• Effect of grazing and cultivation on some chemical properties of soils in the mixed prairie

      Dormaar, J. F.; Willms, W. D. (Society for Range Management, 1990-09-01)
      Components of the organic matter were studied in soil under 3 Mixed Prarie types: grassland dominated by needle-and-thread/blue grama (Stipa comata Trin. and Rupr./Bouteloua gracilis (HBK.) Lag. ex Steud.) in good range condition; grassland significantly modified by grazing, dominated by blue grama and in poor range condition; and grassland, dominated by needle-and-thread/blue grama in good range condition, but converted to cropland and under continuous wheat for 4 years. The soils were sampled on 13 April 1988. Concentrations of total organic carbon in the upper 2 cm were 1.39, 2.70, and 1.87%, respectively. The higher organic carbon under blue grama was caused by an active, ramified, fine rootmass which gave rise to most of the monosaccharides being of microbial origin. The monosaccharides in the lower Ap horizon in the cropland were generally of plant origin from incorporated straw. The Ah horizons of the needle-and-thread/blue grama and blue grama sites and the Ap horizon of the cultivated site yielded 244, 696, and 370 migrograms g-1 organic acids in the alkaline-soluble fraction of the soil, respectively. Although most of the organic compounds identified were present in all 3 soils, the quantitative patterns were quite different. Differences exist due to inputs by different species, and 4 years of cropping also made significant soil chemical changes. This study demonstrated the importance of recognizing the history of the soils studied when describing soil quality.
    • Effects of herbage allowance on defoliation patterns of tallgrass prairie

      Jensen, H. P.; Gillen, R. L.; McCollum, F. T. (Society for Range Management, 1990-09-01)
      Few studies have dealt with measuring individual plant defoliations in the context of intensive grazing management. In May, July, and August of 1987, grazing trials were conducted to quantify the effects of herbage allowance on defoliation patterns of big bluestem [Andropogon gerardii Vitman], Indiangrass [Sorghastrum nutans (L.) Nash], and little bluestem [Schizachyrium (Michx.) Nash]. Herbage allowances of 10, 20, 30, and 40 kg AUD-1 were replicated twice per trial. Tiller height, relative leaf area removed, and lrequency of defoliation were measured every 2 days over 10-day trials. Indiangrass was the most preferred species in all trials. The rate of leaf area removal increased as herbage allowance decreased. Current guidelines which call for rest periods of 30-90 days would result in light to moderate intensity defolia- tion for indiangrass at all herbage allowances. The maximum percentage of tillers grazed only once per trial ranged from 20 to 98% depending on herbage allowance, species, and trial. Selectivity between species was reduced by decreasing herbage allowance but this effect was not large until herbage allowance was below 20-25 kg AUD-1 and selectivity was never completely removed. Grazing all tillers only once in a grazing period, even within a species, is unlikely in a tallgrass prairie community. Leaf area removal was moderate the first time a tiller was defoliated and severe for later defoliations. The goal of grazing any individual tiller at no greater than moderate intensity within a grazing period would he roughly equivalent to grazing any tiller no more than once. However such a goal would require many tillers to go ungrazed.
    • Ermelo weeping lovegrass response to clipping, fertilization, and watering

      Masters, R. A.; Britton, C. M. (Society for Range Management, 1990-09-01)
      A management strategy using short-duration grazing and fertilization was simulated in a study with individual ‘Ermelo’ weeping lovegrass [Eragrostis curvula (Shrad.) Nees] plants. Influence of 2 levels of clipping [unclipped during the growing season and clipped (C) to 10-cm stubble height each time regrowth reached 40 cm]; fertilizer [unfertilized and 70-34-44 kg N-P-K/ha (F)]; and watering frequency [irrigated to field capacity at 7- (WET) and 14 (DRY) day intervals] on cumulative herbage dry matter yield, crude protein yield, and water-use efficiency, and root mass of individual weeping lovegrass plants grown in soil contained in polyethylene tubes was determined. Clipping combined with fertilization improved herbage dry matter yield and water-use efficiency. Fertilized plants yielded at least 5.4 g crude protein/tube as compared to less than 2.0 g crude protein/tube produced by unfertilized plants. Between 1 June and 15 September 1983 WET-F- treated plants provided sufficient regrowth for 5 harvest events with an average of 24-day intervals between harvests. In contrast, regrowth of WET-, DRY-F, and DRY-treated plants was harvested 3 times with intervals between harvests averaging between 32 and 35 days. Clipping had no effect on root mass of WET-, DRY- F-, and DRY-treated plants, but reduced root mass of WET-F- treated plants by 44%. Based on this simulation of a fomge management strategy, periodii harvest of weeping lovegrass combined with fertilization improved herbage dry matter yield and water-use efficiency without adversely affecting root mass when interval between harvest events avenged 32 to 35 days.
    • Genetic variability for elements associated with grass tetany in Russian wildrye

