H: E, 2;; E, : 3: f... .(I ‘ fi fiwot '. A. 4"? j. ‘n‘ bu 3 a: (Org. THESIS A COMPARISON OF FIELD METHODS OF MEASUREMENT OF THE WATER TABLE DRAWDOWN CHARACTERISTICS BETWEEN LATERAL TILE DRAINS IN BROOKSTON CLAY LOAM SOIL AND AN EVALUATION OF AN EMPIRICAL FORMULA FUR PLACING LATERAL DRAINS By Frank Robert Hore ~ AN ABSTRACT Submitted to the School of Graduate Studies of Michigan State College of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Agricultural Engineering Year 1953 59%:wa THE-3‘3 wan-'— ii FRANK ROBERT HORE ABSTRACT Two investigations are reported, both dealing with the design for placing lateral drains. In the first investigation, four field methods of measur- ing the water table drawdown characteristics were compared to determine whether there were any differences in observational data. The procedure used was to install two inch auger holes, three-eighths inch perforated wells, two inch perforated wells, and groups of six three-eighths inch piezometers in parallel rows to a tile drain at varying intervals symmetrically on each side of the drain. One set of nylon resistance blocks was installed in an attempt to evaluate the methods further. A statistical analysis of variance reveals significant dif- ferences between data from all methods except between data from the two inch and three-eighth inch well methods. With the qualification that more replications are necessary, it was concluded that observations from.cased wells lagged those from uncased holes, observations from.small sized wells lagged lthose from larger wells, and, under these study conditions, the piezometer method was more reliable based on results ob- tained from the resistance blocks. . In the second investigation, a newly reported design formula was evaluated under Michigan conditions. Data for the evaluation were taken from.the study site for the first investigation. ’The formula results indicated that the drain 3 1-0979 iii FRANK ROBERT HORE ABSTRACT spacing was too great, which was in agreement with long term field observations. Under the field conditions used, the formula also indicated that small increments of drain depth were critical. It was concluded that more specific information on the minimum recession rate for adequate drainage is needed. and that further field studies should be made to determine mm small increments of depth are as critical as the formula would indicate. It [1 .[7[‘L[ e[ 'lf’fllE-{L’l r a, k[.r'\|.ll|lx [’[IIL[[[[[[[U [I A COMPARISON OF FIELD METHODS OF.MEASUREMENT OF'THE WATER TABLE DRAWDOWN CHARACTERISTICS BETWEEN LATERAL TILE DRAINS IN BROOKSTON CLAY LOAM SOIL AND AN EVALUATION OF AN EMPIRICAL FORMULA FOR PLACING LATERAL BRAINS By Frank Robert Hore A THESIS Submitted to the School of Graduate Studies of Michigan State College of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Agricultural Engineering 1955 ([[l[{illl)llllll[ll [ [fr ACKNOWLEDGEMENTS The author wishes to express his sincere thanks to Professor E. H. Kidder of the Department of Agricultural Engineering for his continued guidance and counsel during the course of this study. His unfailing interest proved to be a source of immeasur- able inspiration. He is also greatly indebted to Drs. A. E. Erickson, and E. P. Whiteside of the Department of Soil Science. To Dr. Erick- son, sincere appreciation is extended for his assistance with the soil tests and his provision of laboratory facilities; to Dr. Whiteside, the writer is deeply grateful for his assistance in classifying the soil. Grateful acknowledgement is also due to Mr. George Moore and to his son, Mr. Robert Moore, for their full cOOperation and permission to make the study on their farm. The author extends his gratitude to ur. James Cawood and all others who provided aid and helpful suggestions during the conducting of the study. TABLE OF CONTENTS Page INTRODUCTION . . . . . . . . . . . . . . . . . . . . . 1 REVIEW OF LITERATURE . . . . . . . . . . . . . . . . . Types and Movement of Soil Moisture. . . . . . . . . Soil PrOperties Affecting Ground Water Movement. . . u) c» .a 9 Criteria for Tile Drainage Design. . . . . . . . . . Methods of Ground Water Measurement. . . . . . . . . 11 Effects of Drainage. . . . . . . . . . . . . °.- . . 13 Previous Comparison of Methods of Water Table Measurement. . . . . . . . . . . . . . . . . . . . . l4 EXPERIMENTAL STUDY . . . . . . . . . . . . . . . . . . 15 The Study Area . . . . . . . . . . . . . . . . . . . 15 Installation of Observation Devices. . . . . . . . . l8 Piezometers. . . . . . . . . . . . . . . . . . . . 21 Two Inch Wells . . . . . . . . . . . . . . . . . . 23 Three-eighths Inch Wells . . . . . . . . . . . . . 24 Two Inch Auger Holes . . . . . . . . . . . . . . . 24 Nylon Resistance Blocks. . . . . . . . . . . . . . 26 Water Level Observations . . . . . . . . . . . . . . 27 Soil Tests . . . . . . . . . . . . . . . . . . . . . 28 Non-capillary Porosity . . . . . . . . . . . . . . 28 Permeability . . . . . . . . . . . . . . . . . . . 29 Volume Weight. . . . . . . . . . . . . . . . . . . 29 Results and Discussion . . . . . . . . . . . . . . . 30 vii Page INVESTIGATION OF AN EMPIRICAL DEPTH AND SPACING FORMULA. . . . . . . . . . . . . . . . . . . . . . . . 42 The Formula. . . . . . . . . . . . . . . . . . . . . 42 Results and Discussion . . . . . . . . . . . . . . . 44 CONCLUSIONS. . . . . . . . . . . . . . . . . . . . . . 47 APPENDIX . . . . . . . . . . . . . . . . . . . . . . . 49 LITERATURE CITED . . . . . . . . . . . . . . . . . . . 56 Figure l. 2. 5. 6. 10. 11. 12. 13. 14. 15. LIST OF FIGURES The steady state. . . . . . . . . . . . . . . Saturated to the surface. . . . . . . . . . Plan view of area around study site . . . Plan view of study site . . . . . . . . . . . Study site looking south. . . . . . . . . . . Field view of an observation point. . . . . Cross sectional view of the four methods and resistance blocks at N5 . . . . . . . . . . . Flushing the piezometers. . . . . . . . . . . Taking a water level reading from a two inch well with the water level indicator . . . . . Drawdown curves for the piezometers . . . . . Drawdown curves for the two inch wells. . . . Drawdown curves for the three-eighths inch wells . . . . . . . . ._. . . . . . . . . . . Drawdown curves for the auger holes . . . . Comparison of drawdowns of the four methods . Schematic diagram showing the location of formula symbols . . . . . . . . . . . . . . . Page 17 19 20 2O 22 25 25 31 32 33 34 38 43 LIST OF TABLES Table Page I. Analysis of Variance of Water Table Measurement Data . . . . . . . . . . . . . . . .1. . . . . . 36 II. Data from the Nylon Resistance Blocks at Observation Point N5 . . . . . . . .. . . . . . 40 III. Comparison of Water Table Elevations Observed by the Four Methods to the Elevation of the Shallowest Saturated Resistance Block at Observation Point N5 . . . . . . . .)t . . . . . 40 IV. Results of Soil Tests. . . . . . . . . . . . . . 50 V. Water Level Observations . . . . . . . . . . . . 51 INTRODUCTION Since 1835 when John Johnston laid the first system of underdrainage in the United States on his farm in New York, increasing numbers of farmers have realized the benefits de- rived from tile drainage of naturally poorly drained soil. Tile drains, however, have been installed at a depth and spac- ing that were arbitrarily felt necessary. The need for know- ledge of the most economical depth and spacing relationship was apparent. To obtain this knowledge, research has mainly taken the form.of comparing the drawdown characteristics of the water table between tile drains installed at various depths and spacings. In making these studies, investigators have utilized various sized observation wells and cpen auger holes. More recently, groups of piezometers have-been used to study ground water and water table movements. Controversy has often arisen over the relative ability of large and small diameter observa- tion holes, cased and uncased, to reflect the position of the receding water table in the finer textured soils. For example, it is felt by some that small diameter wells reflect the water table movement more quickly than large diameter wells. It is the primary objective of this study to compare the