Search

A grain bait treated with an anticoagulant rodenticide (Prolin) and attached by an edible glue to the inside of water repellent paper tubes was tested for the control of meadow mice (Microtus pennsylvanicus) at Sturgeon Bay, Wisconsin, during fall, winter, and spring, 1964-65. In the laboratory, seven of eight mice that could choose other food ate an average of 9.7 g of Prolin-treated bait in 10 days and survived from 4-10 days. Eight mice given no choice ate an average of 8.2 g of bait and survived from 3-7 days. When eight mice were placed in baited caged areas around apple trees all activity ceased in 5 days. In orchard plot tests all trees escaped mouse damage in an area baited for 2 months. In an area baited 18 days, 11 percent of trees were damaged, compared to 39 percent in the unbaited control area. This study was conducted to determine the feasibility and effectiveness of using 5x 1?-inch water repellent paper tubes with an inner lining of bait treated with the anticoagulant rodenticide Prolin (0.05 percent warfarin and 0.05 percent sulfaquinoxaline) to control meadow mice in Wisconsin apple orchards. Wisconsin orchardists estimate they lose an average of 2 percent of their trees to meadow mouse injury each year. In some years entire new plantings of young apple trees are girdled by meadow mice during the winter. Grain or cut apple baits with zinc phosphide and strychnine, or endrin ground sprays have been used in recent years to combat meadow mice. According to the literature, these methods have shown varying degrees of success, but with questionable safety to wildlife populations (Garl ugh and Spencer 1944, Eadie 1950, Horsfall 1951:13, Spencer 1959:25, Mohr 1959, Fitzwater 1961, Wolfe and Durham 1963). This content downloaded from 157.55.39.243 on Thu, 06 Oct 2016 04:44:32 UTC All use subject to http://about.jstor.org/terms RODENTICIDE FOR MEADOW MICE Libby and Abrams 513 Although the anticoagulants have been effective and have achieved a commendable record of safety in commensal rat and mouse control, they have not been used to control meadow mice outdoors. Because a single feeding of anticoagulant rodenticides is rarely lethal to mice, the use of such baits outdoors would require that the bait be applied repeatedly or that protected bait stations be set up and maintained to keep the bait fresh and palatable throughout the baiting period. To establish and service such stations has been considered impracticable in commercial orchards. The development of a water repellent paper tube provides a disposable "bait station" that makes it feasible to use anticoagulant baits in the field. The writers wish to acknowledge the helpful comments and suggestions made during this study by Donald A. Spencer, Chief Staff Officer, Animal Biology, Pesticides Regulation Division, USDA; Walter Dykstra, Research Staff Specialist, U. S. Bureau of Sport Fisheries and Wildlife; William Gusey, Pesticides Regulation Division, USDA, and thank R. S. Ellarson, Department of Wildlife Management, University of Wisconsin, for reviewing the manuscript. LABORATORY FEEDING TESTS Sixteen meadow mice trapped at the Gordon Orchard, Sturgeon Bay, Wisconsin, were received at the laboratories of the Wisconsin Alumni Research Foundation, Madison, on November 11, 1964. These mice were then individually housed in cages and fed whole kernel yellow corn, apples, and alfalfa hay. On November 23, eight mice were continued on this diet plus a Prolin tube on a free choice basis. The other eight mice received the Prolin tube only. The results of the laboratory feeding studies are presented in Table 1. Fig. 1. Caged tree at the University of Wisconsin Experiment Station, Sturgeon Bay.

N April 1, 1948, an adult female Batrachoseps (=Plethopsis) wrighti, heavily laden with eggs, was obtained at the type locality, 8.7 miles southeast of Sandy, Clackamas County, Oregon. Since the eggs of this salamander were unknown, she was brought into the laboratory in the hope that oviposition could be observed. She was placed in a gallon glass container with a metal screw lid. A single paper towel, wetted with distilled water, was put into the bottom of the jar. The container with the salamander was kept in a dark room where the temperature was maintained continuously at approximately 12? C. Thermal data on field animals suggested that this temperature might be commonly experienced under natural conditions. After nearly one month in captivity, the salamander still retained her eggs. Dispairing of obtaining them without stimulation, I employed the pituitary implantation technique described by Noble and Richards (1930) and thereby oviposition was initiated. Implantation was done as follows. On May 25, 1948, at 3:40 P.M., I implanted a single whole pituitary gland freshly removed from an adult male Hyla regilla. The salamander was anesthetized in chlorotone (1 part saturated solution to 3 parts water), whereupon a small incision was made in the skin of the gular area. The Hyla pituitary was removed with forceps and was transferred directly to the site of the incision. A dissecting needle was used to force the gland into place beneath the the skin. By the time the salamander recovered from the chlorotone, several hours later, coagulation of blood and lymph at the site of the operation prevented displacement of the implant. About 42 hours later, a second Hyla pituitary was implanted. On June 1, 1948, at 9:00 A.M. the first egg was laid. Laying was completed June 2 at around 9:00 P.M., approximately 36 hours after the beginning of oviposition. Eleven eggs were deposited. Their chronology was as follows: June 1-3 eggs laid at approximately 21/2 hour intervals, around 9:00 A.M., 11:30 A.M., and 2:00 P.M.; June 2-8:40 A.M., 7 eggs present with the 8th emerging; 1:30 P.M., 9th egg emerging; 7:15 P.M., 10 eggs; 9:00 P.M., 11 eggs. The brief description of the eggs (Stebbins and Lowe, 1949), may now be supplemented [see Fig. 1, (1)]. The most striking characteristic of the eggs is their size. After some hours on the wet paper towel of the container, the jelly layers had swelled enormously, measuring, to their outer surfaces, 9 to 10 mm. The female salamander, only about 5 mm. in greatest width, seemed incredibly small to have deposited such large eggs (see Fig. 2, A). When laid, the jelly layers were thin and the vitellus was elastic, permitting its passage, with some temporary distortion, through the cloacal aperture. The vitellus, or egg proper, was unpigmented, between Cream Color and Pale Ochraceous-Buff.1 It rotated with the animal pole uppermost when the egg

designated a control animal and the feces of these animals were examined frequently for helminth ova. They were consistently found to be free of ova except for a few Strongyloides sp. Twenty-seven to 30 days following the administration of the larvae, feces were collected from all of the lambs. The feces of each of the control lambs was crushed by hand, mixed with charcoal and placed, several millimeters thick, on two thicknesses of moistened filter paper in Petri dishes. The feces from each of the infected lambs were crushed and placed in saturated sodium chloride solution in large centrifuge tubes. With the aid of centrifugation the eggs were separated from the fecal material and were gathered from the surface of the solution. The eggs were then washed once with 0.9 percent sodium chloride solution and three times with tap water. Aliquots of the resulting samples were removed and the number of eggs counted. Approximately 8,000 fertilized eggs were placed on each control lamb's feces in the Petri dishes. The scheme for culturing was as follows

Activity periods of Microtus and Sigmodon were monitored by checking live-traps for 12 consecutive 2-hour intervals. Sigmodon were active nocturnally with peaks of activity at dawn and dusk. Where no Sigmodon were captured, Microtus also showed activity patterns with dawn and dusk peaks. Where Sigmodon and Microtus were both captured, Microtus activity was more evenly distributed throughout the night. Several recent papers exploring the relationships between Sigmodon hispidus and Microtus ochrogaster (Terman, 1974, 1978; Glass and Slade, 1980) or M. pennsylvanicus (Terman and Johnson, 1971) have demonstrated that Sigmodon aggressively inhibit Microtus activities in the laboratory and appear to do so in the field. Glass and Slade (1980) found that the degree of spatial and temporal segregation of S. hispidus and M. ochrogaster varied seasonally and that spatial rather than temporal segregation was the predominant pattern in the summer. They suggested that temporal segregation would have been sufficient for Microtus to avoid Sigmodon aggression, but that hot summer temperatures restricted Microtus activities to nocturnal periods when Sigmodon were also active. However, Glass and Slade could only identify captures as dirunal or nocturnal, so they could not detect temporal separaation within the nocturnal period in the field. Under normal summer temperature regimes in the laboratory, they reported a tendency for Microtus to shift its nocturnal activity away from those hours of highest Sigmodon activity. This paper reports a more complete field study of the impact of Sigmodon presence on Microtus summer diel activity patterns. Two previously established trapping grids on the Nelson Environmental Studies Area, 14.5 km northeast of Lawrence, Kansas, were live-trapped from 13 June to 2 August 1978. They were selected so that grid 1 supported both Microtus and Sigmodon populations, whereas grid 2 had Microtus only. Grid I was 7 stations by 6 stations with 15 m between stations, and This content downloaded from 157.55.39.208 on Fri, 14 Oct 2016 04:07:05 UTC All use subject to http://about.jstor.org/terms VOLUME 84, NUMBER 3 135 was located in an abandoned field overgrown with asters (Aster multiflorus), sunflowers (Helianthus spp.), goldenrod (Solidago sp.) and several other weedy perennials and annuals. One collapsible Sherman (23 cm x 9 cm x 8 cm) and one modified Fitch (Rose, 1973) live trap were set at each station. From 7 to 17 (x = 11; 11.6/ha) individual Sigmodon hispidus and from 34 to 46 (x = 39; 41.3/ha) Microtus ochrogaster were captured during each of 4 trapping sessions, resulting in a total of 58 and 258 captures, respectively. Grid 2 was 8 x 12 with 7.5 m between stations, and was located approximately 1 km northwest of grid 1 in an old hayfield dominated by cultivated brome (Bromus inermis) and little bluestem (Andropogon scoparius), grading into dogwood (Cornus drummondii), Osage orange (Maclura pomifera), and cottonwood (Populus deltoides) along an intermittent stream. One collapsible Sherman trap was set per station. From 12 to 23 (i = 16; 30.1/ha) Microtus were captured per trapping session, for a total of 74 captures, but the only Sigmodon captured was an adult female caught once. Traps were checked every two hours for one 24-hour period each week, alternating weeks between grids. At each capture the species, age, sex, weight and time of capture (military central daylight time) were recorded. Even though most individuals were toe-clipped or ear-tagged, individual activity profiles could not be constructed because few animals were captured more than twice in a 24-hour period. The activity patterns of Microtus and Sigmodon, as indicated by time of capture, are similar to each other (Fig. 1) and to those reported by Calhoun (1945). Both species show bimodal frequencies of capture with evening and morning peaks. The principal alteration in Microtus activity on grid 1 was a third peak, perhaps better described as sustained activity, in the middle of the night. (Glass and Slade (1980) also found an increase in Microtus activity from 2400-0800 hours in the presence of Sigmodon in the laboratory, but they did not observe much activity earlier in the night.) Chi-square tests of independence were used to test for statistically significant differences in these patterns. The captures from 1200-1800 hours were combined in the analysis to insure expected values near 5 for each cell. The differences in activity patterns between Sigmodon on grid 1 and Microtus on grid 2 were of only marginal significance (X27 = 12.24, 0.05 < P < 0.10; 2400 and 200 hours were combined) with the largest differences occurring in relative height and timing of the morning peak. Microtus activity patterns on grid 1 and grid 2 were significantly different (X2, = 18.46, 0.01 < P < 0.025), the principal difference being the aforementioned activity in the middle of the night. Glass and Slade (1980) suggested that high summer temperatures restricted periods of Microtus activity so we divided our Microtus captures into those from cool, necessarily moist days (maximum temperatures 23.328.3oC, minimum temperatures 16.1-18.3?C, 7.4-38.6 mm of rainfall) and This content downloaded from 157.55.39.208 on Fri, 14 Oct 2016 04:07:05 UTC All use subject to http://about.jstor.org/terms 136 TRANSACTIONS OF THE KANSAS ACADEMY OF SCIENCES

Whole and dressed body weights were obtained from 1,009 white-tailed deer (Odocoileus virginianus) from Crab Orchard National Wildlife Refuge, Williamson County, Illinois. Chest girths, hindfoot lengths, lower mandible lengths, and antler dimensions were taken from somewhat smaller samples. The data were analyzed by sex and age, and relationships with body weight were established. Males generally were larger than like-aged females with respect to all measurements. Both sexes reached near-maximum size at 4.5 years of age but females matured earlier than males, especially with regard to body weight. Indices of condition (body weight and antler size) indicated that the previously unharvested Crab Orchard herd compared favorably with nearby hunted populations with the possible exception of yearling males. J. WILDL. MANAGE. 39(1):48-58 This paper describes certain morphological features of a large sample of whitetailed deer from a previously unharvested herd. Body weights, chest girths, hindfoot lengths, lower mandible lengths, and antler measurements are analyzed by sex and age classes and growth patterns are discussed. Body weights and antler measurements are used to compare general condition of the herd with that of deer in nearby hunted range. The sample was obtained during a controlled harvest to reduce herd size on Crab Orchard National Wildlife Refuge (CONWR), Williamson County, Illinois (Roseberry et al. 1969). All deer were taken from a 7,285-ha (18,000-acre) portion of the Refuge that normally is closed to hunting. The herd was first established in 1942 and has been under intensive study since 1960 by the Cooperative Wildlife Research Laboratory (CWRL) of Southern Illinois University at Carbondale. The habitat, which is considered ideal for deer (Hawkins et al. 1971), has been described previously (Autry 1967, Hawkins and Klimstra 1970). Population density prior to the January 1966 special harvest was estimated to be 1 deer/3.2 ha (1 deer/8 acres) (Autry 1967). The authors are indebted to the many graduate and undergraduate workers of the Cooperative Wildlife Research Laboratory who aided in the collection of data. We also acknowledge officials of the Crab Orchard National Wildlife Refuge and personnel of the Illinois Department of Conservation for assistance. This paper is a contribution from Project No. 15 of the Laboratory: Big Game Investigations. METHODS AND MATERIALS All harvested deer were processed at one of two check stations operated by Laboratory personnel. The animals initially were weighed 0.45 kg (1 pound) with a Viking spring balance, eviscerated, then reweighed. The maximum chest girth of all deer with rib cage intact was measured ? 0.32 cm (1 inch) with a flexible tape measure. Length of the hindfoot from hock to the tip of the longest claw was measured by the same procedure (Brohn and Robb 1955, Park and Day 1942). Antler beam diameter was measured to the nearest 1.0 mm with a vernier caliper. The beam diameter was taken 2.54 cm (1 inch) above the burr and, when possible, did not include bony protuberances (Anderson and Medin 1969). Antler length was measured ? 0.32 cm 48 J. Wildl. Manage. 39(1):1975 This content downloaded from 40.77.167.54 on Sat, 14 May 2016 06:18:18 UTC All use subject to http://about.jstor.org/terms MORPHOLOGICAL CHARACTERISTICS OF DEER * Roseberry and Klimstra 49 ('s inch) with a flexible steel tape measure along the outside of the main beam from burr to tip. Length and diameter were recorded from the left beam, but the undamaged right beam was measured when the left was damaged. All antler protuberances over 2.54 cm long, excluding brow tines, were counted (as the number of antler points) on both beams. A lower mandible was removed from most animals for purposes of age determination. Age classes in this study were established on the basis of tooth eruption and wear (Severinghaus 1949). Total length of the mandible, from its most anterior point to the posterior rim of the angle of ramus (dentary length) was measured to the nearest 1.0 mm with a modified meter rule. The distance between the alveola of the anterior premolar and the corner incisor (diastema length) was measured to the nearest 0.1 mm with a vernier caliper (Bergerud 1964). All data were placed on 80-column IBM cards and processed with an IBM 370/155 computer at the data processing facilities of Southern Illinois University at Carbondale.

