Variation In Forage Quality Research Articles

Synchrony in short‐term fluctuations of moose calf body mass and bank vole population density supports the mast depression hypothesis

Inter‐annual variations in body mass of moose, Alces alces, in Norway and Sweden have been considered as most likely due to direct or indirect effects of weather, but so far predictions of autumn body mass of moose calves on the basis of weather data have given a poor fit to data. A striking, but hitherto unnoticed, feature of several time series on body mass of moose calves from south‐eastern Norway is an apparently regular 3–4‐year fluctuation pattern. This short‐term fluctuation could be due to regular variations in forage quality, e.g. caused by a cyclic seed production of some important food plants, as envisaged by the “mast depression” hypothesis. One plant species important as food for moose calves in autumn is bilberry, Vaccinium myrtillus, which usually produces high seed crops (masts) at intervals of 3–4 years. Populations of the bank vole, Clethrionomysglareolus, which feeds on bilberry shoots in winter, are known to peak in bilberry post‐mast years. In two study areas in Norway, there was a positive correlation between the autumn body mass of moose calves and the autumn population index of bank vole in the succeeding year. In the northern area there was an additional positive effect of summer precipitation, whereas in the southern area there was an additional negative effect of summer temperature. In both areas, however, the effect of weather was less pronounced than that of the bank vole index. These results support the mast depression hypothesis.

Predictive Models for Grazing Distribution: A GIS Approach

Grazing distribution and forage use patterns are important influences on rangeland ecosystems. Spatial patterns of grazing by domestic cattle (Bos taurus) were observed over 2 consecutive years under 2 grazing systems, intensive-early stocking and season-long stocking. The purposes were to determine factors influencing observed patterns and develop predictive models for grazing distribution and forage removal. Field-collected data on grazing distribution were linked with associated geophysical properties of pastures utilizing a GIS. Separate models were developed to predict grazing distribution and forage utilization using a backward stepwise regression procedure. The forage utilization model was linked with grazing distribution by utilizing Tobit analysis. Nineteen independent variables were used to interpret the observed variation in grazing distribution. Comparison of predicted probability of grazing values from the model with the observed grazing distribution in a hold-out data set yielded a close fit (R=.99). Eighteen independent variables were included in the forage removal model. Comparison of predicted forage removal with observed values in a hold-out data set yielded a poor fit (R=.28). Lack of forage quality variables probably accounts for the poor performance of the forage removal model. Differences in the success of the 2 models support the hypothesis that grazing distribution and forage utilization operate at different spatial scales and parameters. The use of GIS holds promise as a technique for developing useful predictive models for range management. DOI:10.2458/azu_jrm_v53i1_brock

Predictive models for grazing distribution: A GIS approach

Grazing distribution and forage use patterns are important influences on rangeland ecosystems. Spatial patterns of grazing by domestic cattle (Bos taurus) were observed over 2 consecutive years under 2 grazing systems, intensive-early stocking and season-long stocking. The purposes were to determine factors influencing observed patterns and develop predictive models for grazing distribution and forage removal. Field-collected data on grazing distribution were linked with associated geophysical properties of pastures utilizing a GIS. Separate models were developed to predict grazing distribution and forage utilization using a backward stepwise regression procedure. The forage utilization model was linked with grazing distribution by utilizing Tobit analysis. Nineteen independent variables were used to interpret the observed variation in grazing distribution. Comparison of predicted probability of grazing values from the model with the observed grazing distribution in a hold-out data set yielded a close fit (R=.99). Eighteen independent variables were included in the forage removal model. Comparison of predicted forage removal with observed values in a hold-out data set yielded a poor fit (R=.28). Lack of forage quality variables probably accounts for the poor performance of the forage removal model. Differences in the success of the 2 models support the hypothesis that grazing distribution and forage utilization operate at different spatial scales and parameters. The use of GIS holds promise as a technique for developing useful predictive models for range management.