positions of the receding water table as reflected by four observational methods. - 2 - The four methods compared in the spring of 1953 were two inch diameter auger holes, three-eighths inch diameter per- forated wells, two inch diameter perforated wells, and groups of six three-eighths inch diameter piezometers. Observation devices for each method were installed in a row parallel to a tile drain. The rows were placed at varying intervals sym- metrically on each side of the drain, being closest near the tile. Nylon resistance blocks were installed at one obser- vational point to ascertain their ability to indicate the extent of the saturated zone within the soil profile. Heretofore, drainage engineers have found that field ex- perience has been a necessary supplement to their basic design knowledge. This experience consists mainly of familiarity with the various soil types in their work area, the related crops grown thereon, and the observed response from various tiling systems. For the last two decades or so, the objective of drainage investigational work has been to attain some reliable criterion to determine the preper depth and spacing of tile drains. However none of the previous preposals have gained wide acceptance. The secondary objective of this study is to evaluate Vunder Michigan conditions a newly reported design formula proposed by Walker (44). The criterion is based on core sam- ple permeability measurements of the least permeable layer of a stratified soil. The formula, although mathematical in derivation, is reduced to a simple form and is readily solvable. - 3 _ The required data to evaluate this formula were taken from the results of the soil tests obtained from the study site. REVIEW OF LITERATURE Types and Movement of Soil Mbisture The movement of water through a given volume of soil must take place through the soil pore space. This movement may occur in the liquid or the vapour phase. Briggs (9) in 1897 proposed the following classification of soil water: 1. Hygroscopic water, which is adsorbed from an at- mosphere of water vapor as a result of attractive forces in the surface of the particles. 2. Capillary water, which is held by surface tension forces as a continuous film around the particles and in the capillary spaces. 3. Gravitational water, which is not held by the soil but drains under the influence of gravity. This classification, which has gained the widest acceptance, was based on equilibrium points to express the way that water is held in the soil. Israelson (21) showed that under equilibrium conditions, the water content of the soil decreases with height above the water table. Trullinger (41) stated that results of investi- gations at the Utah Experimental Station indicated the presence of a surface of saturation well above the water table under equilibrium conditions, and that this surface gets lower as the soil texture becomes coarser from clay to sand. Further, Childs(12) recognized and, by electrical analogue, demonstrated the role of this zone of capillary saturation in relation to the moving water table. He concluded that the rate of fall of “VII ‘(flll -5- the water table or capillary fringe boundary depends on the thickness of the capillary fringe particularly when the soil is shallow. If the fringe is thick, the initial rate of fall is lower in comparison to the rate of fall when the water table is at some specified small height above the drain. This is due mainly to the fact that the capillary fringe provides a flow path of appreciable thickness even when the water table is at the drain level. Thus the tail end of the drainage per- formance may be completed more freely than when the capillary fringe is absent. Childs (12) also came to two other conclusions regarding the movement of the water table in homogeneous tile drained soil. First, the water table falls more slowly in a shallow soil than in a deeper soil, other factors being the same. Second, starting from the steady state, Figure l, the water table falls as a whole without appreciable change in shape, after which it naturally falls more slowly over the drain than elsewhere since the distance left to fall is smaller. When saturated to the soil surface, Figure 2, the water table falls faster over the drain than elsewhere, after which the movement is the same as in the steady state. Kirkham (24) and Donnan (16) demonstrated by means of model studies in homogeneous sands that flow lines follow a curved path or streamline of flow mainly in a vertical direction except near the tile. In stratified fine textured soils, the, streamline flow may be sharply restricted. - 5 - Childs (lO), Bodman and Coleman (6), and Marshall and Stirk (27) all concluded that there is no appreciable gradient in moisture content within the transmission zone during down— ward entry of water into uniform columns of soil. Terzaghi (40) illustrated that where there is a strata or pocket of gravel containing discontinuous capillary water in a fine silty sand containing semi-continuous capillary water, the downward entry of water from rainfall causes a temporary perched ground water table above the gravel. This zone of perched ground water is described as a zone of capillary saturation. Ground Surface /—— Ground Surface / and Wafer Tab/e ~Waéer Taé/e 1 I o 7T/e , 077%: , -_J Fig. l. The steady Fig. 2. Saturated state. to the surface. Soil Properties Affecting Ground water Movement Most of the investigations regarding the properties of a nature of movement of ground water have been made by ground water hydrologists. French and German engineers and geolo- gists were among the first to contribute to our knowledge of -7- ground water hydrology. In America, further notable contri- butions have been made by King (23), Meinzer (28), and Veatch (43). In 1856, Darcy (14) established the law of the flow of ground water. Darcy's Law, as commonly referred to, governs the relationship between the velocity of ground water flow, coefficient of permeability, and the hydraulic gradient in a homogeneous water bearing material. The accepted form of Darcy's Law is v = k%? where v = velocity of flow, k : coeffi- cient of permeability, dh = difference between the fluid heads at the inlet and outlet faces of the water bearing ma- terial, and l a length of the flow path through the material. As a result of Darcy's work, much investigational study has been made to determine the permeability of soils. The formal definition of permeability as given by musket (29) may be stated as the volume of a fluid of unit viscosity passing through a unit cross sectional area in unit time under the action of a unit pressure gradient. It is thus a constant -determined only by the structure of the medium in question and is entirely independent of the nature of the fluid. Permea- bility is expressed as cubic feet per square foot per second. Since this unit results in small fractional values, inches per hour or feet per day are more commonly used. The factors affecting the permeability of soils are the number, size, and shape of the non~capillary pore spaces. Other factors such as size, gradation, shape and arrangement (D M to Ca: equ. Call -8- of the soil particles are also governing factors. many at- tempts have been made to equate permeability to the variables which affect it. One of the earliest attempts was made by Hazen (20) who based his formula on the effective grain size. His experi- mental work was done largely with fairly homogeneous river sands and gravels. Slichter (38) based his formula on the mean diameter of the soil particles, while Fair and Batch (18) introduced specific surface and porosity into their equation. Muskat (29) showed striking examples of the failure of these formulae to give even approximately correct values of permea- bility. Childs and George (13) have criticized these formulae relating permeability to porosity and specific surface on the grounds that while porosity and specific surface are uniquely determined by the void space geometry, which also determines the permeability, the converse is not true. It is possible to conceive of different pore-size distribution, with presume ably different permeability, yet having the samm porosity