Southwest Research-Extension Center Field Day 1991

This report is brought to you for free and open access by New Prairie Press. It has been accepted for inclusion in Kansas Agricultural Experiment Station Research Reports by an authorized administrator of New Prairie Press. Copyright 1991 the Author(s).

Climatic conditions were very favorable for crop growth in 1990, particularly from January through late summer. Above normal precipitation from January through May resulted in excellent wheat yields. June and August had below normal rainfall, but above normal rainfall in July contributed to good yields from the summer crops.
Precipitation totaled 18.76 inches or 0.90 inches above normal. Snowfall was above normal, with nearly 15 inches in January and over 16 inches in February. One-half inch fell in March, then 1 inch in November and nearly 5 inches in December.
With the exception of November, precipitation during the last 5 months of the year was below normal. A lack of topsoil moisture in September resulted in some problems in obtaining good stands of wheat.
Temperatures were warmer than normal in January, March, June, September, and November. May, July, and December were cooler than normal. Record high temperatures occurred on 5 days during the year, and record lows on 6 days. The high for the year was 105 o on June 30, and the low was -10 o on December 22 and 23. There were 16 days of 100 o or higher temperatures, and 10 days when the low reached 0 o or below. The coldest day was December 22, with a low of -10 o and a high of 0 o . Single-digit high temperatures of 4 o , 2 o , and 2 o occurred on December 21, 23, and 30, respectively. Average wind speed was 5.1 mph or 1.0 mph below normal. Open pan evaporation was 70.88 inches or 7.44 inches below normal. The frost-free period was from May 10 to October 9, or 152 days, 18 days below normal.
A complete summary of the weather is presented in the accompanying table. Precipitation for 1990 totaled 16.48 inches, which was slightly above normal (Table 1). Precipitation was above normal in 6 months. The wettest months were May, with 2.49 inches, and July, with 3.72 inches of precipitation. The largest single amount of precipitation was 1.30 inches on May 30. Snowfall for the year totaled 29.4 inches. The largest single amount of snowfall was 10.0 inches on January 19 and 20.
Air temperatures were above normal for 6 months and below normal for 6 months. The warmest month was July, with an average temperature of 74.8 o and an average high temperature of 90.1 o . The coldest month was December, with an average temperature of 24.5 o , an average high of 39.1 o , and an average low of 9.8 o . Deviations from the normal were greatest in December, when the average temperature was 10.7 o lower than normal.
The highest temperature was 107 o on June 29 and 30. A total of 16 days reached 100 o or above. The lowest temperature was -21 o on December 30. There were 8 days with sub-zero temperatures. The last frost (30 o ) in spring was on May 10 (29 o ), which was 9 days later than normal. The first frost (30 o ) in the fall was on October 9 (25 o ), which was the same date as the 78-year average. The frost-free period was 152 days or 9 days less than normal. Record high temperatures were set on 12 days; June 19 (105 o  Open pan evaporation from April through September totaled 80.41 inches or 14.27 inches above normal. Wind speed for the same period averaged 5.56 mph; normal speed is 5.63 mph. Preemergent herbicides (usually 3 lbs Ramrod + 1.0 lb atrazine) were used in the WSF-CT and SF treatments for sorghum. Reduced and NT sorghum usually received 4 lbs Ramrod preemergence. In years of light weed pressure, preemergent herbicides probably were not needed in the RT and NT plots.

SUMMARY
Increases in available soil water and yield from a reduction in tillage occurred more often in the WSF system than in the WF system and more often for sorghum than for wheat. Wheat yields from the WF and WSF systems usually did not differ, nor did sorghum yields from the SF and WSF systems. An economic comparison between WF and WSF indicated that the greatest return occurred from reduced-till wheat, no-till sorghum in the WSF system.

INTRODUCTION
A long-term study is being conducted to determine the effects of cropping system and reduced or no tillage on dryland winter wheat and grain sorghum. The effects of reduced and no tillage on available soil water and yield are being determined. This report is a summary of the data collected from 1987 through 1990.

PROCEDURES
The wheat-fallow (WF), wheat-sorghum-fallow (WSF), sorghum-fallow (SF), continuous sorghum (SS), and continuous wheat (WW) systems were studied. Herbicides were used in place of some or all tillage. Treatments varied somewhat from year to year, but the following are currently in use. Wheat was planted with a John Deere HZ drill in 16-inch rows at a rate of 40 lbs/A. Sorghum was planted with a Buffalo slot planter in 30-inch rows at a rate to result in 25,000 plants per acre. Available soil water was measured at 1-foot intervals to a depth of 5 feet at the end of fallow. Grain was harvested with a plot combine, and grain yields were reported at 12.5% moisture. The soil type was a Richfield silt loam with a pH of 7.8, organic matter content of 1.5%, and an available water holding capacity of 10.8 inches in a 5-foot profile. The experimental design was a randomized complete block with three replications.

RESULTS AND DISCUSSION
The use of atrazine in the WF and WSF system (WF-RT and WSF-RT) typically resulted in the elimination of two tillage operations, the one following harvest and the first operation in the following spring (Table 1). Atrazine, particularly at the 1.0 lb rate in WF-RT sometimes did not result in adequate volunteer control, making tillage or the use of postemergent herbicides necessary. The use of Bladex (following atrazine) in the WF system (WF-MT) resulted in the elimination of more than half of the tillage, whereas the use of Bladex prior to sorghum (WSF-NT) eliminated all tillage. Two tillage operations were typically eliminated when Bladex was used in the WSF system prior to wheat (WSF-RT). There were no SS-CT plots, but this treatment would require spring tillage similar to WSF-CT. Reduced or no till in the SF system is not practical, because of the long fallow period and is not currently being studied.