Performance of Romney ewes and their lambs grazing an annual grass-subclover hill pasture was evaluated under both fivepaddock rotational and continuous grazing management treatments during 1977 and 1978. Live weight gains of ewes and lambs tended to be greater under rotational than under continuous grazing during the spring or fall green-feed periods. During the summer dry-feed period, however, sheep maintained thefr body weight better under continuous than under rotational grazing. These data support the hypothesis that rotational grazing most effectively improves animal performance during the green-feed period, perhaps through its reguhtion of pasture production. A common goal of rotational grazing systems is to increase forage production by controlling the frequency and intensity of pasture defoliation. Increased herbage production from rotational compared to continuous grazing management systems has been reported for perennial ryegrass (Ruane and Raftery 1964, Young and Newton 1975) and for perennial ryegrass-white clover (Trifolium repens) pastures (Marsh and Laidlaw 1978). This increase is thought to result from the maintenance of a more nearly optimum Leaf Area Index (LA1 = leaf area per unit ground area) than is achieved under continuous grazing (Smetham 1973). But as Young and Newton (1975) pointed out, increased forage production will result in superior animal production only if the additional forage available is consumed. Lambourne (1956) concluded that where feed supply was insufficient under continuous grazing, the extra feed produced under rotational grazing led to improved lamb growth. When feed supply was already adequate on continuously grazed pastures, the additional forage produced by rotational grazing became rank, causing a decline in forage quality. Lamb growth suffered. This same principle is well illustrated by the stocking rate X grazing management interaction discussed by Robinson and Simpson (1975). They noted that animal performance under rotational grazing only exceeded that of continuous grazing at intermediate to high stocking rates. When stocking rates were either very low or extremely high, rotational grazing tended to reduce animal performance. A second goal of rotational grazing is to improve the efficiency by which forage is harvested and converted to salable animal products. Continuous grazing allows livestock a considerable amount of dietary selectivity (Heady 1961). This selectivity, while allowing animals to choose a high quality diet (Weir and Tore11 1959), often results in overutilization of the most preferred plants allowing less preferred plants to become rank and of low quality (Sharrow and Krueger 1979, Smetham 1973). Undesirable shifts in species composition of the sward may follow as unpalatable species increase and as low growing plants such as clovers are shaded out (Smetham 1973). The high animal density for a short period of time employed by rotational grazing systems tends to increase grazing pressure, thus suppressing dietary selectivity by animals. This favors a more even utilization of all forage produced. The nonuse period immediately following grazing favors the growth of clovers and provides for production of high quality grass regrowth for grazing during the next rotational cycle (Smetham 1973). Since dietary selectivity may be suppressed under rotational grazing, it is important that ample high quality forage be present on each paddock if high animal performance is desired. Heady (1961) expressed concern about the implications of reduced dietary selectivity of livestock if rotational grazing systems were applied to the California annual grassland. He suggests that livestock production would suffer if animals were concentrated on small areas even for short periods of time due to reduced quantity and quality of forage available per animal. U Author is associate professor, Department of Rangeland Resources, Oregon State Although both continuous grazing and rotational grazing sysniversity, Corvallis 97331. terns are widely used in western Oregon, acomparison of these two Paper No. 6210. Oregon Agricultural This riport was submitted as Technical I E: rperiment Station, Corvallis. ~.nn.rrint rwx=ivrd lnnunrv IS 19X2. .._ .._ 1_..r. .__-..__ __..__., __. .._-. systems has not been made on annual grass-subclover pastures. The objective of this paper is to discuss the effects of rotational and JOURNAL OF RANGE MANAGEMENT 36(5), September 1983 593 continuous grazing on the performance of ewes and lambs grazing annual grass-subclover under heavy stocking rates. Table 1. Speeiaa composition (% of weight) and standing crop (kg//ha) of herbage present prior to stocking of pastures grazed continuously (C) and rotationally (R). Materials and Methods Information presented in this paper is taken from the final 2 years of a 6-year study. Animal performance data from the initial 4 years of the study were previously reported by Sharrow and Krueger ( 1979). 18 May 1977 I5 April V978 Research was conducted during 1977 and 1978 at the Soap Creek Ranch, approximately 20 km north of Corvallis, Ore. Elevation is approximately 100 m. The climate is maritime with cool, rainy winters and hot, dry summers. Mean annual precipitation is 103 cm. Species C R C R Vulpia myuros (rattail fescue) 62 61 48 52 Bromus mollis (softchess) 9 9 II 13 L&urn perenne (perennial ryegrass) 4 1 3 T Tk~olium subterraneum (subclover) 20 19 32 29 Misc. grass T 2 2 I

The number of woodchucks (Marmota monax) on an army ordnance depot in south central Pennsylvania decreased by 80% from 1955 to 1970. The depot contained 4,000 ha of former agricultural land that provided excellent sites for burrows. The birth rate was initially low (1.3), but increased and remained high (1.7) after 1958 due primarily to an increase in the percent yearlings pregnant. The death rate of adults changed little, but the loss of young during summer 1964-67 and during hibernation 196267 was high. Movements increased after 1962. Changes in habitat after 1960 resulted from cessation of leasing land for crops of corn, wheat, and barley. A drought occurred from 1961 to 1969, and decreased the quantity of natural vegetation. Thus, a great decline in food supply occurred after 1962. The mechanism for the decrease of woodchucks was failure of young to gain the weight necessary to survive the hibernation period and subsequent emergence. During 1957-60, 50% of young males and females reached a weight considered adequate for survival, compared to only 1-10% during 1962-66; during 1967-70, 50% of males and 80% of females achieved adequate weight. J. WILDL. MANAGE. 45(3):658-668 A population of woodchucks declined drastically from 1955 to 1970 on the Letterkenny Ordnance Depot near Chambersburg, Pennsylvania. Analysis of the data permits identification of the mechanism of population regulation. The methods have been reported in detail in papers that give results of certain aspects of the research (Davis 1962, Davis et al. 1964). This paper interprets the sequence of events that resulted in an 80% decline in woodchucks, and thereby provides an excellent example of the effect of habitat. This research was conducted by persons whose names appear in the Literature Cited, through grants from the National Institutes of Health. The recent data are from Ludwig (1975). The study site consisted of 4,000 ha of agricultural land. Earthen bunkers, constructed in 1942, and distributed throughout the depot, created excellent sites for burrows. Agriculture consisted of grain and hay crops from 1955 until 1960, when the Army stopped leasing cropland. From 1960 to 1970 these fields were in advancing stages of old-field

Nutritional analyses were performed on several fruit species that are eaten by the neotropical frugivorous bat Carollia perspicillata. All essential amino acids assayed were found to be present in fruit pulp. Compared to the relative amounts of essential amino acids required for growth by laboratory rats, methionine and lysine were the most limiting amino acids in fruit protein. Captive bats were fed fruits of either Chlorophora tinctoria or Muntingia calabura so that the digestibilities of pulp nitrogen and gross energy could be determined. Digestibility estimates were lower than those found for Artibeus jamaicensis fed fruits of Ficus insipida, but this was probably caused by methodological differences. The measures of nutrient content and digestibility were used with estimates of daily nitrogen and energy requirements of bats to predict the amount of pulp needed to meet these requirements. Fruits were judged to be adequate nitrogen sources if the amount predicted to satisfy nitrogen requirements was less than that to satisfy energy requirements. By this criterion, most fruits were found to be adequate for maintenance metabolism; only fruit of Piper amalago was adequate for lactating bats. However, when amino acid deficiencies were considered, Piper amalago fruit was adequate only for maintenance metabolism, and no other fruits met this requirement. Insects found in fruit pulp were insufficient to constitute an important nitrogen source. Thus, if insects are important nitrogen sources they must be actively sought by bats. However, this analysis suggests that by selecting certain fruit species, frugivorous bats may not need to supplement their diets with insects. NEARLY ALL SPECIES OF NEOTROPICAL frugivorous bats (Phyllostomidae) studied include insects in their diets (Gardner 1977). The short-tailed fruit bat Carollia perspicillata also ingests insects (Arata et al. 1967, Fleming et al. 1972, Ayala & D'Alessandro 1973, Howell & Burch 1974), but their importance to the diet is unknown. Fleming et al. (1972) found that insects formed on average about 13 percent by volume of stomach contents in a sample of 272 bats whose stomachs contained food, but recently Thomas (1984a) has questioned some of the evidence for voluntary insect eating. An explanation for the inclusion of insects in the diet is that while fruits provide an often abundant and easily handled energy source, they do not provide sufficient nitrogen to meet the dietary requirements of bats (Morrison 1980, Thomas 1984b). Compared with insects, fruits contain very little nitrogen (Morton 1973, White 1974), however it is not clear that fruits are inadequate nitrogen sources. For example, Foster (1978) calculated that some fruits may be adequate nitrogen sources for nestling frugivorous birds. In this paper, I present the results of an investigation of the nutritional contents and digestibilities of some fruits eaten by Carollia perspicillata. The goal of this paper is to assess the value of fruit tissue as the only source of nitrogen for frugivorous bats. Specifically, I test the hypothesis that nitrogen requirements are met before energy requirements when bats ingest enough fruit to meet their energy requirements. MATERIALS AND METHODS The study was carried out from May to August, 1980 and 1981, in Parque Nacional Santa Rosa, Guanacaste Province, Costa Rica. This site was described in detail by Heithaus and Fleming (1978). Many fruit species that are available at this site during the wet season are eaten by Carollia perspicillata (Goodwin & Greenhall 1961, Gardner 1977). I collected sufficient fruit (>20 g dry pulp) for nutritional analyses from the following five plant species: Muntingia calabura (Eleocarpaceae), Chlorophora tinctoria (Moraceae), Cecropia peltata (Moraceae), Ficus ovalis (Moraceae), and Piper amalago (Piperaceae). These species represent roughly half of the wet season fruits reported to be eaten by Carollia in the park and account for about 60 percent of recorded fecal contents (Heithaus & Fleming 1978). Ripe fruits were collected daily and preserved in liquid nitrogen. Individual specimens were randomly sampled from the collected total and the following data were recorded: total fresh weight, seed number, total wet weight of seeds, and the number of insects present in the pulp. Pulp samples were prepared by removing seeds and insects from all fruit except Muntingia. Except for the skins of Muntingia, Ficus, and Cecropia, fruit parts not normally eaten by Carollia, such as the central stems of ' Received 12 November 1984, accepted 25 January 1985. 2 Current address: Department of Ecology and Behavioral Biology, University of Minnesota, Minneapolis, Minnesota 55455, U.S.A. BIOTROPICA 18(1): 39-44 1986 39 This content downloaded from 40.77.167.28 on Fri, 27 May 2016 05:46:25 UTC All use subject to http://about.jstor.org/terms TABLE 1. Physical characteristics of five fruit species. Fresh weight (g) Water content Pulp: seed Fruit species Mean SD N Percenta dry weight ratioa Muntingia calabura 1.41 0.36 120 81.8 (20) 2.33b Chlorophora tinctoria 4.76 1.01 60 79.6 (20) 6.00 (20) Cecropia peltata 4.87 1.61 22 78.1 (22) 0.87 (10) Ficus ovalis 0.82 0.20 130 78.5 (22) 6.53 (22) Piper amalago 1.25 0.69 48 73.0 (20) 0.83 (20) a Overall mean of (N) samples. b Calculated from mean no. of seeds per fruit, mean seed dry weight, and mean whole fruit dry weight. Cecropia and Piper, were also removed (based on personal observations of bats' feeding in captivity and examinations of fruits dropped under night roosts). Samples of Muntingia seeds and skins were analyzed separately and used to correct results from analyses of whole fruits. Pulp and seed samples were dried in an oven to constant weight at 5 5?C. Dried pulp samples were pooled for each species, ground through size 20 mesh in a Wiley mill and analyzed by standard techniques (Association of Official Analytical Chemists 1980). Analyses included: gross caloric content, ash, total nitrogen (Kjeldahl), and neutral detergent fiber. In addition, subsamples (1 g) of all fruit except Muntingia were hydrolyzed and run through an automatic amino acid analyzer using an internal standard of n-leucine. This technique provided quantitative determinations of 17 of the 20 common amino acids. I used the amino acid analyses to calculate the proportion of total nitrogen that is actually in the form of amino acids by summing the amounts of nitrogen present in each amino acid quantity and dividing this by the total (Kjeldahl) nitrogen content. Results of amino acid analyses were also used to assess the quality of fruit protein by comparing the amounts of essential amino acids present (except tryptophan, which was not measured) with the amounts required for a balanced diet. Because no data exist on the amino acid requirements of bats, I used the levels recommended for growth in laboratory rats (Maynard et al. 1979). I conducted digestibility studies on captive bats in order to estimate the efficiency with which bats extract energy and nitrogen from fruit pulp. Eight male bats were captured and held in captivity for 4 days. Six bats were fed fresh fruits of Chlorophora tinctoria and two were fed fruits of Muntingia calabura ad libitum. Each bat was housed in a hardware cloth cage measuring 30 x 30 x 30 cm. To minimize contamination of food by feces, a food dish was suspended on a wire shelf about 15 cm above the cage bottom. Cages were set on aluminum baking pans for the collection of all feces and urine produced each night. The first day of captivity was used as an adjustment period and no feces were collected. On the following three mornings, the feces-urine mixture was collected, weighed to the nearest 0.1 g, and stored in airtight vials in liquid nitrogen. To estimate the wet weight of fruit ingested, uneaten portions of fruit were also weighed and subtracted from the weight of fruit offered. Feces samples were dried to constant weight at 5 50C, pooled for each bat, ground through a size 20 mesh, and subsampled for nitrogen and gross caloric content determinations. The remaining material was pooled among individuals to produce a sample large enough to complete the analysis. Seeds were either removed from the feces before drying (Chlorophora) or were included in analyses and corrected for in later calculations (Muntingia). Dry matter digestibility (DMD) was calculated (on a non-ash-free basis) using the ash tracer method (Johnson & Maxell 1966). Because urine was collected with the feces, corrections for urinary loss of minerals were not necessary. An assumption of this method is that the animals are in mineral balance. This method gave a lower, more conservative estimate than dividing the difference between ingested fruit and feces weights by ingested fruit weight. The partial apparent digestibility (PAD) of a nutritional component was calculated as follows: PD=A, (1 DMD)Af PAD Al(1A, )A (1) A, where A, and Af are the relative amounts of a nutrient in ingesta and feces respectively. These estimates do not yield digestibility per se because urinary losses are included in the feces. Because urinary loss of nitrogen is related to basal metabolic rate rather than intake rate (Pike & Brown 1967), the estimates of nitrogen partial apparent digestibility were corrected as follows: PADcorrected= PAD + EUN/I(N), (2) where PAD = estimated partial apparent digestibility of nitrogen, EUN = endogenous urinary nitrogen loss (g/ Kcal basal metabolic rate), I = observed intake of dry fruit pulp, and N, = the nitrogen content of fruit pulp. These digestibility estimates were used in conjunction with estimated requirements to predict minimum food intake levels for normal and lactating bats.