Forage Quality of 10 Eastern Gamagrass [Tripsacum dactyloides (L.) L.] Genotypes

Ten eastern gamagrass [Tripsacum dactyloides (L.) L.] entries from Kansas, Oklahoma, and Texas, were evaluated for in vitro dry matter digestibility (IVDMD), protein concentration, lignin concentration, and phenylalanine ammonia lyase (PAL) activity. Whole plants were harvested in Spring 1992 and 1995 from 4 replicates of each entry and divided into leaf and stem (plus sheath) components for analyses. Entry significantly affected all measurements except PAL; whereas the entry by year interaction significantly affected stem IVDMD, protein, and PAL. Across entries, stem IVDMD in 1992 (56.3% to 66.4%) and 1995 (55.9% to 64.9%) demonstrated greater variation than leaf IVDMD in 1992 (62.2% to 68.0%) and 1995 (66.7% to 71.0%). In 1992 and 1995, average leaf IVDMD and protein concentration were generally higher than average stem IVDMD and protein concentration. In 1992, lignin concentration was generally higher and varied more in stems (3.27% to 4.99%) than in leaves (3.54% to 4.11%). In 1995, lignin concentration was about the same and varied more in leaves compared with stems. Stem IVDMD was significantly correlated with lignin concentration in 1992 (r = -0.92) and 1995 (r = -0.83) as well as with protein concentration (r = 0.83) in 1995. Activity of PAL in 1995 was correlated with stem IVDMD and protein. When all data were combined, IVDMD was significantly correlated with protein concentration (r = 0.82). These results indicate that gamagrass entries demonstrate significant variation in forage quality, particularly in stems. Potential exists for development of new gamagrass cultivars with high IVDMD and protein.

Liveweight changes in lambs grazing six perennial ryegrass cultivars

Liveweight changes were measured for lambs grazing six high endophyte perennial ryegrasses, in a grazing experiment run by an independent research organisation in Canterbury, New Zealand. Ryegrass cultivars were sown under code in a binary mix with white clover in February 1997. Plots were strategically irrigated to avoid moisture stress, and stocked at approximately 27 Coopworth ewe lambs per ha in May 1997. Common pasture residual mass was maintained across cultivars by strategic addition of extra lambs, with extra grazing days recorded for each replicate. Lamb liveweight gains and botanical composition of replicates were assessed quarterly. Lamb liveweight gains differed significantly between treatments in Years 1 and 2 of this ongoing 3 year study. Lambs grazing 'Quartet' tetraploid perennial ryegrass gained, on average, 37.4 kg and 36.1 kg liveweight during Years 1 and 2 respectively, and together with 'Aries HD' (35.0 and 35.9 kg) produced significantly greater liveweight gain compared with 'Grasslands Nui' (22.0 and 23.7 kg) and 'Bronsyn' (22.5 and 24.1 kg), LSD0.05 = 6.0 and 7.4. Lambs grazing 'Embassy' gained 31.6 kg and 28.7 kg, and 'Vedette' 29.8 kg and 28.0 kg. Extra grazing days expressed as a percentage of base grazing days were between 3.8 and 6.7% across cultivars for Year 1 and between 2.0 and 6.5% across cultivars for Year 2. This study demonstrated significant differences in animal productivity between perennial ryegrass cultivars that produce comparable yields of dry matter. Differences may reflect variation in forage quality, endophyte toxins, clover content in the sward or other unidentified cultivar characteristics. Keywords: digestibility, endophyte, lamb, liveweight, perennial ryegrass, tetraploid

Forage variation in brood‐rearing areas used by pacific black brant geese on the Yukon‐Kuskokwim delta, Alaska