Soil Water
The amount of available soil water (hereafter referred to as soil water) at wheat planting is presented in Table 2. The amounts did not differ between tillage treatments in the WF system. In the WSF system, there was more soil water in the NT plots only in 1989. There was more soil water in the WF-CT and NT plots than in WSF-CT and NT in 1989, also. The advantage for WF in 1989 occurred because of the longer fallow period; much of the storage occurred early in fallow, before the beginning of the WSF fallow period. The WW treatment had less soil water than all WF and WSF treatments in 1987 and 1990; however, the amount did not differ from that in either WSF treatment in 1988 or in WSF-CT in 1989. The amount of soil water at sorghum planting is presented in Table 3. In the WSF system, there was more soil water in RT andNT than in CT in 1987 and1988;in 1989, the amount in RT, but not NT, exceeded that in CT. No significant differences occurred in WSF in 1986. There were no significant differences between RT and NT in any year. Soil water in SS was less than that in all WSF treatments in 1986 and 1988, and less than in WSF-RT andNT in 1990, but more than in WSF-CT in 1987. No difference occurred between any WSF treatment and SS in 1989. The longer fallow period of SF never resulted in more soil water than in WSF-RT or NT but did result in more than in WSF-CT in 1987 and 1989.  Wheat yields Tillage caused no differences in the WF system in 1987WF system in , 1988WF system in , and 1990 (Table 4). In 1989, NT yielded more than the CT and blade-once treatments but yielded the same as RT and MT. The blade-once treatment was included because NT has resulted in hard, dry soil in previous studies. Blading tends to prevent a hard surface layer and makes planting easier. However, in the current study, this has not been a problem. There is no explanation, thus far, for the significantly lower yield of the blade-once treatment in 1989.
In the WSF system (Table 5), RT yielded more than CT in 1989. In December, 1990, several days of extremely cold temperatures reaching -17 o F. occurred before the wheat entered dormancy. This caused abortion of some tillers, and WSF-NT yielded less than WSF-RT or CT. The NT plants were exposed more to the cold because of shallower planting and lost more tillers than the other treatments. Under these same conditions, the yield of WF-NT (Table 4) was not reduced because of insulation from wheat straw remaining from the previous crop. Tillage had no effect on continuous wheat (Table 6). Table 5. Effect of reduced and no tillage on the yield of winter wheat in a wheat-sorghum-fallow system. Garden City, KS. 1987-90. _________________________________________________________ Tillage system 1987Tillage system 1988Tillage system 1989Tillage system 1990 Avg. __________________________________________________________ ------Bu/A-------Conventional 24a 1 26a 12b 57a 30a Reduced 26a 23a 31a 55a 34a No till 23a 19a 23ab 46b 28a __________________________________________________________ 1 Yields within a column followed by the same letter do not differ (P<0.05). Table 6. Effect of tillage on the yield of continuous winter wheat. Garden City, KS. 1987-90. _____________________________________________________ Tillage system 1987Tillage system 1988Tillage system 1989Tillage system 1990 Avg. _____________________________________________________ ------Bu/A----- -Conventional  10  17  7  47  20  No-till  13  14  9  41 19 LSD (.05) NS Figure 1 shows a graphic summary of the effects of tillage on wheat yields when averaged over the WF and WSF cropping systems. Although the data vary from year to year, there is a trend toward increased yield with RT. Figure 2 is a summary of the effects of cropping system on yield, averaged over tillage. Yields of WF and WSF are generally similar. The longer fallow period of WF increased yields only in 1989, because of more stored moisture, discussed previously. The WW system yielded the least in each year, although high rainfall in 1990 resulted in yields greater than those of WF and WSF in the preceding 3 years.  Table 7. Effect of cropping system and tillage on grain sorghum yield. Garden City, KS. 1987-90. ____________________________________________________________ Tillage system 1987198819891990 Avg. ____________________________________________________________ -----Bu/A------WSF CT 49b 1 35a 90a 52a 56 RT 61ab 49b 99a 55a 66 NT 69a 53b 99a 58a 70 SS 56ab ---55b 38b 50 SF 64a ---71b 53a 63 ___________________________________________________________ 1 Yields within a column followed by the same letter do not differ (P<0.05).

YEAR YEAR
Grain sorghum yields Grain sorghum yields, as affected by cropping systems and tillage, are presented in Table 7. Reduced and NT yielded more than CT in 1988, but NT yielded more than CT in 1987. Yield increases were a result of more soil water at planting (Table 3). Figure 3 illustrates the difference between cropping system (WSF is averaged over tillage). Wheatsorghum-fallow yields were similar to those following the longer fallow period of SF in 2 of the 3 years. In 1989, WSF plots matured earlier than those of SF, which were not yet mature by frost. Continuous sorghum yielded less than WSF or SF in each year.

INTRODUCTION
An economic analysis was conducted on the wheat-fallow and wheat-sorghum-fallow systems for the 1987-1990 period. The objective was to determine which cropping and tillage system provided the greatest economic return under several government program scenarios.

PROCEDURES
Cultural practices and treatments used are described on page 5 in the preceding section. For the purpose of this analysis, a 2000-acre, dryland farm was assumed. Wheat-fallow (WF) included 1000 acres of wheat and 1000 acres of fallow each year. Wheat-sorghum-fallow (WSF) included 666.67 acres of wheat, 666.67 acres of sorghum, and 666.67 acres of fallow each year. Conventional tillage (CT), reduced tillage (RT), and no tillage (NT) were compared. All land was owned. Costs were calculated using KSU Farm Management Guides MF 257, MF 903, and MF 904 for the WF and WSF systems. Crop prices used for wheat and sorghum were the averages for southwest Kansas each year. Deficiency payments were based on actual rates in [1987][1988][1989] and an estimated rate for 1990. Payment yields (ASCS yields) used average dryland yields for wheat and grain sorghum in Finney County. Provisions of the 1990 Farm Bill pertaining to cross compliance and building base were used in the calculations. Crop prices, deficiency payment rates, and proven yields are included in Table 1.

RESULTS AND DISCUSSION
Profit or loss from cropping and tillage systems depends on crop yield, price, government payments, Table 1. Yields, prices, and deficiency payments for wheat and grain sorghum. 1987-1990. _______________________________________________________________________________________________________ Deficiency Wheat 24  10 and the costs of herbicides vs tillage. Table 2 is a listing of the number of tillage and spraying operations in the WF and WSF systems and the total costs of production. Costs increase as tillage is decreased, particularly for pure no-till systems. However the WF-RT system is slightly cheaper than WF-CT, and the combination CT wheat, RT sorghum system in WSF is cheaper than all CT. An explanation for the differences in costs can be seen in Figure 1, which is a comparison of the herbicide vs machinery costs. Herbicide costs increase greatly as tillage is decreased. This cost is countered somewhat by lower machinery costs (in Figure 1 machinery costs consist of depreciation, interest, insurance, and housing). The amount and size of machinery decrease as the need for tillage decreases. Figures 2, 3, and 4 are net returns from each system and reflect the average return from the 1987-1990 period, given different government payment assumptions. The notation CNT for WSF means the wheat was conventional-till, but the sorghum was no-till; RNT means the wheat was reduced-till, and the sorghum, no-till. Otherwise the wheat and sorghum were both CT, RT, or NT. The 1990 Farm Bill allows producers to build base, provided they are not receiving deficiency payments on any program crops. Figure 2 compares the net returns of WF and WSF without any government payments. The 1990 Farm Bill also allows producers with no sorghum base to plant both wheat and sorghum and still receive deficiency payments on wheat (cross compliance has been eliminated). Thus, in Figure 3, deficiency payments were received for wheat but not sorghum. In Figure 4, we assumed that the producer had both a wheat and sorghum base; thus deficiency payments were received for both crops. In all three cases, the average returns were greater from WSF than from WF. Reduced-till wheat, no-till sorghum (RNT) returned slightly more   than did CNT, and both of these treatments were substantially more profitable than the other treatments. Figure 5 is a yearly comparison of the returns between the most profitable system WSF system, RNT, and the most profitable WF system, RT. Included are deficiency payments for wheat only or both wheat and sorghum. Although returns varied greatly from year to year (a $100,000 difference between the best and worst system), the profit from WSF exceeded that from WF in each year when deficiency payments were received for both crops. In 1990, the year of the highest wheat yields, the WF return was about the same as the WSF return when only wheat payments were received (no sorghum base). In the event that the producer gets higher than normal wheat yields and lower than normal sorghum yields, the return from WF could exceed that of WSF, particularly if there was no sorghum base. In most years, however, WSF should be more profitable than WF. The 1990 Farm Bill gives dryland wheatfallow producers the opportunity to make the conversion to a wheat-sorghum-fallow cropping system. In this study, WF-RT was the most profitable WF tillage system, and WSF-RNT was the most profitable WSF tillage system. This is not surprising, based on results from this and other studies. What is surprising is that the pure WSF-NT system, in which both wheat and sorghum were no-till, returned more than the CT system. No-till wheat is expensive, and this system was profitable only because of high sorghum yields. Note that the no-till wheat-fallow system was the least profitiable. In years of low sorghum yields, WSF-NT will result in a loss. The possibility of high sorghum yields should not be used to compensate for the high number (at least three and possibly four) herbicide applications that are required for no-till wheat. For this reason, no-till wheat should be avoided. Reduced-till wheat, when only one or two herbicide applications are used, has the potential for higher yields and will often return more profit than conventional tillage. Both reduced and no-till sorghum have a good possibility of producing higher yields and more profit than conventionally tilled sorghum and merit serious consideration for the producer with good management skills.