Northern bobwhites (Colinus virginianus) fed 6 individual foods were able to metabolize a greater portion of the energy in common sorghum, Maximilian sunflower (Helianthus maximilianii), and a prepared mash than in ox-eye heliopsis (Heliopsis helianthoides), common soybean, and smooth sumac (Rhus glabra). Bobwhites gained weight when provided sorghum or the prepared mash but lost weight when fed sunflower, soybean, heliopsis, and sumac. J. WILDL. MANAGE. 50(2):236-238 Bobwhite populations are limited by various environmental factors, primarily food and cover (Rosene 1969). Habitat enhancement programs of most state agencies and private interests concentrate on these 2 habitat parameters in their attempt to increase quail populations. Dietary stress in northern portions of the bobwhite's winter range often results in weight loss and reduced fat reserves (Robel 1972, Robel et al. 1974), both of which reflect constrained metabolizable energy intake. Kendeigh and West (1965) have documented the gross energy content of many important foods of granivores. Robel et al. (1979a) have reported the metabolizable energy content of important foods of bobwhites. Robel et al. (1979b) conducted comparative energetic studies of bobwhites on 6 foods. This paper reports energetic responses and weight changes of bobwhites fed 5 foods (seeds) and a commercially pelletized mash prepared by the Kansas State University Department of Grain Science and Industry. The study was supported by the Kans. Agric. Exp. Stn. and the Natl. Skeet Shooting Assoc. Charities Found., Inc. This paper is Contribution 85-202-J, Div. Biol., Kans. Agric. Exp. Stn., Kansas State Univ., Manhattan. MATERIALS AND METHODS Twenty-five adult male bobwhites were confined individually in 48 x 25 x 13-cm polypropylene cages with 1.3-cm hardward cloth tops and removable floors (Case and Robel 1974). The caged birds were housed in an environmental chamber at 9L-15D photoperiod and 5 C temperature to simulate Kansas winter conditions. A balanced maintenance diet, P-18, and water were provided ad libitum prior to and between feeding trials, which were conducted during late summer 1980. This diet contained 20.5% protein, 2.7% fat, and 3.6% crude fiber; the major ingredients by weight were ground sorghum (46%) and ground corn (15%). Bird were weighed daily to 0.1 g during the last 2 hours of the dark period to ensure minimal weight variation (West 1960). After bird weights stabilized (varied

Physiologic attributes and behavioral response of white-tailed deer (Odocoileus virginianus) in a 252-ha Upper Michigan enclosure were monitored during a 5-year period as the population increased from 23 to 159 animals supported by supplemental feeding. The ration, supplied ad libitum, augmented the natural diet except in winter, when deer ate it almost exclusively. Seasonal feed consumption was lowest at spring greenup, increased in late summer as natural forage matured, and was highest in autumn and winter. As the herd increased, deer ate progressively more of the ration per animal from July to November. Availability of summer forage gradually decreased with higher herd density. Autumn forage supply varied annually, depending upon prevailing weather. Compared to prefeeding years, better nutrition accelerated deer body growth and shortened the time to physical maturity. Except for yearling bucks, antler development improved and casting dates were delayed. In utero productivity of yearling does doubled with supplemental feeding and increased by 50% among 2.5-year-olds and 21% for older does. None of the doe fawns bred. As the population grew, harassment by bucks during the rut apparently caused an energy deficiency which lowered the productivity of yearling does. Yearling does usually conceived a preponderance of male fawns. Under the best nutritional circumstances older does regularly conceived more females. Postnatal fawn mortality gradually increased and became considerable at highest density, particularly among primiparous mothers. Doe territoriality and limited suitable fawning space probably was responsible for lowered maternal success. A marked improvement in physiological parameters after the herd was drastically reduced suggested that the aberrations observed under peak populations were density dependent. We conclude that when properly conducted, supplemental feeding provides a feasible method of maintaining a reasonably large deer herd in good physical condition with minimal damage to the range. Cost factors and recommended procedures for such programs are discussed. J. WILDL. MANAGE. 46(2):281-301 The welfare of deer in northern regions depends primarily upon their winter nutritional status. If the range is inadequate, serious losses to starvation can occur during harsh weather. More importantly, however, acute malnutrition of pregnant does commonly results in heavy neonatal mortality (Verme 1977) which can disrupt herd dynamics for years. Supplemental feeding of deer to prevent catastrophic winter-kill has been tried in many states (Bartlett 1938, Gerstell 1942, Carhart 1943, Doman and Rassmussen 1944, Hesselton 1964, Keiss and Smith 1966). Unfortunately, most of these trials were ill-conceived ventures, done on an emergency basis, offering relatively poor quality foods and/or limited amounts to animals already in starving condition. Generally, severe losses were rarely averted, and in some instances overbrowsing by high deer concentrations caused irreparable range damage. In contrast, feeding of wild cervids during winter is successfully used routinely in Europe, where populations are intensively managed and their numbers closely controlled (Dasmann 1971, Nagy and Bencze 1973, Espmark 1974). Supplemental feeding of whitetails to provide hunting stock has been done in the United States (Woolf and Harder 1979). Dire consequences have occurred where the practice was stopped (Fanter 1977). Artificial feeding of deer is probably impractical under their present management in North America. However, as noted by Dasmann (1971:69), "... it is entirely possible that changing concepts of wildlife values may alter this in the future, at least on some deer ranges. When and if public demand, new and flexible J. Wildl. Manage. 46(2):1982 281 This content downloaded from 207.46.13.132 on Tue, 05 Jul 2016 05:24:50 UTC All use subject to http://about.jstor.org/terms 282 CHARACTERISTICS OF A SUPPLEMENTALLY-FED DEER HERD. Ozoga and Verme laws, and increased funds and manpower make intensive deer management possible, regular feeding may become a standard management procedure." The present study was initiated with this philosophy in mind. Our objective was to evaluate cost: benefit of supplementally feeding white-tailed deer to achieve population levels much higher than the natural habitat could support. This work presented an opportunity to assess expected improvements in the herd's physical traits under the extra ration. Additionally, it allowed us to document changes in deer social behavior as their numbers increased, and to examine the prospective role of density stress at extreme crowding as a population control mechanism. This paper reports the response by deer to a pelleted feed, the influence of increasing herd density on existing vegetation, and changes in the physiologic and reproductive characteristics of deer during a 5-year study. Data obtained before the feeding program was implemented, and that during the 2 years following herd reduction, are included to provide a better perspective on the outcome of this investigation. The senior author thanks his wife Janice, son John Jr., and J. H. Fitch for assistance in collecting behavioral information. We are also indebted to many DNR colleagues for help in different phases of the work, especially D. L. DeLisle, C. S. Bienz, and L. J. Perry. C. L. Bennett was instrumental in originating the study. This paper is a contribution from Michigan Federal Aid in Wildlife Restoration Project W-117-R, and the Cusino Wildlife Research Station. STUDY AREA AND METHODS This study was conducted in a 252-ha enclosure (Fig. 1) near Shingleton in Michigan's Upper Peninsula. The area lies in the cold, deep snow belt along the south shore of Lake Superior, where deer typically face a long winter yarding sea

Egg size variation among female mosquitoes (Aedes aegypti) is documented. Females hatching from large eggs grow faster, attain a larger adult size, take larger blood meals, lay more and larger eggs than females hatching from small eggs. This cycle is self-perpetuating; offspring of large females have a high probability of attaining a large size. Three factors oppose selection for large egg size by this apparently deterministic cycle: (1) Smaller and slower-growing females may also produce large eggs under some circumstances; (2) genetic differences among sibships are more important than differences in egg size; (3) there is no effect of body size on the reproductive success of males, so males pass on genetic material which has not been size-selected. It does not appear that there is an optimal egg size within the limits of observed egg sizes. INTRODUCTION Wilbur (1977) investigated the interrelationship between propagule size and number for two genera of long-lived iteroparous organisms adapted to uncertain reproductive success. Although these two groups (milkweed plants and salamanders) are taxonomically distant, the species considered are similar in that the total energy allotted to reproduction is constant. Variation among species within each group occurs in the way that the total energy is divided among individual offspring. In this article I look at the relationship between propagule (egg) size and number for an insect which differs from the previous examples in that the total energy invested is not constant. The female of the yellow fever mosquito Aedes aegypti L. takes at least one blood meal in order to mature each clutch of eggs. Blood meal mass varies (see Steinwascher, 1982, and references cited therein); for this reason, energy investment is not constant among females or among clutches for a single female. The number of eggs an individual female will lay in the first ovarian cycle depends on its body mass (Steinwascher, 1982), the mass and number of blood meals (Lea et al., 1978; Feinsod and Spielman, 1980; Steinwascher, 1982) and the blood source (Woke, 1937; Bennett, 1970). Furthermore, among eggs collected from a colony of Aedes aegypti, there are visible differences in size. In order to determine the interrelationships among egg size, number of eggs and the probabilities of reproductive success for this short-lived, iteroparous invertebrate, I did four sets of experiments. METHODS Variation in egg size. -A cohort of pupae, part of the seventh generation after colonization, was allowed to emerge in a screened cage (17 x 17 x 35 cm) with 10% sugar water available. The original individuals for the colony were collected from tires in Dade Co., Florida. Week-old females were removed from their emergence cage, anesthetized and weighed to the nearest .01 mg. They were held separately overnight without sugar water and each was allowed access to a mouse the next day. Engorged females were immediately anesthetized and weighed to the nearest .01 mg. Each was placed in a separate cylindrical paper cage (diam = 17 cm, height = 18 cm) with a soaked raisin (as a sugar source) and an oviposition vial containing 10 ml distilled water and lined with paper towelling. After 2 weeks, females were dissected and University of Florida, Institute of Food and Agricultural Sciences, Journal Series No. 4522.

In 1968 and 1969, single capsules containing lactose or lactose and dieldrin were given weekly to pheasants (Phasianus colchicus); hens received 0, 6, or 10 mg and cocks received 0, 4, or 6 mg. Offspring of these birds were bred in 1969 and 1970, and offspring of second generation birds were bred in 1970 and 1971. Direct mortality occurred in breeders at all levels of dieldrin administered, but weights of breeders were not affected by dieldrin. Egg production, fertility, hatchability, viability of chicks at hatching, and weights of chicks from 4 to 6 weeks of age exhibited no relationship to treatment, and were erratic among groups and years. Survival of chicks to 4, 5, or 6 weeks of age was affected (P 0.05). Offspring of parents given dieldrin chose the deep side of a visual cliff more often than control chicks (P < 0.05), as did their progeny in the second and third generations. Offspring of parents given dieldrin were more easily caught by hand (P < 0.01); this effect was also found in the second and third generation. Behavior, both on the visual cliff and in response to hand catching, was generally most affected where both parents or ancestors had been given dieldrin, and less, but still affected, where either parent or ancestor had been given dieldrin. A genetic mechanism may have been involved in effecting aberrant behavior patterns. J. WILDL. MANAGE. 38(2):320-330 Previous research on effects of dieldrin in hen pheasants was reported by Lamb et al. (1967) and Atkins and Linder (1967). Limited work on firstand second-generation hens was reported by Baxter et al. (1969). In order to elucidate possible effects of dieldrin on second and third generations, to continue work through the third generation, and to determine the role of the cock in reproductive and behavioral effects, we initiated the treatment of both cocks and hens with measured doses of dieldrin in 1968. Behavioral effects in first-generation birds have been reported (Dahlgren et al. 1970). This paper reports effects of dieldrin on reproduction, behavior, and survival through the third generation. We are grateful to W. L. Tucker, Agricultural Experiment Station Statistician, for statistical advice, and to C. W. Carlson, Animal Science Department, South Dakota State University, for use of incubator faciliti s. Use of trade names in this paper does not imply endorsement. MATERIALS AND METHODS Breeding stock was purchased from the Ringneck Ranch, Frederick, South Dakota, in the fall of 1967; other stock was purchased from the South Dakota Pheasant Co., Canton, for research in 1969. Breeding hens about 1 year of age were kept in individual cages and pool-mated to cocks kept in larger breeding cages (Atkins and Linder 1967). Birds were placed on 16 hours of artificial light in late January. Treatment was begun in February and eggs were collected beginning in early March. Technical-grade dieldrin, supplied by Shell Chemical Co., was ground, mixed with lactose, and administered in No. 5 gelatin 1 Funds for this study were supplied to the South Dakota Cooperative Wildlife Research Unit by the U.S. Bureau of Sport Fisheries and Wildlife. The Unit is supported jointly by the South Dakota Department of Game, Fish and Parks, U.S. Bureau of Sport Fisheries and Wildlife, South Dakota State University, and the Wildlife Management Institute. 2 Present address: Iowa Coop. Wildlife Res. Unit, I.S.U., Ames 50010. 320 J. Wildl. Manage. 38 (2):1974 This content downloaded from 157.55.39.153 on Mon, 19 Sep 2016 04:59:40 UTC All use subject to http://about.jstor.org/terms EFFECTS OF DIELDRIN IN PHEASANTS * Dahlgren and Linder 321 capsules via a glass tube inserted into the esophagus. Pure lactose was given to control birds. For 17 weeks, in 1968, 0, 4, or 6 mg dieldrin was administered in single weekly doses to cocks and 0, 6, or 10 mg was given to hens. Progeny of these cocks and hens were designated as 0-0, 0-6, 0-10, 4-0, 4-6, 4-10, 6-0, 6-6, and 6-10. (The first number is the weekly dieldrin level given to the cock, and the second is that given to the hen.) In 1969, newly purchased breeding stock was treated in a similar fashion, but for only 16 weeks. Also, in 1969, progeny of birds treated the previous year were either bred without further treatment or were given 6 mg dieldrin once a week for 16 weeks so that progeny were produced as follows: 0-0/0-0, 0-0/6-0, 0-0/6-6, 0-0/10-0, 6-0/0-0, 6-0/6-0, 6-0/6-6, 6-6/0-0, 6-6/6-0, and 6-6/ 6-6. (The first number to the left of the slash indicates the level given to both parents of the cocks in the previous year, and the second number the level of dieldrin currently given the cock, while the first number to the right of the slash indicates the level given to both parents of the hen the year previous except that hens in the 0-0/ 10-0 group were offspring of cocks given 6 mg and hens given 10 mg and the last number the level given to the hen the second or current year.) Second-generation work was similar in 1970 except that progeny of parents given 10 mg dieldrin did not survive to breeding age. For third-generation work in 1970, only controls and offspring of ancestors given 6 mg dieldrin were available as breeders. Part of the birds that had both parents given 6 mg dieldrin for 2 years previous were again given 6 mg dieldrin for 16 weeks so offspring produced were designated 0-0-0/0-0-0, 6-6-0/6-6-0, and 6-6-6/ 6-6-6. In 1971 no dieldrin was administered and offspring in the third generation were designated 0-0-0/0-0-0, 0-0-0/6-6-0, 6-6-0/ 0-0-0, and 6-6-0/6-6-0. Adult cocks and hens were weighed weekly to the nearest gram at the time capsules were administered. Cock breeding groups ranged from 2 to 5 individuals, so that each cock would not have to service more than 4 hens per season. Hen groups were from 4 to 10 for any category of treat

