Journal of the Southeastern Association of Fish and Wildlife Agencies

Yo-yo catch and mortality of caught fish were evaluated at Caddo Lake, Texas, using attended (checked at least once each hour) and unattended fishing techniques. A total of 400 yo-yos (200 attended and 200 unattended) were fished 3,991.75 hours during February and March. Ten species were caught; black crappie (Pomoxis nigromaculatus) and yellow bullhead (Ictalurus natalis) comprised 54% and 20% of the catch, respectively. Catch rates were significantly higher for attended sets (0.12 fish/yo-yo hour) than unattended sets (0.03 fish/yo-yo hour). Unattended sets resulted in high mortality (85% of all fish caught) while no mortality resulted from attended sets.

The effectiveness of gill nets and jug lines were evaluated for selectively harvesting large alligator gar (Lepisosteus spatula) and longnose gar (L. osseus) in Sam Rayburn Reservoir, Texas. Gill nets with bar-mesh sizes ranging from 5.08 to 15.24 cm and jug lines were fished during September and October 1986. Catch per unit of effort (CPUE) in number of fish per man-hour, lengths of gar captured, and relative selectivity by gear were compared. Average effort required to fish I gill net or 10 jug lines overnight was 1 man-hour. CPUE of both gar species was significantly greater by jug lines than by gill nets. Jug lines were more selective for gar than gill nets.

Twenty-seven quantitative fish collections and site habitat analyses were performed from October 1985 to April 1988 in the Little Missouri River in Southwest Arkansas. Longitudinal addition of fish species (from 9 to 30) corresponded to downstream change in physical habitat. Species diversity increased from headwaters to midreaches and was positively correlated with increasing stream size, decreasing particle substrate size, depth, and development of alluvial riffle/pool morphology. The prime factor associated with the increasing fish species diversity was the corresponding increase in habitat diversity downstream. Fish feeding guild structure did not change from headwaters to midreaches. Insectivore/herbivore feeding guilds dominated all riffles and shallow pools. Insectivore/piscivore feeding guilds dominated all deep pools. Differences in species diversity among sample reaches progressing downstream was gradual.

Saugeye (Stizostedion vitreum vitreum X S. canadense) stocked in Thunderbird Reservoir from 1985-1987 were monitored for growth, food habits, and possible diet overlap with largemouth bass (Micropterus salmoides) as they progressed from age-1. Age-1 and age-2 saugeye attained mean length of 445 mm and 543 mm, respectively, by October. Food habits revealed saugeye 301-400 mm began converting from inland silversides (Menidia beryllina) to shad (Dorosma spp.) as their main food item. A predator-prey length relationship showed mean prey length was approximately 30% of predators. Diet overlap between saugeye 301-400 mm and largemouth bass 300-400 mm occurred in the summer season. Saugeye introductions appear to be a desirable management strategy for increasing predation on inland silversides, large shad, and slow-growing white crappie (Pomoxis annularis) populations.

A total of 13 radio-tagged walleyes (Stizostedion vitreum vitreum) were monitored from spring through summer in 1986 and 1987 in Luxapalila Creek near Columbus, Mississippi. Movements indicated pre-spawning aggregation in the lower Luxapalila Creek during January and February, upstream spawning March through early April, and widely varying downstream post-spawning movements. Seasonally high water discharges in March were cues for upstream movements to pool areas below spawning sites at shallow <1.5 m) gravel riffles. High discharges restricted spawning and possibly reproductive success. Three groups of walleyes spawned in Luxapalila Creek: residents of the upper stream, of the Luxapalila Park area, and of the Tennessee-Tombigbee River. Lower Luxapalila Creek was an important summer holding area. Little nighttime feeding occurred during summer high water temperatures (28°-31°C).

Subsampling to obtain length measurements is often necessary when large numbers of organisms are captured with bag seines in a resource monitoring program. Catches of white shrimp were used to examine the representativeness of the subsamples and to construct sample size selection curves for determining the number of length measurements required to detect a given percentage difference between samples. A wide size range of white shrimp (Penaeus setiferus) is often captured in the same sample. Subsampling to measure 19 white shrimp from Texas Parks and Wildlife Department (TPWD) bag seine collections to estimate mean lengths was not always completely random but systematic discrepancies were not evident. A subsample of 19 shrimp can be used to detect a 15% difference in mean lengths among collections 80% of the time (a = 0.01) with a CY of 13.3%, and in the worst case (CY = 37.9%), can detect a 50% difference in mean length 80% of the time (u = 0.01).