ROW SPACING OF WHEAT by Merle Witt
Narrow rows for drill-seeded wheat did not increase grain yields in our study. With 3 years of data, we did not find significant differences between 5", 10", or 15" row spacing (see table 1). There appeared to be a slight trend for both varieties, TAM 107 and Larned, to have lower yields at the widest (15") spacing. However, the variety Larned lodged worst in 1990 with narrow (5") rows and performed better with the 10" and 15" rows.
The narrow rows possibly could times be helpful in hastening fall ground cover, which might reduce erosion and enhance forage production for grazing. However, we did not find a statistically significant grain yield advantage with narrow rows on dryland or under irrigation.

SUMMARY
Dryland soybeans have produced respectable yields, with a 3-year average of 22 bushels per acre. Adjacent, dryland, grain sorghum has had a 3-year average of 52 bushels per acre. Yields averaged 14 bushels per acre in dry 1988, 27 bushels per acre in a wet 1989, and 23 bushels per acre in 1990. Maturity Groups III or IV have produced higher yields than the shorter season Maturity Groups 00, 0, I, or INTRODUCTION Dryland soybeans were initiated in 1988 to determine production feasibility under moisture-limiting conditions and to compare the relative yield potential of various maturity groups.

PROCEDURES
Soybeans of Maturity Groups 00, 0, I, and III were grown on dryland plots planted on May 8 in 1988. Soybeans of Maturity Groups I, II, III, and IV were seeded on May 5, 1989 and on May 2, 1990. Plots were grown on a Keith Silt loam soil type in all 3 years, with Treflan at 2 pints per acre incorporated for weed control. In each of the 3 years, the soybeans followed a year of summer fallow, which had followed a grain sorghum crop.

RESULTS AND DISCUSSION
Yields of dryland soybeans in 1988 ranged from 10 to 19 bushels per acre (Table 1). A nearly full profile of stored subsoil moisture at the start of the season helped overcome a very dry summer. Grain sorghum yields in nearby variety plots averaged 52 bushels per acre by comparison.
Yields of dryland soybeans in 1989 ranged from 20 to 37 bushels per acre (Table 2). This compared with yields from adjacent, grain sorghum, variety plots that averaged 53 bushels per acre.
Yields of dryland soybeans in 1990 ranged from 22 to 26 bushels per acre (Table 2). This compared

DRYLAND PRODUCTION OF SOYBEAN by Merle Witt
with yields from adjacent, grain sorghum, variety plots that averaged 50 bushels per acre.
In all 3 years, yields increased with increasing maturity length of the soybeans. Soil core sampling following harvest in 1988 showed greater rooting depth and greater moisture extraction at depth with the soybeans of later maturity groups. Early plantings were utilized each year when adequate surface soil moisture for emergence was present.

EFFECT OF HYBRID SELECTION AND NITROGEN RATE ON GRAIN YIELD OF DRYLAND CORN
by Alan Schlegel, John Havlin*, and David Frickel SUMMARY Dryland corn can be successfully grown in western Kansas, but yields are highly variable. Grain yields were affected more by hybrid selection than by N rate. Additional information is needed in identifying hybrids best adapted for dryland corn production.

INTRODUCTION
Traditional dryland cropping systems in western Kansas have been restricted to wheat-fallow and wheat-sorghum-fallow rotations. However, recent studies in other areas have indicated successful corn production in areas of 20 inches or less of annual precipitation. Therefore, this research was conducted to determine the production potential of corn grown under dryland conditions in western Kansas.

PROCEDURES
Six corn hybrids were planted at four sites from 1988 to 1990. The varieties ranged in relative maturity from early (90 to 100 days) to late (110 to 120 days). The sites were located in Greeley, Wallace, Sherman, and Thomas Counties. Nitrogen fertilizer was applied at planting at four rates ranging from 0 to 105 lb N/acre. Corn was planted in late April in a wheat-corn-fallow rotation using reduced or no tillage. After corn emergence, all plots were hand thinned to uniform stands. Residual herbicides were used to control weeds in the wheat stubble prior to planting of corn and for in-season weed control in corn. Soil moisture and residual N determinations were made at corn planting and harvest. All plots were harvested in September of each year, and yields were adjusted to 15.5% moisture. In 1989, the Sherman County site sustained considerable hail damage, so grain yields were not determined.

RESULTS AND DISCUSSION
Grain yields ranged from less than 10 bu/acre to over 90 bu/acre, depending on hybrid, location, and year. Hybrid selection affected grain yields more than N rate (Tables 1, 2, and 3). Hybrid selection significantly affected yields in 7 of the 11 site-years. Nitrogen fertilizer had minimal effect on grain yield, with significant differences in yield observed in only 2 of the 11 site-years. Grain yields increased with increased N rates in Greeley County in 1990, and yields decreased with increased N rate in Wallace County in 1990.
*John Havlin, Department of Agronomy, Kansas State University, Manhattan.

SUMMARY
Ridge tillage and conventional tillage produced equal grain yields of irrigated corn and grain sorghum. Preplant irrigation was generally ineffective in increasing grain yields of corn and grain sorghum. Phosphorus fertilizer increased corn and sorghum yields, but P placement had no effect on yield. Soil water at planting, soil water accumulation during fallow, and fallow efficiency were greater with ridge rather than conventional tillage.

INTRODUCTION
This research was conducted to determine the feasibility of ridge tillage for flood-irrigated corn and grain sorghum in western Kansas. Additional objectives were to 1.) determine the benefit from preplant irrigation for ridge and conventional tillage and 2.) determine whether phosphorus placement was affected by tillage practices.

PROCEDURES
Corn and grain sorghum have been grown continuously since 1988 under conventional and ridge tillage. Phosphorus fertilizer has been applied since 1989. Conventional tillage consists of stalk shredding and discing in the fall, followed by spring discing and furrowing prior to planting. With ridge tillage, the only operation between harvest and planting is shredding stalks. Tillage (two cultivations) during the growing season was the same for both systems. Preplant irrigation treatments were applied 2 to 4 weeks prior to planting and averaged 4.3 inches for corn and 5 inches for sorghum. Inseason irrigations were applied uniformly to all plots when needed. Phosphorus was broadcast and band applied at planting at a rate of 40 lb P 2 O 5 /acre, and a zero P check was included. All plots were machine harvested, and grain yields adjusted to 15.5% moisture for corn and 12.5% moisture for sorghum.

RESULTS AND DISCUSSION
With corn, ridge rather than conventional tillage increased soil water accumulation during the period from harvest to planting by 1 inch (Fig. 1), thereby, increasing fallow efficiency from 24 to 45% (Fig. 2) and soil water at planting by over 1 inch (Fig.  3). Preplant irrigation increased soil water accumulation during fallow by 2 inches for conventional tillage and 2.7 inches for ridge tillage. Soil water at harvest was not affected by tillage or preplant irrigation (Fig. 4).   With sorghum, accumulation of soil water during fallow was 0.5 inch greater with ridge than with conventional tillage (Fig. 5). Fallow efficiency was 26% with conventional tillage without preplant irrigation and 32% with ridge tillage (Fig. 6). Soil water at planting and harvest was about 1 inch greater with ridge than with conventional tillage ( Fig. 7 and 8).   Similar corn yields have been obtained with ridge and conventional tillage (Table 1). Preplant irrigation did not significantly increase corn yields. Grain sorghum yields in 1989 were extremely low because of an earlier than normal frost. Sorghum yields in 1990 were not affected by tillage but were increased slightly by preplant irrigation. Both corn and grain sorghum yields were increased by P fertilization; however, placement had no effect on yields.

