varied less than bite dimensions, because of compensatory effects between bite area, bite depth and density. Animals obtained heavier bites in tall sparse swards than on short dense ones of equal mass/area. Even in homogeneous swards, both density and height are necessary to predict bite weight. AbstractEffects of sward height and density on the dimensions and weights of bites taken by cattle were examined. In one experiment, swards of dallisgrass lamina were constructed by hand in a factorial combination of four heights (80, 150, 180 and 300 mm) and three densities (c. 700, 1500 and 2700 g m" ^). In the other, swards of lucerne were constructed in a factorial combination of three heights (70, 150 and 250 mm) and three densities (1500, 2800 and 5900 g m"^). Treatments were replicated on three steers of 750 kg average weight. The first six bites taken from the sward were monitored, and functional relationships between sward characteristics and bite dimensions derived. Results from both experiments were similar. Average bite area was not constant as often assumed, but decreased linearly with density and increased quadratically with height, with slope negatively affected by density and height. In tail swards, bite area reached a plateau of c. 170 cm^, determined by the sweep of the tongue. In contrast with the widely used model, bite depth increased linearly with height, with slope negatively affected by density. Response of bite dimensions was explained by the mechanics of the interaction between tongue and jaw movements, and sward structure. Bite weight

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.. University of California Press and American Institute of Biological Sciences are collaborating with JSTOR to digitize, preserve and extend access to BioScience.

Crop modeling, the computerized simulation of dynamic crop systems, was born about 30 years ago, when systems analysis and modern computers presented a new technique to crop scientists. Since then, crop modeling has gone through a number of developmental stages, similar to those of living organisms. From its infancy, crop modeling seemed to promise a well‐behaved, elegant surrogate for ambiguous and cumbersome field experimentation. Indeed, some of the earliest models proved to be among the most notable achievements to date. During the juvenile stage that followed, there was an impressive increase in complexity and computer sophistication, accompanied by some of the growing pains of childhood. Greater expectations led to more and more detailed descriptions of the functioning of the biotic and abiotic components of cropping systems. The results were often trivial, and the big payoff tended to recede into the future, but the need for predicting future crop performance for management and hypothesis testing, together with progress in crop science and computer technology, spurred crop modeling. The next phase, adolescence, a period marked by intense activity, confusion, and excessive confidence‐sometimes challenged by doubt‐appears to be extending into the present. Not only is the original promise turning out to be elusive, but widely accepted guidelines for scientific modeling, such as greater reductionism, universality, and validation, are being questioned. Maturity may be emerging as expectations become pragmatically adjusted to reality. Crop modeling, like advanced ecological modeling, is proving to be more a heuristic tool than a surrogate for reality. In academic, research, and applied roles, such models can be of great value when used as aids to reasoning about the functioning and response of crop systems under many relevant, nontrivial scenarios.

No abstract

The study determined the species components of the diets of small ruminants grazing mountain ranges of the Montseny Biosphere Reserve (Catalunya, NE Spain). Three mixed flocks of sheep and goats, led by shepherds, were monitored for a year. Aniials grazed a mountain rangeland composed of Quercus ilex woodland and CuUuna-Erica heathland during the day and were returned to their corrals every night. Diet selection was estimated using fecal analysis. Of the 111 species that were identified, 71 were common to both sheep and goat. Of these, 23 were represented in proportions of more than 1% of the annuai diet. Even though goats and sheep grazed together, their diets were significantly different (p < O.OOOl), the animal factor accounting for 18% to 60% of the total variation among the main diet components. Variation between seasons was also a major (5% to 56%) highly significant factor, while differences between flocks accounted for a significant, but relatively small part (3 % to 10 %) of the total variation in diet. The outstanding difference was the avoidance of the tree, Quercus ilex, by the sheep while the goats selected it throughout the year. Sheep selected graminoids throughout the year while goats tended to avoid them. For the rest there was substantial overlap in species composition between the diet of sheep and goats, especially when analysed over an entire cycle.

No abstract

Grazing studies are hampered by difficulty in controlling the characteristics of that part of the sward encountered by animals. A methodology was developed to study the effects of sward height, density and spatial arrangement on bite dimensions, bite weight and instantaneous intake rate by cattle. Hand-constructed swards (HCS) consisted of tillers threaded through wooden modules and attached to a base board. The microphone of a cordless transmitter was attached to the forehead of steers before swards were grazed. HCS were bolted onto a force plate which recorded the forces exerted during grazing. Grazing sessions were videotaped. Height and density of swards were successfully controlled by hand-construction. Bite weight was determined more accurately by dry matter (DM) disappearance than by the reduction of fresh weight corrected for moisture loss and herbage DM content. Bite area and bite depth were precisely measured by counting and measuring the residual height of grazed plant parts. Bites and chews were identified and counted from the soundtrack of video tapes. Visual count of chews was inaccurate because

No abstract