Regression equations were developed to predict 3 mass components of 7 browse species important to moose (Alces gigas) on the Copper River Delta in southcentral Alaska. The accuracy of model predictions was the criterion for model selection. Model accuracy was evaluated using data splitting or jackknife procedures. Annual production of twigs and leaves and available twig mass on a stem were most accurately predicted from stem basal diameter with zero intercept models, zero intercept log-linear models, or log-log models. Twig mass eaten by moose was most accurately predicted from the diameter at the point of browsing of a twig with zero intercept or full linear models. Heteroskedasticity was significant (P < 0.05) in most of the data sets and could not be significantly reduced with log transformations or use of weighted least squares models. Heteroskedasticity appeared to have a relatively minor effect on model predictions. Most of the models gave mean predictions within +/- 20% of the actual values, particularly for the most ubiquitous species that were also the most important to moose. For each species, there were few differences (P < 0.05) in model coefficients between years and among habitat types. Differences in coefficient estimates appeared to be related to differences in stem morphology that were related to both site conditions and past browsing by moose.

This study was conducted in a temperate range of northern Pakistan in 1987 and 1988. The main purpose of the experiment was to determine a suitable residual phytomass level for the moist temperate ranges of Pakistan. Data were collected for 2 consecutive growing seasons. A completely randomized design, with 4 treatments and 2 replications, was used. The treatments were 4 different residual phytomass levels. A rainfall simulator applied rainfall to 48 flexible circular plots (1m2). Analysis of variance and the LSD multiple mean comparisons determined treatment differences, and stepwise multiple regression identified the important vegetation and soil variables affecting infiltration. The control (no grazing) resulted in the highest infiltration while the treatment having the lowest residual phytomass had the lowest infiltration. Among the independent variables, standing phytomass was the most important variable affecting infiltration. Foliar and basal cover were also highly correlated to infiltration.

During 1981-83, we studied 19 grasses for potential improvement of dry tropical (1,001 mm annual rainfall) rangelands in Hainan Island, China. The productivity and persistence of the grasses were tested on sandy (< 5 mg/kg available P) and loam sods (12-25 mg/kg) under cutting in a randomized block design with plots split for sequential sowings in 1981, 1982, and 1983. Grass response to legume-fixed and nitrogen fertilizer was also assessed. Thirteen of the 19 grasses tested under cutting established satisfactorily (density > 5 plants/m2), but only 6 cultivars produced yields > 1.5 metric tonne (t)/ha. Melinis minutiflora Beauv., Brachiaria decumbens Stapf. and Chloris gayana Kunth. yielded more than 1.5 t/ha in 3-year-old swards on both soils, whereas Panicum maximum Jacq. and Setaria sphacelata (Schum.) Stapf. ex Massey only performed well on fertile loam soil. Grass establishment was superior on sandy soil, but plant density did not correlate well with production which was higher on red loam soil due to better waterholding capacity and nutrient status. Compatibility of these successful grasses with companion legumes was low with legumes contributing < 0 .4 t/ha to total yield of 3-year-old M. minutiflora, B. decumbens and P. maximum swards, and having minimal effect on soil nitrogen. Nitrogen fertilizer (50 kg N/ha/yr) more than doubled grass production, but was profitable only where grass response exceeded grass/legume swards > 4 t/ha. S. sphacelata was the only grass to form a stable mixture with companion legumes. [Macroptilium atropurpurem (D.C.) Urban and Stylosanthes guianensis (Aubl.) Sw.], which yielded more than 1.1 t/ha in 3-year-old swards on loam soil. This combination was recommended for improvement of loam soil rangelands. It is concluded from this study that range improvement with perennial grasses is confined to fertile soils, while direct broadcast of Stylosanthes legumes only into grazed native pastures is the practice recommended for improvement of infertile sandy soil until persistent low P tolerant grasses are found for the dry tropics.

Summer habitat relationships of the pyrenean gray partridge (Perdix peridix hispaniensis) were studied in the northern Pyrenees Mountains (France). Six available habitat types were defined, and those selected or avoided were identified. The only habitat type significantly (P < 0.05) selected was at intermediate altitudes, on fairly steep south-exposed slopes, with a moderate cover of woody plants. Two habitat types were significantly avoided. One occurred at low altitudes on mowed plateaus colonized by low woody plants, and the other was at high altitudes on slopes free of low woody plants. We suggest applications of the work in a model that should lead to valid habitat recommendations for restoring partridge populations.