EFFICACY OF STANDARD AND SIMULATED CHEMIGATION APPLICATIONS OF INSECTICIDES FOR SECOND GENERATION CORN BORER CONTROL AND THEIR EFFECT ON SPIDER MITES, 1990
by Gary Dick, Phil Sloderbeck, and Steven Posler SUMMARY Several insecticides were evaluated for control of European corn borers and for their effect on spider mites in furrow-irrigated field corn. Interpretation of the results is complicated by the combined effects of corn borers and spider mites on yield and the differential effect of insecticides on corn borers and spider mites. Numerically, more live corn borer larvae and corn borer-infested plants and significantly more corn borer tunneling were observed in the reduced-mite check plots than in the untreated check plots. All corn borer treatments resulted in a significant reduction in length of stalk tunneling compared to the reduced-mite check. All treatments, except Asana XL, resulted in a significant reduction in the number of live larvae and number of plants infested compared to the reduced-mite check. Capture resulted in the highest level of control of any of the insecticides in this test, with >95% reduction in live larvae, number of tunnels, and length of tunneling compared to the reducedmite check.
The spider mite species composition was 94% Banks grass mites before treatment and 86% Banks grass mites 19 days after treatment (DAT). The effects of corn borer insecticides on spider mites were highly variable and generally not significantly different from the untreated checks.

INTRODUCTION
This test was conducted to evaluate the efficacy of standard ground applications and simulated chemigation applications of several insecticides for the control of second generation European corn borer (ECB), Ostrinia nubilalis (Hubner), on field corn in southwest Kansas. This test historically includes an evaluation of the southwestern corn borer (SWCB), Diatraea grandiosella (Dyar), but natural populations were absent from our 1990 plots. A mixed population of Banks grass mite (BGM), Oligonychus pratensis (Banks), and twospotted spider mite (TSM), Tetranychus urticae Koch, was observed to determine if any of the insecticides were miticidal or caused spider mite numbers to "flare".

PROCEDURES European Corn Borer
This test was conducted using a natural infestation of European corn borer in a furrow-irrigated corn field at the Southwest Research-Extension Center, Finney County, Kansas. Treatments were arranged in a randomized complete block design with four replications. Plots were four rows (10 ft) wide and 50 ft long, with a 4-row (10 ft) border of untreated corn on each side and a 10-ft alley at each end. Simulated chemigation applications were made using three Delavan 100/140, 3/4-in, raindrop nozzles mounted on a high clearance sprayer at tassel height between rows. This system was calibrated to deliver the equivalent of a 0.2-in irrigation on the two center rows (5227 gal/a). Standard treatments were applied with a high clearance sprayer using a 10-ft boom with three nozzles directed at each row (one nozzle directly over the row and one on each side of the row on 18-in drop hoses). The sprayer was calibrated to deliver 20 gal/a at 2.3 mph and 30 psi.
Ample first generation ECB larvae were collected between 27 June and 9 July for use with Kansas State University's European Corn Borer Software model, which predicts the second generation egg laying period. The model predicted 25-50% oviposition to occur during a 10-day period from 19 July to 29 July, which is earlier than normal. The predicted oviposition period coincided with peak light trap catches of European corn borer moths at the SWREC (see report page 32). During this period, we examined local corn fields visually to fine-tune the insecticide application date. All treatments, except the simulated chemigation application of Dipel, were applied on 27 July. Capture was applied at the 0.04 lb[AI]/acre corn borer rate in this test. Treatments were interrupted on 27 July by a 0.88-in rainfall. Another 0.44-in rainfall occurred on the morning of 28 July. The simulated chemigation application of Dipel was completed on 30-31 July. Although our results suggest no statistically significant (and very little numerical) difference between the late, "chemigated", Dipel treatment and the properly timed, standard, Dipel treatment, the lateness of application of Dipel should be considered when interpreting the results of this test. In past tests, simulated chemigation applications of Dipel performed numerically (but generally not statistically) somewhat better than standard spray applications. Comite was applied to one set of plots on 6 Aug to produce a "reduced-mite check" designed to prevent spider mites from rendering corn plants unsuitable as hosts for corn borers and to help us determine the effect of spider mites on corn borer populations and corn yields. The mite-free check plots were purposely treated late in an attempt to avoid a direct effect on corn borer larvae.
Corn borer counts were made between 29 Aug and 10 Sep by dissecting a total of 15 corn plants from the two center rows of each plot (8 consecutive plants left row, 7 consecutive plants right row). This was approximately 2-3 weeks earlier than in 1989 because of unusually high temperatures and rapid maturity of the crop in late August. The number of live corn borer larvae, the number of plants with tunneling (number of plants infested), and the total length (cm) of tunneling were recorded for each plant, and the 15-plant totals were data analyzed using SAS Proc ANOVA (Proc GLM, if data were missing).

Spider Mites
To determine the effect of corn borer insecticides on spider mite populations, two plants were selected from each of the two center rows of each plot and flagged. Naturally occurring populations of BGM were relatively evenly distributed and reached numbers such that artificial infestation was not necessary. Spider mite counts were made before treatment on 26 July by visually searching one-half of every other leaf (one-quarter plant) on the flagged plants for large (adult female) spider mites. Onequarter plant, spider mite counts were repeated on 2 and 15 August (6 and 19 DAT, respectively). Results were converted to mean number of spider mites per one whole plant (n = 4) and analyzed statistically using SAS Proc ANOVA (SAS Proc GLM for analyses with missing data). On each sample date, samples of spider mites were taken from the four flagged plants using a vacuum sampler and mounted on glass slides for microscopic determination of species. There were no significant differences in proportion of BGM (arcsine-square root-transformed) among treatment plots (Table 3).
Because the proportion of BGM remained relatively high during the study, the spider mite population was treated as a single-species complex rather than as separate species. Percent control of mites was calculated using the Henderson & Tilton formula, which adjusts the percent control for increases or decreases in mite numbers that occur in the untreated check.
Harvest yields (bu/acre), adjusted to 15.5% moisture, were estimated by collecting a 1/1000-acre sample of ears (8.7-ft sections of row from each of the two center rows, 30-in rows). Test weights (lb/bu) of samples were determined electronically.

European Corn Borer
The light-trap catch of the European corn borer, reached a 4-year high on the night of 24-25 July (see page 32). Corresponding oviposition was moderate and occurred over a relatively short period of time.
In interpreting the results of this test, the interactive effects of the high spider mite populations and moderate corn borer larval infestations should be considered. Relatively mediocre but statistically significant (p<0.01) spider mite control in the reduced-mite check plot may have contributed to numerically more corn borer larvae and greater number of corn borer-infested plants and significantly more (p<0.01) corn borer tunneling than in either untreated check plot (Table 1). Plants protected from high spider mite numbers may be more suitable for corn borer development.
All treatments significantly reduced (p<0.01) the length of stalk tunneling compared to the reduced-mite check (Table 1). All treatments, except Asana XL, significantly reduced (p<0.01) the number of live larvae and number of plants infested compared to the reduced-mite check. Capture resulted in the highest level of control (p<0.01) in this test, giving 100%, 96%, and 99% reduction of live larvae, number of tunnels, and length of tunneling, respectively, when compared to the reduced-mite check. Furadan resulted in significant (p < 0.01) and generally acceptable (>70%) control in all three damage categories. All but one Dipel and MVP treatment resulted in generally less than acceptable (34-68%) control. The standard application, high rate treatment of MVP resulted in acceptable (75%) reduction of length of tunneling. There is lack of evidence for an MVP rate response.
Capture-treated plots yielded significantly (p=0.05) higher than either untreated check but were not significantly different from the reducedmite check. The level of spider mite suppression achieved in the mite-free check plots did not by itself prevent of yield reduction. Both spider mite sup-pression and corn borer control seemed to be necessary to prevent yield reduction in this test. There was no significant difference (p=0.45) in test weight (lb/bu) among treatments.