The effects of clipping to a 2.54 cm (1 inch) height at 5 stages of development of western North Dakota mixed prairie were investigated. Soil moisture content at the beginning of the growing season had a greater effect on yields than did the clipping treatments. Observations from this two-year study indicate that soil moisture removal was not affected by the clipping treatments. Clipping significantly affected peak yields by needle-and-threadgrass (Stipa comata) and the Carex species during 1977; and the miscellaneous grasses (Agropyron smithii and Agropyron subsecundum) during 1978, at the sandy loam site. No significant differences in yields were observed for the other species and groups at the sandy loam site, or, the species and groups at the loam site for the two-yearperiod. Late season livestock gains in western North Dakota may be improved if the amount of blue grama (Bouteloua gracilis), the dominant palatable warm-season shortgrass, could be increased without decreasing the overall productivity of the range. Whitman (1953) reported that normal grazing usage results in a cover which is probably of greater grazing value than the completely protected cover. Employing a deferred and rotation system, Sarvis (1923) reported greater total gains on fewer acres when compared to a continuous grazing system. Under the deferred and rotation system, the vegetation was allowed to mature periodically before being subjected to grazing again. A new system of ranching, referred to as the Short Duration Grazing or SDG system, was developed in Rhodesia in 1964 (Goodloe 1969). This system consists of a high intensity-low frequency harvesting of the forage, even with intensive harvesting range conditions improved under the semiarid climate where the SDG system was developed. The present study was designed to determine the effect of intensive harvesting on the developmental stages of Northern Great Plains mixed prairie. Clipping results in more severe injury to grass stands than does defoliation from normal grazing use (Whitman et al. 1961). Although clipping does not directly duplicate the effects of grazing, considerable information may be gained by the reactions of species to foliage removal (Sarvis 1923). The effects of clipping on the Northern Great Plains have been reported by different authors. In South Dakota, Black et al. ( 1937) reported that frequency of clipping buffalo grass (Buchloe dactyloides) and blue grama appeared to have little effect on yield. Total production of buffalo grass, blue grama, western wheatgrass (Agropyron smithii), plains bluegrass (Poa arida), and six weeks fescue (Festuca octoflora) was greatest when clipped at 40-day intervals. Total production by all plants was also greatest when clipped at 40-day intervals. Two vegetation types in western North At the time of the research, authors were graduate student and chairman, Botany Department, North Dakota State University, Fargo, 58105. Holderman's present address is Ozark Francis National Forests, P.O. Box 1008, Russellville, Arkansas 72801. This report is a contribution of North Dakota Agricultural Experiment Station. Published with approval of the Director as Journal Paper No. 1091. Manuscript received October 19, 1979. Dakota, a western wheatgrass-blue grama type, and a needle-andthreadgrass (Stipa comata)-blue grama-sedge (Carex spp.) type were studied by Whitman and Helgeson (1946). Average yields for blue grama decreased in both vegetation types when clipped twice during a season to a height of 1.27 cm (1/2 in). Needle-andthreadgrass decreased under all clipping treatments in height, yield, and basal area. Average yields for western wheatgrass were maintained under single clipping to a height of 3.81 cm (1 1/2 in) and 7.62 cm (3 in), but decreased under double clipping. Sarvis (1923) reported blue grama was benefited by frequent clipping while needle-and-threadgrass decreased in relation to clipping frequency. Heinrichs and Clark (1961) in Saskatchewan, and Whitman et al. (1961) at Dickinson, North Dakota, reported that yields of green needlegrass (Stipa viridula) increased when clipped 2 or 3 times a year compared to annual clipping. These studies applied periodic and/or annual clipping treatments. The present paper reports the response of Northern Great Plains mixed prairie to intensive clipping at five stages of development, with respect to peak yields and basal cover of selected species and groups, and the seasonal removal of soil moisture in the profile. Description of Study Locations This study was carried out on two major range sites in western North Dakota. Vegetation of the region is mixed grass prairie as described by Weaver and Clements (1938). Study site I is located approximately 9.6 km (6 miles) northwest of the city of Dickinson on a sandy loam of the Vebar-Parshall-Arnegard toposequence (Larson et al. 1968). Dominant perennial grasses are needle-andthreadgrass, green needlegrass, blue grama, prairie Junegrass (Koeleria cristata), and Sandberg bluegrass (Poa secunda). The miscellaneous grasses are dominated by western wheatgrass and bearded wheatgrass (Agropyron subsecundum). Major Carex species are C. filifolia, C. eleocharis, and C. heliophila. Study site 2 is located approximately 14.4 km (9 miles) southeast of the town of Medora on a loam soil similar to the Shambo series (U.S. Dep. Agr. Forest Serv. 1971; and personal communication with Ken Thompson, Soil Conserv. Serv., Dickinson, N.D.). Dominant perennial grasses are western wheatgrass, blue grama, prairie Junegrass, and needle-and-threadgrass. The miscellaneous grasses are dominated by green needlegrass (Stipa viridula) and Sandberg bluegrass (Poa secunda). Major Carex species are C.filifolia, C. eleocharis, and C. heliophila. Scientific and common names follow Stevens (1963). Abbreviated scientific names follow the Range Analysis Handbook U.S.F.S. (1977). The climate of the region is semiarid with an annual precipitation between 40.13 cm (15.80 inches) and 42.06 cm (16.56 in), 75% of which falls during the growing season April through September. Experimental Procedures The experimental plots were randomly located with three repli188 JOURNAL OF RANGE MANAGEMENT 34(3), May 1981 This content downloaded from 157.55.39.55 on Sun, 12 Jun 2016 05:40:36 UTC All use subject to http://about.jstor.org/terms cates of eight treatments on each of two study sites. Plot size was 10.1 m by 11.9 m (33 ft by 39 ft) with four rows of six plots. Plots were subdivided into six 3.96 m by 5.03 m (13 ft by 16.5 ft) subplots, half of which were randomly selected to be sampled for above ground biomass in alternating years. Six-foot alley ways were left between plots to facilitate movement. Five of the treatments were clipping trials, two were herbicide trials, and one the control which received no treatment. This paper reports only the results of the

C is the total energy content of the food consumed by the fish, F is the energy value of the faeces, U is the energy value of the excretory products, AB is the total change in the energy value of body materials (growth or loss in energy content) and includes any reproductive products released by the fish, R is the total energy of metabolism which is subdivided into three components: R. is the energy equivalent to that released in the course of metabolism in unfed and resting fish (standard metabolism), Ra is the energy required for swimming and other activity, Rd is the energy required for the processes of digestion, movement and deposition of food materials (including specific dynamic action). The extensive literature on energetics in fish has recently been reviewed by Beamish, Niimi & Lett (1975). Few workers have published complete energy budgets and have studied the effects of varying body size, temperature and ration size. The chief purpose of the present paper is to provide this information for brown trout and to develop equations to estimate the various components of the energy budget. Previous papers have dealt with various aspects of feeding and growth in brown trout (Elliott 1972, 1975a,b,c,d, 1976a,b) and the information in these papers is used extensively in the construction of the energy budgets. All the methods used in the experiments are described in detail in these previous papers. The methods used in the construction of the energy budgets are described in the appropriate part of the results.

Losses of domestic livestock to coyotes (Canis latrans) and dogs were investigated from claims filed at county courthouses in Iowa between 1960 and 1974. A total of 5,800 claims was examined, representing losses of 18,309 sheep, 826 cattle, 2,257 swine, and 6,839 chickens, turkeys, geese, and ducks. For all reported sheep losses, 49 percent were attributed to dogs and 36 percent to coyotes. The proportions of cattle, swine, and poultry killed by dogs also were greater than those attributed to coyotes. Since 1960, the proportion of the predator loss attributed to coyotes has increased, although there has been no change in the total magnitude of predation losses. Overall, predators killed a mean of 1.5 percent of the lambs born and a maximum of 3.5 percent of the total sheep production on the 14-county area. Sheep, swine, and poultry losses were greatest in the summer or early fall and cattle losses were greatest in the spring. The magnitude of depredations appeared to be more a function of livestock availability than of predator density. Traditional management tools that result in the taking of coyotes primarily in the fall and winter (hunting, trapping and bounties) have little effect on coyotes during the season when the greatest livestock losses are occurring. J. WILDL. MANAGE. 42(2):362-372 In recent years, wildlife managers have recognized the need for factual data on predator-livestock relationships if predators are to be managed intelligently. To date, studies concerned with livestock losses to predators have centered on the western sheep-raising industry in areas where coyote densities are high and human populations are relatively sparse. Presnall (1948) first concluded that a large proportion of livestock depredations in the West are attributable to coyotes. Balser (1974), however, pointed out that total sheep losses in the West do not differ much from those in the eastern United States where, he speculated, losses to dogs and heavy parasite loads may offset coyote losses. The relationship between livestock predation by native versus domestic or feral predators is poorly understood in areas where both types occur. Wildlife management agencies commonly are presented with conflicting, and often extreme, viewpoints concerning the relative severity and magnitude of depredations by domestic versus native predators. The major sources of data on livestock depredations in the western United States have been reviewed by Leopold (1964), Wagner (1972), Cain et al. (1972), Balser (1974), Nesse et al. (1976), and Gee et al. (1977). These include the records of government agencies (most of these data were collected incidental to other activities), data obtained from mail questionnaires and personal interviews, and data collected in field assessment studies designed to locate and necropsy dead sheep. All these studies have been concerned primarily with coyote predation on sheep and have been conducted almost exclusively in areas where some predator control was in effect. Denny (1974) pointed out that few data, or even rational estimates, have been available on the scope and magnitude of property losses attributed to dogs. This is despite the fact that domestic dogs number about 34,100,000 in the United States 'Journal Paper No. J-8718 of the Iowa Agriculture and Home Economics Experiment Station, Ames. Project No. 2031. 2 Present Address: Kansas State University, Southwest Area Extension Office, 1501 Fulton Terrace, Garden City 67846. 362 J. Wildl. Manage. 42(2):1978 This content downloaded from 157.55.39.183 on Thu, 26 May 2016 05:20:55 UTC All use subject to http://about.jstor.org/terms LIVESTOCK LOSSES TO COYOTES AND DOGS" Boggess et al. 363 (American Humane Association 1972, cited by Denny 1974). Feral and free-ranging dogs, as well as coyote-dog hybrids, also exist in many areas of the United States (Young and Jackson 1951, Gipson 1972, Scott and Causey 1973, and Nesbitt 1975). Denny (1974) also conducted a nationwide survey of damage caused by dogs and found that, from an economic standpoint, state agricultural and wildlife agencies ranked livestock damages as the most important problem posed by dogs. This paper represents the results of an investigation into the magnitude and timing of depredations by coyotes and dogs on domestic livestock. The study was conducted in an area where no organized predator control program was in effect. Domestic animal loss claims filed by the producer at county courthouses in Iowa provided a unique opportunity to compile data on livestock losses incurred over a period of several years in a wide area. From these data we were able to determine not only the relative magnitude of reported losses, but also the yearly and monthly loss trends, factors of great potential significance to any applied predator management program. This paper is a portion of a M.S. study (Boggess 1975) financed by the Iowa Conservation Commission, Federal Aid Project W-115-R-2 and by the Iowa Agriculture and Home Economics Experiment Station. M. K. Petersen, J. F. Heer, L. D. Wing, and A. Farris gave valuable criticism of this manuscript. K. Augustine aided in collection of the data and D. F. Cox and T. Bubolz provided much help in statistical analysis of the data.

William E. Smith (Fairleigh-Dickinson Univ., Madison, N.J. 07940): Findings of fibers in water supplies bring up the question of whether they present any hazard to people drinking the water. To develop experimental information on this question, we maintained 600 hamsters from the age of about two months throughout their lives on drinking water with and without addition of some mineral fibers. We found some tumors that may be related to treatment in hamsters that drank water containing fibers of naturally crystallized amosite asbestos from South Africa. No tumors attributed to treatment occurred in hamsters that drank water containing tailings from milling of taconite ore rich in cummingtonite/grunerite mineralogically related to amosite. Since carcinogenicity of mineral fibers has been related to their dimensions by results of intrapleural injections in experimental animals, it is of interest to look at dimensions of fibers in our drinking water exposures. In considering dimensions of fibers, I was glad to see that Dr. Millette's data did not show merely mean dimensions. Mean dimensions were not impressively different in our sample of amosite as compared to our samples of tailings as measured by electron microscopy at 2500x. However, differences in fiber length distributions were obvious, as better seen in measurements at 600x, which show that the percent of fibers longer than 10 gum was 14% in the amosite as compared to 4% in the tailings. Among fibers measured as longer than 10 ,m, 25% were longer than 20 Atm in the amosite; none were longer than 20 ,um in the tailings. Following Dr. Millette's paper, a question was asked as to how mineral fibers were suspended in water to assure ingestion by animals. I responded as follows: To maintain suspension of insoluble mineral fibers in water, and thus assure their ingestion by animals drinking the water, we designed drinking fountains in which the water was put in funnels and agitated with a stream of air. We described this method in a paper last year [Am. Ind. Hyg. Assoc. J. 39: 583 (1978)].

Late spring grazing by sheep altered the amount of several forage categories available to deer the subsequent autumn and winter. Total herbaceous plant material was much reduced by spring-time sheep grazing, but regrowth following fall precipitation increased the proportion of green herbaceous material available. Current year's growth of bitterbrush was also increased relative to the nongrazed situation due to the release of moisture and nutrients accompanying the removal of herbaceous plants by sheep. Subsequently winter diets of mule deer on the sheep-grazed area were higher in herbaceous components but lower in shrub components than on the adjacent area where sheep had not been previously grazed. Implications of these findings are that quality of deer diets was not detrimentally affected where sheep had grazed during the preceding spring and a much greater animal yield is possible through dual use. A deficit of winter forage apparently limits mule deer (Odocoileus hemionus hemionus) population over much of their range (Aldous 1945; Doman and Rasmussen 1944). This can be viewed in terms of both extent of winter rangeland and quantity of forage (principally shrubs) produced there. The Utah Division of Wildlife Resources estimates that there are approximately 7,424,000 ha of mule deer winter range in Utah, including some 1,149,000 ha dominated by the sagebrush complex, primarily big sagebrush (Artemisia tridentata). However, big sagebrush is viewed as only moderate quality winter forage for deer because of its low acceptability (Smith and Hubbard 1954). This is a particular problem where sagebrush exists in stands devoid of more palatable shrub species. Winter deer losses in Utah appear to be inversely related to the amount of palatable browse species available (Robinette et al. 1952). The grazing of deer winter ranges by domestic livestock is common throughout the Intermountain West. Such ranges are grazed in spring when forage is typically in short supply for the livestock industry. Hence, the generally low state of productivity of these ranges is viewed as a limitation to livestock production (Cook and Harris 1968) as well as to deer production. Recent research indicates that with properly designed grazing strategies, livestock-big game competition can probably be minimized (Jensen et al. 1972; Jensen et al. 1976). Moreover, these same studies suggest that livestock may be used to manipulate vegetation on deer winter ranges to effectively Authors were research assistant, associate professor, and research assistant, respectively., Department of Range Science, Utah State University, Logan 84322. Smith and Fulgham are now, respectively, assistant professor of animal science, Angelo State University, San Angelo, Texas 76901, and assistant professor of animal and range sciences, New Mexico State University, Las Cruces 88003. This report is Journal Paper No. 2249 of the Utah Agricultural Experiment Station, the supporting institution for this research. Authors also acknowledge the Utah Division of Wildlife Resources (UDWR) for providing the study site, the experimental animals, and logistic support. Charles H. Jensen, game biologist, UDWR, assisted in site selection and study design. -Dr. Charles Romesburg and Mr. Kim Marshall, US/IBP Desert Biome Project, were both especially helpful in the data analysis phase of the project. Manuscript received November 14, 1977. increase quantities of browse available to wintering deer. However, the specific responses of mule deer to such grazing systems have not been well established. Thus, the study reported in this paper was designed to determine: (1) the plant species present and available to wintering mule deer following spring-time sheep grazing and (2) the relative proportions of the various plant species in the diets of mule deer during the winter following the spring sheep grazing treatment.

Various authors have studied hatching of cestode oncospheres infecting animals and man (Heath and Smyth, 1970, Parasitology 61: 329343). Silverman (1954, Ann. Trop. Med. Parasitol. 48: 207-215) reported up to 74% activation of oncospheres of Taenia saginata. Jones et al. (1960, J. Parasit. 46: 170-174) described the hatching of T. hydatigena; Berntzen and Voge (1962, J. Parasit. 48: 110-119) reported on the hatching in vitro of Hymenolepis citelli and H. nana. Berntzen and Voge (1965, J. Parasit. 51: 235-242) also described the requirements for hatching of oncospheres of four hymenolepid cestodes and noted that each of the four species of the genus Hymenolepis had slightly different requirements for hatching in vitro. Smyth (1963, J. Parasit. 51: 235-242) proposed the uniformity of requirements to be furnished, not only for the rupture of the membranes surrounding the eggs, but also for the activation of the oncospheres. Laws (1968, Exp. Parasitol. 23: 1-10) reported on the hatching in vitro of Taenia hydatigena, T. pisiformis, and Echinococcus granulosus. Hatching is here defined as the liberation of the oncosphere from all enclosing coats and membranes, with the activation of the embryo. There is no evidence from published papers that by our definition, eggs of T. solium have been hatched successfully in vitro. We report here the hatching in vitro of oncospheres of Taenia solium. Ten terminal gravid proglottids were obtained from a patient treated with niclosamide. The segments were passed by the patient 8 hr after treatment. They were washed in 0.85% NaCI and the uterine branches were counted for appropriate identification (Brown and Neva, 1983. Appleton-Century-Crofts, pp. 186187). The segments were placed in Petri dishes lined with filter paper moistened in saline, and were covered with moist cotton and filter paper. The dishes were sealed with Parafilm to prevent drying and kept for 1 wk at room temperature until experiments began. The following procedures were used as precautionary measures. Containers sufficiently large to accommodate Petri dishes, Pasteur pipettes, slides and other small glassware were filled with 95% ethyl alcohol. All material possibly contaminated with eggs was discarded into the container, taking care that the items were fully submerged in the liquid. The work area was swabbed periodically with 70% alcohol, and, at the end of the experiment, hands were rinsed in alcohol before washing with soap and water. Observations on eggs subjected to 50% ethyl alcohol (final concentration) or higher showed that, on a slide, embryophores fell apart within 5 min of addition of the alcohol and within 1 min when preparation was covered with a coverslip. The embryos appeared black and shrivelled. However, if the final alcohol concentration was 35%, the eggs remained intact during a 15min observation period under the microscope. Eggs were dissected from proglottids using a dissecting microscope and were transferred into Solution 1 consisting of: (a) 120 mg pepsin, 1: 10,000 (DIFCO Laboratories, Detroit, Michigan); (b) 2 ml HC1, 1 N; and (c) 20 ml distilled H20. The pH of this solution was 2.0. Ten ml of his solution was added to eggs in a Petri dish, and placed in an incubator at 41 C for 15 min. The eggs were then washed three times in 0.85% NaCl to prevent transfer of ingredients from the first solution into the second. Eggs were then pipetted into a 10-mm-diameter screw cap tube, containing Solution 2, consisting of: (a) 10 ml 0.85% NaCI; (b) 35 mg trypsin, crystalline (Type XII, Sigma Chemical Company, St. Louis, Missouri); (c) 0.5 ml whole bile from pig (kept frozen until use); and (d) 1 ml NaHCO3, 10% solution in distilled water. The pH of this solution was 7.2 to 7.4. The tube was then tightly capped and placed in an incubator for 15 min at 41 C, removed to room temperature (21-22 C) for observation and kept for another 15 min.

