• Impact of prescribed burning on vegetation and bird abundance at Matagorda Island, Texas

      Van’t Hul, J. T; Lutz, R. S.; Mathews, N. E. (Society for Range Management, 1997-07-01)
      We measured the impact of prescribed summer and winter burns on vegetation characteristics and spring abundance of birds in a Spartina/Paspalum grassland at Matagorda Island National Wildlife Refuge and State Natural Area, Texas, 1993-94. We burned 8 (4 summer burn, 4 winter burn), 122-ha plots. We estimated bird abundance by surveying once a week from March through May at 12-16 fixed-radius point count stations in each plot. We measured forb and grass foliar cover, litter depth, visual obstruction, and woody and residual stem density at each point count station 6-10 months after burning and 18-22 months after burning and found few differences in vegetation between summer and winter burns. Litter depth, visual obstruction, and woody stem density values were greater on control plots 6 to 10 months post-burn. By 18 to 22 months post-burn, only litter depth and visual obstruction remained higher on control plots than on either burn treatment. At 6 to 10 months after burning, wrens were more abundant on control plots and sparrows were more abundant on the burned plots. By 18 to 22 months post-burn, wren abundance had increased on the burned plots, but was still highest on control plots. Sparrow abundance remained highest on burned plots 18-22 months after burning. Precipitation was higher in 1993 than 1994; we believe blackbirds responded more to annual precipitation differences than to burning treatment. In this coastal island grassland, wren abundance was highest on unburned plots and sparrow abundance was highest on burned plots. We suggest that land managers could burn at > 2 year intervals in this grassland without negatively impacting most resident bird species.
    • Vegetation response to increasing stocking rate under rotational stocking

      Taylor, C. A.; Ralphs, M. H.; Kothman, M. M. (Society for Range Management, 1997-07-01)
      This 10-year study was designed to evaluate vegetation response to increasing stocking rates under rotational stocking (3 days graze, 51 days rest) and long-term rest. The 4 stocking rate treatments ranged from the recommended rate for moderate continuous grazing to 2.7 times the recommended rate. Common curly-mesquite [Hilaria belangeri (Steud.) Nash] increased (P = 0.05) in all grazed treatments and decreased in the livestock exclosure. Sideoats grama [Bouteloua curtipendula (Michx.) Torr.] along with other midgrasses decreased (P = 0.07) in all grazed treatments and increased in the livestock exclosure. Because the midgrasses were palatable species and not abundant, they were defoliated too intensively and too frequently. Rotational stocking was not able to sustain initial species composition at any of the stocking rates tested.
    • Viewpoint: Are grazing rights on public lands a form of private property?

      Raymond, L. (Society for Range Management, 1997-07-01)
      Some have argued that federal grazing preferences or permits are a form of private property and should be recognized as such by the federal government. This viewpoint studies the grazing controversy from 2 perspectives; legal and theoretical. A strict analysis of statutes and case law reveals some ambiguity in the law, but little that clearly supports the private property rights argument. A second analysis of several more theoretical approaches to the issue reveals a stronger case for private property based on the idea of customary use, as embodied in certain interpretations of the public trust doctrine and other alternative views. Because the non-legal argument is much more persuasive, it should be utilized more frequently by private property advocates. Opponents of the private property argument should consider that simple legal victory is often an inadequate solution to conflicts such as the grazing rights controversy. Advocates on both sides, as well as range managers and others simply wanting a better understanding of the issue, should make an effort to view the controversy from both of the perspectives presented here. Failure to do so will likely result in more lawsuits, more damaging controversy, and a continued lack of resolution to the conflict.
    • Structure and causes of vegetation change in state and transition model applications