Mite Species Complex
Unlike the situation in 1989, the spider mite species composition remained predominantly Banks grass mites (94% before treatment and 86% at 19 DAT) throughout the test period (Table 3). The spider mite species composition in other area fields ranged from high percentages of BGM to high percentages of TSM. Capture (remember-at the corn borer rate) did not result in significant (p=0.17) spider mite control at 6 DAT, even though it resulted in considerable numerical suppression of spider mites (Table 2). At 6 DAT, Capture was the only treatment resulting in reasonable suppression (<300 mites per plant). By 19 DAT (10 DAT for the Comite application), the only treatment that significantly (p<0.01) reduced mites was the reduced-mite check, but remaining numbers of mites were unacceptably high (>1000/plant). At 19 DAT, the Capture plots had significantly higher spider mite numbers than the mite-free check.
Asana did not appear to "flare" mites as sometimes occurs. In 1989, some treatments of MVP (MYX 7275) resulted in significant control of BGM; in 1990, MVP and Dipel generally had a neutral effect compared to the untreated checks. Means in the same column followed by the same letter do not differ significantly (DMRT). 1 Except Dipel ES, which is given in pints/acre, and MVP, which is listed in quarts/acre. 2 This product not currently registered for use on field corn. 3 Applied on 6 Aug, much later than other treatments in an attempt to prevent severe spider mite damage, which makes corn plants less suitable hosts for corn borer larvae. Note: This rate is intentionally higher than the label rate. Means in the same column followed by the same letter do not differ significantly (DMRT). 1 Except Dipel ES, which is given in pints/acre, and MVP, which is listed in quarts/acre. 2 This product not currently registered for use on field corn. 3 Applied on 6 Aug, much later than other treatments in an attempt to prevent severe spider mite damage, which makes corn plants less suitable hosts for corn borer larvae. Note: This rate is intentionally higher than the label rate. 4 Because of the late simulated chemigation application of Dipel ES, the corresponding DAT should be 3 and 16. ************************************************************************************************* We would like to acknowledge the very able assistance of Todd Staats, Steve Sandoval, Mike Sandoval, and Mary Belland. Their patience, persistence, and attention to detail resulted in a very useable, complete, and reliable data sets for all 8 tests we conducted this year. We would also like to express our appreciation to Bill Spurgeon for allowing Janet Baier and Thom Makens to assist us with two of our tests. Don Dick and Randy Dick were kind enough to contribute some of their vacation time to help with our plot harvest. The SWREC Farm Crew contributed greatly to the success of all our tests this year. ************************************************************************************************* SUMMARY Hot and dry weather during the last half of the growing season contributed to a heavy infestation of spider mites in our miticide plots. Several miticides resulted in both a significant reduction in numbers of spider mites and generally acceptable control at 6 days after treatment (DAT). None of the miticides resulted in acceptable control at 18 DAT.

INTRODUCTION
This trial was conducted to evaluate the efficacy of several miticides against the Banks grass mite (BGM), Oligonychus pratensis (Banks), and the twospotted spider mite, Tetranychus urticae Koch. The proportion of BGM remained relatively high throughout the test, so the species complex is treated as a single unit in the following discussion.

PROCEDURES
This experiment was conducted in a furrowirrigated corn field at the Southwest Kansas Research-Extension Center, Finney County, KS. Treatments were arranged in a randomized complete block design with four replications. Plots were four rows (10 ft) wide and 50 ft long, with a 4-row (10 ft) border of untreated corn on each side and a 10-ft alley at each end. Thio-Sul, an ammonium thiosulfate fertilizer solution often applied to the growing corn crop through center pivot irrigation systems, was included in this test in order to attempt to substantiate popular press reports (The High Plains Journal, March 12, 1990, pp. 1A-2A) that it had miticidal properties. Simulated chemigation applications of Thio-Sul were made on 27 July using three Delavan 100/140, 3/4-in, raindrop nozzles mounted on a high clearance sprayer at tassel height between rows. This system was calibrated to deliver the equivalent of a 0.2-in irrigation on the two center rows (5227 gal/a). All standard-volume treatments, except the experimental insecticidal soap (applied 31 July), were applied on 27 July with a high clearance sprayer using a 10-ft boom with three nozzles

EFFICACY OF MITICIDES AGAINST SPIDER MITES IN CORN
by Gary Dick, Phil Sloderbeck, and Steven Posler directed at each row (one nozzle directly over the row and one on each side of the row on 18-in drop hoses). The sprayer was calibrated to deliver 20 gal/a at 2.3 mph and 30 psi.
Two plants were selected from each of the two center rows of each plot and flagged. Prior to the test, we determined that a naturally occurring population of predominantly BGM was fairly evenly distributed and had reached numbers such that artificial infestation was unnecessary. Spider mite counts were made before treatment on 26 July by visually searching one-half of every other leaf (onequarter plant) on the flagged plants for large (adult female) spider mites. One-quarter plant, spider mite counts were repeated on 2 and 14 August (6 and 18 DAT, respectively). Results were converted to mean number of spider mites (n=4) per one whole plant and analyzed statistically using SAS Proc ANOVA (SAS Proc GLM for analyses with missing data). On each sample date, samples of spider mites were taken from the four flagged plants in each plot using a vacuum sampler and mounted on glass slides for microscopic determination of species. The percent BGM values were transformed using the arcsine-square root transformation before analysis and back-transformed for presentation. Percent control of mites was calculated using the Henderson & Tilton formula, which adjusts the percent control for increases or decreases in mite numbers that occur in the untreated check plots. The value used for the untreated check in the Henderson & Tilton formula was the mean of the two untreated checks. Examination of the numerical differences in number of mites and yield between the two untreated checks (Table 1) shows why seemingly large differences in resulting numbers of mites in treatment plots are often not statistically different.
Harvest yields (bu/acre), adjusted to 15.5% moisture, were estimated by collecting a 1/1000-acre sample of ears from an 8.7-ft section of each of the two center rows, 30-in rows. Test weights (lb/bu) of samples were determined electronically.
Unlike the situation in 1989, the spider mite species composition remained predominantly Banks grass mites (95% before treatment and 77% at 18 DAT) throughout the test period ( Table 2). The spider mite species composition in other local fields ranged from high percentages of BGM to high percentages of TSM.

RESULTS AND DISCUSSION
There was no significant difference (p=0.40) among treatments in numbers of mites per plant before treatment (Table 2). There was no significant difference (p=0.4, 0.3, and 0.13 for respective sample dates) among treatments in proportion of BGM (Table 2). At 6 DAT, 8 of the 14 miticide treatments resulted in a significant reduction (p<0.01) in number of spider mites to generally acceptable levels (<300 per plant) ( Table 1). Although there was no significant difference in spider mite numbers among the eight treatments, Capture and Supracide resulted in >80% control. At 18 DAT, only Avid, Capture, and Kelthane resulted in significant reduction (p=0.01) of mite numbers (depending on which check plot used for comparision), but the level of control acheived was generally unacceptable.
Despite failure to acheive acceptable spider mite control for more than a few days, there were significant differences (p = 0.01) in yield among treatments. Capture resulted in a significantly higher yield than either untreated check plot. Cygon, Asana plus Cygon, and Furadan plus Disyston resulted in significantly higher yields than one, but not both, of the untreated check plots. Higher yields with these four treatments may be attributable to their combined effect on spider mites and corn borers. Unfortunately, none of the materials used in this test was specific for corn borers. Future miticide tests should include a specific corn borer insecticide, such as one of the Bacillus thuringiensis products.
Asana did not "flare" mites in this test. The addition of Asana to Cygon did not significantly improve the performance of Cygon in this test. Furadan did not perform as well in 1990 as it has in the past, but the combination of Furadan and Disyston performed reasonably well at 6 DAT. Thio-Sul did not have a significant effect on spider mite numbers in this test. Means in the same column followed by the same letter do not differ significantly (DMRT). 1 Except Sunspray, which was applied as a 2% (vol:vol) solution; the experimental soap, which was applied as a 1% (wt:vol) solution; and Thio-Sul, which is an ammonium thiosulfate fertilizer solution (12-0-0-26) and was applied at 18 gallons of product per acre. 2 Percent control was calculated using the Henderson and Tilton formula from the mean of the two untreated checks. 3 This treatment was not applied until 31 July, so counts are at 2 and 5 DAT. ** These products not labeled for use on field corn. Infestations of first generation European corn borer larvae ranged from 0 to 61% of plants. There appeared to be an inverse visual correlation between level of first generation European corn borer infestation and planting date (later planting = lighter first generation corn borer infestation). Based on samples of first generation larvae collected on 12 and 27 June and 7 July, the model predicted second generation oviposition to occur from 19 to 29 July. Field scouting was begun on 19 July just prior to the estimated oviposition period and was used to verify second generation oviposition and to fine-tune insecticide applications on experimental plots and bulk corn at the Southwest Research-Extension Center. Bulk corn fields were treated for a combination of heavy spider mites and moderate second generation corn borers on 26 July (uppercase b, Figure 1). Experimental plots were treated on 27, 30, and 31 July (uppercase a, Figure 1).