Discriminant analysis was used to classify animals into ration groups based on blood composition. Blood samples were drawn from groups of 3 bison (Bison bison) steers and 4 Hereford cattle steers receiving 4 rations each in summer and winter (N = 56). Four discriminant analyses, 1 for each combination of species and season, were used to classify blood samples into respective ration groups. Rates of correct classification, evaluated using a jackknife procedure, ranged from eleven of 12 winter bison samples to six of 16 summer cattle samples using blood urea nitrogen (BUN), alkaline phosphatase (ALP), and cholesterol (CHOL) as discriminating variables. Classification of summer cattle samples was improved to eleven of 16 by using BUN, total serum protein (TSP), 3-globulin (fl-GL), and hemoglobin (HGB) as variables. The overall correct classification rate was 79% (44 of 56 samples). J. WILDL. MANAGE. 51(4):893-900 Although it has been demonstrated frequently with wild and domestic animals that the levels of many blood components vary significantly with ration composition (Preston et al. 1965; Seal et al. 1972b; LeResche et al. 1974; Torell et al. 1974; deCalesta et al. 1975; Kirkpatrick et al. 1975; Manston et al. 1975; Parker and Blowey 1976; Seal et al. 1978b; Bahnak et al. 1979, 1981; Warren et al. 1981, 1982; Card et al. 1985), attempts to evaluate nutritional status from blood composition have had uncertain success (Adams et al. 1978, Franzmann and LeResche 1978, Lee et al. 1978, Rowlands 1980, Warren et al. 1981, Kopf et al. 1984). Most of these studies have involved primarily univariate analyses of blood data. A multivariate approach might be more successful, and blood data lend themselves to multivariate analyses because many components are routinely measured in a single blood sample. This paper is a statement of approach, involving the use of multivariate methods to analyze and interpret blood data. I used discriminant analysis as an example to illustrate the classification of samples into ration groups using blood data from bison and cattle. Discriminant analysis has been used widely to assign individuals to groups in taxonomy (Thorpe 1983, Gerasimov 1985), ecology (Stiteler 1979, Williams 1983), medicine (Lachenbruch and Clarke 1980, Brown 1984), and wildlife management (Cowardin and Johnson 1973, Hanson and Jones 1976, Crawford et al. 1981, Hale et al. 1982). I thank D. G. Peden for his support and for suggesting discriminant analysis to classify these blood samples, and gratefully acknowledge the statistical reviews of this manuscript by B. Collins and A. Sharma. This study was supported by Can. Wildl. Serv., Contract OSZ5-0285, and by the Alberta Environ. Cent. MATERIALS AND METHODS Sampling, handling, and feeding methods were described by Hawley and Peden (1982). Briefly, rations were formulated to contain high protein-high energy, high protein-low energy, low protein-high energy, or low protein-low energy. Groups of 3 yearling bison steers and 4 yearling Hereford steers, housed at the University of Saskatchewan, Saskatoon, were fed these rations ad libitum in late summer (24 Aug-4 Oct 1976) and late winter (26 Feb-9 Apr 1977). The Ist of a series of blood samples drawn from each animal in each ration group was selected for this study. Thus, there were 3 bison and 4 cattle samples for each of 4 rations in 2 seasons (N = 56). All samples were analyzed as described by Hawley and Peden (1982) to determine the levels of 20 blood components. These data were subjected to discriminant analysis in the present paper. Discriminant analysis is a multivariate technique for developing linear combinations of predictor variables in order to maximize differences among groups. In our data a single blood sample represented the analytical unit, or case, for which 20 dependent variables (the blood components) were measured. The different rations formed 4 groups, within which the cases were clustered. Discriminant analysis maximized the distinction among these groups. Classification functions were developed, 1 for each group, to determine in which group a case belonged. The analysis calculated the probability

The daily mean energy expenditure of free-ranging heifers grazing crested wheatgrass rangeland was estimated as 161 kcal/kgBW,75/day using the carbon dioxide entry rate technique (CERT). This was 46% greater than the mean 110 kcal/kgBW 75/day estimated for stall-fed heifers consuming similar forage. Under laboratory conditions, CERT can provide a mean estimate of energy expenditure that has a standard error of 5% and a coefficient of variation of 20%. In contrast to the stall-fed animal, the range animal requires additional dietary energy to meet the demands of its free-roaming condition. For sheep on pasture, Young and Corbett (1972) reported a 60-70% increase in daily energy expenditure above the maintenance costs of penned sheep. Young (1970) and Wallace (1956) have reported 60% and 50% increases, respectively, for grazing cattle. These substantial increases in energy expenditure have several possible origins. Osuji (1974) reviewed the literature on the energy expenditure of the pastured ruminant and suggested that the increased energy costs might be due in part to an increased amount of time spent obtaining food and the locomotion associated with grazing. For example, 40% of the energy expended by grazing sheep has been attributed to their activities of standing, walking, ruminating and eating as compared to 10% for a penned animal consuming roughages (Graham 1964). Each hour spent grazing by cattle represents a 2% increase in daily energy expenditure (Holmes et al. 1978). Cattle on crested wheatgrass rangeland typically spend 7.7 to 10 hours daily grazing (Scarnecchia 1980). Additionally, the daily time spent grazing increases as the quantity of available forage declines (Nastis 1979), although cattle seldom graze longer than 13 hours a day, irrespective of forage conditions or availability (Arnold and Dudzinski 1978). The energetic cost of travel on level terrain has been estimated as 0.45 to 0.78 kcal/kgBW/km (Blaxter 1962, Graham 1964, Osuji 1974). Although travel involves a negligible expense for the stallfed animal, various researchers have reported traveling distances of 5 to 12 km a day for cattle on western U.S. ranges (Dwyer 1961, Herbel and Nelson 1966, Malechek and Smith 1976). Thus, travel could easily add 10% to the daily energy expenditure of the grazing animal. Additional unknown stress-related factors also may contribute to increases in energy expenditure when animals are forced to search for food under conditions of scarcity. The extent to which Authors are assistant professor, Department of Animal and Range Science, Montana State University, Bozeman, and professor, Department of Range Science, Utah State University, Logan. At the time of this research. Havstad was research assistant, Range Science Department, Utah State University. This paper was approved by the Director, Utah Agr. Exp. Sta. as Journal Paper No. 2727. Manuscript received in 1980. free-roaming conditions increase the energy expended by the grazing ruminant warrants a close examination of factors affecting energy expenditure. This knowledge would be important in planning grazing management systems, particularly those that have as their objective uniform grazing, which may tend to stress animals. Thus, the purpose of this study was to measure the energy expenditure of grazing heifers at several different levels of available forage, and to compare these quantities to the energy expended by similar stall-fed animals consuming a diet of similar quality and quantity. Precise estimates of energy expended by the free-roaming animal are limited byavailabletechnology. Techniques for indirect estimation of energy expenditure include time budget analysis and food intake balance and energy balance models. These techniques are either still being tested or may be limited by numerous assumptions involved in their application. For direct estimation of energy expenditure, accepted and applicable methods of indirect calorimetry include respiratory gas exchange (Young et al. 1975), heart rate and respiratory rate estimation (Yamamoto et al. 1979), and the carbon dioxide entry rate technique (CERT) (Whitelaw et al. 1972). Typically, gas exchange methods require fistulation of the trachea, and the associated equipment required for gas measurement is cumbersome. Young and Corbett (1972) reported that this technique estimated energy expenditure with a standard mean error of 12%. The heart rate measurement technique has been used with increasing frequency during recent years. Each animal's heart rate must be calibrated with their oxygen consumption, and this relationship varies with cardiac stroke volume and responses to emotional factors (Brockway 1978). Yamamoto et al. (1979) reported a mean standard error of 16% when heart rate was used to estimate energy expenditure of cattle. CERT has been suggested as the most suitable measurement for estimating energy expenditure of the grazing ruminant (Webster 1972). However, mean standard errors of 7 to 20% for energy expenditure estimates have been reported by users of CERT (Young et al. 1969, Whitelaw et al. 1972). Methodology is apparently best chosen on the basis of availability and suitability of a particular technique and its associated equipment to conditions at hand rather than upon attainment of a prescribed limit of statistical reliability. Due to the availability of the necessary equipment, CERT technology was selected for the experiment reported here.

The many factors affecting the growth and feeding of fishes have been reviewed by Brown (1957), Ivlev (1961), Paloheimo & Dickie (1965, 1966a, b) and Phillips (1969). These reviews include the relevant work on salmonids and the only important addition is the work on sockeye salmon (Oncorhynchus nerka Walbaum) by Brett, Shelbourn & Shoop (1969) and Brett (1971). The present paper is one of a series dealing with feeding and growth in brown trout (Salmo trutta L.). From the results of previous work, it is now possible to estimate the rate and time for the gastric evacuation of different meals (Elliott 1972), the maximum weight of food consumed in a meal (satiation amount) and the time taken to eat this meal (Elliott 1975a), the maximum weight of food consumed in a day and the maximum rate of feeding (Elliott 1975b). These estimates are applicable to trout of different weights (approximate range 10-350 g live weight) at different water temperatures (range 3-821-6? C). The next stage was to examine the growth rates of trout fed on maximum rations. These experiments are described in the present paper and a mathematical model is developed to estimate maximum growth rates of different sized trout at different temperatures.

Information on the exact time of the daily resting and feeding periods of incubating ring-necked pheasant (Phasianus colchicus) hens and the daily time of egg laying is rather meager. Leopold (1933) gave dawn and 4:00 p.m. as the times at which incubating pheasant hens leave the nest. Baskett (1947) found no definite time at which he could predict with accuracy that any individual hen would be away from the nest. He states that eggs were generally laid in the interval from about two hours after sunrise to two hours before sunset, or between the two daily feeding periods. Difficulty in obtaining accurate data is largely responsible for this lack of information. Direct observation of the nest can be carried on only during daylight hours and may cause desertion of the nest or abnormal behavior by the hen. The need for a method of automatically recording the daily movements of nesting pheasant hens was evident. A number of methods have been devised for recording nesting activities of song birds. Baldwin and Kendeigh (1927) described the use of a thermocouple and recording potentiometer to record nesting activities. The application of heat to the eggs registered the presence of the bird, and a drop in temperature depicted its departure. The thermograph could be placed in the nest with little difficulty, but the recording potentiometer required an electric current, a limiting factor in its use. Kendeigh and Baldwin (1930) detailed the use of an itograph, an electromagnetic device particularly adapted for use at nests with but one approach. Another electromagnetic instrument, the terragraph, was invented by a German, Count von Hegendorf, for studying the habits of foxes and badgers, and was applied to bird activities by Bussman (1933). The apparatus consisted of a paper dial-plate marked at half-hour intervals on a revolving disk, with a carbon paper on top. A pencil attached to a telegraph magnet pressed on the carbon paper when the circuit was made, making a short line on the paper dialplate below. Two small pieces of wood fastened at one end by a flat steel spring comprised the nest contact. The contact was placed in circuit by means of an insulated copper wire with two pocket batteries and the telegraph magnet. Kuusisto (1941) used a glass tube containing mercury and with one end expanded like a thermometer as a thermo-contact.

Early mention of bats in Iowa is found in publications by Allen (1869), Goding (1883), Osborn (1890, 1892), Van Hyning and Pellet (1910), Cory (1912), and Ruthven and Wood (1912). The occurrence in the state of some species listed in these early accounts evidently was not substantiated by specimens. Scott (1937) in his "Mammals of Iowa" reviewed the status of chiropterans in the state. In this paper, six species (Myotis keenii, Myotis lucifugus, Eptesicus u.rcus, Lasiurus horealis, Lasiurus cinereus, and Tadarida macrotis) were authenticated by actual specimens. Scott (1939) later added Pipistrellus subfiavus to the list of Iowa bats, and also identified as Myotis subulatus the one specimen assigned to M. keenii in 1937. This bat was incorrectly identified by Scott and is discussed in the account of Myotis lucifugus. In the past decade, Muir and Polder (1960) recorded Myotis sodalis from Iowa, Kunz (1966) reported Nycticeius humeralis, and Easterla and Watkins (1967) mentioned the first actual specimens of Lasionycteris noctivagans from the state. The first authentic record of Myotis keenii septentrionalis is reported in this paper. Thus, at this writing, a total of 10 species of bats is known certainly to occur in Iowa. In the past few years there has been increased interest in the study of bats in Iowa. We began investigations independently in the state in 1965 and combined our efforts in 1966. One of us (Kunz) is continuing research on the natural history of several species. In the course of recent work, many bats have been banded and released in an attempt to learn more of movements and longevity. Additionally, workers in adjacent states have banded thousands of bats, some of which already have been recaptured in Iowa. Persons finding banded bats are requested to read the band number carefully, and, if possible, release the bat unharmed. Band numbers should be reported to the Bird and Mammal Laboratories, U. S. Fish and Wildlife Service, U. S. National Museum, Washington, D. C. 20560, along with the exact date and place of capture. This paper represents a progress report of our studies and includes specimens obtained by us, those deposited in other institutions, records of

In an earlier paper (1944) we corroborated and extended the work of one of us (Chandler, 1943) showing that Hymenolepis diminuta grows in a normal manner in the absence of any vitamins in the diet of the host except something associated with the G complex, absence of which causes a marked stunting in growth in female but not in male rats. This stunting becomes intensified as the depletion period in the host prior to infection is increased. In the present paper are reported a series of experiments designed to determine, if possible, what factor or factors in the G complex is responsible for the stunting effect. The hosts used were female white rats three to four months old, weighing between 150 and 200 gm at the beginning of the experiments. The rats were infected by oral administration of artificially reared cysticercoids as described in our earlier paper. The complete diet for control animals was the same as described in that paper. The basic vitamin-deficient diet was also the same except that the fat-soluble vitamins were added in the manner previously described, for the benefit of the host animals. The G complex, when used, was provided in the form of 9 gm of autoclaved brewer's yeast per rat per day. Other vitamins or nutritional factors, added in synthetic or concentrated form, were dissolved in 20% alcohol, and the proper amount added to the basic diet each day. Riboflavin (B2), pyridoxine hydrochloride (B6), calcium pantothenate, nicotinic acid, p-aminobenzoic acid, inositol, and choline were used in pure synthetic form and provided at the rate of 100 micrograms per day per rat. Biotin was used in the form of a concentrate containing 20 micrograms per cc, and provided at the rate of 30 micrograms per day per rat. Folic acid was provided in the form of a concentrate at the rate of 90 micrograms per day per rat.