and specific surface. Structural fissures contribute negligibly both to porosity and specific surface and yet they dominate the permeability. Roe and Park (33) concluded from their work that there is no mathematical relationship reliable enough for direct appli- cation to practical problems between the centrifuge moisture equivalent and permeability. Klute (25) obtained a statisti- cally significant correlation between the non-capillary pore -9- space and permeability as measured by air. Baver (4) con- cluded that texture alone may be used to estimate pore space relationships when the soil is primarily sandy or silty in character. In compacted soils, as well as those soils con— taining large amounts of clay, it is necessary to evaluate the pore space situation by means other than texture. Granu- 1ation and structure play significant roles in such soils. Terzaghi and Peck (39) have outlined formulas to evalu- ate the cver-all permeability of a stratified soil based on the thickness and permeability of each of the layers. How- ever, Luthin (26) disclosed that the theoretical value was 16 to 20 percent higher than the experimental value obtained. The possibility of a reduced permeability existing at the interface between the layers was checked, with negative re- sults. Criteria for Tile Drainage Design Norling (31) advocated that texture is one of the most important factors in drainage design. In 1934, Neal (30) summarized in tabular form the recommendations of fourteen previous investigators for tile drain spacings and depths. Most United States recommendations were only general, while most German recommendations were given by a formula. These investigators based their recommendations on some physical property of the soil such as hygrosOOpicity, permeability, or effective diameter of soil particles. Neal (30), however, -10- derived formulae and nomographs for depth and spacing of tile based on such simple soil characteristics and constants as moisture equivalent, plasticity and percentage of clay. Also, he observed that the lowering of the water table is due to deep seepage alone after the hydraulic gradient of the water table becomes less than one to five percent depending upon the soil and spacing of the tile. Limited observations indicated that crops were not seriously injured if the water table was held at least six inches below the surface and was lowered at the rate of one foot per day through the second six inch depth interval and at the rate of seven-tenths foot per day through the third six inch depth interval. The more recent attempts to formulate depth and spacing of tile have been based on the laboratory measurement of permeability of field samples. Aronovici and Donnan (l) and later Slater (3?) advanced a simplified approach by applica- tion of Darcy's Law to lateral conditions of flow. Assumptions for their formulae included homogeneous soil conditions with a barrier below. Suggestions were made regarding the method of application of the formulae to stratified soils. Aronovici and Donnan (l) utilized a weighted average of laboratory determined variations of aquifer permeability to obtain a comparable value to the field measurement. Slater (37) stated that the permeability of the soil layer at the depth at which the tile is placed governs to a marked degree the rate of flow. However, Baver (3) pointed out that the permeability of the -11- entire profile is determined by the permeability of the least pervious horizon. Kidder and hytle (22) suggested the probable use of field and laboratory permeability data and non-capillary pore space data as determined by a drainage tension in tile drainage de- sign. The most promising formula to date that utilizes the above suggested data and accounts for a soil profile composed of several layers of varying permeability and porosity was derived by Walker (44). The procedure used in developing his formula was to measure the drawdown characteristics of the soils, to measure the permeability of the different types of soil, and then to determine the relationship between the two. Also, he assumed that the permeability of the least pervious layer in that part of the soil profile through which the water passes determines the rate of flow to the drain. methods of Ground Water Measurement Donnan and.Bradshaw (17) stated that the use of obser- vation wells is necessary to obtain positive data on the position and fluctuation of the water table. The water level in the well reflects the strongest hydrostatic pressure pro- duced in the entire soil profile traversed by the well. The most common type are cpen holes, and if used for any length of time, they should be cased to prevent caving. Casing materials range from one-quarter inch to six inch diameter sheet metal pipe, stove pipe, and drain tile to commercial - 12 - well casing. Norling (31) and Neal (30) reported using auger holes lined with four and five inch diameter tile on end. Schlick (35) used three-quarter inch diameter galvanized pipe with three-sixteenths inch diameter perforations. Kidder and Lytle (22) used three-quarter to two inch diameter perforated wells; Walker (45) used three-quarter inch, one inch, and ten inch diameter perforated wells; Schwab (36) used three-eighths inch diameter perforated wells; and Ayers (2) used four inch diameter auger holes. Schwab (36) stated that their reason for using the small diameter pipe was that the water level will come to equilibrium.more quickly than in a larger hole. King (23) noted fluctuations in shallow wells twelve to fourteen feet deep due to temperature and atmospheric pressure variations. Schlick (34) pointed out that such variations were negligible in shallow wells used for drainage investiga- tion. Childs (ll) concluded from his studies by electrical analogy that the water table is lower at the observation point and that the indicated level of the water table can have little relation to the unperturbed level before the hole was made. The use and method of installation of piezometers for water table fluctuation and ground water flow pattern deter- minations has been described by Donnan and Christiansen (15). The ground water piezometer (17) is an unperforated small diameter pipe so designed and installed that after it is driven into the soil the ground water enters it only at the - 13 _ bottom. It measures the hydrostatic presure of the ground water only at the bottom of the pipe. Ayers (2) showed one of the difficulties encountered in the use of piezometers in studying ground water flow. A photo illustrated the condition where a piezometer was installed to such a depth that the bottom open end terminated in an artesian aquifer, thus caus- ing the piezometer to act as a flowing well. In 1940, Bouyoucos and Mick (7) described the use of a plaster of Paris resistance block for moisture determination under field conditions. The sensitivity to moisture changes between field capacity and saturation was, however, quite low. Later in 1948, Bouyoucos (8) reported on the develOpment and refinement of a nylon absorption electrical resistance unit for making measurement of soil moisture under field con- ditions. The unit was designed to minimize the effects of external soil factors such as degree of compaction, texture, structure, salt content, and electrical lines of force. At the saturation range, the unit was found to be very sensitive and responsive to changes of moisture cOntent. Bethlahmy (5), reporting on preliminary studies of the fiberglass resistance unit showed that valid calibration curves can only be obtained if the original structure of the soil is maintained. Effects of Drainage Baver (3) stated: The primary influence of adequate drainage on soil structure lies in the removal of excess water, with a -14- corresponding increase in the air capacity. This increase in aeration is accompanied by greater root development, more intense bacterial activity and the promotion of oxi- dation processes. The combined effects of these factors lead to better granulation. Schlick (34) pointed out that additional direct effects of drainage were to raise the average temperature, reduce heav- ing, and reduce erosion. Haswell (19) stated that