trees was observed seasonally and before and after group selection harvest. Plots averaged 12.2 cavity trees/ha and 25.5 snags/ha before harvest. Densities did not vary between plots with northeast and south aspects. Species composition of cavity trees and snags did not reflect overall stand composition.

Direct-seeding is no longer commonly used for reforesting bottomland hardwood habitat because of low seedling survival due in part to predation of seeds by animals. We investigated the impact of seed predation by rodents on direct seeded willow oak (Quercus phellos). Acorns of willow oak were planted at a rate of 5,982 seeds/ha on low-lying farmland in the Ouachita Wildlife Management Area, Louisiana. A 200 live-trap grid and randomly placed seedling survival plots were used to study the relationship between density and activity of rodents, and predation rate of seeds and seedling survival. Rice (Oryzomys palustris) and cotton (Sigmodon hispidus) rats were the most likely acorn predators. Average survival rate of seedlings was negatively correlated with rodent activity. The phenological maturity of vegetation and the amount of vegetative cover was positively correlated with rodent activity.

We examined plant species richness and species overlap among 3 site preparation treatments (roll-chop and burn, imazapyr and burn, and imazapyr only) and a mature pine-hardwood forest during 2 years following site preparation in east-central Mississippi. Treatments were applied beginning June 1990. Inventory of plant species in 2 to 5 1.6-ha plots/treatment was completed in spring 1991 and 1992. Species richness did not differ among treatments or between treatments and pine-hardwood forest (P = 0.31). Similarity indices showed no apparent grouping among treatments. All values were approximately equal and were similar within years for all treatments and growth forms. All treatments were more similar to other treatments than to pine-hardwood forest. Intercommunity similarity increased from 1991 to 1992 (P = 0.001).

We measured reproductive parameters of 2,833 female white-tailed deer (Odocoileus virginianus) collected from 1978 to 1986 and 1989 to 1993 in 3 geographical regions of Missouri. Fawn pregnancy and fetal rates were lower (P < 0.001) than those of yearling and older does. Ozark region fawns had lower fetal rates than fawns from the Glaciated Plains region; no other regional differences were found. Fetal sex ratios approached 1:1 and were not influenced by year of collection, litter size, region, or age of doe. However, yearling and older does that conceived late in the annual breeding season produced a higher proportion (P = 0.001) of males than those conceiving earlier. Fawn does breed later in the fall than yearling or adult does, but there were no regional or annual effects on mean conception dates. Most (75%) yearling and adult breeding occurred during a 2-week period while 75% of fawn breeding covered 7 weeks.

Our objectives were to determine population and ecological characteristics for an unmanaged white-tailed deer (Odocoileus virginianus) herd on Chickamauga Battlefield National Military Park, Georgia, and to predict the potential for overpopulation in this herd in the future. We radio-collared 67 deer on the park over 4 years (1991-1994). An additional 295 deer were killed by vehicles on the park's roads, and 36 were collected for herd health analysis in August of each year (8-10 deer per year). The range of estimates for population density was 10-41 deer/km2. The deer herd was in good nutritional condition: reproductive rates were high and kidney fat indices (KFI) closely reflected seasonal and sex-related physiological stresses (P = 0.0001), although the animals were never nutritionally deficient to the point that humerus marrow fat (HMF) declined (P = 0.082).

Thirty-seven white-tailed deer fawns (Odocoileus virginianus) born during summer and autumn 1993 were used to study effects of early weaning on survival and growth. Fawns at birth (date recorded) were weighed (kg), measured (cm), and tagged for identification. Fawns were randomly assigned to early weaned (treatment) or control groups at 60 days of age. Treatment animals were separated from their dams at this time, and control animals remained with their dams until 6 months of age. Both groups were fed a pelleted ration containing a medium protein level (11.6%). Study animals were sedated, weighed, and measured at 6,12, and 18 months of age. Number of points, inside width, main beam length, and main beam basal circumference of antlers was recorded at 18 months of age. Eleven animals died during the course of the study, but no deaths were treatment-related.

The welfare of white-tailed deer (Odocoileus virginianus) fawns orphaned during an early doe harvest is of management interest when the hunting and fawning seasons are temporally close. A 2-year study on early-weaned, captive white-tailed deer fawns was conducted to evaluate potential effects of early orphaning on growth and survival. Comparisons were made between survival of 60-day-old weaned fawns (TV = 28), 90-day-old weaned fawns (N = 21), and control fawns left with their does (N = 21). No significant differences were found between survival rates in 1994 (P = 0.68) and 1995 (P = 1.00). There were no significant differences between groups for any growth measurement at 7 months of age. These data demonstrate fawns orphaned at >60 days of age are capable of normal development with proper nutrition.

Resource managers require accurate estimates of hunter success rates with various weapon types to predict annual harvests and design management strategies. We obtained harvest data for white-tailed deer (Odocoileus virginianus) on the McAlester Army Ammunition Plant in southeastern Oklahoma during periods of compound (1983— 1988) and traditional archery (1989-1995) hunting to compare hunter success rates and the sex and ages of deer harvested. Hunter success was greater (P = 0.001) with compound (¯x = 17.8%; SE = 1.3) than traditional archery equipment (¯x = 10.7%; SE = 0.9). Total harvest (P = 0.002), number of bucks harvested (P = 0.001), and number of does harvested (P = 0.027) was also greater during compound archery hunts. Deer population estimates (P = 0.484) and fawn:doe ratios (P = 0.148) were not different between periods of compound and traditional archery.

The impact of raccoon (Procyon lotor) hunting with trained hounds on movement, daily use area, and harvest potential of white-tailed deer (Odocoileus virginianus) was studied on Westvaco Corporation's North Whitener Tract in Jasper County, South Carolina, in winter 1993-94 and 1994-95. Two groups (treatment and control) of radioinstrumented deer were selected from a larger group of radio-collared deer. Minimum daily total distance moved (MTD), 4 subsets of MTD, and daily use area were calculated and/or plotted from radio-locations during the 24-hour period before and the 24-hour period during and after a raccoon hunt for each radio-instrumented deer. There were no differences (P ≥ 0.05) between treatment and control areas in pre-hunt and post-hunt periods for any of the movement parameters. The percent overlap for each deer's prehunt daily use area and its corresponding post-hunt daily use area did not differ between treatment and control areas.

White-tailed deer (Odocoileus virginianus) damage to cotton (Gossypium hirsutum) was evaluated during 2 growing seasons in east-central Alabama. Deer began browsing cotton as soon as cotyledons emerged, and all plant parts were browsed during the growing season. Browsing of cotton cotyledons may kill plants and will reduce yields if it is extensive. However, browsing on cotyledons was rare in this study. Most feeding was done on cotton leaves, and occurred too late to reduce yields. Similarly, square and terminal removal after August and small boll removal after September occurred too late to impact harvestable yields. The most serious damage to cotton occurred when deer fed upon small squares and associated terminals in July and August and on small and medium (<27 days of age) bolls in August and September.