Visual observations on group composition of white-tailed deer (Odocoileus virginianus) on the Fred T. Stimpson Wildlife Sanctuary, Alabama, were conducted from November 1975 through March 1976. Group composition was divided into 3 periods (pre-rut, 6 Nov - 6 Jan; rut, 7 Jan - 29 Feb; post-rut, March) to determine effects of reproductive behavior. During the study, 937 groups containing 2,391 deer were sighted. Sightings of single males and single females increased during the rut, and mean group size decreased. Data suggest a temporary dissociation of adult does from family groups as a result of reproductive activities. Based on observation of 2 radiotagged does the dissociation occurs primarily between adult deer or yearlings and not fawn-doe associations. The temporary dissociation of a doe from the family unit could facilitate breeding encounters without disrupting the movements and feeding activities of her social group.

Winter densities, age structure, and sex ratios of 9 populations ofraccoons (Procyon lotor) in western Tennessee were compared to 16 habitat variables and between aquatic-associated and upland habitats for 3 pairs of sites. Densities ranged from 0.8 to 18.3 raccoons/km2 • Highest densities and proportions of juveniles and females occurred in bottomland deciduous forests; lowest estimates were in upland, pine-deciduous forests. The proportions of females and juveniles and the density estimates were correlated with habitat variables that reflected a bottomland-upland gradient, such as the distance to deciduous shoreline and the basal areas of conifers and large deciduous trees. In addition to supporting larger raccoon densities, aquaticassociated habitats may also have relatively more on-site reproduction than upland habitats.

The removal of 50% annual alligator production over a 6-year period on 3 central Florida lakes did not appreciably change population size structures. On Lake Jessup, increases were found in the >0.6 m size classes, while on Lake Griffin no changes were found in any size classes. Lake Apopka experienced significant (P <0.05) declines in all size classes, but these declines resulted from unexplained and unrelated mortality and reproductive failures and not from the early age-class harvest. No change in size structure was found for the control area, Lake Woodruff.

Food habits were determined for a black bear (Ursus americanus) population in Great Dismal Swamp on the Atlantic Coastal Plain. A total of 535 scats was collected from May 1984 to August 1986. Seasonal shifts in diet composition were similar to previously reported findings for black bear food habits in the southeastern United States. Diets changed from succulent, herbaceous material in the spring to soft mast and corn in the summer. Early fall diets were largely black gum (Nyssa sylvatica) (33%) and oak (Quercus spp.) (32%) mast. Late fall and winter diets were dominated by fruits of evergreen shrubs and vines. Animal foods comprised 3% of the annual diet. Four scats contained evidence of cannibalism. Management plans for Coastal Plain bear populations should include maintenance and enhancement of stands of mature gum, oaks, pocosins, and forest openings.

Harvest data constitute an important source of information for the deer manager, but interpretation can be complicated by reporting bias, hunter selectivity, differential vulnerability, and aging errors. Daily harvest records for >165,000 white-tailed deer (Odocoileus virginianus) taken during 1979-85 Illinois firearm seasons were examined for evidence of bias. Antlered males were apparently killed at a higher rate than other classes of deer; consequently, they became relatively less numerous in the herd and in the harvest as the season advanced. Yearling males, in particular, were highly vulnerable early in the season. Declining availability of antlered males and reduced hunter selectivity shifted pressure toward fawns and females as the season progressed. Fawns were apparently underrepresented in the total harvest but not the antlerless harvest. Male fawns were more vulnerable than their female cohorts.

Evaluations of health status were made on 119 pen-raised wild turkeys (Meleagris gallopavo) by complete necropsy, serological, and microbiological testing, blood smear examinations, subinoculation trials, and parasite identification. At least 33 species of parasites including 9 protozoans, 14 helminths, and 10 arthropods were found. Infectious disease agents isolated or identified histopathologically were avian pox virus, Mycoplasma gallisepticum. and Aspergillusjumigatus. Serologic testing disclosed antibodies to infectious bursal disease virus-2, M. gallisepticum. M. meleagridis. and Salmonella spp. Based on an epidemiologic evaluation of the disease risks, we conclude that the release of pen-raised wild turkeys without proper consideration for disease prevention should be discouraged or prohibited.