      Rodríguez Iglesias, R. M.; Kothmann, M. M. (Society for Range Management, 1997-07-01)
      State and transition (ST) descriptions of rangeland vegetation dynamics provide information on current perceptions of explicit causes of change in dominant vegetation. Structural attributes of ST applications allow an evaluation of the complexity of the ST model and comparisons with the organization of the traditional succession-retrogression model of secondary succession. An analysis of 29 applications of the ST model revealed consistent trends. The number of transitions connecting states showed a less-than-expected increase with the size of the application. This is probably associated with limitations to interpret complex relationships and a need to produce relatively simple applications. Larger applications exhibited a shift towards stable states with pivotal positions within structures less connected (ie., with fewer transitions) than expected by chance for a given number of states. Thus, some stable states assume key intermediary roles as the number of states considered increases. It is debatable whether this is a property of larger systems or an effect of modeling bias. The analysis of causes of vegetation change confirmed current perceptions about the importance of man-related sources of disturbance. Grazing, fire, and control of woody plant species are visualized as the most relevant man-related agents of change. Some ST applications retain autogenic behaviors embedded in transitions in spite of the event-driven nature of the approach. However, the ST model removes autogenic processes from their central role as general causes for vegetation change. This approach is theoretically very limited because no general properties or attributes of the components (e.g., plant species assemblages, individual species) or processes (e.g., growth, reproduction, mineralization) of the system are used in any comprehensive way to generate predictive rules of wider than local relevance. Alternative approaches are suggested that would allow ecological generalizations and comparisons across systems.
    • Agricultural land use patterns of native ungulates in south-eastern Montana

      Selting, J. P.; Irby, L. R. (Society for Range Management, 1997-07-01)
      Mule deer (Odocoileus hemionus), white-tailed deer (Odocolieus virginianus), and pronghorn antelope (Antilocapra americana) use of 6 agricultural land use categories in southeastern Montana were monitored to identify use patterns at specific sites. Alfalfa (Medicago sativa L.), bottom rangeland, Conservation Reserve Program (CRP) lands, upland rangeland, wheat (Triticum aestivum L.) stubble, and growing wheat were observed during dawn, day, dusk, and night hours over a period of 12 months. Mule deer densities on alfalfa peaked in fall and again in spring. The CRP lands were selected in all seasons. Rangeland sites were most heavily used in winter and summer. White-tailed deer used CRP lands in all seasons except fall. Alfalfa was selected in fall, spring, and summer. Antelope densities on alfalfa were highest in spring and fall, while growing wheat fields were used most in spring. Antelope in the northern study area selected CRP land in all seasons except fall. Densities of animals and patterns of use observed during this study would be unlikely to produce significant impacts on forage or crops at most of our study sites.
    • Creating aversions to locoweed in naive and familiar cattle

      Ralphs, M. H.; Graham, D.; Galyean, M. L.; James, L. F. (Society for Range Management, 1997-07-01)
      The objective of this study was to determine if cattle that were familiar with white locoweed (Oxytropis sericea Nutt, ex T&G) could be aversively conditioned to avoid eating it. In the first preliminary trial, we tried to aversely condition native steers that were already eating locoweed. Six of 12 steers were penned, fed fresh-picked locoweed, then dosed via a stomach tube with lithium chloride (LiCl, 200 mg/kg BW). When released into the locoweed-infested pasture, they gradually increased locoweed consumption over the next 5 days. The conditioning procedure was repeated with a lower dose (100 mg/kg BW), but locoweed consumption increased within 10 days until they were consuming as much as the non-averted controls. In the second trial, we compared the strength and longevity of aversion between steers that were familiar with locoweed (n = 6) and naive steers (n = 6). Both groups were averted to locoweed as described in Trial 1 and returned to locoweed-infested pasture. The Familiar group decreased locoweed consumption for the first 2 days, then gradually increased locoweed consumption and extinguished the aversion. The Naive group subsequently refused to graze locoweed. In the third trial, aversions were reinforced following grazing locoweed in the pasture. Three steers from the Familiar group were allowed to graze locoweed for 30 min. periods, then were returned to the pen and dosed with LiCl (100 mg/kg BW). These steers were kept in the pen and allowed to recover for 36 hours. This reinforcement process following grazing was repeated 4 times. Steers in the Reinforced group abstained from eating locoweed when released into the locoweed-infested pasture for the remainder of the trial. Reinforcement of aversions following field grazing of locoweed prevented cattle that were familiar with locoweed from grazing it.
    • In vitro regeneration of fourwing saltbush [Atriplex canescens (Pursh) Nutt.]