sometimes several years are required before tile drains reach their maximum efficiency. Roe (32) cited three cases of circumstantial evidence of in~ creased permeability of soils as influenced by tile drainage. In two cases, drainage seemed to improve after four and seven years, and in the third case, gradually over a thirteen year period. Schlick (35) concluded that although tile installed an average of four feet deep will not remove surplus moisture as rapidly during the first few years as shallower tile, the soil will become "open" as time goes on. In the lifetime of the drainage system it will be more economical to hold to the four foot depth. Previous Comparison of Methods of Water Table Measurement To the author's knowledge, only one comparative study on methods of water table measurement has been made. Walker (45) reported that a statistical analysis of variance of data col- lected on one inch and ten inch wells and three-eighths inch piezometers in the lighter soils in Virginia revealed no sig- nificant difference. There were, however, indications that differences may occur in wells in heavier soils. EXPER IMENTAL STUDY The Study Area The site chosen for these investigations was in the north-east corner of Section 3, T8N, R1W, Clinton County, two miles north and one mile west of Elsie, Michigan. The soil type was mapped in 1942 as Brookston clay loam. A mechanical analysis verified this textural classification. However, the revised United States Department of Agriculture classification chart of 1949 would classify it as a loam. This soil type belongs in the Humic-Gley great soil group. The topography was nearly level and the original drainage was poor. Whiteside (46) described the horizons of the profile as follows: Ap - From 0 to 8 inches in depth, very dark gray‘ clay loam, indistinct coarse granular to cloddy structure, firm consistency, and slightly acid in reaction. G1 - From.8 to 18 inches in depth, light olive gray to light olive brown splotched with moderate yellowish brown to strong yellowish brown1 clay loam, medium sub- angular to angular blocky structure, firm.consistency, and neutral in reaction. G2 - From 18 to 42+-inches in depth, light olive ‘ ISCC - NBS color names. - 16 - gray to light olive brown splotched with moderate yellow- ish brown to strong yellowish brown1 clay loam, coarse angular blocky structure, firm consistency, and slightly alkaline in reaction. According to the owner of the property, the tiling sys- tem was laid approximately forty years ago. It was desirable to locate the study site on an older system so that the full effects of artificial drainage would be reflected in the soil. The four inch lateral drains of the system ran from.west to east and were spaced approximately six rods apart. They joined with a ten inch main drain that ran northward in the road ditch bottom along the east side and emptied into an open ditch at the north-east corner of the field (Figure 3). The drain at the study site was 27 inches deep with a grade of three-tenths foot per hundred feet. Three-eighths of an inch of sediment was observed in the bottom of the tile. The adjacent drains were one hundred feet to the north and one hundred and ten feet to the south. The crOp on the field at the time of the study was winter wheat. The crOps for the previous five years were as follows: 1952 - white beans 1951 - oats 1950 - sugar beets 1949 - winter wheat 1948 - red clover hay ‘ ISCC - NBS color names. -17- Road ‘7 1‘ u? N ____4.”.E£.r§./Z7_-____.____ ._.—:2- ' '1 )h I F ‘8 8: -_. +9 \ E" G l ifgza‘a____ii____- __:—::*_-_‘9J V x I ft ‘1 I O o I A J 3 E: , J SI __4_d_”_‘3’_’.7 ____________ ::=_—!:...J l I l F9. 3. P/aw w’ew of area around .sz‘zm’y 5/‘7‘3. -17- Road 3° I 4"dra/r7 h ._.... II N _________________________ _‘ ¢ I I ‘8 s: -I— I? ‘ 'f' u I ifgrflr1___.k______ -_z-za-ll ‘3 >. I It \I I 0 ,E 3| 0: tn ‘ ‘b o I A a : El t I . i 2! __4_CL”_‘?’L7 ____________ ::::-P:.__I I | | \ F13. 3. P/c'm 14st of area around .sflm’y 5/‘7‘e. - 13 - Prior to 1948, the cropping system followed a two year rota- tion of beans and grain, with all grains being seeded down to clover. At the time the observation devices were being installed, a sandy loam pocket was encountered near the south end of the study site. After all the field work was concluded, the area was augered and the approximate extent of the pocket is shown in Figure 4. The pocket was approximately three and a half feet deep. Installation of Observation Devices A plan view of the location of the piezometers, two inch wells, three-eighths inch wells, two inch auger holes, nylon resistance blocks, and rain gauge is shown in Figure 4. Figure 5 is a view of the site looking south. The observa- tional devices for each method were installed in a row parallel to the drain. The spacing between the observation devices was three feet (Figures 5 and 6). Each row was designated as an observation point. These observation points were labeled numerically starting with zero at the drain and ending with nine at the most remote point. Points north of the drain were designated by the letter N before the number, and points south of the drain were designated by the letter S. The intervals both north and south between observation points beginning at zero were 2, 3, 5, 5, 5, 10, 10, 10, and 20 feet. Thus N9 and S9 were each seventy feet to the north and south of the drain respectively. 19 IQNIQ\ILAI Qx I’TIQx I+W+W+W1+M0NN 324 amass .HI at... «\m ..+ 3 in? “\xfiiz N. sum! 0 o 3 wxmmecNmQ Wk: _ as he NA w I m 3 he *2 at me +6.. mm mm t. aw. Sin 5’: : life/7: e bloc/r: . PM? WWW? ......_ _.__.__: _.__ _ __ ._.. -5)..- _ I e 4 ' dram Y flpproxfrnaf'c clef/inc ofsand /oam pocket ~~~ I ....) / \Ii i/ I: :4. \ 1) F/y. 4. F/é’f? View of sfady s/fe. Fig. 5. Study site looking south. -Fig. 6. Field view of an observation point. - 21 - All elevations were determined from a reference bench mark near the study site.~ An engineer's level was used to set the wells and piezometers at the desired elevation. The elevations of the bottoms of each observation device were all the same and equal to the drain bottom elevation. This was done to guard against the possibility of artesian pressure from below the drain depth affecting the water level in the wells and piezometers. A cross sectional view of the four methods and resistance blocks at N5 is shown in Figure 7 as they would appear viewed from the south. Omitting the blocks, all the other cross sections at observation points were the same. Figure 6 is a cross sectional field view looking south, of the four methods.‘ From left to right they are the auger hole, the three-eighths inch well, the two inch well, and the piezometers. All the wells and piezometers, including metallic covers to exclude rain and foreign matter, were prepared for instal- lation during October, 1952. Piezometers. The piezometers were installed on December 13, 1952. They consisted of nineteen sets of 18, 24, 30, 36, 42, and 48 inch lengths of three~eighths inch water pipe. The 18, 24, 30, 36, 42, and 48 inch piezometers were designated as P1, P2, P3, P4, P5, and P6 respectively. At each observa- tion point, the bottom end of P6 was installed at the desired elevation. The other five piezometers were installed relative to P6 as indicated in Figure 7. -22- anxxb Minsky m6Q%\M\%h\ Bum MUQVXMS \SQ TVS“ \c kmi \mtobaumm. mmsxb N “Aux wkuo\fl Nutrcxuwwm I I mw\n\acuok\km~\k I X ‘9 W) ‘s \I 0—, . k? 03 AL. I is V~ I‘é‘ e E E'I'9"'I H u -u n u 5. I N X I "I . ~17» a I E ..E‘Itfitflh‘? *I I, j I? I". X SL| V) I I75 I. '43 T 1 3‘ Uh. teem. QSOx IN 32x kwmas. s a M a To? :J *8 (d (J /J 3‘00)» Uukbxfiawax Sm} : um \xeie N 33.