We studied the effects of vegetation characteristics in southern Texas on site selection by mature, male white-tailed deer (Odocoileus virginianus Raf.). Thirteen, radio-collared animals were monitored during winter, spring, summer, and fall of 1986-87 and 1987-88 to determine area-usage patterns within each animal's respective seasonal home range. After each season, structural vegetation attributes were measured with transect-oriented data collection techniques inside the most heavily used and unused areas of each animal's home range. Comparisons were made between these areas to determine whether site selection by deer was in response to differing vegetation characteristics. In general, the most heavily used areas possessed a greater amount of woody canopy cover (greater than or equal to 85%), woody species richness (18-20), and horizontal screening cover than areas with no use. In contrast, herbaceous densities did not differ between the most heavily used and unused areas. Consequently, habitat management manipulations conducted specifically for mature male white-tailed deer in southern Texas, should include provisions for creation or maintenance of sites possessing dense woody canopy cover, a high number of woody species and dense horizontal screening cover.

Plains pricklypear cactus (Opuntia polyacantha L.) is abundant on the Central Great Plains with dry matter yields from 1,500 to 2,000 kg/ha. Cactus spines prevent cattle from grazing as much as 50% of the herbage around the plant. Pricklypear pads are quite palatable once spines have been removed. The possibility of simultaneously controlling and feeding plains pricklypear led to development of machinery for harvesting cactus. The harvesting machine is a side-delivery rake modified to uproot and windrow pricklypear which is later despined and fed to cattle. Machine harvesting was compared to hand harvesting on both a sandy loam and a clay loam site. There was no significant difference in cactus removal between hand and machine harvested plots or significant damage to desirable forage species. Pricklypear phytomass removal by the harvester averaged 89% and 88% on the sandy and clay loam sites, respectively. This provided an average of 1,166 kg/ha cactus as potential feed and increased availability of desirable forage species.

The slow and uncertain rate of recovery of plant communities after severe disturbance is a major problem on rangelands. Earlier studies sketched the outline of secondary succession on mixed-grass prairie, but were based on 1 or 2 years of observation on different areas disturbed at different times in the past, or several years of observation of a single area. To provide a more complete picture of succession over decades, we began observations in 1977 on 4 areas disturbed from 1 to 51 years previously, and on undisturbed areas of the same 2 soil types with and without grazing. Observations continued for 11 years. Secondary succession proceeded through the usual stages: annual forbs, perennial forbs and annual grasses, short-lived perennial grasses, and long-lived grasses. Western wheatgrass [Pascopyrum smithii (Rydb.) A Love] was an exception because it appeared much earlier and in much greater abundance than other long-lived perennial grasses. Blue grama [Bouteloua gracilis (H.B.K.) Lag ex. Steud.) may be another exception; total recovery of this grass may require centuries. Time of appearance in succession seemed to be related to availability of propagules and ease of establishment; persistence of species was related to competitive ability. Abundance of many species fluctuated widely from year to year, but fluctuations did not appear to be related to precipitation. After 61 years, secondary succession had not returned plant communities to the climax state.

Improved regression models were developed to predict winter forage production from big sagebrush (Artemisia tridentata Nutt.) through consideration of the subspecies variation among mountain big sagebrush (A.t. ssp. vaseyana [Rydb.] Beetle), Wyoming big sagebrush (A.t. ssp. wyomingensis Beetle and Young), and basin big sagebrush (A.t. ssp. tridentata). Changes in shrub morphology from browsing were also accommodated in our models. Colinearities among some variables used in previous studies were found and avoided in our models. Models used easily measured objective variables of which major axis and average cover of shrubs were most useful. Multivariable models without colinearities were evaluated on the basis of their R2a values which increased by an average of 10% to near 0.90, with taxa and browse form class included, compared to a model ignoring these differences.

Abandoned bentonite mine spoils are extremely difficult to revegetate because of their high clay content, salinity, sodicity, low permeability, and the semiarid climate of the area where bentonite mining occurs. Recent research has led to the development of technology utilizing sawmill wastes (chips, bark, and sawdust) to enable the successful revegetation of these lands. The use of wood residue amendments increased water infiltration, leaching of soluble salts, and vegetation establishment; however, sodicity continued to be a problem and threatened to destroy the established vegetation. Surface application of gypsum was evaluated to determine its effectiveness in ameliorating the spoil sodicity and its effect on plant growth. In a 3-year field study, surficial gypsum amendment resulted in significant increases in perennial grass biomass (150%) and canopy cover (140%). These changes were not evident until the second or third year after gypsum amendment. Annual forb biomass did not respond to gypsum amendment; however, canopy cover did exhibit a significant increase in the second year at lower wood residue amendment rates. This research demonstrates that surface applied gypsum can be effective in ameliorating bentonite spoil sodicity when applied to established plant communities.