      Mei, B.; No, E. G.; McWilliams, E. L.; Gould, J. H.; Newton, R. J. (Society for Range Management, 1997-07-01)
      A successful, in vitro regeneration system for fourwing saltbush [Atriplex canescens (Pursh) Nutt.] has potential use for mass propagation and genetic engineering of this important range species. This system could be used as a vehicle for gene transfer and for genetic modification of forage quality, stress tolerance, and biomass. The objectives were to develop an in vitro system for shoot production from both leaf discs and axillary buds, and induce root production. Shoot organogenesis (265 shoots) from leaf disc explants was accomplished at rates of 12.3 shoots/disc or 1.7 shoots/mm2 of leaf disc explants. Root organogenesis was induced in 63% (168) of the shoots, using indolebutyric acid (IBA, 0.5 mg liter-1) and gibberellic acid (GA3, 0.1 mg-1 liter) in a Murashige and Skoog (MS) medium. Roots were also induced on shoots placed in soil. Survival of both shoots and plantlets transferred to soil was 65%. Plantlets transferred to a hydroponic solution were twice the height of plantlets grown in soil. For axillary bud multiplication, the number of shoots generated ranged from 0.7 to 9.1 per explant. Roots were induced on 63% of the shoots, and 58% survived transfer to soil. The successful production of plantlets using both leaf-disc and axillary-bud methods demonstrates that this important range species can be mass propagated in vitro.
    • Habitat selection patterns of feral horses in southcentral Wyoming

      Crane, K. K.; Smith, M. A.; Reynolds, D. (Society for Range Management, 1997-07-01)
      Feral horse habitat selection patterns and forage attributes on available habitats were studied on public rangelands of southcentral Wyoming. Environmental assessments preceding roundup of excess horses requires resource data to justify the number of horses removed. Randomly selected bands of horses were followed for 24-hour observation periods during the spring and summer to determine if they utilized habitats in proportion to their abundance. We also determined if forage abundance, succulence (an index to forage palatability), percent utilization, and dietary composition were related to habitats selected. Streamsides, bog/meadows, and mountain sagebrush habitats were preferentially selected (p less than or equal to 0.05). Lowland sagebrush habitats were avoided and no apparent selection behavior was shown for grassland and coniferous forest habitats. Forage abundance, palatability, and percent utilization were higher (p less than or equal to 0.05) in streamside and bog/meadow habitat classes. Diet composition indicated that sedges (Carex sp.), common in streamsides and bog/meadows, were an important forage of feral horses. Palatability and abundance of graminoid vegetation and proximity to preferred habitats seemed to be the primary influences on habitat selection by feral horses.
    • Ownership and management changes on California hardwood rangelands: 1985 to 1992

      Huntsinger, L.; Buttolph, L.; Hopkinson, P. (Society for Range Management, 1997-07-01)
      Results of a 1985 survey of California hardwood rangeland landowners were used to develop a multi-agency research and extension program known as the Integrated Hardwood Range Management Program. In 1992, the same properties were re-surveyed. Although the results cannot prove the program is the sole or direct agent of change, program-sponsored education and research aimed at encouraging oak rangeland owners to change oak management practices is reflected in changes in key landowner behaviors. Program-sponsored research showed that intermediate levels of oak canopy cover did not significantly reduce forage production. Concurrently, landowners have significantly reduced the clearing of oaks for forage production. Other significant changes are reduction of cutting of living oaks for any reason, reduced cutting of oaks for fuelwood, increased use of oak promoting practices, and a growing awareness of the need to be concerned about the status of oaks. Landowners who were aware of the resource benefits of having oaks, or who believed oaks were threatened, or who had been in contact with a natural resource advisory service were significantly more likely to carry out oak-promoting practices. Between 1985 and 1992, many properties changed hands: 24% of parcels were sold but remained intact, while an additional 11% were subdivided. As was found in 1985, owners of smaller properties manage for different and more diverse goals than those of larger properties. The changing pattern of hardwood rangeland land ownership will have an impact on education and conservation programs.
    • Biomass and carbohydrates of spotted knapweed and Idaho fescue after repeated grazing