SUMMARY
Irrigation frequency did not affect yields. Therefore, switching to a LEPA system and applying smaller amounts to minimize runoff should not affect yields adversely. Yield is significantly reduced by underirrigation and is not significantly increased by overirrigation.
LEPA is easier to justify when purchasing a new sprinkler, because the cost difference is smaller (approximately $5,000). Converting an existing system to LEPA is much harder to justify, unless water costs are high and the producer is currently underirrigating the crop.

INTRODUCTION
A Low Energy Precision Application (LEPA) sprinkler system was installed at the Southwest Research-Extension Center in 1989. This report summarizes the results and procedures for 1989 and 1990.

PROCEDURES
Corn was planted in a circle on May 19. The late planting was due to a wet spring and high residue cover. The system was run around once to establish the tower tracks, used as a marker. The corn was planted from the even towers (i.e. towers 2, 4, and 6) out to the odd towers. A total of 200 lbs of nitrogen was applied in 50 lb increments through the system four different times during the growing season.
The flexible drop hose initially installed was replaced with PVC pipe in late June, 1990. Dual nozzles also were installed in some locations at this time. The dual nozzles allowed the amount of water applied to the research plots to be varied. Once the center pivot was out of the plots, the desired rate for the bulk corn could be applied.
Aluminum access tubes were installed for use with a neutron probe to determine soil moisture. Measurements were taken weekly to verify estimates of crop water use and were used to calculate the change in soil water over the season.
The field was furrow diked to help prevent runoff. Dikes or deep ripping are used with LEPA systems to store water for infiltration and prevent excessive runoff.
Irrigation treatments of 0.4, 0.7, 1.0, and 1.3 times evapotranspiration (ET-estimated crop water use) were used. The rated flow was changed for the nozzles by the respective percentage. Irrigation frequencies of 3, 6, and 9 days were also used. Each treatment was replicated four times. A typical replication is shown in Figure 1. Each of the first three irrigations (June 15, 23, and 25) was used to apply 50 lbs of nitrogen. Each plot received the same total amount of water. Plots were then irrigated every 3, 6, or 9 days with the desired fraction of ET. At the end of each time interval, we replenished the amount of water used during that interval. A final 50 lbs of nitrogen was applied on July 25 during corn pollination. Irrigation amounts for each plot varied by treatment and frequency. Application amounts ranged from 0.4 to 3.8 inches per irrigation event. The 3-day frequency was used to study the effects of high frequency applications. LEPA systems will probably require amounts less than 1 inch because of high runoff potential. The 9-day frequency resulted in very high water applications for LEPA, but the plots were bordered to contain the water. Thus, the 9-day treatment resembled low frequency irrigation like furrow irrigation.
Forty feet of row were hand harvested from each plot on October 9. Yields were adjusted to 15.5% moisture and reported in bushels per acre.

RESULTS AND DISCUSSION
This study was patterned after a study at Texas A & M conducted by Dr. Bill Lyle. The Texas study used the same amount and frequency treatments but added a 12-day frequency.
These data (Figure 2) and the Texas data show that irrigation frequencies of 3, 6, and 9 days are not significantly different. The yields for 12-day frequency were significantly lower than yields for the 3-, 6-, and 9-day treatments. Yields for all treatments are given in Table 1. These data indicate no yield losses when high frequency irrigation is required, such as for a LEPA system.  Figure 3 shows that yields level off for amounts greater than 1.0 ET. This presents a case for using irrigation scheduling to help the producer obtain optimum yield without wasting water. As expected, yields increase significantly with irrigation amounts up to 1.0 ET. The combined data from 1989 and 1990 showed significant differences among all amount treatments, except 1.0 and 1.3 ET. The seasonal soil water change is given in Table  2. A negative value shows that water was extracted from a 5-ft. profile between June 30 and September 22 (1989) and June 27 and October 3 (1990). Soil water was monitored in the 3-, and 9-day treatments for each replication in 1990. In 1989, only one replication was monitored. In the underwatered irrigation treatments, water was generally extracted from the soil profile to help meet the crop's water needs. 14.1 (6.5) -(7.5) (7.0) 0.7 11.2 12.3 11.2 11.6 (7.3) -(7.0) (7.2) 1.0 9.9 9.8 9.0 9.6 (7.3) -(6.6) (7.0) 1.3 8.0 8.3 7.3 7.9 (6.3) -(5.7) (6.0) Avg. 10.7 11.1 10.6 (6.9) -(6.7) Similar results were obtained for each year, despite the difference in rainfall. We received 15.4 inches of rainfall during the 1989 growing season and 7.2 inches in 1990. The irrigation amounts applied were 11.9 inches in 1989 and 22.2 in 1990 for the 1.0 ET treatment. This results in a total of 27.3 and 29.4 inches, respectively, for the 1.0 ET treatment.
Total water use is shown in Table 3, including seasonal soil water change, irrigation, and rainfall amounts.
The total water use and irrigation water applied were used to calculate total water use efficiencies (TWUE) and irrigation water use efficiencies (IWUE). Both are shown in Table 4. Water use efficiency is defined as the corn yield divided by the appropriate water quantity (bu/A-in).
The LEPA concept is to keep every other row dry to reduce evaporation losses. Slopes greater than 0.5 to 1.0 percent will produce significant runoff and reduced yield. Therefore, furrow diking is recommended for all LEPA systems. The plots were not furrow diked in 1989, because fields were too wet from excessive rainfall during June. This may be the reason yield was lower in 1989. Improved corn yields might have resulted from using the flat spray mode rather than the bubble mode. Only 0.42 inch of rainfall fell between June 1 and July 19 (1990), but 3.06 inches of rain fell between July 19 and August 2. During a hot dry year like 1990 and using the above rainfall amounts, the soil profile (4 ft) would have approached 3 inches depletion for a system capacity of 5 gpm/ac at 100 percent efficiency (Figure 4). Over 4 inches would have been depleted in a 4-ft profile with a capacity of 4 gpm/A. Both maximum depletion levels would have oc-curred around July 19, near or after pollination, the most critical growth stage. Assuming that LEPA is 98 percent efficient and the soil holds at least 2 inches per ft in a 5-ft soil profile, fully irrigated corn may be possible with 5 gpm/ac or less. However, that would not leave any extra capacity for system repair, and the grower would assume more risk in meeting the crop's water needs.
The current cost to convert an existing system to LEPA is approximately $10,000. It is hard to justify conversion unless fuel costs are high and water is limiting (i.e., the producer is currently underirrigating). It is possible, however, to pay off the difference in cost between spray heads and LEPA heads (approximately $5,000) for new installations in a 3-to 5year period, depending on fuel costs and corn prices. Diking with ripping worked best on the slopes studied (1 to 6 percent). The flat spray mode was more effective than reservoir tillage. The combination of flat spray mode and reservoir tillage produced the highest yields.