      Asay, K. H.; Mayland, H. F. (Society for Range Management, 1990-09-01)
      Grass tetany (hypomagnesemia) may be an important factor limiting productivity of animals grazing Russian wildrye [Psathyrostachys juncea (Fisch.) Nevski]. This malady is associated with relatively low concentrations in the forage of Mg and Ca, and high values for K and K/(Ca+Mg). We studied the genetic variability in a Russian wildrye breeding population for mineral elements that relate to grass tetany in ruminants. Forty-five progeny lines, estblished as spaced plants in a randomized complete block, were sampled at the pre-boot and boot stages in each of 2 years and analyzed for Mg, Ca, K, and P. Although seasonal variation was evident, K/(Ca+Mg) of the progeny lines ranged from 3.2 to 4.6, well above the 2.2 level at which a 5% incidence of grass tetany has been found in dairy cattle. With few exceptions, progenies differed for all traits evaluated. Differences among progenies were relatively consistent over harvests for all traits. A reduced tetany potential (RTP) was computed as the sum of normalized Mg and reciprocal of K/(Ca+Mg) values, providing an estimate of the grass tetany risk for individual progeny lines. The variation among progenies, and the magnitute of broad-sense heritability estimates for RTP (0.48) and K/(Ca+Mg) values (0.31), indicate that mineral ion composition of this breeding population can be altered through breeding. The high K/(Ca+Mg) values in the population suggest that it may be helpful to introduce genetic factors conditioning lower grass tetany potential from other sources. Intercharacter correlations suggest that breeding for higher levels of Mg will be accompanied by increased Ca and, to a lesser extent, increased K.
    • Overstory-understory relations in pinyon-juniper woodlands in New Mexico

      Pieper, R. D. (Society for Range Management, 1990-09-01)
      Herbage biomass for blue grama (Bouteloua gracilis [H.B.K. Lag]), pinyon ricegrass (Piptochaetium fimbriatum [H.B.K.] Hitche.), New Mexico muhly (Muhlenbergia pauciflora Buckl.), other grasses and forbs was estimated on 25 pinyon-juniper stands of varying overstory cover on the Fort Stanton Experimental Ranch in south-central New Mexico. Negative 2nd degree polynomial curves best expressed the relationships between total understory and blue grama biomass and overstory canopy cover. Positive polynomial relationships were shown for cool-season graazs, New Mexico muhly, and pinyon ricegrass. Reducing pinyon-juniper canopy cover would likely increase blue grama production and reduce production of New Mexico muhly and pinyon ricegrass.
    • Picloram release from leafy spurge roots

      Hickman, M. V.; Messersmith, C. G.; Lym, R. G. (Society for Range Management, 1990-09-01)
      Picloram (4-amino-3,5,6-trichloro-2-pyridineacarboxylic acid) release from leafy spurge (Euphorbia esula L.) roots was not affected by application rate, root system temperature, or addition of 2,4-D [(2,4-dichlorophenoxy)acetic acid]. Release of 14C from leafy spurge roots was detected 12 hours after 14C-picloram foliar application and increased linearly over a 120-hour period. Over all experiments, 72% of the recovered 14C remained unabsorbed on the treated leaf and 22% remained in the leaves and stems. Less than 7% of the recovered 14C was in the root zone (roots plus nutrient solution), but over 60% of this portion was in the nutrient solution. Adding 2,4-D at rates up to 1.1 kg/ha to 14C-picloram at 0.14 kg/ha did not affect 14C release from leafy spurge roots. The experimentally determined temperature coefficient (Q10) for 14C-picloram release from leafy spurge roots was 1.3+/-0.8. A linear rate of picloram release with time and a Q10 of 1.3 support the hypothesis of passive release of picloram from leafy spurge roots.
    • Role of irrigation and fertilization in revegetation of cold desert mined lands