1 We investigated the effects of grazing by black brant geese on Carex subspathacea lawns on the Yukon‐Kuskokwim delta, Alaska. 2 We compared variation in growth and forage quality in both grazed and temporarily exclosed sites to determine responses of C. subspathacea to grazing at landscape scales within two nesting colonies that had experienced different population dynamics over recent decades. 3 Landscapes differed in forage quality, grazing patterns, and in the effect grazing had on C. subspathacea forage characteristics. We found no effect of grazing on net above‐ground primary productivity (NAPP) over a wide range of natural grazing intensities at the landscape scale. 4 No differences in forage quality, NAPP, or response of C. subspathacea growth rates to grazing pressures could be detected between colonies. This suggests that goose grazing does not have deleterious effects on C. subspathacea in this ecosystem. 5 It has been suggested that gosling growth rates are sensitive to seasonal declines in forage availability and quality. Spatial variation in forage quality and availability per sampled area exceeded seasonal variation in these characteristics and is likely to have dramatic effects on gosling growth and recruitment rates.

Relationships between graminoid growth form and levels of grazing by caribou (Rangifer tarandus) in Alaska.

Herbivores and their forage interact in many ways, in some instances to the benefit or detriment of herbivore and vegetation. Studies of wildebeest (Connochaetes taurinus) in Africa and snow geese (Chen caerulescens) in the Arctic have suggested that these grazers enhance graminoid production in certain sites by repeatedly using them. Other studies have concluded that herbivores are sensitive to local variation in forage quality and quantity, and preferentially use those sites that are intrinsically more productive. In this study, caribou (Rangifer tarandus) were observed foraging at different densities on two adjacent Alaskan ranges, within which particular feeding sites contained predictably high, medium, or low densities of caribou. Vegetation from one high- and one low-use site on each of the high- and low-density ranges was sampled and monitored for productivity, measured as re-growth following clipping, with the objectives of determining which forage characteristics influence usage by grazers and whether the productivity and nature of graminoid growth after clipping were related to grazing history. Forage biomass density (g/m3), shoot density (number/m2), stand densities of nutrients and minerals (g/m3), and forage concentrations of nutrients and minerals (g/100 g tissue) correlated positively with use of sites by caribou. Productivity was independent of previous use by grazers, but consistent within ranges. These results indicate that caribou are sensitive to local variation in forage quality and quantity, preferentially use those sites with higher returns of nutrients and minerals, and have the potential to enhance graminoid growth on sites that are inherently more productive.

Quality-related characteristics of forages as influenced by plant environment and agronomic factors

The most important factor influencing forage quality is herbage maturity. For example, a 1-week delay in harvesting of alfalfa ( Medicago sativa L.) will decrease digestibility and crude protein concentration by about 20 g kg −1 and increase cell-wall concentration by approximately 30 g kg −1. Forage quality also is influenced by the environment in which forages are grown and by soil fertility and these cause year-to-year, seasonal, and geographical variation in forage quality even when herbage is harvested at the same stage of maturity. High temperatures normally increase rate of plant development and reduce leaf/stem ratios and digestibility. Moderate water stress usually delays plant maturation and causes forage quality to be maintained at higher levels. Excessive amounts of forage protein nitrogen may be excreted as urea in animal urine, which can limit animal production. Protein in forages with high ruminal escape is used more efficiently. The amount of escape protein is higher for red clover ( Trifolium pratense L.) than for alfalfa or white clover ( Trifolium repens L.). As animal production continues to improve through genetic gain, the need for high quality forages will become more important. There is good potential for producing new cultivars with improved forage quality and several are now on the market. Additional cultivars with improved forage quality will be available in the near future.