Field parties from the University Museum of Vertebrate Paleontology worked in Meade County, Kansas in the summers of 1940 and 1941. During July and August of these two years there was so much rainfall at times that it was impossible to work on fossils. Consequently, a considerable amount of time was spent collecting and studying the recent mammals of that area. Part of the data thus obtained is presented in this paper. A general discussion of the mammals of the Meade County State Park was presented by Joe A. Tihen and James M. Sprague in "Amphibians, Reptiles and Mammals of the Meade County State Park," Trans. Kans. Acad. Sci., 42:499-511, 1939. Claude W. Hibbard and George C. Rinker published a more specific paper on "A New Bog-lemming (Synaptomys) from Meade County, Kansas," Univ. Kans. Sci. Bull., Vol. 28, Pt. 1, pp. 25-35. 1942. References have been drawn from both of these papers in the following discussion. Cryptotis parva (Say). One female of this species with three embryos was taken on July 22, 1941, and a second, with five embryos, on July 25, 1941. In each case, the embryos measured approximately 10 mm. in length. Citellus spilosoma major (Merriam). In all, five adult females of C. s. major were taken, three on August 3, 1941, one on August 4, 1941, and one on August 9, 1941. Of the three taken on August 3, two were carry.ng embryos. One of them had 8 embryos and the other 6. The female taken August 9 was captured after being ploughed up in a wheat field. With her were 4 young, two of which were captured and proved to be about two-thirds grown. Evidently two litters of young are born during the summer. Onychomys leucogaster arcticeps (Rhoads). Only one female with embryos was taken. This specimen, taken August 9, 1941, had five embryos. A number of immature and sub-adults were taken at the same time as the female mentioned above, which indicates that that litter represented at least the second and possibly the third litter of the season. This is in accordance with the conclusion of Tihen and Sprague (op. cit.) who wrote of this species, "The grasshopper mouse in Kansas has at least two litters a season, for females bearing full time fetuses were taken June 18, and 30, 1938, and July 20, 1938, while a female bearing fetuses approximately one-fifth grown was taken July 16, 1938. Four to five young is the common number." Reithrodontomys megalotis (Baird). One female of this species, with embryos, was taken July 19, 1941. The number of embryos in this specimen was four. They were five mm. in length. The total number of females taken was five.

The relative infestation of small mammals by fleas was studied in northwestern Illinois. Eleven host species numbering 718 individuals were collected from February to October in 1958. Three hundred eighty-six fleas of 11 species were taken from these hosts. More fleas per host were collected in the cooler months than in the warmer months. Orchopeas leucopus, a flea often considered to infest Peromyscus without exhibiting an apparent preference as to species, may actually be more common on species which occupy woodland habitats. Mean numbers of fleas of each of three species on Peromyscus leucopus, a host species taken in all seasons, indicated that peak populations of the three species do not coincide. A relatively large number of shrews and associated small mammals were trapped during a survey for rabies in shrews between February 14 and October 3, 1958. Fleas were removed from these specimens and submitted for identification to the Section of Faunistic Surveys and Insect Identification, Illinois Natural History Survey. Because relatively little is known about fleas in this section of the Midwest these records are believed to represent a contribution. Acknowledgment.-These observations were supplementary to an investigation of sylvan rabies which was supported by a grant (No. E-1349) from the Division of Research Grants, U. S. Public Health Service, Department of Health, Education, and Welfare. Thanks are due Dr. C. 0. Mohr for suggesting the study and for a critical reading of the manuscript and to Dr. L. J. Stannard for identifying the fleas. Also, Drs. T. G. Scott and R. E. Yeatter, and Mr. G. C. Sanderson gave help and advice which is gratefully acknowledged. METHODS Ninety-five trap lines, each consisting of 20 stations 10 yards apart with two break-back traps per station, were set along secondary road rights-of-way in Jo Daviess, Carroll, and Whiteside counties. The traps, baited with liver sausage to attract shrews, were allowed to remain in place for two consecutive nights. The traps were examined each morning, captured animals removed, and the bait replaced when necessary. Specimens obtained were placed in individual No. 1 paper bags. In the laboratory, the contents of each bag were dumped into the center of a piece of brown wrapping paper. The sack was torn apart, and fleas clinging to the bag or to debris were collected with forceps and placed in a vial of 70 per cent alcohol. The mammal carcass was then brushed and combed thoroughly with the points of the forceps, and additional fleas clinging to the host were collected. Mammals which were badly deteriorated or partially eaten by other animals were not examined for fleas.

Analysis of retrapped Canada geese (Branta canadensis maxima) at Seney National \Vildlife Refuge showed that only 30 percent of the birds retained flexible plastic collars after 6 years. Retention was higher among females than among males. Fading, brittleness, and loss of the number-letter combination of the collars are discussed. J. WILDL. MANAGE. 37(2):1 7S178 During a study of limiting factors and behavior of Canada geese at the Seney National Wildlife Refuge between 1963 and 1965, 518 geese were marked with flexible plastic collars (Sherwood 1966a). Although the study officially ended in 1965, many of the geese survived long past that date and some of the collars were retained for several more years. The purpose of this paper is to report on the long-term retention rates of these collars and to describe the condition of the collars after up to 6 years of use. The author thanks D. H. Johnson, of the Northern Prairie Wildlife Research Center, for his assistance in the statistical analysis. MATERIALS AND METHODS Sherwood (1966b), who conducted the behavior study, designed the collars; their construction has been described in detail in an earlier paper. Each collar was 2.S inches tall and had a number-letter combination routed into the plastic and painted with black lacquer. Orange collars were used in 1963, white in 1964, and yellow in 1965. Ends of the collars were not bonded with adhesive, so a birds could pull the collar off if it became caught. Determination of retention rates was nlade through the analysis of retrap data 1 Present address: Lacreek National Wildlife Refuge, Martin, South Dakota 57551. for the years 1966-69. When a collared or otherwise color-marked goose was retrapped, the marking was recorded on a field sheet. A check of band numbers of all retrapped geese was made to determine those birds that had originally been lllarked with collars. Retention rates were determined by comparing collars still in place with collars lost for each age-group of collars. Only birds trapped were used in determining the retention rates. This has resulted in a smaller sample, but has eliminated the bias of field observations tovards marked birds. RESULTS AND DISCUSSION During the 4-year period, a total of 194 retraps was made, 105 with the collars in place and 89 with the collars missing. A breakdown of the recoveries by the number of years since collar placement and by sex is given in Table 1. The smaller number of retraps of the 1and 2-year-old collars is because orange collars were already 3 years old and white collars 2 years old during the first year of data collection in 1966. The same is true in the older groups where only orange collars were present in ffie 6-year-old group and orange and white collars in the 5-yearold group. The data in Table 1 were analyzed to determine whether the sex of the bird and nurnber of years after placement had an J. Wildl. Manage. 37(2):1973 176 LONG-TERM RETENTION OF PLASTIC OLLARS ON CANADA GEESE Table 1. Collar retention rates of Canada geese at Seney National Wildlife Refuge. YEARS NUMBER OF RETRAPS AFTER RETENTION COLLAR CO11aIS CO11arS RATE PL XCEMENT Sex On Off ( percent ) M 11 5 69 1 F 15 3 83 Total 26 8 76 M 12 6 67 2 F 19 8 70 Total 31 14 69 M 15 13 54 3 F 14 19 42 Total 29 32 48 M 4 9 31 4 F 7 9 44 Total 11 18 38 M 2 5 29 S F 3 5 38 Total 5 10 33 M 0 3 0 6 F 3 4 43 Total 3 7 30 PLASTIC OLLAR ETENTION ON GEESE * Fjetland 177 both their collars and their service bands would have also been overlooked in this study, but this number of birds is small and would not greatly affect the retention rates. The significantly higher retention of collars by females was supported by field observations of marked birds. Of the 518 collars placed, 262 were on males and 256 (49 percerlt) on females. Observations of 24 collared geese in 1969 revealed that 20 were females and only 4 were males. Similar observations in 1968 revealed that 21 out of 28 collars seen were on females. Part of this unbalance in field observations can be attributed to a higher mortality rate for males than females in the Seney flock. Sherwood ( 1966a ) found that of yearlings and older geese in the Seney flock 43 percent s^7ere males and S7 percent females. He offered evidence that males were being shot more heavily than fenlales as one possible explanation for the unbalanced sex ratio. Dispersal of males to other flocks would also account for some of the difference in field observations. An unbalanced mortality rate and dispersal do not explain the different retention rates by sex in Table 1, however, where the rates are independent of the sex ratio of the flock, of field observations, of mortality rates, and of the nulnber of collars originally placed. Sherwood (1966b) reported two principal causes of collar loss. The first, collars slipping over the smaller heads of goslings, was solved by marking the birds after they were 2 months old. The second, collars pulling off on wire duck traps, 11as largely disappeared. Wire duck traps are still used at Seney Refuge, but only one collar svas found on a trap in 1968 and none in 1969. Those birds that tend to feed under the traps have in large part already lost their

Mandibles (plus teeth) of 68 white-tailed deer (Odocoileus virginianus) of various age-classes were collected in December, 1959, at Aberdeen Proving Ground, Maryland, to study strontium-90 distribution within a deer herd. No statistically significant difference was observed in the mean strontium90 accumulated by males and females, but there was a significant difference between age-classes. Further, there was no significant age-class-sex interaction, implying that the difference between age-classes was independent of sex. The amount of strontium-90 in the mandible was correlated with year of birth and the annual strontium-90 increment and total strontium-90 in the soil. There were no statistically significant differences between the variances of the age-classes of either sex. Considerable variation in the strontium-90 content of mandibles within age-classes was observed. Since the advent of the atomic age there has been concern about the accumulation of radionuclides by wildlife populations and the effect of such accumulation on wildlife and on persons consuming harvested game animals. Atmospheric nuclear testing and underground testing, where venting occurs, results in introduction of radionuclides into the biosphere. As a result of short physical half-lifes, many of these nuclides are transitory. Historically, major interest has been directed toward carbon-14, cesium-137, iodine-131, and strontium-90. With the exception of iodine-131, these radionuclides have relatively long physical half-lifes. Strontium-90 has a physical half-life of 27 years and a biological half-life in mammals of about 7.5 years (Hawthorn 1959: 1294). It emits only beta particles, has chemical properties similar to calcium, and is deposited in bone where it has a long residence time. This paper is a report on strontium-90 concentrations observed in deer mandibles (plus teeth) collected from a herd of whitetailed deer at Aberdeen Proving Ground, Maryland, December, 1959. The paper deals specifically with the concentration of strontium-90 in mandibles by sexes and ageclasses of this herd and with the variability in strontium-90 deposition observed. T Aberdeen Proving Ground herd is unique in Maryland. According to Flyger (1958), the herd is apparently at the limit the habitat can sustain. He has observed that it is not unusual to count 100-200 deer in one morning and that a distinct browse line is evident on the area. Instances of mass mortality have occurred in this herd and the average weight of the deer is 20 pounds less than for deer of the same age-class from western Maryland (Flyger 1958). The only published commentary on the past history of deer in this study area is that of Bullene (1949). He states that when the U. S. Government acquired the area in 1917 there were a few deer. We have received unconfirmed reports that deer were not present on the area at that time, that deer were released in 1932 and 1933, that the stock apparently came from Pennsylvania-probably Indian Town Gap Military Reservation -and consisted of one buck and three does. Regardless of the source of deer, it is quite evident that the initial population was very low and that, as a result of fencing, a mimimum influx into the area occurred, thus resulting in a highly inbred herd. The literature on strontium-90 accumulation in deer bone is rather limited. French (1960) analyzed a deer bone collected in the tropics of Ecuador in 1958 which contained

BETWEEN 1939 and 1948, the writer camped over four months on Big Black Mountain in Harlan County, Kentucky, making a survey of the amphibians, reptiles, and mammals of the region. Previously published reports deal with the reptiles (Barbour, 1950a) and the mammals (Barbour, 1951). For a general discussion of the area, reference is made to these papers. Big Black Mountain reaches a maximum elevation of 4150 feet above sea level, and is the highest point in Kentucky. It lies largely in Harlan County, Kentucky, but occupies a part of adjacent Wise County, Virginia. The majority of the collections discussed in this paper were made within a four-mile radius of the

Waletzky, Hughes, and Brandt (1949) reported that the continuous feeding of nitrophenide in the mash at a concentration of 0.05 per cent for eight to ten days, starting on the day of inoculation or one or two days before inoculation, checked mortality from severe cecal coccidiosis in chicks. They also referred to unpublished data by themselves which indicated that nitrophenide was effective in controlling Eimeria tenella infections when administered 48 to 96 hours after inoculation; the drug was ineffective, however, when administered at time of inoculation to 48 hours after inoculation. In the same paper they presented preliminary evidence to show that nitrophenide is also effective against Eimeria necatrix. Bottorff (1950) presented data from field tests to show that the feeding of nitrophenide in the mash to chicks, from one day of age through 11 weeks, reduced mortality from cecal coccidiosis. Seeger (1950) fed nitrophenide continuously in the mash at concentrations of 0.0125 and 0.00625 per cent and reported good results. Swales (1950) fed nitrophenide continuously in the mash at 0.0125 per cent concentration, starting 24 hours before exposure to contaminated litter, and obtained little protection from cecal coccidiosis. In continuous feeding at 0.05 per cent concentration, starting at time of exposure, he got widely varying results. However, he found that the feeding of 0.1 per cent concentration for four days, starting 48 hours after exposure, resulted in a definite reduction in mortality. In a more recent paper (1951), Swales reported a reduction in mortality among birds on field tests, treated continuously with 0.0125 per cent nitrophenide mash. Peterson and Hymas (1950) found that a concentration of 0.05 per cent was necessary to suppress mortality in experimental Eimeria necatrix infection. Brackett and Bliznick (1949) reported that chicks given nitrophenide continuously for the duration of the experiment discharged fewer oocysts of Eimeria tenella than did untreated chicks, and that the percentage of sporulation of the oocysts passed by treated birds was much lower than that of the oocysts passed by untreated birds. Dickinson, Babcock, and Osebold (1951) found that continuous feeding of nitrophenide in the mash at a level of 0.05 per cent suppressed oocyst development of both cecal and intestinal coccidia, and gave protection from the disease during the period of administration of the drug.

