PDF version
for printing

Impact
Statement

© 2009 Plant Management Network.
Accepted for publication 13 July 2009. Published 20 August 2009.

Year-round Irrigation and Fall Dormancy Affects Alfalfa Yield in a Semiarid, Subtropical Environment

L. M. Lauriault and R. E. Kirksey, Agricultural Science Center at Tucumcari, New Mexico State University, Tucumcari, NM 88401; and D. M. VanLeeuwen, Agricultural Biometric Service, Agricultural Experiment Station, New Mexico State University, Las Cruces, NM 88003

Corresponding author: L. M. Lauriault. lmlaur@nmsu.edu

Lauriault, L. M., Kirksey, R. E., and VanLeeuwen, D. M. 2009. Year-round irrigation and fall dormancy affects alfalfa yield in a semiarid, subtropical environment. Online. Forage and Grazinglands doi:10.1094/FG-2009-0820-01-RS.

Abstract

Fall dormancy (FD) group selection and irrigation management may be valuable tools for increasing alfalfa (Medicago sativa L.) yield in semiarid, subtropical regions. Dry matter (DM) yields of moderately dormant, semidormant, moderately nondormant, and nondormant alfalfa as subplots (each represented by four varieties) were compared for four years under two irrigation regimes (furrow-irrigated once before each harvest or 11 times throughout the year) as whole plots in the Southern High Plains of the USA. Variety, year, and harvest effects also were tested. There were four randomized complete blocks. Year-round irrigation increased (P < 0.05) first harvest yield in 10 of 16 comparisons by > 0.39 Mg/ha, affecting each FD grouping in at least 2 of 4 years. Warmer spring temperatures magnified the difference. Typical dormancy effects on yield were observed in late summer and autumn, but they were inverted in spring and early summer. Differences existed between FD groups in average annual production over the four years of the study (17.65, 18.66, 19.62, and 18.89 Mg/ha for moderately dormant, semidormant, moderately nondormant, and nondormant, respectively; 5% LSD = 0.95). Year-round irrigation may have value during a warm winter when continued warm temperatures are forecast throughout spring.

Introduction

Alfalfa varieties are grouped into fall dormancy categories (1 to 11) based on their autumn height (1,4,20), being further grouped as very dormant varieties (those going dormant earliest and having a shorter fall growth habit), moderately dormant, semidormant, moderately nondormant, and very nondormant varieties (those continuing to grow, thus, being the tallest) (5,6,9). Shorter growth during fall and winter by more dormant varieties is largely triggered by shorter days (4,8).This fall growth characteristic also has been associated with winter survival and drought tolerance for dormant varieties and early initiation of spring growth, more rapid regrowth after harvest, and heat tolerance for less dormant varieties (6,7). The possibility of winterkill exists when using less dormant varieties in areas having winter conditions to which they are not adapted (6,7). The relationship between fall dormancy and winter hardiness in alfalfa has been challenged, however, because of observed winter survival by less dormant varieties in more northern latitudes (4,7).

Alfalfa producers in any environment desire to produce the highest possible yields. An option often employed is to risk winter kill by using less dormant varieties because their late summer and autumn yields tend to be higher than more dormant varieties (6,7).

Irrigation management also has been explored for increasing alfalfa yield in semiarid, subtropical environments. Most of the work in both semiarid and higher precipitation areas has focused on the effects of in-season irrigation modifications (10,11). Some research has been conducted in semiarid regions to evaluate the effect of irrigation termination during winter (11,15). Only the study by Malinowski et al. (14) compared irrigation effects across a broad and inclusive range of FD groupings.

Fall dormancy category selection and irrigation management may be valuable tools for increasing alfalfa yield in semiarid, subtropical regions. Using DM yield measurements, the objectives of this research were to determine which FD groups would be best suited to semiarid, subtropical regions similar to the Southern High Plains of the USA and to evaluate the effects of growing season only or year-round irrigation on DM yield production and seasonal distribution across FD groups.