      Powell, K. B.; Vincent, R. B.; Depuit, E. J.; Smith, J. L.; Parady, F. E. (Society for Range Management, 1990-09-01)
      This study determined responses of vegetation and soils to different rates and seasonal schedules of first-year irrigation in combination with varied N-P fertilization on cold desert mined lands. Certain irrigation treatments increased soil water content initially, but had no appreciable effects on soil salinity or fertility. Specific rates and schedules of irrigation temporarily benefited total stand and dominant perennial grass establishment and productivity, but treatment effects diminished or reversed over time. Subdominant shrubs and perennial forbs were more persistently enhanced by specific irrigation treatments. Fertilization did not modify plant response to irrigation regimes. Although annual species were positively influenced by fertilization with heavier rates of irrigation, such stimulation proved ephemeral and perennial species never responded to fertilization under any irrigation regime.
    • Root length leaf area, and biomass of crested wheatgrass and cheatgrass seedlings

      Svejcar, T. (Society for Range Management, 1990-09-01)
      Cheatgrass (Bromus tectorum L.) dominates large tracts of rangeland in the western United States. Previous research has demonstrated the competitive nature of this species; however, the mechanisms contributing to its dominance have not been well elucidated. It is often suggested that cheatgrass outcompetes perennial seedlings because it germinates and grows at lower soil temperatures. However, even in studies where temperatures are not limiting, cheatgrass outcompetes perennial seedlings. Therefore, I conducted a study to compare accumulation of root length, leaf area, and biomass of cheatgrass and crested wheatgrass (Agropyron desertorum (Fisch. ex Link) Schult. cv. Nordan) seedlings under non-limiting conditions. Seedlings were grown in pots in the greenhouse for 60 days post-sowing. There were 4 seedlings per pot, and sampling was conducted weekly at 24-60 days after sowing. Maximum and minimum temperatures were 280 degrees C and 4 degrees C, respectively, and plants were watered twice a week. Cheatgrass had greater root length density and leaf area than created wheatgrass, especially during the later samplings. For the last 2 samplings, cheatgrass averaged about 12% more root mass and 56% more shoot mass, yet bad more than twice the root length and leaf area of crested wheatgrass. Cheatgrass was more efficient (per unit of biomass) in producing leaf area and root length, which helps explain its ability to quickly become established and exploit soil nutrient and moisture reserves.
    • Seedling growth rate of 3 subspecies of big sagebrush

      Booth, G. D.; Welch, B. L.; Jacobson, T. L. C. (Society for Range Management, 1990-09-01)
      Differences in growth rate among 3 subspecies of big sagebrush (basin big sagebrush, mountain big sagebrush, and Wyoming big sagebrush) for mature plants have been reported by a number of workers. Little has been reported on comparisons of seedling growth rate among these 3 subspecies. Results of this study indicated that (1) over an extended period and in a non-water-limiting environment, the rate of seedling growth In Wyoming big sagebrush approached zero and was less than that of seedlings of basin and mountain big sagebrush; (2) basin and mountain big sagebrush continued to have nonnegligible growth rates even at the end of the study; and (3) Wyoming big sagebrush reached its point of maximum growth rate approximately 2 weeks earlier than did the other 2 subspecies. It appears that Wyoming big sagebrush has evolved, placing 2 important growth characteristics under genetic control: (1) the maximum growth rate is attained earlier (when more water is available to sustain such growth) than in the other 2 subspecies, possibly enhancing its ability to survive on xeric sites during the early stages of growth; and (2) top growth produces smaller aboveground parts, enhancing survival after the early stages.
    • Sheep grazing behavior as affected by supplementation

      Hatfield, P. G.; Donart, G. B.; Ross, T. T.; Galyean, M. L. (Society for Range Management, 1990-09-01)
      Behavioral characteristics of supplemented (S; 23% crude protein, 3.41 mcal DE/kg) and nonsupplemented (N) ewes grazing native New Mexico rangeland were monitored from December 1983 to August 1984. Four marked ewes were observed per treatment group. Behavioral data wera collected by simutaneous observations of both treatment groups by 2 observers starting 0.5 hour before sunrise, and continuing until 0.5 hour after sunset. Trials were conducted during the winter breeding season while ewes grazed dormant forage, during spring lambing that coincided with the onset of active forage growth, and during summer lactation, during both a quiescent period and active forage growth. Supplement was not fed during the summer. Total grazing (min/- day) was less (P=0.01) in winter, equal in spring, and greater (P=0.01) in summer for S than N ewes. Mean daily periods of loafing for S ewes was greater (P=0.01) in winter; equivalent in spring, and less (P0.01) in summer than for N ewes. Daily travel time did not differ between S and N ewes during the winter; however, S ewes traveled more (P=0.02) in spring and less (P=0.01) in summer than N ewes. Supplemented ewes tended (P=0.06) to weigh less after the winter breeding season, and weaned similar, but slightly heavier, lambs (P=0.14) than N ewes. There were no differences (P>0.22) in ewe live weights during the remainder of the study. Mean fleece weights were not different (P=0.52) between S and N ewes. Supplemental feeding of range ewes, under the conditions of this study, did not appear to improve overall production by S ewes. Slower weight gains for S ewes during breeding may have resulted from reduced grazing time noted for this period.
    • Supplementation and monensin effects on digesta kinetics. I. Cattle grazing summer range