Sampling forage corn for quality

Plant-to-plant variation for corn (Zea mays L.) forage-quality traits has not been adequately described. A field study was conducted for 2 yr, using hybrids with a range in maturation, each grown at two sites. Precipitation and temperature deviated considerably from normal in both growing seasons. Ten consecutive plants were harvested from each plot and separately analyzed for forage-quality traits. Differences among the forage-quality variables were consistent for the hybrid maturity groups, and data were pooled for analysis. When the product of variance × total cost was minimized, sample size was not greatly different for any of the measured variables when the cost ratio was low. If the cost per plot is 10 times that of the cost per plant, the product of variance × total cost is minimized at 9 plants plot−1 for whole-plant digestibility. When detection limits were considered but not costs, whole-plant digestibility was the least variable trait and grain proportion was the most variable trait among the traits measured. If a detectable difference of 20 g kg−1 in whole-plant digestibility between hybrids is desired, it would require sampling 4 plants plot−1, while a difference of 20 g kg−1 in stover digestibility could not be detected with an infinite number of plants per plot. Grain proportion explained a relatively small portion of the variation in whole-plant digestibility and was better related to stover-quality traits. Key words: Corn, maize, forage quality, variation, stover, grain

Forage Quality for Ruminants: Plant and Animal Considerations

Forage quality is a function of nutrient concentration, intake, nutrient availability, and partitioning of metabolized products within animals. Of these factors, intake potential is the major determinant of production by animals fed forage-based diets; however, it is one of the most difficult aspects of forage quality to determine or predict because variation among animals has a large influence on intake. Physical fill limits intake of forages with high cell wall concentrations when fed to animals with high energy demand. Hence, grasses, with their higher cell wall concentration, typically have lower intake than legumes. Energy availability of forage is also limited by cell wall concentration because cell contents are almost completely digested, whereas forage cell walls are slowly digested. Thus, the proportion of cell walls to cell contents is a major determinant of energy availability in feeds. Protein digestion by ruminants is complex. When crude protein concentration in herbage drops below 7% of dry matter, ruminal fermentation of forages may be limited and protein requirements of animals may not be met. Additionally, inefficient use of protein in high quality forages may limit performance of high producing animals. Usually, only about 25% of the forage protein escapes degradation from the rumen. Efficiency would be improved if a larger portion of forage protein passed from the rumen undegraded so that it can be degraded in the intestines where absorption is more efficient. Another important plant factor influencing forage quality is herbage maturity. Systems are now available for determining maturity of both legumes and grasses that will become more important as aids for predicting forage quality before forages are harvested or grazed. Forage quality is also influenced by the environment in which forages are grown and by soil fertility and these cause year-to-year, seasonal, and geographical variation in forage quality even when herbage is harvested at the same stage of maturity.

Sex, age, and seasonal variation in the foraging dynamics of muskoxen, Ovibos moschatus, in Greenland

The relationship between resources (forage availability and quality) and the foraging dynamics of muskoxen, Ovibos moschatus, was studied in Angujaartorfiup Nunaa, western Greenland, during spring (calving, post calving), summer, autumn (rut), and midwinter. Muskoxen did select among vegetation types in all seasons but calving. During the postcalving period, summer, and rut, the vegetation types meadow, moist dwarf shrub heath, and grassland were preferred, whereas meadow, moist dwarf shrub heath, and steppe were selected during midwinter. The relative availability of graminoids in vegetation types did not correlate with selection of vegetation types. Seasonal patterns of daily feeding time were negatively correlated with forage quality, whereas no correlation with forage availability was found. Seasonal rumination times were negatively correlated with both availability and quality of forage. The length of rumination time is apparently not solely explained by variations in forage quality. Results presented here indicate that rumination time is also a behavioural strategy of muskoxen. Movement rates did not correlate with forage availability. During midwinter, muskoxen did not respond to low forage quality by increasing daily feeding time as was seen during calving. Instead, an energy-conserving strategy was observed, where a relatively larger proportion of time was allocated to resting, i.e., nonruminating. Bachelor herds and mixed herds differed in daily feeding time and habitat use during the prerutting and rutting season. Prior to the rutting season, muskox bulls increased energy intake to compensate for the subsequent allocation of time to reproductive activities. During the short arctic growing season, muskoxen in western Greenland decreased daily feeding time, whereas those in eastern Greenland increased feeding time. This difference in foraging dynamics is hypothesised to be a result of differences in biotic and abiotic constraints imposed on muskoxen in these two regions.