Experimental Procedures

The study was conducted from 1998 to 2001, at the New Mexico State University Agricultural Science Center at Tucumcari, NM, USA (35.20°N, 103.68°W; elevation 1247 m) as a split-split-split-split-split plot over years and harvests with four randomized complete blocks. Main plot treatments were irrigation treatment (IRR) [furrow-irrigated usually once before each of six cuttings as soon as canal water became available (26, 19, 26, and 24 April 1998 to 2001, respectively, through 26, 30, 20, and 20 October 1998 to 2001, respectively) and year-round irrigation using ground water monthly November through March in addition to the canal water irrigations described for the first treatment]. Subplots were FD groupings (GROUP) [moderately dormant (FD 2 and 3), semidormant (FD 4 and 5), moderately nondormant (FD 6 and 7), and nondormant (FD 8 and 9)]. Sub-subplots were FD within GROUP and sub-sup-subplots were alfalfa varieties (VAR) within FD [Viking I representing FD 2; DK127, Garst 645, and Rainier representing FD 3; Jade II and Landmark for FD 4; Archer and Baralfa54 within FD 5; Tahoe and Wilson representing FD 6; Dona Ana and Helena 7000 for FD 7; 13R Supreme and WL525HQ within FD 8; and Salado and WL612 for FD 9 (1)]. Commercially available seed of each variety was acquired from standard variety test entries submitted by sponsoring companies. Any seed not already inoculated was uniformly treated before planting with a product including Sinorhizobium meliloti and Rhizobium leguminosarum biovar trifolii.

The test was planted after winter wheat (Triticum aestivum L.) cover crop into a Canez fine sandy loam soil (Fine-loamy, mixed, thermic Ustollic Haplargid) with initial soil test levels of P of 75 mg/kg (NaHCO₂ extractant), K of 151 mg/kg (NH40Ac extractant), and 8.4 pH within the surface 30 cm. The seedbed was conventionally tilled and formed into beds on 0.9 m centers for furrow irrigation. Plots, 7.0 m × 1.8 m, were sown 30 April 1997 using a disk drill (20-cm drill spacing) fitted with a seed-metering cone. The seeding rate was 22.4 kg/ha commercial seed product. After planting the study area was managed to promote establishment with the first harvest taken on 21 July when the alfalfa was full bloom. Irrigation and harvest management during the seedling year was otherwise similar to that to be described for the remaining years of the study.

Irrigation water was delivered through gated pipe to achieve field capacity, applying approximately 150 mm with each application at a 50 to 60% efficiency rate. The first growing season irrigation in 2001 was delayed until after the first harvest because 60 mm precipitation fell between 24 April, when irrigation water became available, and 10 May, when the first harvest was taken that year. Soil water content was not measured; however, no evidence of moisture stress was observed during the study period. Experimentation was discontinued after the 2001 growing season because irrigation water became unavailable due to drought in the watershed.

No fertilizers were applied in 1997. In 1998, P at 56 kg/ha and K at 34 kg/ha were applied after the first harvest. From 1999 to 2001, before the initiation of growth, N and P at 25 and 117 kg/ha, respectively, were uniformly broadcast over the area to replace estimated removal during the previous season and to offset some fixation due to high pH. During the study period, N was applied because monoammonium phosphate was the most widely used and available P source; triple super phosphate was completely unavailable in the region. Lack of application of K likely had no effect on productivity as no deficiency symptoms were observed and soil test K levels when the field was renovated in 2005 were 188 mg/kg even though forage had been removed two to three times annually from 2002 through 2004. Soil P levels measured at that time had declined to 28 mg/kg.

Grassy weeds, mostly field sandbur (Cenchrus incertus Curtis), were controlled each year using Clethodim or Sethoxydim at labeled rates. Alfalfa weevil (Hypera postica Gyll.) was not a problem in the area during the test years. In early 2001, cowpea aphid did infest; so, all plots were protected with permethrin (38.4% a.i.) at 584 mL/ha on 3 April. Re-infestation did not occur. A 0.5-m alley between plots was chemical fallowed as needed with 2.5% glyphosate solution with ammonium sulfate at 15 g/liter to facilitate harvesting.

All plots were harvested on the same date for each harvest. Harvests were scheduled when the first flower was observed and were executed after all plots had reached bud stage usually within 5 days of the observation of first flower. Prior to harvest, maturity was rated as a visual estimate of the percentage of stems over the whole plot having open flowers at the canopy surface with bud being rated as zero. Dry matter yields were measured five times mid May through mid September, based on first flower, and near the end of October, allowing at least 6 weeks of fall rest prior to anticipated first fall temperature of -5°C (3,18). Exact harvest dates within each year for 1998 through 2001 are shown in Table 1 along with the averages for the four years. Also included in Table 1 are maturity ratings by GROUP.

Table 1. Harvest dates for the entire study and harvest maturity by fall dormancy (FD) grouping for alfalfa grown at Tucumcari, NM, from 1998 to 2001.