“ match. v.3. -23- The driving technique for installing the piezometers was to hold the pipe vertically with a bolt loosely placed in the lower end to prevent soil from entering during driving. A bolt was placed in the upper end to protect the top, and the piezometer was driven to the desired depth with a hammer. The bolt was then punched from the lower end with a length of one-quarter inch rod to form a cavity 1 1/2 inches in length below the piezometer. The cavity was flushed with water by inserting to the bottom of the piezometer a one- quarter inch plastic hose connected to a hand pressure pump. Figure 8 shows the flushing Operation. Flushing was continued until the effluent ran clear and the rate at which the water level drOpped in the piezometer was reasonably rapid consider- ing the permeability of the soil. This flushing operation was delayed until March 26, 1953, as it was felt that the piezometers should be flushed just prior to their service. Two inch wells. The two inch wells were constructed from 48 inch lengths of two inch water pipe. Perforations, one- quarter inch in diameter, were made in the lower portion of the pipe by drilling four holes equally spaced around the pipe at one inch intervals along the pipe from the bottom. The position of the four upper holes was calculated to be two inches below the ground surface. The wells were installed on November 23, 1952. A hole, slightly larger than the outside diameter of the well, was bored with an auger to a depth one inch higher than the - 24 - desired elevation. An inch of undisturbed soil was left in the bottom to give the well a fairly solid footing. The well was placed in the hole, and tapped to the desired elevation. A small amount of bentonite was mixed with the surface inch of soil directly adjacent to the well, heaped up, and tamped by hand. This was done to prevent surface water from running down the outside of the well and entering the well directly through the perforations just below the ground surface. Three-eighths inch wells. The construction of these wells was nearly the same as that of the two inch wells. The only differences were that three-eighths inch water pipe was used instead of two inch pipe, and the perforations were three- sixteenths of an inch in diameter instead of one-quarter inch. These wells were installed on December 11, 1952. The installation technique was identical to that used on the two inch wells. Two inch auger holes. The technique for installing the auger holes was rather sumple. A hole was bored with a two inch auger to the desired depth. A two inch hole was drilled in the centre of an 8 x 8 x 1 inch wooden cover and plugged with a rubber stopper. The stopper could be removed to make a water table measurement. The loose tOpscil was removed from around the hole; the cover was placed squarely over the hole and pressed firmly into place. Soil was then packed around the sides of the cover. The cover not only prevented rain and foreign matter from Fig. 9. Taking a water level reading from a two inch well with the water level indicator. -269 entering the hole, but provided a datum plane from which the water level could be measured. Elevations of the covers were taken one week after installation at which time it was felt that they should be stable. The installation of the auger holes was left until March 23, 1953, so that there would be less Opportunity for the holes to deteriorate. Emlon resistance blocks. The eight resistance blocks were installed on December 11, 1952, at observation point N5. The location of each block is shown in Figure 7. A hole was bored with a two inch auger to the desired depth, taking care to lay out the soil from the hole on a canvas in the same se- quence that it was taken from the hole. The block was satur- ated with water, and wet soil taken from the same area in which the block was to be embedded was pressed firmly around the block. This was done to ensure a good contact between the block and the soil. The block was lowered into the hole, pressed gently but firmly in a horizontal position on the bottom, and the soil was packed on top of the block in the reverse sequence that it was removed. The soil was packed as tightly as or a little more tightly than the undisturbed soil. Four thermocouples were installed in the deepest hole (Figure 7). Soil temperature measurements were required in order to apply a temperature correction to the block resist- ance readings. - 27 - Water Level Observations To prevent compaction of the soil at each observation point, two planks to walk upon were placed on tOp of cement blocks. Figure 8 shows the planks in place. The planks were moved from one observation point to the next when measurements were being made, and placed to the side of the study site when not in use. ‘ The elevations of the water table were measured with an electric water level indicator from the tops of the wells and piezometers and the wooden covers on the auger holes. A probe containing two electrodes at the bottom end was attached to a slider that was mounted in grooves on the sides of a section from a surveyor's rod (Figure 9). The two electrodes were connected in series to a 45-volt battery and a voltmeter. When the indicator was placed on top of a well, piezometer, or auger hole, the probe was lowered until the electrodes touched the water. The water completed the circuit and the contact was registered on the voltmeter. The depth from the top to the water level was read directly from the rod section at the point indicated by a hair line on the slider. This reading was subtracted from the known elevation of the well, piezometer, or auger hole top to obtain the water table elevation. Readings were taken to the nearest hun- dredth of a foot. The resistance readings of the nylon blocks were taken with a Bouyoucos bridge. Immediately after taking the .. 28 _ resistance readings, potential readings were obtained from the thermocouples with a potentiometer and converted to t emperature read ings . Soil Tests After obtaining the last water table drawdown measure- ments, undisturbed soil core samples were taken from near the two inch wells at each observation point except S1 and N1. At 81 the samples were taken from near the three-eighths inch well, and at; N1 no samples were taken because of its proximity to adjacent observation points. The core sampler and sampling technique used were very similar to those described by Uhland and O'Neal (42). The soil cores were contained in metal cyl- inders three inches in diameter and three inches in length for easy handling during the tests. Samples were taken from each horizon at 2 to 5 inch, 13 to 16 inch, and 23 to 26 inch ‘10P the. Soil core samples contained in 1 3/8 inch diameter by 3 inch long metal cylinders were also taken from the centre of the repacked soil in the holes containing the resistance b100ks. Two samples each were taken at 2 to 5 inch, and 13 to 16 inch depths. Non-capillary pore space. The first determination made On the core samples was the percent by volume of the non- capillary pore space. A filter paper next to the core and a Piece of cheese cloth were secured to the bottom of the cyl- inder and core with a rubber band. The core was saturated - 29 - with distilled water, weighed, placed on a tension table set at a tension of sixty centimetres of water, and allowed to drain for at least 48 hours. At the end of this time, the core was reweighed. The loss of weight in grams divided by the total volume of the core in cubic centimetres, and mul- tiplied by one hundred gave the percent of non-capillary pore space. Permeability. After the cores were reweighed, they were again saturated with distilled water. A ring, one inch in height and three inches in diameter, was taped onto the tOp of the cylinder so that the joint was waterproof. A filter paper was placed on top of the core. The core was placed on a screen to give free drainage below, and one hundred cubic centimetres of distilled water were added to the top. Water in.fifty cubic centimetre increments was added when required in order to maintain approximately a half inch of water on top of the core. The amount of water passing thrOugh the core during a two hour interval was measured. The number of cubic centimetres of water passing through the core per hour divided by a factor of 115.7 gave the permeability rate in inches per hour. Volume weight. At the completion of the permeability determinations, the cores were oven dried at 105 degrees cen- tigrade for four days and then weighed. The oven dry weight of the core in grams divided by the total volume of the core in cubic centimetres gave the volume weight in grams per cubic centimetre. IIIIIIIIJIIIIII I III . ILI. - 30 - Results