      Olson, B. E.; Wallander, R. T. (Society for Range Management, 1997-07-01)
      Spotted knapweed (Centaurea maculosa Lam.), an aggressive Eurasian forb, is replacing many native perennial grasses such as Idaho fescue (Festuca idahoensis Elmer.) on foothills of the Northern Rocky Mountains. We assessed biomass allocation, carbohydrate reserves (total nonstructural carbohydrate concentrations - TNC), and carbohydrate pools (TNC X biomass) as indicators of cumulative effects of 3 summers (1991-1993) of repeated sheep grazing on spotted knapweed and Idaho fescue. In early May 1994, we excavated 30 spotted knapweed and Idaho fescue plants previously exposed to repeated grazing and 30 ungrazed plants of each species. On grazed Idaho fescue plants, shoot (P < 0.02) and root (P < 0.06) biomass were 38 and 27% less than on ungrazed plants. In contrast, shoot (P = 0.26) and root biomass (P = 0.85) of grazed and ungrazed spotted knapweed plants were similar. Although grazing resulted in some minor differences in total nonstructural carbohydrate concentrations and carbohydrate pools of shoots, total nonstructural carbohydrate concentrations and pools of crowns and roots were similar for grazed and ungrazed plants of each species. Thus, carbohydrate concentrations or pools were not sensitive indicators of the response of Idaho fescue or spotted knapweed to the cumulative effects of repeated grazing. In contrast, aboveground biomass could be used to indicate the response of Idaho fescue to repeated grazing. By reducing shoot and root biomass of Idaho fescue but not spotted knapweed, repeated grazing may reduce the ability of Idaho fescue to compete with spotted knapweed when both species are grazed.
    • Degradability of Andrean range forages in llamas and sheep

      Genin, D.; Tichit, M. (Society for Range Management, 1997-07-01)
      In sacco dry matter degradability (DMD) of the most commonly consumed range forages by llamas and sheep in the arid highlands of Bolivia was measured during the wet and dry seasons to determine if llamas exhibit a higher digestive ability than sheep. Results showed that degradability of low quality forages (DMD below 60% in sheep) was 20 to 30% higher for llamas than sheep, while no significant differences were found for highly digestible forages. There was a high correlation between DMD in llamas and sheep with a coefficient of determination of 0.96. Parameters of degradation curves indicated that llamas did not have higher microbial activity than sheep, since there was no consistent difference in degradation rates of the studied forages. Nonetheless, significantly higher potential degradability and effective degradability found in this study suggested that the longer retention time in the forestomach of llamas may be responsible for higher digestibility of poor quality forages.
    • Effects of sheep grazing on a spotted knapweed-infested Idaho fescue community

      Olson, B. E.; Wallander, R. T.; Lacey, J. R. (Society for Range Management, 1997-07-01)
      Spotted knapweed (Centaurea Maculosa Lam.), a Eurasian perennial forb, is replacing many native perennial grasses, such as Idaho fescue (Festuca idahoensis Elmer.), in foothills of the Northern Rocky Mountain region. Our objective was to determine if 3 summers of repeated sheep grazing would reduce spotted knapweed without impacting the dominant, associated native perennial grass. Each summer, small pastures were grazed for 1-7 days in mid-June, mid-July, and early September. Areas repeatedly grazed by sheep had lower densities of seedlings, rosettes, and mature spotted knapweed plants than ungrazed areas. In addition, the proportion of young plants in the population was less in grazed than ungrazed areas. Basal areas of spotted knapweed plants were greater in grazed (8.2 cm2) than ungrazed areas (4.0 cm2). There were fewer spotted knapweed seeds in soil samples from grazed areas (12 seeds m-2) than from ungrazed (49 seeds m-2). Idaho fescue plant density increased 40% in grazed areas from 1991 to 1994, but leaves and flower stems on these plants were 38% and 17% shorter, respectively, than in ungrazed areas. By 1994, frequency of Kentucky bluegrass (Poa pratensis L.) was 35% greater in grazed than ungrazed areas. Grazing did not alter the amount of litter; however the amount of bare soil increased from 2.2 to 5.6% in grazed areas, while it decreased from 4 to 1% in ungrazed areas. Three summers of repeated sheep grazing negatively impacted spotted knapweed, but minimally affected the native grass community. A long term commitment to repeated sheep grazing may slow the rate of increase of spotted knapweed in native plant communities.
    • Grass hay as a supplement for grazing cattle. II. Ruminal digesta kinetics