A total of2,l89 pen-reared northern bobwhites (Colinus virginianus) was banded and released when 14 weeks old on a hunting club in north-central Texas from May 1984 through February 1985. Recovery of banded birds released during the 5 months preceding the hunting season was 7.5%, whereas 38.6% of birds released during the first 3 months of the hunting season were recovered compared to 55.4% of those released during the last 2 months. Release of birds just prior to the hunt decreased the cost per bird recovered to $6.5 as compared to $18.65 for birds released 14 days before the hunt.

From December 1980 to June 1988,66 nuisance black bears (Ursus americanus) were captured and released at beeyards to create an aversion to beeyards. In most instances, bear depredations at a beeyard stopped after a bear was trapped, handled and released at the site. Of 63 bears released with tags, 14% were recaptured 1-3 times after causing additional apiary damage. These repeat offenders were mostly adult males. Two bears that continued to raid apiaries after being traped 3-4 times in a year were relocated when it became apparent that these bears were not deterred by trapping. When used in conjunction with a working electric fence, trap and release may further condition bears to prevent apiary depredation.

Wood duck (Aix sponsa) nest boxes (N = 190) were placed along the lower Holston River and the lower French Broad River in eastern Tennessee and were maintained and checked from 1976 to 1979. Wood ducks used 1.3% of the boxes inspected during the 4-year study. Night brood counts ranged from 0.42 to 0.68 broods/km of river and did not indicate an increase in brood production. Nest box use and brood production in this study were significantly less than reported for a study on the upper Holston River. We believe that aquatic vegetation, valuable as a food base for wood duck broods, was a more limiting factor in our study than the availability of nest cavities.

A stratified random survey design proposed to increase the efficiency of estimating numbers of waterfowl wintering on Mississippi catfish ponds was evaluated. The optimally allocated sample generally produced estimates with coefficients of variation <50%, similar to those obtained from a completely random design used previously. Coefficients of variation were not associated with survey date. Stratified random sampling reduced the number of catfish pond clusters surveyed and flight time, compared to completely random sampling. We recommend using the stratified random design to estimate numbers of waterfowl on aquacultural impoundments when sampling of impoundments must be limited.

Reproductive success of 94 ruffed grouse (Bonasa umbellus) reintroduced into northern Arkansas during September and October 1984 was apparently low. Four broods were found during 2 reproductive seasons after release. Dispersal of all located male ruffed grouse was limited to a 1.6 km radius of the initial release site. Brood sightings in an area 2.4 to 2.8 km northeast of the initial release site were direct evidence of female dispersal. Survivorship of male grouse up to the first courtship drumming period was at least 25%.

The strip transect and fixed-area circular plot methods of sampling birds were compared in 4 even-age pine-hardwood stands (seedling, sapling, pole and sawtimber tree-size classes) during winter and spring of 2 years. During spring the circular plot method resulted in more species and more individuals than the transect method. Most differences were significant (P > 0.05). Winter samples showed the same pattern, however most differences were not significant (P >0.05). For selected species and species assemblages, fixed-area circular plots generally resulted in higher numbers during both seasons, except for high canopy inhabitants.

We determined the incidence of lead, steel, and total shot ingestion in 4 species of dabbling ducks on a major wintering area of the Atlantic Flyway. Gizzards (N = 1,771) were collected from hunter-harvested mallards (Anas platyrhynchos) , northern pintails (A. acuta), blue-winged teal (A. discors), and American greenwinged teal (A. crecca) during the 1985-86,1986-87, and 1987-88 waterfowl hunting seasons on the Santee River Estuary (SRE), South Carolina. Total shot ingestion rates were high in northern pintails (15.9%) and mallards (8.4%), and low in American green-winged teal (1.3%) and blue-winged teal (0.8%). Because gizzard analysis can underestimate the extent of lead shot exposure in waterfowl, the SRE is implicated as a serious contributor to lead shot ingestion by wintering northern pintails and mallards.