INTRODUCTION
Low Energy Precision Application (LEPA) sprinkler systems produce high application rates because of the small wetted diameters of the nozzles. On sloping ground, this can cause considerable runoff. A study was initiated in 1990 to provide the producer with effective guidelines for managing LEPA systems on slopes greater than 1 percent.

PROCEDURES
Corn was planted on May 19 in a circle. Various tillage treatments and spray modes were used to determine which combination reduces runoff the most. Slopes ranged from 1 to 6 percent and averaged 3 percent.
Tillage treatments included furrow diking (forming basin reservoirs between rows), in-furrow ripping, and a combination of ripping and furrow diking ( Figure 1). Dikes, small reservoirs dug into the soil surface, and deep ripping are used to hold water until it can infiltrate into the soil.
All treatments were irrigated by the "bubble" and flat spray modes. The "bubble" mode concentrates the water into a small area directly beneath the nozzle (approximately 1.3 ft. in diameter). The flat spray spreads the water out over a greater area (approximately 10 ft.).
Aluminum access tubes were installed for use with a neutron probe to determine soil water content. Soil water measurements were taken weekly to calculate the change in soil water over the season.

Dike/Rip
The first irrigation was on June 15, and plots were irrigated approximately once a week thereafter. The irrigation application amount was kept below 1 inch, the current recommendation for flat slopes. Borders were installed across the field to prevent water from one treatment from running onto any treatment further downhill.
Forty feet of row were hand harvested from each plot on October 8. Yields were adjusted to 15.5% moisture and reported in bushels per acre.

RESULTS AND DISCUSSION
Runoff rates were so high in the "bubble" mode that corn yields were reduced (Figure 2). Ripping and furrow diking increased yields slightly (Table  1). Diking with ripping increased yields the most (Figure 3). Furrow diking by itself did little to increase corn production on these slopes.
The furrow dikes may have been too shallow (not properly installed) to hold the water applied and washed out early in the season.   Diking with ripping had the greatest effect on yields when the "bubble" mode was used. This could be because of the increased intake rate from ripping and because this treatment had the best reservoirs.
The flat spray mode showed less sensitivity to tillage treatment because of the larger area wetted as compared to the "bubble" mode.
The seasonal soil water change for the period between June 27 and October 10 is given in Table 2. -4.6 -2.9 * A negative value shows soil water was extracted from the profile by the crop.
Total water applied is shown in Table 3, including the seasonal soil water change, irrigation (21. 1 inches), and rainfall (7.2 inches) amounts. Not all of the water applied was available for use by the crop because of runoff from the plot area. The total water and irrigation water applied were used to calculate total water use efficiency (TWUE) and irrigation water use efficiency (IWUE). Both are shown in Table 4. Water use efficiency is defined as the corn yield divided by the appropriate water quantity. A study of drip-line spacing and plant population for corn was conducted in 1989 and 90. Twoyear corn yield averages were180 and 201 bu/a for line spacings of 7.5 ft and 2.5 ft, respectively. Yields from the 7.5 and 10 ft. spacings were lower than that from the 2.5 ft. spacing. The soil water content decreased in the upper 2 to 3 ft as close as 15 inches from the drip line. Yields from population treatments were different and peaked at 199 bu/a for the 32,000 plants/a treatment.

INTRODUCTION
Water tables in southwest Kansas are declining; therefore, producers want to use their water efficiently to allow the resource to last as long as possible. Producers might consider drip irrigation to save water, if production were profitable.
A drip irrigation study was initiated at the Southwest Research-Extension Center in 1989.
Objectives of the study are: (1) to determine optimum plant population, (2) to determine the effect of drip line spacing on yield, and (3) to determine the effect of drip line spacing on water movement.

Plot Layout
The field was fertilized with 200 lbs of nitrogen and 60 lbs of phosphorous. Drip lines were buried 16 inches below the ground surface and spaced 2.5, 5.0, 7.5, and 10 ft apart in a silt loam soil. Corn was planted on May 1 in 30-inch rows perpendicular to the drip lines and thinned to populations of 38,000, 32,000, 26,000, and 20,000 plants/a. Each plot consisted of four crop rows. Populations were replicated four times.

Soil Water Monitoring Method
Aluminum access tubes were installed in increments of 7.5 inches from a drip line in each spacing replication. The access tubes were installed in the 32,000 plants/a population treatment. A neutron probe was used weekly to determine the soil water status.

Irrigation Method
All spacing treatments were irrigated to apply 100 percent of evapotranspiration (ET -crop water use). Therefore, each plot received the same gross average depth. The wide spacing treatments received enough water to cause deep percolation. This was done so that maximum horizontal water movement was not hindered.
The drip lines were 195 ft long and were rated at 0.3 gpm per 100 ft. Set times for the various spacings needed to apply an average depth of 0.5 inch over the plot area were: 4.3 hr for 2.5 ft, 8.6 hr for 5.0 ft, 13 hr for 7.5 ft, and 17.3 hr for 10 ft. Set times were reduced slightly by operating the system at 15 psi rather than the suggested pressure of 10 psi.
June 1990 was hot and dry, which enhanced spacing effects. There was a measurable decrease in plant height, about 15 inches, between drip lines for the wide spacings.
The first irrigation occurred on June 16. Plots were irrigated by replenishing ET when the soil water deficit reached 0.5 inches. A total of 21.9 inches of irrigation water was applied.

Harvest Samples
Each plot consisted of four corn rows and four drip lines. The two middle corn rows were used for yield samples. One row was used for bulk yield samples and the other row for individual plant yield. Because the drip lines were perpendicular to the corn rows, the length of row harvested was equal to two times the drip line spacing. The sample began halfway between the first and second drip lines and spanned across the two middle drip lines.

Data Analysis
Both bulk yield and individual plant yield samples were taken. An analysis of variance was performed on the bulk yield samples for population and drip line spacing treatments. Individual plant yield (mass of grain per plant) was collected but has not been analyzed. Figure 1 shows the 2-year average yield for the various populations for each spacing treatment. Also, the 2-year average population treatment yields are shown in Figure 2. Yields for 1989 and 90 are given in Table 1. Yield differences were statistically significant and peaked for 32,000 plants/a.   Two-year average yields for the spacing treatments are shown in Fig. 3. Yields were higher for narrow drip line spacing, although they stayed relatively high for the wider spacing. We received 10.2 inches of rainfall, with 2 in. on July 19, during the 1990 growing season.

Soil Water Movement
Soil water content was monitored weekly to a depth of 8 ft. Access tubes were placed at 15-inch increments away from the drip lines in 1989 and 7.5 inches in 1990. This was done for all of the spacing treatments in the 32,000 plants/a population treatment.
The average volumetric soil water content for the upper 4 ft at 3, 15, 30, 45, and 60 inches from the drip line for one of the 10 ft spacing treatments is shown in Figure 4. Also, rainfall and irrigation events are shown. This figure shows that we were able to maintain high soil water contents 15 inches from the drip line. Soil water content decreased as the distance away from the drip line increased and approached a value dependent on rainfall rather than irrigation.
Our data show that the volumetric soil water content approached 60 percent of the available water content at 45 and 60 in. from the drip line. This dry region extends 2 to 3 ft below the soil surface for both the 7.5 and 10 ft spacing treatments. Corn height was about 1.5 ft shorter in-between drip lines for the 7.5 and 10 ft spacings.