      Ward, M. G.; Adams, D. C.; Wallace, J. D.; Galyean, M. L.; Knapp, B. W. (Society for Range Management, 1990-09-01)
      Sixteen ruminally cannulated beef steers grazing native summer range in the Northern Great PIains were assigned to 4 treatments in a 2 X 2 factorial arrangement. Main effects were barley grain and monensin. Rolled barley (RB) containing 7.5% molasses was fed at 0 and 1.36 kg head-1 day-1. Steers received no monensin (M) or M released at 101 mg/d via a ruminal delivery device. Forage intake and digestibiIity, ruminal fermentation, and ruminal passage rate were measured during trials in: (1) June, (2) July, and (3) August. Diet samples were collected from esophogeally fistuIated steers during each trial. Dietary crude protein was greater (P<0.05) during trials 1 (15.2%)and 3 (14.3%) than in trial 2 (10.2%). In vivo organic matter (OM) digestibiiity, ruminal fluid passage rate, and fermentation variables varied by trial (P<0.01). Forage OM intake was reduced (P<0.10) by RB, but was not influenced (P>0.10) by M or the M and RB combination. In vivo OM digestibility was increased (P<0.05) by M, while RB had no effect. Particulate passage was not affected by M or RB but gastrointestinal tract fill was reduced by monensin (P<0.05). Ruminal fluid passage rate was affected by the RB X M X Trial interaction (P<0.05). Within June and July, fluid passage rate was similar among treatments and ranged from 14.0 to 11.3 %/h, respectively. During trial 3, a RB X M interaction (P<0.05) increased fluid passage rate. Ruminal ammonia-N concentration was similar among treatments. Barleey lowered (P<0.05) ruminal pH and increased (P<0.10) total volatile fatty acids. A RB X M X Trial interaction (P<0.05) was noted for molar proportions of acetate, propionate, and butyrate. Within trials, RB, M, and their combination affected (P<0.01 to P<0.10) acetate, propionate, and butyrate. We conclude that barley, monensin, and forage quality infiuence ruminal fermentation, passage rate, and intake traits of steers grazing summer range.
    • Supplementation and monensin effects on digesta kinetics. II. Cattle grazing winter range

      Ward, M. G.; Adams, D. C.; Wallace, J. D.; Galyean, M. L.; Volesky, J. D. (Society for Range Management, 1990-09-01)
      Sixteen ruminally cannulated steers grazing Northern Great Plains native winter range were assigned to 4 treatments in a 2 X 2 factorial arrangement. Main effects were protein and monensin. A soybean meal-barley pellet (P; 26% crude protein) was fed at 0 and .8 kg head-1 d-1. Steers either received no monensin (M) or M was released at 101 ml/day via a ruminal delivery device. Forage intake, rmninal fermentation, in vivo organic matter (OM) digestibility, and ruminal fluid passage and particulate digesta kinetics were measured during trials in November and January. Esophageally fistulated steers were used to collect diet samples during each trial. Dietary crude protein was greater (P<0.01) in November (8.3%) than January (4.9%). Forage OM intake was not (P>0.10) influenced by either P or M. In vivo OM digestibility was increased (P<0.05) by P (60.6 vs 57.4%) and not affected (P>0.10) by M. Particulate passage rate increased (P<0.05) when P was combined with M. Ruminal fluid flow characteristics, fluid volume and pH were not affected (P>0.10) by either P or M. Ruminal ammonia-N was increased (P<0.01) by P (2.9 vs .6 mg/100 ml) and not affected (P>0.10) by M. Total ruminal volatile fatty acid concentrations, along with molar proportions of ruminal propionate and butyrate, were not affected (P>0.10) by P or M. Ruminal acetate was decreased (P<0.10) by P and not influenced (P>0.10) by M. We conclude that supplemental protein, through ruminal modifications, has beneficial effects on OM digestibility, and can thereby provide cattle grazing winter range with additional energy at a time when it is most crucial.
    • Survival and growth of globemallow [Sphaeralcea] species in dry land spaced-plant nurseries