      Villalobos, G.; Klopfenstein, T. J.; Adams, D. C.; Shain, D.; Long, T. E. (Society for Range Management, 1997-07-01)
      This study evaluated the effects of supplementing a diet of range hay (5.7% crude protein, 68% NDF) with grass hay from subirrigated meadows (16.5% crude protein, 53.5% NDF), or with a 70% soybean meal:30% wheat grain supplement (40% crude protein) on intake and ruminal digesta kinetics. Twelve ruminally fistulated steers were assigned to 3 treatments (4 steers/treatment) at 2 levels of intake. Treatments were: control, range hay; range hay supplemented with meadow hay (meadow hay was 20% of intake); and range hay supplemented with soybean meal:wheat supplement (supplement was 8% of intake). Intake levels were: ad libitum and equal intake (1.5% of body weight). Range hay was Yb-labeled, and meadow hay and soybean meal:wheat supplements were Er-labeled to measure passage. Intake and digestibility of range hay was not affected by supplementation (P > 0.05). During ad libitum intake, total intake (range hay + supplement) was greater (P > 0.05) for supplement treatments than for the control. No supplement treatment X level of intake interactions were detected (P > 0.05). Total digestibility (range hay + supplement) was greater (P < 0.01) for the soybean meal:wheat treatment than for the control or meadow hay treatments. Total digestibility was similar (P > 0.05) for control and meadow hay treatments. Ruminal passage rate (% hour-1), total tract mean retention time, and intestinal transit time of range hay did not differ among treatments (P > 0.05), but ruminal passage rate, total tract mean retention time, and intestinal transit time were greater (P < 0.01) with ad libitum than equal intake. We conclude that a meadow hay supplement produced similar effects on ruminal kinetics and intake of range hay as a soybean meal:wheat supplement.
    • Grass hay as a supplement for grazing cattle. I. Animal performance

      Villalobos, G.; Adams, D. C.; Klopfenstein, T. J.; Nichols, J. T.; Lamb, J. B. (Society for Range Management, 1997-07-01)
      Regrowth grass hay produced on subirrigated meadows in the Nebraska Sandhills was evaluated as a supplement for gestating beef cows grazing winter range. Ninety-six crossbred spring calving, gestating beef cows were used in a winter supplementation study on upland Sandhills range from 5 November to 27 February in 1990 and again in 1991. Cows were divided into 4 treatments (24 cows/treatment): 1) control (range forage only, no supplement); 2) range forage and 2.2 kg cow-1 day-1 of meadow regrowth hay (15.5% crude protein); 3) range forage and 1.2 kg cow-1 day-1 of a 30% wheat grain and 70% soybean meal:30% wheat supplement (36.0% crude protein); and 4) range forage with supplements in treatments 2 and 3 fed on alternate days. Meadow hay and soybean meal:wheat supplements provided 0.32 kg of crude protein/cow daily. Supplemented cows gained 3 to 53 kg body weight/year and maintained body condition, while control cows lost an average of 24.5 kg body weight/year and lost body condition. Intake of range forage was less (P < 0.05) by cows fed meadow hay and soybean meal:wheat supplements on alternate days than by cows on other treatments. Digestibility of range forage was lower (P < 0.05) for supplemented cows than control cows, but differences were small (avg. = 2%). Calving date, birth and weaning weights, and pregnancy rate were similar (P > 0.05) for all treatments. We concluded that subirrigated meadow regrowth grass hay was an effective alternative to traditional soybean meal-based supplements for maintaining body weight and body condition of gestating beef cows grazing winter range.
    • Damage from the larkspur mirid deters cattle grazing of larkspur

      Ralphs, M. H.; Jones, W. A.; Pfister, J. A. (Society for Range Management, 1997-07-01)
      The larkspur mirid (Hopplomachus affiguratus) is host specific to tall larkspur (Delphinium barbeyi Huth). These insects suck cell solubles from leaves and reproductive racemes, causing flowers to abort and leaves to desiccate. Cattle generally prefer lark spur flowering heads, pods, and leaves, and are frequently poisoned. The objective of this experiment was to determine if cattle would avoid eating mirid-damaged larkspur. A 2-choice cafeteria feeding trial was conducted to determine preference for mirid-damaged and undamaged larkspur. Four cows were offered a choice of the 2 types of larkspur in 10 min. feeding trials in the morning and evening for 5 days. Cows preferred the undamaged larkspur plants (0.8 kg +/- 0.08, SE) over mirid-damaged plants (0.1 kg +/- 0.03, SE). The cows were then turned out into a larkspur-infested pasture and consumption of mirid-damaged and undamaged larkspur was quantified by bite count. The cows did not select any mirid-damaged larkspur. Consumption of undamaged larkspur peaked at 17% of bites on the second day of the grazing trial, then declined as mirid damage on the plants increased. If the density of mirids on larkspur is sufficiently high to damage most of the leaves and flowering racemes, grazing by cattle may be deterred, and subsequent poisoning avoided.
    • Early season grazing by cattle of tall larkspur-(Delphinium spp.) infested rangeland