and Discussion Two sets of water table drawdown observations were ob— tained between the periods of April 27 to April 30, 1953, and May 3 to May 6, 1953 (Table V, Appendix). The rainfall just prior to and during these periods was as follows: April 25 to April 27 (intermittently) 1.52 inches April 29 0.11 inches April 30 0.12 inches April 30 to May 2 (intermittently) 0.55 inches Graphs showing the drawdown curves from.uay 3 to May 6 as measured by the piezometers, two inch wells, three-eighths inch wells, and two inch auger holes are illustrated in Figures 10, ll, 12, and 13 respectively. The highest water level elevation observed from the group of six piezometers was considered to be a measure of the water table reflected by the piezometer method at the observation point. The failure of the auger holes to stand up in coarser soils is evident in Figure 13 at observation points S6, S7, and $8. The sandy loam interfered with accurate water table measurements at S6 in the three—eighths inch well and, al- though no sandy loam was observed near the two inch well at $7, there appeared to be some type of interference present (Figures 11 and 12). There were also varying amounts of fine sediment starting to fill most of the other auger holes and the occasional well, causing interference in water level -31- 6ka WSONOxQ mix (3.x c.9290 3:38:ka .Q\ .N\ . v.39: £63 93* met \ok sexing S: OMEACQ \\ SIR 4 .Il « u a H . .. pun! . . . . I .1.. OIvv . r. .vv.i .ov. .io. ...! the: IoIfuc-nu «- u e Q-.r.r.v . n n ----‘I-T ._- h I . ‘ru.-‘>—..I.-1.0*—*—1"— . 1. . . . .. .I.J . . .. . . ..I. .. .p. I... ... .... .i-»i...... .... I.. .I . .v-.I. . ... .y-,.. ,. -....,. . e . ... . . .I . . . .1.. .- ..... . . . e c ——. ”._.—.1 ‘a—u—fi q—n—o—Q-q I . I I . . . . . I I I , . 1 I u . . . . . . q . a . n . n O b . . a . 0 A; ‘1 . v q I o . . . n . d . , l . , . .. . . . I .. . . 4 .1 .. .. . 1. o . 5... 5 -~ «4 W ._.... a ‘ E I o I 4 . . W I. fl, . w 1, . . I ... ....+ H .w . . ... .. i. I . . . a. .«I. II. ......I . . ... .A h ...-A ..I a . .I I .11.- .. ... - I,” IN“ . -32- IIYIIL... . I -I I .m , . r .3 Lynx. . g 1.11.» J 13.... H I I. 1.9“” HVig; H 3.2.. 33.11: PM .... QONKAE- firm ”13.3..“ .H3.. 4 ....b.... 2:33.; 3, 3, .g. ... V2.5.“ {Mann}... 13111.“ I]. 3 . “3...”-.. 31;...“ ..3 J. . ..... .. .3 .J.g...... .... r H.......HH..-I3..I...$ : 33.3. .1 :.1 . . u H. ...HEHULTIHMWM.........._Hfi_.....I.I.3.3-11““... 1.3”..me 33.3mm”. .. . hunQKOQ. . g h ......H ....... .3“ 3x 313 .. 3. 3.x 3.3 “U. 3 S... . 1 MIN? : . . e .. ... . : .... 3. ..xxr...lmI....g-3r. :33...3.3.... : 3.... 32......HHGVMAHX? .I . I! 1.. .-. e... . i . .. I. ... .. .. . I .I. . I. .. . .I.. -Ii .-..M. . I o . . H—e . . r l a . I . . I . b '- Aa+~b~ ‘ O .I , ' . .q I . .. ‘4 r1 . I .. . . . . . . v. Iu.w .u . e -.(p—«m . O—-4-—>—)—+ r4ve-e o-wHWo-ne v4 0: t I . I m . . _ ., 3.31. .....h H . . ,. . w . A. . . . . . . I. H.“ . .. I. .. . .. ....fi . I”. .. H 5. .4. . ... , I .. . . I - H ... . I-H .1 H A . .. . . H ... . ..HL H.H.;.H....:.3WPL.J H . H .0 ..I h . —. . bIF . ..H.... ..dlu. .I o o. i m. I a . _ . EH .4 .1. . a ... 3......1i... I» I ..w\\m\: $63 MQLNEHSIMMKFN mfix keg“ «Ming QiQBkMIxQ N\.%\I\ mmvsfi \mS. 5% 93 tax mmxISu 8:33:ka .m.\ .&.K - 35 .. measurements near the end of the drawdown periods. The inter- ference due to the fine sediment in the cased wells probably could have been reduced considerably by cleaning out the wells in the spring Just prior to taking measurements. A statistical analysis of variance to evaluate differences between the water table data obtained by the four methods is shown in Table I. The data compared in the analysis were taken from observation points N9 to 0 for the period May 3 to May 5. Data from $1 to S9 and the water table observations for May 6 were excluded from the analysis to eliminate the source of error in measurement due to mud in the bottom of the observation devices. The results of the analysis showed a highly significant difference between methods. The probable cause of this high significance was the piezometer method whose mean was close to two inches higher than the average of the other means. The necessary differences between the means for significance at the one percent and the five percent probability levels are 0.034 feet and 0.025 feet respectively. A re-oheek on the datum elevations used to determine the water level elevations disclosed an occasional maximum variation of 0.01 feet. For practical purposes, this amount of error could be expected. Because of this known possible variation and lack of repli- cates, the necessary difference was based at the one percent level. A comparison showed that the mean of the piezometer method was significantly higher than the mean of any of the - 36 - TABLE I ANALYSIS OF VARIANCE OF WATER TABLE IMEASUREMENT DATA '5 Degrees of Sum of Mean Source of Variation Freedom. Sguares Square Locations 9 12.29 1.361“ Days 2 3.29 1.65n Methods 3 0.63 0.211“ Locations 1 methods 27 0.36 0.0131“ Locations 1 days 18 0.51 0.028n Methods x days 6 0.07 0.0121“ Locations x methods x days1 54 0.36 0.0024 “Necessary Mean of Methods Difference ‘2 inch’ 13'8‘Inch Au er Piezometers wells 43118 hoIes p : 0.05 p = 0.01 4.075 3.924 3.956 3.881 0.025 0.034 ‘ Best error term. ‘* Significant at the 1% level. -37- other three methods, while the mean of the auger hole method was significantly lower than the mean of the piezometer, two inch well, or three-eighths inch well methods. There was no significant difference between the means of the two inch well method and the three-eighths inch well method. These comparisons seem to indicate that size relation- ship, which in this study was 16 to l, was not necessarily the cause of significant differences. However, at the five per- cent probability level, there is a significant difference between the means of the three-eighths inch well and two inch well methods. Although too much confidence cannot be placed on this significance, it does indicate that observations from the small sized wells lagged those from the larger wells. The significant difference observed between the means of the auger hole method and the two inch or three-eighths inch well method indicates that observations from cased wells lag those from.uncased wells. However, it must be realized that in order to obtain more conclusive evidence regarding these differences, similar studies should be replicated at other locations in similar, and still finer textured soils. A graph of the drawdown curves on May 3 and may 5 for each method is shown in Figure l4 to illustrate the order of magnitude of the discrepancies observed between methods. From a practical point of view, the lag in the piezometer method is quite significant, becoming most apparent toward the end of the drawdown period. It is, however, doubtful whether the 630%me {econ mat s5 WEE ohkbkk Kc Qemkauxkxob . ,, l..-l,--.l,--.l ewes mess .I .11. ..1I. Edi $3.. weakeamIMWofih IIIIIIIII mu dc: 403. SE Mensa m3 Nmfi. .1 33:] - A/O/i’ V1373 - 39 - statistical significant difference between the means of the auger hole method and the two inch or three-eighths inch well method is of practical importance when it is considered that there is no verification that any of the methods reflect at any time the actual elevation of the water table in the soil. Further, it is not known preciSely to 0.075 feet, the differ- ence between the means of the auger hole and three-eighths inch well methods, the recession rate of the water table re- quired for adequate drainage. Accurate measurement of the upper limit of the zone of saturation by means of the nylon resistance blocks could not be obtained for two reasons. First, the blocks were installed at a minimum of three inch vertical intervals. Therefore, continuous measurements through the soil profile could not be made. Second, the laboratory calibration of resistance at saturation was not too closely related in some cases, to what appeared to be the saturated resistance of the blocks in the field. This is shown in Table II where, for example, the cal- ibrated resistance reading at elevation 3.87 feet was higher than any of the field resistance readings. However, by assum- ing that a state of saturation existed if any field resistance readings at any elevation remained fairly constant from day to day, a general picture of the movement of the upper limit of the zone of saturation was obtained. Those readings below the dashed lines in Table II were considered to indicate a state of saturation. - 40 - TABLE II DATA FROM THE NYLON RESISTANCE BLOCKS AT OBSERVATION POINT N5 Elevation of Block Resistance (ohms)In Eggisggzzzdat the Block (feet) may 3’ may 4 May‘5 may 6’ Saturation 5.12 710 ----t ---i ---i 190 4.87 558 752 890 ---i 242 4.62 __434__ _522___ 570 684 215 4.37 234 235 .385... 