      Pendery, B. M.; Rumbaugh, M. D. (Society for Range Management, 1990-09-01)
      Globemallows (Sphaeralcea spp.) have potential in rangeland seedings. Thirty-seven accessions of globemallow were grown at 2 sites in northern Utah and southern Idaho to quantify their agronomic attributes. Data for transplant survival, standing crop, and seed yield were collected in 1987 and 1988. Total globemallow survival (mean = 92%) and seed weights (mean = 0.8 g/plant) differed significantly (P is lesser than or equal to 0.05) among locations. Plant weight (mean = 102 g/plint) differed significantly (P is lesser than or equal to 0.05) among locations, species (S. grossulariifolia, S. coccinea, S. parvifolia, S. munroana, and interspecific hybrids), and years. In a second study, 5 globemallow accessions of 2 species and ‘Spredor 2’ alfalfa (Medicago sativa) were grown with ‘Hycrest’ crested wheatgrass (Agropyron desertorum X A. cristatum) to determine forage yields from 1985-1988. Globemallows produced significantly (P is lesser than or equal to 0.05) less forage (62 g/ml) than alfalfa (389 g/m2). Forage yield of S. munroana (76 g/m2) did not differ significantly (P>0.05) from that of S. grossulariifolia (48 g/m2). Forage yield of crested wheatgrass (mean = 101 g/m2) did not differ significantly (P>0.05) when grown with globemallow versus alfalfa. Plant breeding and selection could probably improve these agronomic attributes for globemallows seeded on rangelands.
    • Technical Notes: Germination of 2 legumes in leachate from introduced grasses

      Fulbright, N.; Fulbright, T. E. (Society for Range Management, 1990-09-01)
      Kleberg bluestem [Dichanthium annulatum (Forsk.) Staph and buffelgrass (Cenchrus ciIiaris L.) may produce pbytotoxic chemicals that inhibit germination and growth of legumes planted in seeding mixtures with grasses. We determined the effects of leachate from these introduced grasses on seed germination of Illinois bundleflower [Desmanthus illinoensis (Michx.) MacM.] and partridge pea (Cassia fasciculata Michx.). Percent germination of Illinois bundleflower seeds on substrata moistened with Kleberg bluestem root or buffelgrass leaf leachate was lower than that of seeds placed on substrata moistened with distilled water. Buffelgrass root Ieachate reduced germination of partridge pea more than did root leachate from Kleberg bluestem or leaf leachate from Kleberg bluestem or buffelgrass. Results of these laboratory experiments indicate that field studies are warranted to determine the effects of buffelgrass on establishment of partridge pea and Illinois bundleflower in order to help land managers select the optimum combination of species for rangeland seeding.
    • Timahdit sheep production and behavior at three stocking rates in the Moyen Atlas of Morocco

      El Aich, A.; Rittenhouse, L. R.; El Khamkhami, S.; Mhand, T. A. (Society for Range Management, 1990-09-01)
      Liveweight (LW) changes and behavior of growing rams were measured in 1981, 1982, 1985, and 1986 at the Timahdit Experiment Station in the Middle Atlas, Morocco, at stocking rates of 2.78, 4.17, and 6.67 rams/ha during the growing season. Maximum animal production would have occurred at a stocking rate greater than any imposed in the study in all years except 1982. Increased variable costs should drive stocking rates down while increased selling price would provide an incentive to increase stocking rates. Mean grazing time of animals stocked at tight, moderate, and heavy rates was 410, 436, and 504 minutes/day, respectively. Mean resting time was 206, 174, and 106 minutes/day, respectively. Rams under light, moderate, and heavy stocking rates walked 2.1, 2.4, and 3.1 km/day, respectively. Diet IVDMD was highest early in the grazing season and lowest in the nongrowing season, and was not affected by stocking rates. Early in the spring, diet diversity was low, increased during active growth, and then deciined as the season advanced. Selectivity was lowest when forage was abundant. Dry matter intake varied with season in 1985 but not 1986. There was less opportmdty to be adaptive under heavy than under light stocking, resulting in a stocking rate by season interaction in 1986. Under heavy stocking, forage intake (G/kg LW/min) was 61 and 83% of intake under light and moderate stocking, respectively, and forage intake per km walked was 71 and 84%, respectively
    • Utilization of linear prediction procedures to evaluate animal response to grazing systems