 ^x A temperature of < -5°C occurred on 13 April and is considered to have had no effect on alfalfa growth as the minimum soil temperature remained > 10°C.

 ^y Harvest was delayed to allow recovery from pea-sized hail on 30 April.

 ^z Harvest maturity is the estimated % of stems having open blooms near the canopy surface; 0 = bud. The 5% LSD for harvest maturity within any column is < 1.

For each harvest, topgrowth was collected from a 1.37-m wide swath through the center of each plot using a self-propelled forage plot harvester equipped with a reciprocating blade and electronic scales leaving 7.5-cm stubble on the bed tops. Immediately after weighing forage from each plot, a subsample was collected and placed in a paper bag and sealed inside a plastic bag. The subsamples were weighed and plastic bags removed before drying for 48 h at 70°C. Subsamples were then reweighed to determine DM concentration, which was used to convert fresh harvest weights to DM yield. After each harvest, stems remaining between the harvested swaths were cut at approximately 7.5 cm above the bed tops and removed. Although data from the seedling year (1997) were collected they were not included in the analysis because irrigation treatments had not been applied. Weather data were collected from a National Weather Service station located within 1 km of the study area (Table 2).

Table 2. Monthly air temperature and total precipitation at Tucumcari, NM, from 1997 to 2001 and the long-term (1905-2002) means.

The test was analyzed as a split-split-split-split-split plot across time (13) with IRR as the main plot, and GROUP, FD, VAR, year, and harvest as the sub, sub-sub, sub-sub-sub, and sub-sub-sub-sub-, and sub-sub-sub-sub-subplots, respectively. Dry matter yield (harvest yields, annual total yields, and average annual yields for the four years of the study) were subjected to SAS PROC MIXED to test the main effects of IRR, GROUP, FD, VAR, year, and harvest and all possible interactions (16). Rep, all interactions that included Rep as a term, and residual mean squares were considered random and appropriately used by PROC MIXED as denominators for tests of significance (13). All differences reported are significant at P ≤ 0.05. When an interaction was significant, an LSMEANS statement with the PDIFF and SLICE options were used to determine where differences occurred between IRR and/or GROUP within years and/or harvests. Varietal differences have been published elsewhere (12) and will not be presented in this article. Additionally, GROUP differences were of greater interest than FD differences; therefore, there will be no discussion of specific FD differences.

Winter Survival Observations

Significant stand loss due to winterkill was not observed, likely because of mild winters throughout the years of this study (4). Proper fall harvest management (3) also likely permitted the alfalfa to cold harden (4). Still, all GROUPS but moderately dormant remained green throughout winter [personal observation; Brown et al. (5)], but stem elongation was not sufficient at any time to justify harvest after November (15). Brown et al. (5) in the humid southeastern USA, concluded that the long growing season and mild winters may reduce the need for high root carbohydrate concentrations to maintain stand and promote spring growth (18). Cunningham et al. (7) also concluded that continued growth of nondormant alfalfa during fall and winter did not prevent the accumulation of root carbohydrates or consume reserves necessary for spring growth.

Sledge et al. (19), comparing FD 2, 4, and 7, concluded that FD 7 and less dormant varieties were too winter-sensitive in the southern Great Plains of the USA (34.18°N, 97.13°W); however, Malinowski et al. (14) found that FD 8 could survive slightly to the west (34.15°N, 99.33°W) of Sledge’s location. Varieties having FD 9 were not included in that study (14).

Dry Matter Yield

Year, harvest, and year × harvest effects were significant for harvest DM yield, but these effects also interacted with IRR, GROUP, FD, and VAR. The only interaction between IRR and GROUP was the year × harvest × IRR × GROUP interaction (Fig. 1). Differences among years are apparent with 1998 and 2001 being the most consistent to other studies at this location (11). The first harvest in 1999 was delayed until 21 May to permit recovery from pea-sized hail on 30 April (Table 1). Harvested forage was mostly regrowth after the hail, although it also included living and dead stems from the first growth, likely elevating yield. Yield of the second harvest in 1999 (Fig. 1) was likely depressed because of a shortened harvest interval (Table 1). Exceptionally high yields for the first two harvests of 2000 are likely due to higher late winter and early spring temperatures (Table 2).