Food Intake and Foraging Energetics of Elk and Mule Deer

The energetic cost of grazing was measured using indirect calorimetry with a tracheotomized elk (Cervus elaphus nelsoni). The cost of eating herbaceous forage averaged 0.32 kcal/kg body weight/ hour, an energetic increment of 26% over standing costs. Forage intake was quantified using esophageal fistulated elk and mule deer (Odocoileus hemionus hemionus). Dry matter consumption rate and mean bite size increased curvilinearly as functions of forage biomass for both species. Asymptotic grass intake rates were 2.22 and 14.04 g dry matter/minute for deer and elk, respectively. Consumption rate and bite size were greater in shrub-forb communities than on grass pastures of comparable biomass. Biting rate ranged from about 15 to 60 bites/minute and was inversely related to bite size. Rate of travel during foraging decreased exponentially with increasing forage availability. A computer model was constructed to evaluate the interaction of time and energy constraints on foraging ungulates. The effects of variation in forage quality, forage abundance, and snow cover on the ability of animals to meet daily energy requirements are discussed. J. WILDL. MANAGE. 48(4): 1285-1301 Foraging is the dominant activity of free-ranging ungulates. Wild ungulates typically devote 40-60% of each day to finding and consuming food (Craighead et al. 1973; White et al. 1975; Schwartz 1977; Belovsky and Jordan 1978; Collins et al. 1978; Owen-Smith 1979, 1982; Gates 1980; Hanley 1982). The energetic cost of these activities represents an important component of the animal's energy budget (Young 1966, Osuji 1974, Chappel and Hudson 1978). However, in spite of the significance of foraging in the life strategy of wild ungulates, relatively little is known about the energetics of this behavior. The literature is replete with food habits information, but such observations are of limited value without an understanding of the efficiency with which the animal harvests the food resource and extracts energy from its environment. This study was undertaken to quantify the cost-benefit functions of foraging by captive elk and mule deer relative to food availability. Specific objectives were to: (1) determine forage intake rate as a function of the biomass of acceptable food items; (2) determine the relationship of intake rate to mean bite size and biting rate to better understand the behavioral adjustments that regulate energy intake; (3) quantify locomotion rate during foraging relative to food availability; (4) measure the energetic cost of eating as distinct from expenditures for standing and locomotion; and (5) model the effects of variation in forage quality and abundance on the ability of free-ranging cervids to meet daily energy requirements. This project was funded by the U.S. For. Serv. and Washington State Univ. We thank K. L. Parker for support and assistance in data collection and M. A. Reisenauer for the care of experimental animals. L. D. Bryant and J. W. Thomas provided facilities at the U.S. For. Serv. Range and Wildl. Habitat Lab. in La Grande, Oregon. J. Wildl. Manage. 48(4):1984 1285 This content downloaded from 157.55.39.29 on Tue, 12 Apr 2016 10:03:48 UTC All use subject to http://about.jstor.org/terms 1286 INTAKE AND FORAGING ENERGETICS OF ELK AND DEER * Wickstrom et al. METHODS AND MATERIALS Elk calves and mule deer fawns were bottle-raised and habituated to the experimental protocol. Training included acceptance of gentle restraint with a head halter, manipulation of the esophageal and tracheal cannulae, and transport in a trail-