Defoliation patterns of little bluestem [Schizachyrium scoparium (Michx.) Nash] on tallgrass prairie were compared using continuous and rotational grazing systems on six 24-ha pastures for each system over a range of stocking rates (0.28 to 0.49 AU ha-1) in 1991 and 1992. We tested the generalization that rotational grazing provides greater managerial control over the frequency, intensity, and uniformity of tiller defoliation compared to continuous grazing. Rotational system pastures were subdivided into 8 paddocks with 4 grazing cycles (3-7 day graze periods) per grazing season. Tillers were sampled biweekly in continuous system pastures and at the beginning, midpoint, and end of each grazing period in rotational system pastures. Multiple regression prediction equations were developed for grazed height, number of defoliation events in a grazing season, percent of tillers defoliated per sampling period (of continuous system) and grazing cycle (for rotation system), and number of defoliation events within a grazing cycle (rotational system). Grazed height decreased as stocking rate increased, but was not influenced by grazing system. The number of cumulative defoliation events per tiller increased with increasing stocking rate over the grazing season. Under similar stocking rates, a higher percentage of tillers were defoliated during the grazing season in the continuous than in the rotational grazing system. Within both grazing systems, percentage of tillers defoliated increased with increasing stocking rates. The percentage of tillers defoliated biweekly in continuous system pastures was similar over the grazing season; the percentage of tillers defoliated per cycle increased as grazing periods lengthened in rotational system pastures. A large number of tillers were defoliated during the second half of each grazing period. Less than 10% of tillers were regrazed within a grazing cycle, even at the highest stocking rate and longest grazing period. Rotational grazing provided greater managerial control over the frequency and uniformity of tiller defoliation; intensity of tiller defoliation was similar between the 2 grazing systems. We hypothesize higher range condition will be maintained over the long-term in rotational system pastures as little bluestem will remain more competitive and productive resulting from fewer defoliation events throughout the grazing season.

Studies of harvest date by fertilizer interactions on hay meadows are rare and none have been published for prairie meadows. Our objective was to evaluate the effects of initial harvest date (15 June, 15 July, and 15 August) and spring-applied N (0, 45, 90, and 135 kg ha-1) on first cutting and regrowth dry matter yield and forage quality from native and interseeded wetland meadow sites. Regrowth was harvested on all plots in late September. 'Garrison' creeping foxtail (Alopecurus arundinaceus Poir.) was interseeded on plots 4 years prior to application of treatments. Native vegetation was dominated by sedges (Carex spp.). Interseeded plots were dominated by Garrison creeping foxtail. Yield and quality on different dates and response to N were similar for vegetation types despite differences in duration of spring flooding between years. Harvest date by fertilizer interactions occurred for first cutting yield and crude protein concentration. Yield response to applied N ranged from 8.5 to 31.2 kg ha-1 kg-1 N. Fertilizer had no effect on digestibility and increased crude protein concentration only in herbage harvested on 15 June. Within levels of N, first cutting yield was about 60% of peak standing crop on 15 June and 90% on 15 July compared with 15 August. Greater plant growth rates and response to N after prolonged spring flooding compensated for initial differences between years by 15 July. Regrowth dry matter yield was not affected by spring-applied N and increased by about 43 kg ha-1 day-1 after initial harvest in both years. Sedge-dominated, prairie meadows are productive and provide predictable forage and wildlife habitat management alternatives.

Botanical composition of llamas and sheep diets were quantified monthly during 1 year in the arid highlands of Bolivia to identify competition between these species for forage resources. Results indicated higher proportions of coarse bunchgrasses in llamas diets (48 to 75%) than in sheep (37 to 68%), while sheep consumed more soft herbs and grasses than llamas (25 to 45%, and 8 to 25%, respectively). Llamas had higher (P < 0.05) digestion coefficients than sheep for organic matter, dry matter, crude protein, and fiber fractions of the principle bunchgrass paja brava (Festuca orthophylla) during the vegetative phenological stage. Shrubs represented less than 20% of the diet components in both llamas and sheep. A canonical discriminant analysis showed that there was not a strong dietary overlap between these species, and suggested that mixed herds could allow a better utilization of the overall available forage.