Fig. 1. Dry matter production among harvests, years, fall dormancy groupings [moderately dormant (FD 2 and 3), semidormant (4 and 5), moderately nondormant (6 and 7), and nondormant (8 and 9), respectively], and irrigation regimes (alfalfa irrigated only during the growing season or receiving supplemental winter irrigation monthly) at Tucumcari, NM. Data are the least squares means of four varieties and four replicates. Isolated bar indicates the LSD (0.39 Mg/ha, P < 0.05) for fall dormancy grouping comparisons within any cutting in any year or irrigation regime. Harvests 1 through 6 were taken on approximately 15 May, 16 June, 15 July, 13 August, 11 September, and 29 October, which are average dates for the four years of the study.

Regarding the year × harvest × IRR × GROUP interaction (Fig. 1), the more important component of the interaction was related to differences in IRR within GROUP for the first harvest of each year as indicated by the SLICE option, the results of which are presented in Table 3. Differences between irrigation treatments occurred every year and there were several comparisons that approached significance. The least number of differences occurred in 1999 when the first and second harvests were affected by hail, which may have moderated the treatment effects (Table 3). Also of note is the general decline in first harvest yield for both irrigation treatments as dormancy level progressed from semidormant to dormant.

Table 3. The year × fall dormancy (FD) grouping × irrigation treatment effect on first harvest alfalfa yields (Mg/ha) at Tucumcari, NM.

 ^x No winter irrigation and winter irrigated signify irrigated once prior to each harvest and irrigated once prior to each harvest as well as monthly November through March, respectively.

 ^y P-value is the result of the SLICE = year × harvest × GROUP option within the PDIFF analysis for the year × harvest × IRR × GROUP in SAS PROC MIXED. Data for irrigation treatments are the least squares means of 4 fall dormancy groupings each represented by 4 varieties and 4 replicates.

 ^z cdefMeans within a column within a year followed by the same letter are not significantly different at P < 0.05 based on the PDIFF analysis in SAS for that treatment × year interaction.

The year × harvest × IRR interaction also was significant because the difference between irrigation treatments occurring in the first harvest of each year was greater in 2000 when warm temperatures in late winter and early spring encouraged earlier activation of growth and a longer growth period prior to the first harvest. This allowed increased soil moisture from the winter irrigation to enhance growth potential for that first harvest even more so than the other years of the study (Fig. 1; Tables 2 and 3). Sharratt et al. (17) also found that spring yields of alfalfa were positively correlated to spring temperatures. Bolger and Matches (2), in the southern Great Plains of the USA, observed decreased yield due to warm, dry spring conditions and delayed first irrigation. Additionally, Field et al. (9), in the desert southwestern USA, measured yield increases for alfalfa up to FD 7 (moderately nondormant) by increasing the length of the irrigation season.

The magnified difference in the first harvest in 2000 (Fig. 1; Table 3) also led to a significant harvest × IRR interaction (data not shown) and a year × IRR interaction for total annual yield, but not for average annual yields for the four years of the study even though winter irrigated alfalfa had numerically higher yields each year of the study (Table 4). While the only year in which the difference in the first harvest yield caused a difference in annual yield was 2000, there also was a strong trend in 2001. Although not are warm as early as it was in 2000, April 2001 was exceptionally warm, likely contributing to the treatment effects in that year. Consequently, the maximum benefit of winter irrigation may only be realized when a warm spring (March through May) is forecast and water is available during that time for irrigation. Still, in another study at this location (11), conducted in the same years with similar irrigation treatments as the present study, but with fewer varieties, winter-irrigated alfalfa outyielded alfalfa not irrigated during the winter every year in the first and, sometimes, subsequent harvests leading to differences between treatments each year and in the 3-year total.

Table 4. The differences between irrigation treatments and among fall dormancy (FD) groups in total annual and 4-year average annual yields (Mg/ha) of alfalfa harvested six times per year from 1998 to 2001 at Tucumcari, NM.

 ^x No winter irrigation and winter irrigated signify irrigated once prior to each harvest and irrigated once prior to each harvest as well as monthly November through March, respectively.
Data for irrigation treatments are the least squares means of 4 fall dormancy groups, each represented by 4 varieties, and 4 replicates.
Data for fall dormancy groupings are the least squares means of 4 varieties, 2 irrigation treatments, and 4 replicates.

 ^y bc: Means within a row, excluding column for the 4-year average yield, followed by the same letter are not significantly different at P < 0.05 based on the PDIFF analysis in SAS for that treatment × year interaction.

 ^z ABC: Means within a column within the fall dormancy grouping treatment followed by the same letter are not significantly different at P < 0.05 based on the PDIFF analysis in SAS for the fall dormancy grouping × year interaction or average yield.