      Winder, J. A.; Beck, R. F. (Society for Range Management, 1990-09-01)
      Best linear unbiased prediction (BLUP) procedures were used to separate genetic merit from environmental effects on 205-d weight (205-d wt) of calves produced by cows grazing 2 pasture systems. Phenotypic measures of 205-d wt were statistically partitioned into genetic effects (breeding value) and environmental effects. Means were regressed on year of birth of calf. Analysis of covariance was used to test difference in slope and elevation (means) of the regression lines. The continuously grazed pasture (CC) produced higher 205-d wt than did the rotationally grazed pastures (RG) (P<.10). Rate of change in 205-d wt was similar in the 2 grazing systems. Genetic merit was similar among the animals in the 2 grazing systems. The rate of change per year in genetic merit (genetic trend) was also similar. Means tended to vary sharply from year to year, indicating inequality of genetic merit should be taken into account in this type of data. Mean environmental effects resulted in greater (P<.10) 205-d weight in CG than in RG. Rate of change of environmental quality was similar in the 2 systems. These results indicate, from the animals perspective, the RG system did not improve productivity when compared to CG. The CG system was of higher nutritional quality, but the rate of change was similar to that of the RG system.
    • Warm-season grass establishment as affected by post-planting atrazine application

      Bahler, C. C.; Moser, L. E.; Griffin, T. S.; Vogel, K. P. (Society for Range Management, 1990-09-01)
      Atrazine [6-chloro-N-ethyl-N’-(l-methylethyl)-1,3,4-diamine] provides effective weed control during big bluestem (Andropogon gerardii Vitman) and switchgrass (Panicum virgatum L.) establishment. However, most other desirable warm-season grasses are susceptible to atrazine injury at establishment. The objective of this study was to determine if atrazine application after seeding would affect susceptible warm-season grass establishment. Big bluestem, switchgrass, indiangrass [Sorghastrum nutans (L.) Nash] sideoats grama [Bouteloua curtipendula (Michx.) Torr.], and little bluestem [Schizachyrium scoparium (Michx.) Nash] were seeded into greenhouse flats or field plots and 2.2 kg a.i. atrazine/ha applied at 0 (atrazine control), 7, 14, or 21 days after planting. An untreated control was used also. In greenhouse experiments, Indiangrass and sideoats grama plant survival increased when atrazine applications were delayed. Switchgrass, big blue stem, and little bluestem plant survival was not affected by atrazine application. Field studies were conducted in 1983, 1985, and 1986 using the same soil type, grass species, and application periods as the greenhouse study. Delaying atrazine application 7 or more days after planting generally favored survival of lndlangrass and sideoats grama. Big bluestem, switchgrass, and little bluestem were not affected by atrazine treatment. Delaying the application of atrazine may favor the survival of atrazine sensitive species. However, further research needs to be conducted on various soil types and environmental conditions before this can be a recommended practice.
    • Water relations of honey mesquite following severing of lateral roots: influence of location and amount of subsurface water

      Ansley, R. J.; Jacoby, P. W.; Cuomo, G. J. (Society for Range Management, 1990-09-01)
      Location and amount of subsurface water may ifiuenee the degree of dependence of honey mesquite (Prosopis glandulosa Torr.) on shallow lateral roots to supply water. The objective of this study was to determine influence of lateral roots on water relations of honey mesquite on 2 sites which differed in location and amount of subsurface water. Lateral roots were severed with barriers placed to 1.5 m depth and completely surrounding individual trees in February 1985, during mesquite winter dormancy. Stomatal conductance and predawn leaf water potential were signifiicantly reduced in root-severed trees during the following growing season (May-September) at both sites, but reduction was greater on the site with less subsurface water. Daytime leaf water potential was bigger in root-severed than control trees on tbe site with less subsurface water, but not on the other site. By mid-summer 1986, no difference in stomatal conductance between treatments were detected at either site, yet daytime leaf water potential remained higher in root-severed than control trees at the site with less subsurface water. Predawn leaf water potential was greater in root-severed than control trees in 1986, which was a reversal of 1985 trends. Leaf abscission was not observed in either treatment during either growing season. These results suggest that: (1) when less subsurface water was available, trees were more dependent on lateral roots to supply water, (2) treatment effects were minimized by the second growing season following root severing, possibly from new root growth within or below the root barrier region, and (3) the lateral root system may play a significant role in regulating leaf water relations on sites with limited subsurface water.