INFLUENCE DES CULTIVARS, DES DATES ET DES DOSES DE SEMIS SUR LE RENDEMENT ET LA QUALITÉ DE LA LUZERNE SEMÉE EN FIN D’ÉTÉ

The effects of cultivars, date and rate of seeding on the yield and quality of alfalfa (Medicago sativa L.) seeded at the end of summer and harvested the year after was studied from 1971 to 1975 on a Kamouraska clay. Seeding in early August gave the best results; seeding should not be carried out after mid-August. Differences between cultivars were more important on the third than on the first cut. Aftergrowth was an important factor on yield during the first year of harvest. Seeding rate did not always affect the yield, but the 6,7 kg/ha rate yielded less than the 13,4 kg/ha rate at the first cut and when the cuts were summed. The combined effects of the three factors studied indicates the importance of the seeding date on the height, cover and yield of alfalfa. Digestibility and crude protein content of the dry matter were affected more by seeding dates than by cultivars. The effect of cultivars on quality was variable and seeding rates had no effect. The variation in forage quality may be attributed to a difference in maturity of the plant at harvest, rather than only to the factors studied.Key words: Medicago sativa L., cultivar, date, rate, yield, quality

Forage Quality Response of Four Lovegrasses to Stage, Interval, and Season of Harvest1

AbstractDespite the importance of weeping lovegrass, Eragrostis curvula (Schrad.) Nees, in Argentina, South Africa, and to a lesser extent in the United States, much remains to be learned about its forage quality. The objectives of this research were to evaluate: a) forage quality of four weeping lovegrass entries, b) seasonal patterns of forage quality, and c) effect of stage of first growth and age of regrowth on forage quality.Field experiments were conducted on an Enterprise loam (Thermic Typic Ustochrepts) soil at Woodward, Oklahoma. ‘Morpa’, ‘Ermelo’, common, and 673 were harvested under three managements: a) jointing stage and four subsequent 28‐day intervals, b) heading stage and three subsequent 34day intervals, and c) anthesis stage and two subsequent 45‐day intervals. In vitro dry matter disappearance (IVDMD), crude protein, cell contents (CC), hemicellulose, cellulose, and permanganate lignin (PL) were determined. Results were analyzed by ANOVA and regression procedures.Morpa and Ermelo were similar and higher in IVDMD than common and 673. This higher IVDMD was correlated with lower PL and dry matter contents. Maturity differences among entries did not account for all entry variability in forage quality. Nongenetic correlation coefficients showed that high IVDMD was associated with high crude protein and CC and low cellulose, PL, and dry matter. Seasonal trends in forage quality showed that from jointing to anthesis IVDMD declined 0.46 percentage units per day. Following the initial harvests, IVDMD continued to decline at 0.02 percentage units per day during the remainder of the summer, despite equal interval regrowth harvests. Similar patterns were found for crude protein and CC. Hemicellulose also decreased from jointing to anthesis but then increased during regrowth harvests. Cellulose and PL increased from jointing to anthesis, but showed only small seasonal changes during regrowth harvests. Increasing age of regrowth resulted in lower IVDMD (—0.28 percentage units per day), crude protein, and CC, and higher cellulose and PL. However, changes were more gradual than from joint to anthesis.

Influence of Nitrogen, Phosphorus, and Plant Population on Yield and Quality of Forage Corn1

AbstractThe acreage of forage corn (Zea mays L.) produced in Kansas has steadily increased to meet the demand for corn silage required by commercial feedlots, Fertilization practices and plant populations are major factors involved in corn production, but their combined effects on yield and quality of forage have not been thoroughly evaluated. Therefore, irrigated forage corn was grown at five plant populations and five levels each of N and P at five field locations in central Kansas during a 3‐year period to determine the influence of each variable on yield and quality of forage produced. An incomplete factorial arrangement of 23 treatments was used in a randomized complete block design. Multiple regression analyses indicated both forage and grain yields were significantly affected by N but not by P or plant population treatments. Treatment variables contributed little to variability in forage quality as measured by in vitro fermentation and forage fiber analyses. Nitrogen, cellulose, and lignin concentrations in the forage contributed up to 84% of the variation in in vitro dry matter digestibility (IVDMD). Simple correlation coefficients indicated a significant negative relationship between N and cellulose concentrations, a significant positive relationship between cellulose and lignin concentrations, and no consistent relationship between lignin and N concentrations in the forage. These results indicate that in central Kansas, irrigated corn yields are primarily a function of the rate of N application, and that a rather wide range of plant populations can be tolerated without significantly affecting yield or quality of forage produced.