Further research is needed to establish appropriate rates and timing of irrigations to promote greater spring growth and to measure the increase in growth relative to the amount of water applied because it was not determined in either study which of the five individual winter irrigations was most likely to provide the greatest benefit or if several or all were equally necessary. Ottman et al. (15) terminated irrigation from November through February with no reduction in first harvest yield. It is, therefore, possible that irrigation beginning in early March at this and similar environments is all that is necessary, even for less dormant varieties.

The year × harvest × GROUP interaction was significant (data not shown) but was mainly caused by differences in magnitude over the harvest × GROUP interaction, which is shown in Figure 2. Yield differences in late summer and autumn were as expected with yield increasing as dormancy decreased. Early season yields, however, were inverted such that the more dormant groupings had higher yields at that time. There was no effect of GROUP in mid summer. The demarcation between FD groupings was initiated by the end of summer, becoming complete in autumn (Fig. 2). Grimes et al. (10) in a semiarid Mediterranean environment (winter precipitation pattern), observed a similar response with slightly higher yields by a more dormant variety in spring followed by a reversal during summer, such that, nondormant varieties had higher yields in the autumn. This led to a lack of difference in annual yield between CUF101 [FD 9 (1)], Moapa 69 [FD 8 (1)], and WL318 (FD 4, Mike Peterson, personal communication) in the study by Grimes et al. (10). The phenomenon of lower spring production by less dormant varieties also has been observed in areas having continental (summer precipitation pattern) but higher precipitation at more northern latitudes. Brummer et al. (6), in the central Great Plains of the USA, measured more pronounced yield reductions by nondormant alfalfa than those measured in the present study (Fig. 2, Tables 3 and 4). Earlier initiation of growth in the spring by nondormant varieties (7) in cooler continental climates, such as those similar to this latitude, elevation, and precipitation pattern, might cause reduced harvestable yield in the first and possibly the second harvests.

Fig. 2. Dry matter production among harvests and fall dormancy groupings [moderately dormant (FD 2 and 3), semidormant (4 and 5), moderately nondormant (6 and 7), and nondormant (8 and 9), respectively], at Tucumcari, NM. Data are the least squares means of four years, two irrigation regimes (alfalfa irrigated only during the growing season or receiving supplemental winter irrigation monthly), four varieties, and four replicates. Isolated bar indicates the LSD (0.24 Mg/ha, P < 0.05) for fall dormancy grouping comparisons within any cutting in any year or irrigation regime. Harvests 1 through 6 were taken on approximately 15 May, 16 June, 15 July, 13 August, 11 September, and 29 October, which are average dates for the four years of the study.

Yield differences within groupings across years were consistent although there was a difference in magnitude leading to a significant year × GROUP interaction for annual yield and a difference in average annual yields over the four years of the study (Table 4). Generally, moderately nondormant FD’s (6 and 7) were the consistently high yielding categories across years at this latitude and climate. Moderately dormant FDs (FD 2 and 3) always yielded less than the moderately nondormant grouping. Semidormant alfalfa (FD 4 and 5) yielded equally well to moderately nondormant alfalfa only for the first two years and, after the first year, there was no difference between the moderately nondormant and nondormant (FD 8 and 9) groupings (Table 4).

Conclusions

Alfalfa producers in semiarid, subtropical climates near latitude 35° North or South and in climatically similar areas can maximize yield for at least four production years by using varieties in fall dormancy categories 6 or 7. Studies such as this could be useful in any environment to determine the most likely adapted range of FD categories for that area. Standard alfalfa variety tests already being conducted also could serve this purpose. Using the least dormant varieties that will survive at a particular latitude will not necessarily maximize production because of possible reduced spring yield.

Year-round irrigation increased first harvest alfalfa yields compared to winter termination of irrigation and even more so when spring temperatures were warmer than average. In irrigation water limited circumstances, producers could likely benefit from late winter or early spring irrigation near anticipated initiation of growth to increase yield in years when warm spring temperatures are forecast; however, use of less dormant varieties to take advantage of available soil moisture earlier in the spring in this environment, even when spring temperatures were warmer, did not increase spring yield over moderately dormant varieties.

Acknowledgments

We gratefully acknowledge the technical and field assistance of George Arguello, Eutimio Garcia, and Leslie Robbins; secretarial assistance of Terri Warren, Doris Hight and Patty Cooksey; and our coworkers at the NMSU Library Document Delivery Service.