Fetal counts of muskrats (Ondatra zibethicus) yielded an average of 2.97 ± 1.06 fetuses per litter (range: 1 to 5). Placental scars averaged 7.97 ± 4.34 per breeding female (range: 1 to 22) and indicated production of 2.7 litters per breeding female per year. Corpora lutea counts disclosed that litter sizes ranged from I to 7, averaged 3.54 ± 1. 15, and differed significantly from fetal counts. Litter size as determined by lodge surveys (i = 2.18 ± 0.25) was less than that determined by fetal counts and represented post-partum mortality. Mortality estimates indicated a loss of 16.1% between ovulation and fetal counts; an 18.5% loss occurred between the time fetal counts were made and the time that muskrats reached the 1- to 5-day age class. Subsequent losses of 7.4%,5.4%,9.9%, and 9.4% occurred between 1-5, 6-10, 11-15, 16-20, and 21-24 day old age classes, respectively.

The quality of spring hunting for eastern wild turkey (Meleagris gallopavo silvestris) gobblers may decline with increased road access. A questionnaire mailed to turkey hunters assessed response to road closures for walk-in hunting on a Mississippi wildlife management area during spring 1986. Because hunters felt road closures improved hunting quality and reduced interference from other hunters, they strongly supported (92%) walk-in turkey hunting. Hunters who disapproved of road closures were significantly older than those who approved. Hunter involvement in the road closure decision likely was important to its success.

In 1984 North Carolina opened the first tundra swan (Cygnus columbianus) hunting season in modern times in the Atlantic Flyway. During this first season, 1,000 permits were issued. Subsequently, 6,000 permits/year were allowed and issued. A 4-year study using aerial, ground, and hunter surveys was established in 1984 to monitor this hunting season. Estimated total annual kill beginning with the 1984-85 season was 334, 2,783, 2,579, and 3,007 for each year of the study, respectively. The harvest averaged slightly over 5% of the state's and about 3% of the Flyway's midwinter population. The ratio of permits issued to swans harvested was 2.2, and was comparable to results reported from Utah. Crippling rates (x = 11.3) were less in the North Carolina season than those reported elsewhere. Hunter success rates were not related to immature/adult ratios.

A 1985-86 project in southeastern Georgia was conducted in an effort to study the effectiveness of several methods of applying a sardine bait-station technique for indexing black bears (Ursus americanus). A total of 120 stations were set out in 20 lOO-ha experimental replicates during each of 3 months (May, July, and November) with random sampling of equal numbers of stations according to method of hanging baits (hanging or nailing) as well as equal numbers of stations being checked after 4, 8, and 12 days. Chi-square analysis indicated that a bear visit to a bait-station was not dependent on method of presentation. Overall visitation rates were proportionally higher though not statistically different for stations checked after 8 and 12 days than for stations checked after 4 days. Analysis by month sampled indicated significantly more bear visits during May than during July or November.

Selected characteristics of forest industry hunt-lease programs were determined for II southern states. Mail surveys were obtained from 77 of 109 (71 %) delivered questionnaires that were completed and returned. Respondents reported owning a total of more than 9.4 million ha within the study area in 1984. Most (83%) charged hunters for access to corporate lands. A majority of respondents indicated that non-monetary benefits were gained from hunt-lease programs. Annual lease fees ranged from $2.47 to $26.88/ha and varied by state, timber type and location. Administration was the highest corporate cost of hunt-lease programs. Local tradition of free public hunting was the most frequent reason for not leasing hunting rights on forest industry lands.

Movements of coyotes (Canis latrans) (N = 6) and bobcats (Felis rufus) (N = 4) on the La Copita Research Area in southern Texas were determined by radio-telemetry from April 1985 through September 1986. Mean home range sizes of resident individuals were 3.04 km2 for coyotes and 2.88 km2 for bobcats. These predators frequently traveled outside their home ranges and exhibited extensive interspecific home range overlap. Frequent travel outside the home range seemed related to subsequent dispersal. Several individuals captured on La Copita proved to be non-residents or temporary residents of the ranch. Thickets and drainages were important habitat types in bobcat home ranges. Coyotes were less selective in their habitat use patterns. Drainages were used as travel corridors by both coyotes and bobcats.