Factors Affecting Growth and Body Size of Roe Deer

Differences in body size of roe deer (Capreolus capreolus) in four study areas in Denmark were investigated in relation to climate, soils, vegetation types and forest cultures, quality of available forage, population density of roe deer, and other environmental factors. Rumen samples, body weights and measurements, reproductive tissues, and lower jaws were obtained from 73 deer during the study. Body weights and measurements and sex and age data were available from several hundred deer from earlier studies. Field-dressed weight, length of the lower jaw, and femur length showed significant differences among the study areas. The largest deer and those with the most rapid growth were from the two areas that had lower soil fertility, lower ratios of agricultural land to the forest area, lower ratios of perimeter to total forest area, and lower densities of deer. Chemical ana]yses of washed and unwashed rumen contents indicated that the forage being consumed by the deer was of essentially equal quality on the four study areas. In Denmark, soil quality appears to be a primary factor determining density of roe deer or biomass but does not appear to directly regulate body size. Body size is most directly related to population density. The mechanisms by wllich density influences body size are apparently social pressures, which affect energy expenditure and food intake, and competition during spring and early summer when foods of high nutritive quality may be limited and physiological demands of the deer are greatest. Regional differences in growth and body size of roe deer have been observed throughout Europe. Selective hunting, practiced for long periods on numerous large estates in Europe, has failed to bring about increases in size of roe deer and their antlers the specific objectives of the prevailing system of game management. Bubenik (19i59), Hennig (1962), Stubbe (1966), Sagesser (1966), and others have described regional differences in body size and antler quality of roe deer. Ueckermann (1957) attempted to relate size differences to forest composition and to-the relationship of forest edge to field. He also emphasized the importance of the characteristics of the substrate as did Hesse (1921) and Sagesser ( 1966 ) . Bubtenik and Bubenik (1967) have pointed to the significance of environmental stress. Little attention, however, has been directed toward the effect of nutrition on the growth and size of roe deer, and little information is available in 1 Present address: Alaska Cooperative Wildlife Research Unit, University of Alaska, College. 64 comparison with the information available for North Ameriean deer. Speeulation by European workers on the possible eauses of regional differences in body size has usually been based on information about only one or a few of the environmental variables that might be involved (for example, vegetative coimposition, substrate, density, and similar factors). Suffieient data have not been available to allow a eonelusive appraisal of the faetors that may influenee body size. This study in Denmark was planned to investigate the influenee of variations in forage quality on the growth and size of roe deer, although it was neeessary to eonsider other faetors of the environment, sueh as vegetative eover types and roe deer density. Because it later beeame apparent that faetors other than forage quality might have a dominant influenee on bo!dy size, the seope of the study was broadened. We aeknowledge the eoolperation and assistanee of Count Ahlefeldt-Bille, J. Due, and P. Thaarup in the use of the study This content downloaded from 157.55.39.215 on Wed, 31 Aug 2016 04:23:09 UTC All use subject to http://about.jstor.org/terms GROWTH AND SIZE OF ROE DEER * Klein and Strandgaard 65 areas. We also appreciate the help of gamekeepers E. Holm and O. Poulsen and foresters O. Jacobsgaard, J. Nielsen, and A. Karlskov. H. Berthelsen did the food habits analyses of the rumen samples and the laboratory work. Chemical analyses of the rumen samples were done by Professor F. Sch,dnheyder and Grete Risgaard at the Department of Biochemistry, University of Aarhus. F. Christoffersen did much of the collecting of the specimen material, and B. Kristiansen assisted in the statistical analyses of the data. Christoffersen, Kristiansen, and J. Andersen were useful sources of information and ideas throughout the study.