      Pfister, J. A.; Ralphs, M. H.; Manners, G. D.; Gardner, D. R.; Price, K. W.; James, L. F. (Society for Range Management, 1997-07-01)
      A series of summer grazing studies were conducted to evaluate cattle consumption of preflowering tall larkspur (Delphinium barbeyi [Huth] or D. occidentale [Wats.] Wats.) on mountain rangeland in Utah, Colorado, and Idaho during 1987 to 1995. Cattle ate little or no larkspur before flowering racemes were elongated. Consumption of tall larkspur by cattle was also generally low during the early flower stage, with some notable exceptions at the Salina and Manti, Utah sites. These grazing studies indicate that risk of losing cattle to tall larkspur is low if plants have not flowered. Even though concentration of toxic alkaloids is typically much higher in immature compared to mature tall larkspur, toxicosis is unlikely to occur because consumption by cattle is low. Many livestock operations can gain 4 to 5 weeks of low-risk grazing on tall larkspur-infested rangeland early in the grazing season, and this should be considered in developing grazing management plans.
    • Effect of seed moisture on Wyoming big sagebrush seed quality

      Bai, Y.; Booth, D. T.; Roos, E. E. (Society for Range Management, 1997-07-01)
      Seed germination and seedling vigor of Wyoming big sagebrush (Artemesia tridentata Nutt. ssp. wyomingensis) were evaluated following manipulation of seed moisture, a practice benefitting many species. At the time of harvest, seed moisture ranged from 2.3 to 9.0% for 5 collections tested and seeds with moisture between 5 to 6% had the highest and most rapid germination. Seed moisture changed during storage, but germination percentage was not affected by post-harvest seed moisture change, indicating that germination is related more to habitat or genetic variations than the initial moisture content. Seedling vigor increased after storage, suggesting that after-ripening may be required. Seeds of 2 commercial collections were subsequently humidified at 2, 5, 10, and 15 degrees C for up to 15 days, or to 60% moisture content. Seed moisture increased most gradually at 2 degrees C and seeds held at 10 degrees C attained a higher moisture level than at other temperatures. Germination percentage, germination rate, and seedling vigor were similar between treatments and controls regardless of seed moisture change. Imbibition temperature did not affect germination percentage or seedling vigor, but the time to 50% germination decreased with increasing imbibition temperature. We conclude that artificial seed moisture management did not affect germination percentage, germination rate, or seedling vigor of this species when tested under optimum moisture conditions. Germination is more related to habitat or genetic variables than initial seed moisture content.
    • Persistence of aversions to larkspur in naive and native cattle

      Ralphs, M. H. (Society for Range Management, 1997-07-01)
      The objective of this study was to create and maintain a long lasting aversion to tall larkspur (Delphinium barbeyi Huth) under field grazing conditions. Two grazing trials were conducted: the first used cattle naive to larkspur, and the second used cattle that were familiar with larkspur. In the first trial, 12 mature cows that were not familiar with larkspur were randomly allocated to Averted and Control groups (n = 6). They were fed larkspur (a novel food) and the Averted group was dosed with lithium chloride (200 mg/kg body weight) to create the aversion. Both groups were then taken to larkspur-infested mountain rangeland where they grazed in 2 separate pastures during the late summer of 1993, 1993, 1994, and 1995. The aversion was not reinforced in 1994 or 1995. Diets were quantified by bite count. The Averted group abstained from eating larkspur for 3 years. The Control group grazed larkspur for an average of 14% of bites. Three Control cows died from larkspur poisoning in 1993, and another cow was poisoned, but survived in 1994. During the last 2 weeks of the 1995 trial, the Averted and Control groups were placed together, and the social influence of the Control cows eating larkspur caused the Averted cows to sample larkspur and gradually extinguished the aversion. In the second trial (1994 and 1995), 5 native cows that had grazed on the allotment and were familiar with larkspur were averted to larkspur by the procedure described above. They abstained from eating larkspur while grazing separately, but extinguished the aversion when placed with non-averted Control cows at the end of the study in 1995. Aversions are retained in long term memory and may last indefinitely if averted cattle graze separately.