535 280 4.12 382 388 360 .438“. 352 3.87 205 202 180 182 348 3.62 --- --- --- --- ...:x 3.12 244 258 245 247 242 * Resistance readings over 1,000 ohms not recorded. “ In0perative block. ‘t‘ Corrected to 60° F. TABLE III COMPARISON OF WATER TABLE ELEVATIONS OBSERVED BY THE FOUR METHODS TO THE ELEVATION OF THE SHALLOWEST SATURATED RESISTANCE BLOCK AT OBSERVATION POINT N5 4* ‘fii :Elevafions (feet) Observation Method MEXI3 ‘May 4 NBI:5' IM§1_5 Piezometer 4.34 4.27 3.86 3.69 Two inch well x 4.26 3.94 3.77 3.63 Three-eighths inch well 4.25 3.94 3.77 3.65 Auger hole 4.20 3.89 3.77 3.8111 Resistance block 4.38 4.38 4.12 3.87 maximum differences 0118* 0.49 0T35 0.24 * Mud level. - 41 _ Table III shows that for any day between May 3 and May 6, the blocks indicated that the soil was saturated to a higher elevation than the water table as indicated by any of the other four methods. On May 4, the maximum difference was 0.49 feet. Differences for the other three days ranged from 0.l8 feet to 0.35 feet. The soil surrounding the blocks in the G1 horizon where these differences were measured was, of course, disturbed and had a permeability rate approximately one-eighth the permeability rate of the undisturbed soil adja- cent to it (Table IV, Appendix). This reduced permeability could have been a factor causing part of the discrepancy be- tween the two elevations. It is suggested that the remainder of the discrepancy was caused by the presence of a zone of capillary saturation immediately above the water table, and the failure of open hole types of observation methods to re- flect accurately the true position of the water table in the soil. The significant differences found between methods seem to indicate that inaccuracies, though small, do exist. It appears that under these conditions the piezometer method is the most reliable of the four methods. Although there is no certainty that this method measures accurately the position of the water table, it does however follow the upper limit of the zone of saturation as measured by the blocks, more closely than the other three methods. In the final analysis, a measure of the recession rate of this upper lhmit for ade- quate drainage should be the criterion sought. INVESTIGATION OF AN EMPIRICAL DEPTH AND SPACING FORMULA It was desirable to carry this study further by utilizing the data obtained from the experimental work to evaluate a new depth and spacing formula prOposed by Walker (44). The and where, £9 ‘Ay The Formula formula consists of two equations which are as follows: cos 9 = Ayp / 2Kt s = 2 conJ y tan 6} maximum angle between the vertical and the radius to the drain that passes through the midpoint or crest of the drawdown curve. minimum water table recession increment per day which permits satisfactory plant growth. Assumed to be 0.7 feet. non-capillary porosity between the 1.0 foot and 1.5 foot depth expressed as volume of water per unit volume of soil. minimum permeability in the soil profile to depth, D, in feet per day. time interval of one day. - 43 _ s := total lateral spacing between drains in feet. con.) y = "i = (y. + 5'2) / 2 = difference in elevation between the midpoint in .ay and the drain in feet. Figure 15 is a schematic diagram showing the location of the various dimensions utilized in the solution of the above expressions. /——— Ground Sur {a ce 4_______5/2 a 5/2 _ ,. ‘ S Fig. 15. Schematic diagram.showing the location of formula symbols. The.minimum permeability, K, selected was the lowest average of the 18 permeability rates obtained from core samples of each horizon of the soil profile. The percent non-capillary porosity, p, was the average of the 18 porosity determinations on the G1 horizon (Table IV, Appendix). These 51 horizon determinations were representative of the porosity between the 1.0 foot and the 1.5 foot depths. The regular usage of the formula is to assume two differ- ent drain depths and solve for the drain spacing at each depth. -44- A comparison of the two solutions will indicate the most econ- omical design. However, the evaluation procedure for this study was to use the drain depth of the experimental drain and the adjacent drain to the south, solve for the drain spacing, and compare this computed spacing with the actual field spac- ing. Results and Discussion The computation for determing the spacing of the two lat- eral drains is as follows: Depth of drain from.ground surface 2.0 ft Recession increment per day, ay 0.? ft Porosity x unit volume, p 0.087 Time interval, t 1.0 day Minimum permeability, K 1.50 fpd Cos 9 = Ayp / ZKt 0.0203 Tan e} 49.38 Average depth to drain, conj y 0.65 ft Spacing, 3:: 2 con: y tan £9 64.2 ft The computed spacing is 64.2 feet whereas the field spac- ing is 110 feet. The owner of the preperty has observed the drainage of the field since the time that the drains were installed. His Opinion was that the drains were spaced too far apart because during a prolonged wet spring, the drainage has not been adequate. The present drainage has been adequate, however, during a normal year. -45.. Since the minimum recession rate in the formula was ob- tained from the investigational work of Neal (30), his obser- vations were evaluated. He observed that crops were not seriously injured if, in essence, the water table receded to a depth of 18 inches below the surface in 29 hours. This observation was evaluated against the water table position as indicated by the two inch and three-eighths inch wells south of the experimental drain after the cessation of rainfall at 8:00 P.M., May 2, 1953. In both cases the conditions of the observation were met only to a distance of ten feet to the south of the drain. This indicated that a drain spacing of twenty feet would be required under these conditions. Thus, it seems quite evident that more specific knowledge of the .minimum recession rate of the water table for adequate drain- age is needed at this time in order to know first what criter- ion should be used for depth and spacing design. It is, however, peculiar that the formula gave a result that is in agreement with the unbiased estimate that the author would make based on field experience in drainage design. Walker (44) has made this same observation from other tests of the formula in soils of the Southern Atlantic Coastal Plain. An observation regarding the mechanism of the formula .might be pointed out. As stated previously, to compute the mmst suitable depth and spacing, two depths are assumed and, by means of the formula, the spacing for each depth is calcu- lated. The most economical design is then selected. However, -46.. where a shallow outlet or flat topography necessitates a shal- low installation, and there is no change in the minimum perm- eability of the soil to the two assumed depths, a slight change in depth, D, can have an appreciable effect on the spacing, s. For instance, if the two study drains had been installed 0.35 feet deeper, conj y would equal 1.0 feet, and the calculated spacing would be increased from 64.2 feet to 98.7 feet. Under the above mentioned conditions, the formula indicates that depth is a critical factor. It would seem that there should be further study of the effects of a small increment in drain depth on the water table drawdown charac- teristics under conditions where the minimum permeability does not vary in that increment. l. CONCEUSIONS The results of the analysis of variance made on a portion of the data revealed significant differences between the means of all four methods except between the means of the two inch well and the three-eighths inch well methods. The water level observations from the piezometer method show a greater lag than observations from the other three methods. Size relationship was not necessarily the cause of these differences. It appears that observations from cased wells lag those from uncased holes, and that observations from the small sized wells lag those from the larger wells. However, more replicates of a similar study are necessary for more conclusive results. It is doubtful whether the necessary difference for statistical significance is of any practical significance. Only a general picture of the movement of the upper limit of the zone of saturation was obtained by_means of nylon resistance blocks. A maximum difference of 0.49 feet between this upper limit and the lowest indicated water table at an observation point was observed. This discrep- ancy is believed to be caused by a combination of disturbed soil surrounding the blocks, presence of a zone of capil- lary saturation immediately above the water table, and -48.. failure of open hole types of observation methods to re- flect accurately the position of the water table. 