The germination ecology of antelope bitterbrush [Purshia tridentata (Pursh) Nutt.] seed has probably been investigated more than any other range shrub. Seeds of this valuable browse species are known to require moist prechilling before they will germinate. Our purpose was to investigate the nature of this dormancy breaking by placing packages (2 X 2-mm mesh screen) of seeds on the surface and buried in the seedbed at several locations in Idaho and Nevada and to recover the seeds monthly through the winter. The seeds were categorized based on their being: (a) capable of germinating; (b) dormant; or (c) dead at each recovery. The seedbeds of the 2 sites in Nevada, during 2 years of drought, were not sufficiently wet to bring large amounts of the antelope bitterbrush seeds out of dormancy. The seeds did not rot in the field, and being protected from predation, they remained dormant in the seedbed. The highest elevation site in Idaho had as high as 80% of the seeds lose dormancy. If seedbed microenvironmental conditions were satisfactory, the inherent seed dormancy was lost by midwinter. Snow cover, as it influences seedbed moisture and temperatures, apparently is an important factor in the prechilling of antelope bitterbrush seeds.

Incorporation of crop residues into the soil prior to planting has been shown to reduce the growth of subsequent crops. Information is limited on the allelopathic effect of sericea lespedeza [Lespedeza cuneata (Dum. Cours) G. Don.] residues on multiple cropping and rotational systems. Experiments were conducted to determine (1) if sericea lespedeza residues affected seed germination and plant growth of rye (Secale cereale L.), ryegrass (Lolium multiflorum Lam.), and tall fescue (Festuca arundinacea Schreb.); (2) if cultivars of these species varied in response to phytotoxins from sericea lespedeza residues; and (3) if N fertilization nullified the effects of residues. Germination experiments were conducted by using water extracts from low- or high-tannin sericea lespedeza residues, distilled water (control), and topsoil and subsoil obtained from areas in which low- or high-tannin sericea lespedeza plants had grown for 4 years. Greenhouse experiments showed that germination, emergence, seedling growth, above-ground biomass, and N concentration of rye and tall fescue were reduced by sericea lespedeza residues. Although ryegrass germination was not affected by the residues, biomass and N concentration were reduced. Rye and tall fescue germination were not affected by soils where sericea lespedeza previously had grown, but ryegrass germination and seedling growth of all 3 species were reduced. Immobilized N was the main factor limiting plant growth. Fertilizer-N more than compensated for the negative effects of the residues on all species. Establishment of rye and tall fescue in a sericea lespedeza field is likely to require higher seeding rates than normal to compensate for reduced germination, whereas ryegrass would not be affected. Fertilizer-N may be needed to enhance growth of grasses that otherwise would be curtailed by sericea lespedeza residues.

Leafy spurge (Euphorbia esula L.) represents a serious threat to the productivity and profitability of many cattle ranches in the midwestern U.S. Sheep and goats will graze the weed, but cattle generally do not. We hypothesized that cattle avoid leafy spurge because it contains 1 or more chemicals that elicit a conditioned flavor aversion when consumed. First, we tested cattle to determine if they reduced their intake of a novel feed on subsequent days if we modestly increased rumen fill by introducing additional feed or additional feed plus an aversive agent (lithium chloride, LiCl) after they had consumed the novel feed. We observed that cattle became averted (P = .0001) to the novel feed only when LiCl was administered with additional feed. Simply increasing rumen fill by a small amount did not cause cattle to reduce their intake of the novel feed the following day. Secondly, we tested cattle to determine if they reduced their intake of a novel feed on subsequent days if we introduced leafy spurge into their rumina following consumption of the novel feed. We also tested cattle to determine if a spurge-induced aversion to a novel feed was preventable by inoculation with rumen microbes from sheep with spurge in their diets. We found that introducing spurge into cattle after their intake of novel feed reduced (P < .01) their intake of the novel feed on subsequent days. Cattle inoculated with rumen microbes from spurge-adapted sheep had similar (P > .40) aversions to a novel feed paired with spurge introductions. Apparently, cattle avoid leafy spurge partly or wholly because they develop a conditioned aversion after first ingesting some threshold amount of it.

The effects of soil water deficit on root growth, carbon allocation, and plant mortality of broom snakeweed (Gutierrezia sarothrae (Pursh) Britt and Rusby) were studied during the spring-summer growing season in plants subjected to different soil water regimes. As soil water deficit developed, root length density decreased, indicating that water deficit reduced root proliferation. Root/shoot ratio remained unchanged (p > 0.05) as soil water potential decreased from -0.023 Mpa to -2 MPa; but it became higher (P < 0.05) in extremely stressed plants (-3.4 MPa), indicating that root growth was favored over shoot growth. Root length density was more closely correlated with green tissue dry weight/-stem dry weight ratio (r = 0.82, p < 0 .0001) than with root/shoot ratio (r = 0.52, p < 0 .05). This suggests that (1) expansion of photosynthetic area was more sensitive than stem growth to water deficit and (2) carbon allocation within the shoot was more sensitive to water deficit than allocation between root and shoot. Plants died when gravimetric soil water decreased to around 0.03 g g-1 (equivalent to a soil water potential of -7.5 MPa). The leaf relative water content just before death was about 0.50. Soil water content in the top 20 or 30 cm of the profile is the single most important factor determining mortality of the snakeweed plant and can be used in making decisions in snakeweed control programs.