The percentage of calcium in the ash of white-tailed deer mandibles was determined for 68 specimens collected from a Maryland herd. There was no statistically significant differences among ages or between sexes. The overall average percent, 37.8, was comparable with presented values for North American big game. During an investigation of strontium-90 accumulation in a Maryland herd of whitetailed deer (Odocoileus virginianus) (Schultz and Flyger 1965), I collected data on the calcium content of mandible ash. These data were collected because strontium accumulates in the bone mineral where it behaves like calcium. Furthermore, calcium data are used in an expression known as the strontium unit (1 micromicrocurie of 90Sr per gram of calcium) or in a strontiumcalcium ratio. The purpose of this paper is to record these data and to evaluate the relationship of calcium content of mandibles to sex and age of deer in this Maryland herd. In addition, I desire to summarize the data known to me on the calcium content of bones of big game. The deer mandibles were collected at Aberdeen Proving Ground, Maryland, in December, 1959. Details concerning the study area, sample characteristics, and statistical procedures are presented in Schultz and Flyger (1965). The calcium determinations for this paper and those reported by Schultz and Longhurst (1963) were conducted by contractors of the U. S. Atomic Energy Commission's Health and Safety Laboratory, utilizing procedures recommended by the laboratory (Harley and Whitney 1957). The average difference of the percentages between seven blind duplicate determinations was 0.71 ? 0.12 (sE). I am indebted to personnel of the U. S. Atomic Energy Commission's Health and Safety Laboratory for supervising the calcium analyses and to J. L. Carmon and J. Fortson who so kindly did the machine computations for the least-squares analysis. K. J. Chiavetta, R. D. McDowell, C. W. Severinghaus, and J. E. Tanck estimated the age of the specimens. I am also indebted to F. B. Turner who commented on

PIERSOL (1907) was the first to describe the spawn of Ambystoma jeffersonianumn, showing that the egg-masses can readily be distinguished from those of A. maculatum. Smith (1911) compared the eggs of the two preceding species and those of A. tigrinum. Noble (1927), in considering the relation of breeding habits to phylogeny, states that both Allen and Bishop noted the males embracing the females. The reference is probably an error and should read, Wright and Allen. No one, however, has heretofore described the spermatophores of A. jeffersonianum. Smith (1911) considered the extremely early spawning season suggestive of an autumnal fertilization, while Piersol (1929), noticing that both sexes hide on land near the ponds during the day, suggested the likelihood of a terrestrial mating. In a previous paper (1930) the writer has reported field observations of seven embracing pairs of salamanders and the discovery of a wellformed spermatophore in the cloaca of a male A. Jeffersonianum, examined in the laboratory last year. No description of the spermatophore was given in that paper. During the present season observations have been made every night for a period of ten days after the appearance of the first A. jeffersonianum, and trips have been made to the ponds at frequent intervals since that time. The only mountain ponds which are readily accessible are near Woodward, Centre County, at an altitude of 1475 feet, and nearly thirty miles to the west of Bucknell University, Lewisburg, Pennsylvania, where the writer was located. The first individuals of A. jeffersonianum, in 1931, were seen on the night of March 25. At this-time a spermatophore, believed to have been deposited by this species, was found attached to a broken twig in about three inches of water close to the edge of the pond. About a dozen of the fifty odd salamanders seen were collected. On transferring them to an aquarium the next morning, two spermatophores were found in the collecting can. The next trip was made primarily to hunt spermatophores and an hour's intensive search revealed thirteen. On returning to the laboratory it was found that the salamanders collected on the first night had deposited five spermatophores in the aquarium. On the third night, March 27, fully twenty-five spermatophores were located in a second pond. On this night the first eggs were seen, and three females were observed in the process of egg-laying. The fact that no A. maculatum reached the ponds until the following night, March 28, as

A total of 285 lesser prairie chickens (Tympanuchus pallidicinctus) were mist-netted in the spring on 16 booming grounds (display grounds) in eastern New Mexico during 1962-70. Life tables, based on capture-recapture data from 3 consistently trapped grounds, were constructed for males only, as comparatively few females were trapped and none was recaptured. The male population underwent virtually a complete turnover in about 5 years. The mean annual overall mortality rate of males was calculated at about 65 percent. Inefficiencies inherent in the trapping method (probable failure to recapture all banded birds present on booming grounds each year) presumably magnified the calculated mortality rate, possibly by about 5 to 10 percent. Age and sex ratios indicate that adult females had a higher mortality rate than adult males. The removal by hunting of about 1,100 birds per year, on the average, over a 12-year period had no observed harmful effect on the population. Recaptures of banded birds suggest that adult males are faithful throughout life to the same booming grounds where they initially established territories, but hunter recoveries show that at least some males shift for feeding purposes in fall and winter to harvested grain fields that may be several miles from their accustomed booming grounds. This paper reports the results of a field study of the lesser prairie chicken on its native range in eastern New Mexico. The work was done in the heart of that range in southern Roosevelt and northern Lea counties. Life tables, based on capturerecapture data, were constructed for the male segment of the population. Biological and behavioral data derived from the hunting harvests are also presented. Several general accounts have been published concerning the lesser prairie chicken in New Mexico (Bailey 1928:207-209, Ligon 1961:89-92, Sands 1968), but published reports of specific research studies on the local populations have been few. The first of these reports appears to be that by Lee (1950); it contains data on age and sex ratios, weights, criteria for determining age and sex, habits, and habitats. Campbell (1950) described a typical instance of the 1A contribution from Federal Aid Project W104-R, New Mexico. often-seen persistent harassment of prairie chickens by wintering marsh hawks (Circus cyaneus). Snyder (1967:121-128), in addition to a general account of the species in the state, summarized some of the technical data from a series of unpublished New Mexico P-R reports by W. S. Huey, G. W. Merrill, L. G. Frary, W. A. Snyder, J. F. Johnson, M. G. Wischnofske, and J. L. Sands. The present paper is based chiefly on these reports, and on several of my later unpublished P-R reports. Portions of Colorado, Kansas, Oklahoma, and Texas also have populations of lesser prairie chickens, and investigators in at least the latter two states have published important work on the species. Copelin (1963) published a comprehensive account of lesser prairie chickens in Oklahoma, and Jones (1963) compared in great detail the habitats of the lesser and greater prairie chickens in Oklahoma. Jones (1964) also studied seasonal behavior of various plant species in

Experiments were conducted in the laboratory to investigate the ability of trichostrongylid larvae to migrate vertically on various grasses under controlled environments. Only a small proportion, 2 to 3%, of the larvae given opportunity to ascend the grass blades, under conditions considered favorable for migration, climbed. Of these, 59.2, 26.7, 9.9, 3.4, and 0.8% were found on the 1st-, 2nd-, 3rd-, 4th-, and above the 5th-inch height level from the soil, respectively. It was concluded that though infective trichostrongylid larvae can climb grasses, the greatest portion remain at the base of the vegetation. The larvae were comparatively more successful on the tall fescue than on the smooth brome. The external morphology of grasses apparently influences migration of the larvae. A film of moisture is obviously advantageous to migration. On an average 0.65% of the test larvae climbed wetted blades while only 0.04% climbed unwetted blades. Upward migration was favored by higher relative humidity. On the average 1.36% of the test larvae were successful at 95% relative humidity while only 0.06% were successful at 56% relative humidity. Lower temperature adversely affected upward migration. Only 0.13% of the test larvae climbed at 40 F while 2.54% did so at 79 to 80 F. A direct influence of light intensity was not clearly established since migration in light and darkness was essentially similar. Taylor (1938), Rogers (1940), Kauzal (1941), Dinaburg (1944), Crofton (1948), and Rees (1950) studied vertical migration of infective trichostrongylid larvae on herbage. Their techniques essentially involved placement of known numbers of infective larvae, contained in a known volume of water, on the soil at the base of the test herbage in experimental plots in the field or laboratory. Subsequent recovery of larvae on the test herbage was interpreted to indicate vertical migration of the larvae from the soil surface onto the herbage. However, except for Crofton (1948) and Rees (1950), most of these investigators did not establish the actual height on the herbage the larvae ascended. Furthermore, only in a few cases (Crofton, 1948; Rees, 1950; Rogers, 1940) was the ability of the larvae to migrate vertically on the herbage correlated with environmental factors. The present study was undertaken in order to obtain an extensive series of observations Received for publication 20 August 1963. * From the Department of Veterinary Science, Paper N.S. 400, with the approval of the Director, Wis. Agr. Exp. Sta., supported in part by the University Research Committee. on the actual height of vertical migration by trichostrongylid larvae as well as the portion of the population which does migrate, and to observe the influence of various environmental factors including moisture, temperature, relative humidity, light, type of grass, and the number of grass blades available upon movement of the larvae. MATERIALS AND METHODS The larvae for these studies were obtained from "seed" calves primarily infected with Haemonchus placei with minor populations of Cooperia onchophora and Trichostrongylus colubriformis. The larvae were cultured in sterilized sphagnum according to the method of Cauthen (1940) and recovered by using Baermann funnels. Differential counts of larvae were not made for individual tests, and the larvae, therefore, have been referred to as trichostrongylid. The larvae were 7 to 20 days old when used in the tests, but in a given trial all larvae were the same age. Grass blades were grown from seeds planted in a light silt soil or a muck soil in paper pots. The soils were sterilized to render them free of nematode life prior to the tests. The grass seedlings emerged in a tuft of blades and were used when 3 inches or more tall. Numbers of larvae used in a test were estimated by dilution counts. By restriction of volume of the suspension, a known quantity of larvae could be placed in position on the soil surface at the

Field sampling equipment consisted of a supply of No. 12 or slightly larger Kraft paper bags for holding young birds and keeping those already bled separate from those to be bled during sampling activities; a supply of sterile 1or 2-c.c. hypodermic syringes (the writer preferred using a 2-c.c. syringe when bloods from several birds in a nest were to be pooled); sterile 1-inch, 27-gauge needles; small (10-c.c.) serum bottles with 10% rabbit serum in buffered saline diluting fluid;' a bottle of 70% alcohol; tightly corked sterile 10 x 75 m.m. soft glass tubes with adhesive tape labels, for blood samples; and a pint thermos flask partly filled with wet ice for immediate refrigeration of blood samples and for carrying these to the laboratory. A bottle of clean water for rinsing syringes and needles after use facilitated later cleaning of this equipment. A supply of Fish and Wildlife Service bands was carried for marking the birds when possible. The young birds were removed from a nest and placed in a Kraft paper bag. To prepare the syringe for procuring the blood sample, 0.1 c.c. or 0.2 c.c. diluting fluid was drawn into it. This enabled the operator to obtain better initial suction with the syringe and to waste less blood when only 0.1 c.c. or 0.2 c.c. was taken from a nestling. A young bird to be sampled was then placed on its back in the left hand, with its head toward the finger-tips. The neck of the bird was placed between index and middle fingers and the head pulled back slightly. Legs and posterior part of the bird were held down between little finger and thumb, which tended to arch the breast of the bird upwards for easier needle penetration. The ring finger was used to hold down the left wing when it was of any size and tended to get in the way. The operator worked from a sitting or kneeling position and steadied his hand holding bird by resting it on his knee.

Whole-body gamma-irradiated (250r) cottontails (Sylvilagus floridanus) confined in 25x 60-ft outdoor pens produced litter sizes significantly smaller than normal cottontails kept as controls. Animals irradiated in 1964 tended to recover from irradiation and produce litters approaching normal size in 1965. Early papers on the effects of artificial radiation on wild animals are Schnell (1963, 1964) and Willard (1963). The objective of this paper is to report on the effects of whole-body gamma irradiation on reproduction of penned cottontail rabbits. I am indebted to R. A. McCabe, L. B. Keith, and J. J. Hickey of the Department of Wildlife Ecology of the University of Wisconsin, for guidance and assistance throughout this project. K. H. Clifton of he Radiology Department was very helpful in showing me the technique of wholebody irradiation. MATERIALS AND METHODS Cottontails were obtained by livetrapping at the University of Wisconsin at Madison campus and at the State Game Farm near Poynette, Wisconsin. Pairs were kept in separate rearing pens, 25 x 60 ft, at the State Game Farm. The pens were made of 1 This paper is based on an M.S. Thesis in Wildlife Ecology completed under a Graduate Research Assistantship, University of Wisconsin, Madison. 2 Present address: Box 1435, New Mexico State Health Department, Roswell, New Mexico. This content downloaded from 207.46.13.127 on Wed, 12 Oct 2016 04:34:46 UTC All use subject to http://about.jstor.org/terms REPRODUCTION OF IRRADIATED COTTONTAILS * Forcum 181 1-inch mesh poultry wire buried in the ground 12 inches. They were covered and connected to a brooder house at one end. Commercial rabbit pellets were made available weekly in addition to the natural food and cover which was predominantly quack grass (Agropyron repens) and Kentucky blue grass (Poa pratensis). Water was provided during the summer. The procedures of rearing cottontails in pens were similar to those described by Rongstad (1966). Females were palpated frequently to determine the stages of pregnancy and of lactation. By carefully searching the pen area, I was able to locate 85-95 percent of all known litters. Females and litters were tolerant of handling. Rabbits were irradiated with gamma rays from a 2,000-Curie cesium-137 source at the University of Wisconsin General Hospital. Each rabbit was placed on its left side in a sock within a lucite cage 25 cm in diameter and 10 cm deep. The cage had a 5-mm-thick adjustable lid, which was placed against the animal's right side. The cage was placed in the center of an adjustable shelf and suspended from the head of the cesium unit. The distance from the cesium source to the floor of the cage was a constant 55.5 cm. The upper surface of the animal was approximately 49 cm from the source. The energy of gamma rays emitted by 137Cs was 662 kev, and the unit was fitted with a copper field flattening disc which aided in maintaining uniform dosage over the entire bottom surface of the cage. The dose rate to the upper surface (right side) of the animal was 30.7 rontgens in air per minute and to the center of the floor of the empty cage with the lid in place was 25.4 r per minute. Dosimetry indicated a significant difference in the dose rate to the upper and lower surfaces of the radiation chamber. This resulted in a 25 percent reduction of radiation to the abdominal cavity as compared to the animal's upper surface. In terms of total dose (measured in air at the upper surface of the body) the first series of 10 treated rabbits received 750 r, which was 90 percent lethal within 8-10 days. A second series of 26 rabbits received 250 r which was 4 percent lethal in 10 days, and 20 percent lethal in 30 days. Animals were individually irradiated, then paired with normal animals. Equal numbers of each sex were irradiated on one date, usually five of each sex, until the entire series was completed. Normal pairs served as controls. Reproductive capacity was assessed by appraising complete pedigrees (reproductive records) of all animals. RESULTS AND DISCUSSION Results indicated that whole-body gamma irradiation of either parent decreased litter size of the cottontails. Average sizes of all groups of litters with one irradiated parent were numerically smaller than the average sizes of litters produced by normal parents (Table 1). Average sizes of litters with one irradiated parent (either sex) in 1964, and litters born in 1965 to females irradiated in 1964 were significantly smaller (P 0.05). Some recovery was indicated among the animals irradiated in 1964 which bred in 1965 (Table 1). The first five litters in 1964 were born April 8-11, and the first six in 1965 were born April 10-26. The last litters in both years were born during late August. A few of the animals irradiated in 1964 were still paired and breeding in 1965 when a number of additional animals were irradiated. A decrease in the litter sizes of mammals This content downloaded from 207.46.13.127 on Wed, 12 Oct 2016 04:34:46 UTC All use subject to http://about.jstor.org/terms 182 Journal of Wildlife Management, Vol. 32, No. 1, January 1968 Table 1. Sizes of litters produced by normal and irradiated cottontails (first litters excluded). Statistical analyses according to Duncan's multiple range test, Kramer's (1956) modification.