SUMMARY
A study of water requirement for corn using buried drip lines was initiated in 1990. Corn yielded 176 bu/A for the full ET irrigations and 134 bu/A for the dryland plots. Yields were reduced by a hail storm on July 19. The horizontal movement of water was adequate to supply water to corn rows 15 inches away, yet provided little water for weed growth in the furrows.

INTRODUCTION
This study was designed to evaluate the use of buried drip line irrigation for corn in Holcomb, Kansas. The corn was irrigated at various fractions of evapotranspiration (ET).
The objectives of this study are: 1) to determine the water requirement of corn grown with drip irrigation and 2) examine the feasibility of largescale adoption of drip irrigation for row crops in Southwest Kansas.

PROCEDURES
Corn was planted on May 9 in 30-inch rows on 60-inch beds. Each bed was irrigated by a drip line running through the center of the bed, 16 inches deep. Each drip line watered 2 corn rows, 15 inches to either side of the line. There were 6 irrigation treatments. They were 0, 0.25, 0.5, 0.75, 1.0, and 1.25 times ET.
Access tubes were installed in every plot in the corn row, 15 inches from the drip line, for use with a neutron probe. The neutron probe was used to determine soil water to a depth of 8 ft. Also, access tubes were placed at 3, 7.5, 15, 22.5, and 30 inches from the drip line in the 1.0 ET treatment only. This enabled us to study the horizontal movement of water away from the drip line.

RESULTS AND DISCUSSION
Differences in yields were observed for the 1990 harvest (Table 1 and Figure 1). Timely rains contributed to high yields for the low water-amount treatments. A hail on July 19 reduced yields for the high water-amount treatments.

40
The 1.0 ET treatment received 19.3 inches of irrigation water (Table 2). The rainfall was 9.2 inches between June 7 and September 24. The 1.0 ET treatment had the highest yields, 176.1 bu/A. The 1.25 ET treatment yielded 174.1 bu/A. The increased amount of water did not increase yields. This may have been due to loss of aeration and/or the leaching of fertilizer below the crop root zone.
Irrigations were frequent, using small amounts (0.75-1.25 inches). Soil water status by ET level throughout the season is shown in Figure 2.
The lateral movement of water did not appear to be hampered by the subsurface drip system ( Figure  3). The soil water was maintained at 15 inches from the drip line. However, the soil water level dropped off beyond 15 inches. As expected, 30 inches from the drip line, the soil water was lowest. Of course, this is the furthest point from the drip line and has the advantage of keeping the furrow dry and suppressing weed growth. * The change in soil water content is for 4 ft. of soil profile, from June 7 to September 24. A negative value means soil water was extracted from the profile by the crop. ** The water use efficiency = yield + total water use.

INTRODUCTION
Subsurface drip was used to irrigate corn in Holcomb, Kansas. This is a method of supplying low volumes of water to the root zone, thus minimizing evaporation losses and potentially reducing deep percolation losses. Eight different frequencies of irrigation were used, and the yields were compared.
The objective of this study is to determine the effect of frequency and amount of irrigation on crop yield and soil water content.

PROCEDURES
Drip lines were buried 16 inches deep in the center of each bed, running parallel to the crop rows. Therefore, each drip line supplied water to 2 corn rows 15 inches away. The corn was planted on May 9, on 60-inch beds. The study consisted of eight watering treatments. The treatments were 1-, 3-, 5-, and 7-day watering intervals and 0.5-, 1-, 1.5-, and 2-inch depletion levels. The evapotranspiration (ET) was calculated to determine the amounts to be watered for each treatment. The depletion-level treatments were watered when the depletion reached the stated amount, and frequency plots received the amount of water used during the specified interval.
Access tubes were installed in every plot in the corn row, 15 inches from the drip line, for use with a neutron probe. The neutron probe was used to determine soil water to a depth of 8 ft.

RESULTS AND DISCUSSION
Statistically, there were no differences among the yields of the different treatments ( Table 1). The highest yielding plot was the 7-day study, with 181 bu/A. The lowest was the 3-day treatment, with 162 bu/A. Therefore, it is difficult to draw any conclusions because of the lack of statistical difference and the uneven damage caused by the July 19 hail storm ( Figure 1).  The total water applied showed differences among treatments (Table 2). Rainfall was 9.2 inches between June 6 and September 24. The less frequent and the larger amount treatments allowed the soil to dry out in between irrigations and thus have the ability to store rainfall. Because irrigating 1 inch takes 21 hours, there is a limit to how frequently 1 inch may be applied. All treatments were brought back to field capacity at each irrigation. Also, we continued to irrigate during rain storms to stay consistent and to avoid the error that would be caused by variations in the irrigation amounts. However, practical management, i.e., leaving a deficit for the storage of rainfall, could reduce irrigation amounts.
The soil water was monitored weekly. Access tubes were usually read just prior to irrigation. The larger amounts and less frequent irrigation treatments generally dried out more between irrigations ( Figure 2). However, treatments did not fall below 75% of available water content. Preliminary results reveal inconsistent variation in corn yield with respect to line lengths of 330 and 660 ft. and irrigating up-or downgrade. This inconsistency might have occurred because of variable hail damage. Smaller variation is expected under normal conditions, indicating that line length can be increased without reducing yield. Increasing line length will reduce installation costs of drip irrigation by 10-20 percent.
Objectives of the study are: (1) to determine the effect of length of drip lines on corn yield and (2) to determine the effect of water flow upgrade and downgrade on corn yield.

Plot Layout
Drip lines were buried 16 inches below the ground surface and spaced 60 inches apart in a silt loam soil. There were four drip lines per length treatment. Lengths of 330 ft. and 660 ft. were used, and for each length, the water flowed from the up-or downslope end. Also, one of the 660-ft. treatments had water pumped in from both ends. The slope was about 0.15 percent. Corn was planted May 9 in 30 inch rows parallel to the drip lines. Each plot consisted of eight crop rows.

Soil Water Monitoring Method
Aluminum access tubes were installed in the corn rows and were 15 inches from the drip line. They were read to a depth of 8 ft. A neutron probe was used weekly to determine the soil water content.

Irrigation Method
All treatments were irrigated to apply 100 percent of evapotranspiration (ET -crop water use). Plots were irrigated when the depletion reached 0.75 to 1.25 inches. The first irrigation occurred on June 16. Plots were irrigated by replenishing ET. A total of 19.3 inches of irrigation water was applied. Rainfall was 9.2 inches from June 6 to September 24.
The drip lines were rated at 0.25 gpm per 100 ft. A pressure of 10 psi was maintained on all plots.

Harvest Samples
Each plot consisted of eight corn rows and four drip lines. The two middle corn rows were used for yield samples, and 20 ft. of row was harvested in each. The 660ft. length was harvested at both ends and two places along its length. The 330-ft. length was harvested on both ends.

Data Analysis
Corn yield per acre was calculated from each of the sample areas. The yields were adjusted to a 15.5 percent moisture content. An analysis of variance showed no difference among corn yields.

RESULTS AND DISCUSSION
One season (1990) of data has been collected. Figure 1 shows differences in yield, but they were not consistent with our expectations. Furthermore, we would have expected the 330-ft. downslope flow to have the highest yield and the 660-ft. upslope flow to have the lowest.
A hail storm on July 19 affected yield by location A portion of the cost of drip installation is in in the field. Lower yields occurred on the west end feeder lines, which supply water to the drip lines. of the field (the upslope end), causing the 330-ft.
Therefore, assuming less yield difference when hail downslope treatment to have lower than expected damage is not present, longer lengths of drip line yield. Yield by position is given in Table 1. may be used to reduce installation costs.