Literature Cited

1. Alfalfa Council. 1999. Fall dormancy and pest resistance ratings for alfalfa varieties, 1999/00 edition. The National Alfalfa Alliance, Kennewick, WA.

2. Bolger, T. P., and Matches, A. G. 1990. Water-use efficiency and yield of sainfoin and alfalfa. Crop Sci. 30:143-148.

3. Bootsma, A., and Suzuki, M. 1985. Critical autumn harvest period for alfalfa in the Atlantic region based on growing degree-days. Can. J. Plant Sci. 65:573-580.

4. Brouwer, D. J., Duke, S. H., and Osborn, T. C. 2000. Mapping genetic factors associated with winter hardiness, fall growth, and freezing injury in autotetraploid alfalfa. Crop Sci. 40:1387-1396.

5. Brown, L. G., Hoveland, C. S., and Karnok, K. K. 1990. Harvest management effects on alfalfa yield and root carbohydrates in three Georgia environments. Agron. J. 82:267-273.

6. Brummer, E. C., Moore, K. J., and Bjork, N. C. 2002. Agronomic consequences of dormant-nondormant alfalfa mixtures. Agron. J. 94:782-785.

7. Cunningham, S. M., Volenec, J. J., and Teuber, L. R. 1998. Plant survival and root and bud composition of alfalfa populations selected for contrasting fall dormancy. Crop Sci. 38:962-969.

8. Fick, G. W., Holt, D. A., and Lugg, D. G. 1988. Environmental physiology and crop growth. Page 163-194 in: A. A. Hanson, D. K. Barnes, and R. R. Hill, ed. Alfalfa and Alfalfa Improvement. Agron. Monograph 29. ASA-CSSA-SSSA, Madison, WI.

9. Field, L., Melton, B., and Currier, C. 1987. Effect of fall dormancy of alfalfa cultivars on forage yield under different rates of irrigation. Res. Rep. 607. New Mexico State Univ. Agric. Exp. Stn., Las Cruces, NM.

10. Grimes, D. W., Wiley, P. L., and Sheesley, W. R. 1992. Alfalfa yield and plant water relations with variable irrigation. Crop Sci. 32:1381-1387.

11. Lauriault, L. M., Kirksey, R. E., and VanLeeuwen, D. M. 2008. Perennial cool-season forage legume performance in diverse soil moisture treatments, Southern High Plains, USA. Bull. 796. Online. New Mexico State Univ. Agric. Exp. Stn., Las Cruces, NM.

12. Lauriault, L. M., Ray, I. M., English, L. M., Murray, M. W., and O’Neill, M. K. 2002. The 2001 New Mexico Alfalfa Variety Test Report. Online. New Mexico State Univ. Agric. Exp. Stn., Las Cruces, NM.

13. Littell, R. C., Milliken, G. A., Stroup, W. W., and Wolfinger, R. D. 1996. SAS system for mixed models, 633 pp. SAS Inst., Inc., Cary, NC.

14. Malinowski, D. P., Pinchak, W. E., Kramp, B. A., Zuo, H., and Butler, T. J. 2007. Supplemental irrigation and fall dormancy effects on alfalfa productivity in a semiarid, subtropical climate with a bimodal precipitation pattern. Agron. J. 99:621-629.

15. Ottman, M. J., Tickes, B. R., and Roth, R. L. 1996. Alfalfa yield and stand response to irrigation termination in an arid environment. Agron. J. 88:44-48.

16. SAS Inst. 2001. The SAS system for Windows. Release 8.02. SAS Inst. Inc., Cary, NC.

17. Sharratt, B. S., Baker, D. G., and Sheaffer, C. C. 1986. Climatic effect on alfalfa dry matter production Part I. Spring harvest. Agric. Forest Meteorol. 37:123-131.

18. Sholar, J. R., Stritzke, J. F., Caddell, J. L., and Berberet, R. C. 1988. Response of four alfalfa cultivars to fall harvesting in the Southern Plains. J. Prod. Agric. 1:266-270.

19. Sledge, M. K., Hopkins, A. A., and Bouton, J. H. 2006. Grazing alfalfa in the southern Great Plains. CD-ROM. Proc. AFGC, Vol. 15, Georgetown, TX.

20. Teuber, L. R., Taggard, K. L., Gibbs, L. K., McCaslin, M. H., Peterson, M. A., and Barnes, D. K. 1998. Fall dormancy. Online. North American Alfalfa Improve. Conf., Stillwater, OK.