The antioxidant used in commercial animal feeds, 6-ethoxy-1,2-dihydro-2,2,4-trimethylquinoline (ethoxyquin, Santoquin, or EMQ) had an inhibitory effect on larval nematode development. Increasingly fewer larvae were recovered from fecal cultures as the quantity of EMQ fed to infected calves, or mixed with feces from infected calves, was increased from the amount commonly used as a feed preservative (1/4 lb/ton) up to 100 times that amount. For example, the number of Cooperia pectinata larvae recovered from fecal cultures from a calf fed untreated chopped alfalfa hay was 40.4% as compared to 9.5% when 100 X EMQ was mixed with the hay. Similarly, 25 X EMQ resulted in reductions as follows: Trichostrongylus colubriformis from 33.2 to 13.5%, T. axei from 28.3 to 10.8%, and Oesophagostomum radiatum from 45.3 to 4.8%. Comparing commercial pelleted alfalfa hay with and without EMQ showed reductions, varying from 17% for C. punctata to 64% for T. colubriformis. Larval development in feces-vermiculite cultures treated with various concentrations of EMQ showed reductions as follows: T. axei from 27.9 to 0.0% with 500 X, C. pectinata from 22.8 to 0.8% with 30 X, and T. colubriformis from 24.6 to 15.3% with 4 x. Studies on the effect of host diet on the bionomics of the free-living stages of some cattle nematodes showed a reduction in the number of larvae recovered from cultures when the diet was changed from alfalfa hay to dehydrated alfalfa pellets. Further investigations indicated that inhibition of larval development could be due to the chemical preservative or antioxidant used in the preparation of commercial animal feeds. Of the antioxidants used in the food industry, 6-ethoxy1,2-dihydro-2,2,4-trimethylquinoline (known as ethoxyquin, Santoquin, or EMQ) is probably the most prevalent because of its high activity and desirable physical properies (Bickoff et al., 1954). According to Machlin et al., 1957, and Wilson et al., 1959, this chemical is rapidly absorbed and its metabolites are readily and nearly completely excreted in urine and feces. This paper considers the effect of EMQ on larval development when mixed with various feeds, and when added to feces obtained from donor calves monospecifically infected with various cattle nematodes. Received for publication 28 September 1966. * Resigned 13 August 1966. t In cooperation with the Georgia Agricultural Experiment Station. Station Journal Series Paper No. 546. MATERIALS AND METHODS Commercial dehydrated alfalfa pellets (16 or 17% crude protein) and alfalfa meal (17% crude protein) were compared with alfalfa hay and with chopped alfalfa hay, to which various amounts of EMQ (Santoquin-Mixture 6, Monsanto Company, St. Louis) had been added. The chopped hay was passed through a hammer mill with a 1/2-inch screen. The standard concentration of EMQ, designated as 1 X, was based on the amount commonly used in the manufacture of commercial pellets and other animal feeds (1/4 lb/ton, or 125 Mig/g). Higher dosages were in multiples of the standard dosage, and were designated, for the sake of brevity, from 1 X to 100 X. Dairy calves, monospecifically infected with various cattle nematodes, were used as experimental hosts. The species present were Cooperia punctata, C. pectinata, Trichostrongylus axei, T. colubriformis, and Oesophagostomum radiatum. Feces were collected from the hosts' rectums hourly during the day (8 AM to 4 PM) for various days while the particular host calf ate, ad lib., one of the rations. No collections were made 3 or 4 days following a change in diet in order to allow the elimination from the calf of all waste products of the previous diet and for the rumen flora to become adjusted to the new diet. The total daily output was mixed thoroughly, samples were taken for nematode egg counts, and the remainder was cultured in plastic pans. The cultures, each containing 1,000 g of feces, 166 g vermiculite, and any necessary water, were incubated for 9 days in constant temperature cabinets set at 25 C. All cultures were weighed daily and opened to aerate and to replace any water lost through evaporation. At the end of the incubation period, the cultured material was placed in Baermann funnels, left over

Investigations with captive and free-ranging cow elk (Cerous elaphus) have demonstrated that urinary allantoin;creatinine (A;C) ratios can be used as an index of metabolizable energy intake. Practical management applications of the technique, however, would require random collections of snow-urine, resulting in unknown proportions of samples from various sex-age classes. The goals of this research were to investigate sampling protocols and statistical methods for analyzing A:C ratios from random collections of snow-urine to develop an index of cow elk nutrition, and to evaluate these methods by comparing the winter nutrition of cow elk in several free-ranging herds. Snow-urine samples (n = 2,051) were collected from cow, calf, and bull elk in the Madison-Firehole Drainage of Yellowstone National Park during 5 consecutive winters from 1991-92 through 1995-96. Significant differences were detected in mean A:C ratios among sex-age classes. Computer simulations using these data demonstrated that mean A:C ratios derived from random sample collections were highly variable and consistently overestimated the true mean A:C ratio of cow elk. These problems were alleviated by the use of trimmed means, which we calculated by trimming 15% off the right tail and 20% off the left tail of the ordered sample distribution and calculating the mean of the remaining samples. Trimmed estimates of mean A:C ratios were consistently less biased with lower variability of sample means, and hence smaller confidence interval widths than were untrimmed means. Additional simulations and field experience suggested that a realistic sampling goal would be the collection of 20 samples at 2-week intervals throughout the winter. We tested this sampling protocol and the use of trimmed means during the winter of 1996-97 by collecting 1,008 samples from 6 elk herds that were managed for 3 different goals and were expected to have different nutrition. One difference in average overwinter A:C ratios was detected with trimmed means that was not detected with untrimmed means. Several differences in average overwinter A:C ratios were also detected that were consistent with expected differences in nutrition among the 6 herds, indicating that the sampling protocol and statistical methods detailed in this paper for urinary A:C ratios may provide a useful technique for indexing winter nutrition of ungulate herds.

Livestock-guarding dogs are an effective way of protecting rangeland sheep from predators. However, open mountain/forest range and widely ranging sheep are factors that may make adaptation to Norwegian conditions difficult. This paper focuses on the dogs' working patterns and effectiveness under different working regimes. A 3,500 ha. unfenced forest/mountain range pasture in bear habitat comprised the research area in which 624 sheep from 2 herds grazed. The field trial lasted 3 months, and a total of 10 Great Pyrenees participated for various time intervals. Three different working regimes were evaluated. 1) loose dogs without the command of a dog handler (Method A); 2) loose dogs under the command of a dog handler (Method B); and 3) loose dogs guarding sheep inside a fenced, 1 km(2) forest pasture (Method C). Nocturnal behavioural activity patterns and data on predation were recorded. Method A proved too uncontrolled for Norwegian conditions, because sheep dispersed too widely and dogs ranged too far, causing conflicts in nearby settlements with wildlife, and with livestock. Pasture dogs (C) were > 3 times less active and were engaged in guarding activities 15 times more frequently, and no sheep carcasses were found inside the fence. Therefore, Method C probably had the best preventive effect.

Prior to migration, many migratory birds accumulate large stores of lipids as the prime energy for their migratory flights. Although this phenomenon is well known, comparatively little is known about the mechanisms of fat accumulation, and the nutritional demands and nutritional adaptations of fattening migrants. This paper shows recent findings on the role of nutrients in fattening, and how migrants are likely to achieve optimal rates of fattening by diet selection. Dietary energy and nutrient composition of foods clearly influenced the rate of daily body mass gain during migratory fattening. However, it was neither energy nor a single nutrient alone which accounted for migratory fattening. Most of the variation in body mass gain was explained by the combined effects of high energy intake, where the energy was provided by fat, and by protein intake, whereas the role of carbohydrates is less clear. The data are discussed with respect to the adaptive role of seasonal shifts in diet selection during fattening periods known for many species, and the consequences of diet for fat deposition time, fat deposition rates, and fat loads at departure as major predictors in optimal migration models.

1. With changes in the agricultural policy for rural areas in the UK, there is increased pressure to develop management protocols for indigenous and sown swards based on grazing by ruminant species other than sheep and cattle, e.g. goats, red deer and South American camelids. This can only be achieved through a greater understanding of the feeding behaviour of these species grazing on vegetation communities typical of the hills and uplands. 2. This paper describes the organic matter intake, diet digestibility and activity time of goats, red deer and South American camelids grazing three vegetation types typical of upland ecosystems in Scotland: a sown grassland (Lolium perenne-dominated), an indigenous grassland (Nardus stricta-dominated) and a dwarf-shrub community (Calluna vulgaris-dominated). Intake and digestibility of organic matter (OM) were measured by the n-alkane technique, and activity time was measured using vibracorders. 3. For all species OM digestibility of the diet was higher on the sown sward than on the indigenous communities, and higher on the indigenous grassland than on the dwarf-shrub community. There were no consistent patterns to between-species differences in OM digestibility. 5. When expressed on a metabolic liveweight basis, there was no consistent pattern to the seasonal differences in OM intake (OMI) between species or within species between vegetation types. 6. Irrespective of season, grazing session or vegetation type, activity time of the goats was significantly lower than that of either the red deer or camelids. 7. The conclusions of the study are that these species differ in their foraging strategy, particularly on indigenous communities where heterogeneity in resource dispersion and availability is higher. Goats and camelids maintain diet digestibility at the expense of intake, whereas red deer maintain intake.

1. A sustainable, low-input land use might be achieved by the use of grazed grass-legume mixtures. However, maintaining the species balance has proven to be difficult, especially as grazing provides differential defoliation and regrowth. Since changes in the sward botanical composition can be ascribed to species differences in growth, senescence or intake, a method that yields independent estimates of these three fluxes for both components of a grazed grass-legume mixture is needed. 2. The objective of this paper is to present and assess such a detailed tissue turnover method based on non-destructive measurements of individually marked population units. This method was applied during two spring and summer growing seasons to a ryegrass-clover mixture continuously or rotationally grazed by sheep and is assessed using these experimental results. 3. Defoliation was measured on grazed population units (i.e. grass tillers; elongated and axillary shoot growing points of clover). To calculate the growth rate of defoliated leaves, a sample of ungrazed leaves of each defoliation status (with or without previous defoliation) was used. Additionally, for clover, a sample of ungrazed leaves of each 'botanical number' was also required. The mean extension rates of leaves that underwent defoliation during an observed period, and of leaves that did not, was not significantly different during the preceding period. 4. The estimates of the mixture growth flux varied by less than 10 and 17% when the mean values of the gravimetric coefficients and of the density of the population units, respectively, were increased by one standard deviation. The net recruitment of new population units led to an underestimation of the growth flux by 8 and 5% with grass and clover, respectively. 5. The method was partially validated, by comparing independent estimates of the mean dry-weight of the population units, based either on tissue turnover or on destructive measurements. For vegetative tillers and elongated clover growing points, both estimates were highly correlated with no significant bias. 6. Although this method is labour intensive, it yields reliable estimates of the growth, intake and senescence fluxes, at three different levels (square metre, population unit and plant part). Hence, a better understanding of the impacts of herbivory on the grass-legume balance in a grazed mixture can be gained.

1. This paper documents the results of the first 4 years of an experiment which has involved frequent applications of ammonium sulphate within a 1-km 2 tract of nitrogen-poor dry heathland in southern England. Experimental additions of ammonium sulphate at deposition rates of 7.7 and 15.4kgha -1 year -1 of N plus background deposition give total nitrogen deposition rates similar to the critical load suggested for the conversion of dry heathland to grassland. By investigating long-term biological and chemical changes in the experimental system it was hoped to determine whether the proposed critical loads are soundly based. 2. Despite an increase in shoot nitrogen content of Calluna vulgaris following only 1 year's addition of ammonium sulphate at a rate of 7.7 kg ha -1 of N, no statistically significant effects were found in the subsequent 3 years. However, upward trends in shoot nitrogen from the control to the high-nitrogen treatments were apparent in the third and fourth years of the experiment. 3. Application of ammonium sulphate at rates of 7.7 and 15.4 kg ha -1 year of N resulted in significant stimulations in shoot growth, flowering and litter production from the second year of the experiment onwards. However, given the low nitrogen status of the experimental site, the positive effects demonstrated in this study may not be wholly typical of responses which may be expected in heathlands of higher nitrogen status. 4. Continuation of the experiment is necessary to cover a greater part of the life cycle of Calluna in order to investigate how continued nitrogen inputs modify responses to natural stresses, and ultimately to determine the validity of the critical load suggested for dry heathlands.

This paper describes the large-scale spatial and temporal patterns of grazing on salt-marsh by Dark-Bellied Brent Geese Branta bernicla bernicla in winter and examines them in relation to grazing optimization theory. The numbers and spatial distribution of geese feeding on salt-marsh were recorded, on average, every 2.5 days from October to April 1991/92, and the data analysed for regular cyclic patterns of grazing within areas of the marsh. The mean percentage ground cover of the main food plant, Puccinellia maritima (Poaceae), was estimated in all areas and the proportion of each area composed of short vegetation was measured. The effects of grazing on the age structure of leaves of P. maritima, and on the patterns of regrowth following grazing, were monitored in swards of high, medium and low mean P. maritima cover. Eleven of 23 defined areas of salt-marsh were grazed by geese infrequently (fewer than five records) or never. These areas were clearly distinguished from those used frequently for at least part of the winter by their lower coverage of short vegetation (average 9%, compared with 40% for frequently used areas). Of the 12 frequently grazed areas, six showed evidence of cyclic patterns of grazing, with the interval between grazing bouts ranging from 7 to 15 days. The remaining six areas showed only weak evidence or no evidence of cyclic grazing patterns. The available evidence suggested that disturbance, position in relation to roost and inland feeding areas, and the occurrence of night-time feeding may influence whether or not grazing cycles occur. The time required for the complete regrowth of grazed leaf tissue was greater than 50 days but on cyclically grazed areas, regrazing intervals were much shorter than this. The geese were therefore grazing partially depleted swards. The lengths of cycles were unrelated to the intensity of grazing (measured in goose min m -2 ) but positively related to the percentage cover of P. maritima in the sward. Grazing resulted in a younger leaf age structure on tillers of P. maritima, particularly in swards of low cover. Because geese showed no selection for younger leaves or unbitten tillers, it is suggested that the shorter cycle lengths on lower biomass swards represent a preference for the higher nutrient quality provided by the younger leaf age structure following grazing.

Efforts to determine quality of ruffed grouse (Bonasa umbellus) winter diets from food habit studies have been hampered by lack of knowledge of nutritional values for potential forages. We measured food preferences and digestive efficiencies of 8 grouse fed quaking aspen (Populus tremuloides) staminate flower buds, willow (Salix discolor) buds, and American hazel (Corylus americana), paper birch (Betula papyrifera), and speckled alder (Alnus incana rugosa) catkins. We collected forages in winter condition and fed them to grouse that had acclimated to winter foods for 31-41 days prior to mass balance trials. We collected winter and spring aspen buds from trees used by grouse in the field. Grouse preferred spring aspen flower buds and avoided birch and alder catkins (P = 0.008). Apparent metabolizable energy coefficients (MEC N * ), corrected for nitrogen balance, were hazel (0.30), spring aspen (0.26), mixed browse (0.25), winter aspen (0.21), and willow (0.18). Plant secondary metabolites in birch and alder may have limited palatability of these forages to grouse. Aspen bud MEC N * increased (P < 0.001) in spring as buds began to swell and emerge. All winter browse tested had lower (P < 0.001) MEC N * than predicted by a nutritional model. Preferences of ruffed grouse for winter browse are influenced by palatability and digestible nutrient content of forage.

This paper documents the daily activities and feeding rates of Barnacle Geese Branta leucopsis through the autumn and winter in relation to changes in biomass of their food supply. Energy intake and expenditure were estimated and energy surplus or deficit for each month calculated. A net surplus was achieved in the months of October (+ 235 kJ d-1) and November (+ 247 kJ d-1) when most food was available. Geese suffered a net deficit in December (225 kJ d-1) and January (113 kJ d-1) when biomass and food quality were lowest. In February, when grass began to grow again, the birds achieved a surplus once again (+ 167 kJ d-'). Body mass changes through the winter were modelled using observed intake rates and estimated energy expenditure. Body mass predictions were consistent with observed mass and body condition changes. It is argued that mass loss in midwinter in this species is enforced by the limited feeding opportunity and declining food supply rather than an adaptive response to reduce predation risk. Geese gain substantial body reserves in autumn prior to this predictable mass loss. As soon as daylength and food biomass allow, lost reserves are replaced and mass increases rapidly in preparation for migration and breeding.

In the Soay sheep population of Hirta (St. Kilda), high winter mortality occurs every 3-4 years following summers when population density exceeds 2.2 sheep ha −1 . During these die-offs, more than 50% of adults, 70% of yearlings and 90% of lambs die and population density falls by around 65%. This paper investigates the extent to which density-related changes in early growth might be responsible for fluctuations in over-winter survival and population size

An Amazon from another collection died acutely and was brought in for necropsy. The proventriculus was markedly thickened, the proximal bowel showed gross evidence of bleeding and the ventriculus was full of crushed walnut shell bedding. The owner determined that the bird had access to the wanut shells for less than two hours approximately one month before it died. Another Amazon in the flock with access to the same walnut shell bedding had previously been presented to the clinic for "off and on eating" and "on and off depression." Although the owner had declined radiographs at the time, the bird responded to treatment with Milk of Magnesia and piperacillin. Avian veterinarians need to be aware that ingestion of some bedding materials such as crushed corn cobs and walnut shells (in addition to kitty litter and excess grit) can result in gastrointestinal obstruction. I am not aware of problems with the use of Crown bedding paper pellets, which seem to be digestible if ingested.

1. This paper explores the relationship between energy expenditure, body size and the capacity for homeothermy in young of a precocial mammal. 2. Metabolism and rectal temperature (T b ) were measured during postnatal development in 4-81-day-old white-tailed jackrabbits (Lepus townsendii). 3. Whole-animal metabolism increased with age, mass-specific metabolism decreased with age, and mass-independent metabolism showed no age-related change. 4. Jackrabbits 4-5 days old responded to low ambient temperature (T a ) by increasing metabolism, but nonetheless, their T bS declined at low T a