3. Under the conditions of this study, it appears that the piezometer method was more reliable than the other three methods as it followed more closely the upper limit of the zone of saturation. 4. The results of the evaluation of an empirical depth and spacing formula on drains at the experimental site showed that the drain spacing was too great. This is in agreement with the opinion of the preperty owner who has observed the field drainage since the time that the drains were installed. 5. More specific information on the minimum.recession rate for adequate drainage is needed. in evaluation of a recession rate that has been observed by Neal (30) to give adequate drainage indicated that this rate is not the minimum that can be permitted. 6. Under these field conditions, the formula indicated that drain depth was a critical factor in determining the drain spacing. Further field studies should be made to determine whether small increments of depth are as critical as the. formula would indicate. APPEN D IX _ 5Q - TABLE IV RESULTS OF SOIL TESTS ...—A 0bservation Permeability %:N0n-capillary Point (in/hr) Porosity Volume Weight (adjacent £_ 0' G to 2 inch G A G G A G well) p l 2 p 1 2 p l 2 N9 1.34 0.53 0.64 15.11 6.86 7.21 1.33 1.60 1.62 N8 3.03‘0.32 0.76 18.62‘ 7.50 10.55 1.32 1.67 1.59 N7 0.30 0.34 2.17 10.84 6.46 7.01 1.39 1.70 1.73 N6 1.37 1.16 3.57 19.17 9.37 7.61 1.25 1.59 1.66 N5 0.86 0.85 1.55 14.01 9.37 7.84 1.32 1.57 1.65 N4 2.28 0.99 2.54 17.30 8.22 8.77 1.25 1.60 1.62 N3 1.09 0.84 0.64 13.06 9.72 7.01 1.34 1.52 1.67 N2 1.87 0.44 1.07 20.47 7.58 7.04 1.19 1.58 1.63 0 1.36 0.76 0.10 15.19 6.72 4.56 1.34 1.58 1.60 31* 0.92 0.97 0.77 13.02 9.92 7.67 1.39 1.55 1.63 S2 1.58 0.64 0.27 16.86 8.45 7.18 1.36 1.59 1.65 S3 1.52 0.64 0.84 17.81 10.73 8.19 1.33 1.54 1.61 S4 1.22 1.35 1.31 15.11 10.61 9.54 1.36 1.64 1.64 S5 0.74 0.29 1.90 12.86 8.33 6.66 1.50 1.73 1.76 S6 1.62 1.68 0.46 13.11 7.79 7.53 1.49 1.65 1.82 S7 1.03 0.29 0.45 15.73 10.73 11.59 1.48 1.78 1.71 88 0.32 0.78 0.79 10.22 9.08 8.82 1.55 1.74 1.73 S9 0.99 0.59 1.09 13.75 10.32 9.03 1.35 1.52 1.56 Average 1.30 0.75 1.16 15.13 8.75 7.99 1.36 1.62 1.66 ns“ 0.41 0.11 ---- 22.40 8.86 ---~ 1.12 1.44 ---- * Adjacent to 3/8 inch well. 1‘ Centre of repacked soil above the resistance blocks. _ 51 - TABLE V WATER LEVEL OBSERVATIONS Observ. Observ Ground Elevations (ft.) Point Devicei Elev. April May (ft.) 27 28 29 30 3 4 5 6 N9 Pl 5.28 4.61 -—-- --—- ---- 4.68 4.68 ---- ---- " P2 " 4.33 4.33 4.31 ---- 4.51 4.48 4.32 --~- 3' P3 fl 4053 4032*4022 4012 4.64 4036 4012 ""-" " P4 " 4.27 4.28 4.25 4.23 4.63 4.37 4.12 3.92 " P5 " 4.03 4.05 4.05 4.05 4.16 4.17 4.17 4.16 " P6 " 4.56 4.23 4.12 4.05 4.62 4.32 4.10 3.92 " LW " 4.51 4.17 4.10 4.03 4.55 4.23 4.00 3.88 " SW " 4.53 4.18 4.10 4.00 4.57 4.26 4.02 3.87 " AH " 4.55 4.19 4.06 4.03 4.54 4.21 3.99 3.84 N8 AH 5.35 4.52 4.15 4.07 3.94 4.59 4.23 3.96 3.78 " SW " 4.60 4.26 4.11 4.03 4.65 4.33 4.07 3.87 " LW 7 4.62 4.21 4.11 3.99 4.67 4.30 4.02 3.82 " P6 " 3.58 3.71 3.76 3.81 4.14 4.19 4.19 4.15 " P5 " 3.67 3.78 3.83 3.87 4.17 4.23 4.21 4.15 " P4 " 4.67 4.27 4.14 4.04 4.70 4.34 4.06 3.87 " P3 " 4.17 4.21 4.19 4.17 4.48 4.48 4.37 4.04 7' P2 ll "-"" "“" "" “"‘" 4037 4038 4036 4031 " Pl " 4.68 4.56 ---- ---- 4.75 4.56 ---- ---- N7 Pl 5.34 4.60 ---- ---- ---- 4.71 ---- ---- -~-- " P2 " - 4.55 ---- ---- ---- 4.68 4.31 ---- ---- " P3 " 4.10 4.14 4.11 4.08 4.37 4.37 4.30 4.21 " P4 " 4.50 4.23 4.17 4.10 4.71 4.30 4.01 3.84 " P5 " 4.49 4.18 4.07 4.00 4.65 4.32 4.13 3.97 " P6 " 4.54 4.15 4.06 3.92 4.61 4.24 3.97 3.78 “ LW " 4.59 4.16 3.98 3.92 4.63 4.23 3.97 3.79 " SW " 4.53 4.11 3.97 3.88 4.58 4.19 3.93 3.75 " AH " 4.47 4.07 3.96 3.87 4.51 4.14 3.90 3.73 N6 AH 5.39 4.31 3.93 3.82 3.77 4.34 4.00 3.79 3.66 " SW " 4.33 3.95 3.80 3.78 4.35 4.00 3.80 3.67 " LW " 4.38 3.95 3.81 3.79 4.36 4.01 3.81 3.66 " P6 " 3.66 3.69 3.70 3.70 3.84 3.87 3.87 3.86 " P5 " 4.18 3.99 3.85 3.80 4.43 4.1813.94 3.74 " P4 " 4.36 4.00 3.86 3.83 4.39 4.06 3.85 3.73 " P3 " 4.09 4.09 ~--- ~--- 4.47 4.34 4.16 4.05 " P2 " ---- ---- ---- ---- 4.30 4.27 --—- --—— 1! Pl 7! ...--- ...-._- ._-“- -_.-..- _-__ ____ -____ _-__ ‘ LW-—2" well; SW—-3/8" well; AH—-Auger hole; P—-Piezometer. - 52 - TABLE V (continued) Observ. Observ. G2123? ApriIElevations (TE°XM§y P°1nt DeVice‘ (ft.) 27 28 29 30 3 4 5 6 N5 P1 5.38 ~--- ---- ---- ---- ---- ---— —-~- ---- " P2 " 4.37 ---- ---- ---- 4.33 4.27 ---— ---- " P3 " ---- ---- ~--- —--~ 4.07 4.07 —--- ---- [1 P4 [1 3.97 4000 3088 ‘“'“ 4034 4002 3086 ""'""" " P5 " 3.67 3.79 3.79 3.78 4.36 4.03 3.83 3.69 " P6 " 4.22 3.87 3.77 3.73 4.25 3.94 3.76 3.63 " LW " 4.27 3.88 3.75 3.73 4.26 3.94 3.77 3.63 " SW " 4.29 3.89 3.79 3.75 4.25 3.94 3.77 3.65 " AH " 4.17 3.82 3.76 3.75 4.20 3.89 3.77 3.81 N4 AH 5.38 4.13 3.78 3.68 3.66 4.14 3.83 3.67 3.62 " SW " 4.19 4.01 3.79 3.72 4.26 3.99 3.82 3.69 " LW " 4.24 3.83 3.74 3.74 4.21 3.90 3.76 3.67 " P6 " 3.89 3.85 3.84 3.83 3.93 3.94 3.93 3.91 " P5 " 3.57 3.59 3.61 3.52 3.91 3.94 3.92 3.89 " P4 " 4.25 4.09 3.93 3.86 4.26 3.93 3.84 ---- " P3 " 4.08 ---- --—— --~- 4.10 ---- ---- -—-- " P2 " 4.35 ---- ---- ---- 4.33 4.30 ---- ---— ll Pl H -___ ____ -_-- ____ ____ ____ _____ ____ N3 Pl 5.35 ---- ---~ ---- ---- ---- ---- —--- ---- " P2 " 4.35 ---~ ---- --—- ---- ---— ---- ---- " P3 " 4.12 ---- ---- ---- 4.13 ---- ---- --—— " P4 " 4.01 3.97 3.85 ---- 4.15 3.88 3.79 ---- " P5 " 3.81 3.80 3.79 3.77 3.86 3.86 3.85 3.82 ' P6 " 3.82 3.80 3.80 3.78 3.87 3.87 3.85 3.83 " LW " 3.99 3.79 3.66 3.62 4.08 3.87 3.66 3.55 " SW " 4.03 3.77 3.77 3.66 4.08 3.81 3.66 3.64 " AH " 4.02 3.70 3.58 3.58 4.00 3.73 3.60 3.57 N2 AH 5.34 3.73 3.50 3.45 3.42 3.73 3.51 3.45 3.38 " SW " 3.81 3.58 3.48 3.47 3.81 3.59 3.50 3.46 " LW " 3.67 3.46 3.44 3.42 3.66 3.49 3.45 3.41 " P6 " 3.64 3.63 3.61 3.60 3.65 3.65 3.63 3.61 " P5 " 3.63 3.65 --~- ---- 3.74 3.59 ---- ---- " P4 " 3.76 —--— ---- ---- 3.80 ---- ---- ---- 7' P3 '1 _____ ___~_ ______ ____ ____ ___- ______ --__ fl P2 ll ___-_ __-_ ____ _-__ ._-..- -___ ____ ____ " Pl 7' _____ ____ -_-- ____ ____ ._--- ___- _____ ‘ LW—-2" well; SW-—3/8" well; AH-—Auger Hole; P-Piezometer. - 53 - TABLE V (continued) Observ. Point Observ. Devicei Ground Elev. (ft.) Elevations (ft.) C" 27 Aprif— 28 29 3O 3 hwy 4 N1 77 H 30 33:33:: U) it“ 3:33: 333333!“ a: P1 P2 P3 P4 P5 P6 LW SW AH AH SW LW P6 P5 P4 P3 P2 P1 P1 P2 P3 P4 P5 P6 LW 5.38 " fl 1! II II H fl " ---- ---- ---- 3.62 3.48 3.61 3.68 3.58 3.50 3.69 3.59 3.44 3.62 3.51 3.54 3.87 3.61 3.77 3.76 3.76 3.37 3.69 3.35 3.39 3.48 3.39 3.36 3.63 3.43 3.43 3.50 3.34 3.65 3.39 3.57 3.50 3.51 3.39 3.33 3.39 3.39 3.36 3.36 3.56 3.40 3.41 ---- ---- ---- ---- 3.49 3.32 3.53 3.37 3.49 3.45 3.42 3.39 3.32 3.38 3.37 3.36 3.36 3.51 3.42 3.41 3.47 3.31 3.46 3.37 3.48 3.43 3.40 3.39 3.57 3.48 3.55 3.67 3.53 3.57 3.67 3.61 3.48 3.59 3.52 3.53 3.75 3.60 3.77 3.73 3.70 3.53 3.67 3.37 3.40 3.53 3.38 3.45 3.61 3.48 3.47 3.51 3.39 3.63 3.46 3.63 3.53 3.49 3.53 3.34 3.38 3.43 3.37 3.40 3.55 3.46 3.46 3.50 3.38 3.53 3.38 3.51 3.47 3.42 3.51 3.31 3.38 3.38 3.37 3.36 3.49 3.44 3.44 ...-0 cu..- ---- ---~ 3.48 3.36 3.48 3.35 3.44 3.44. 3.39 3.47 ‘1 LW--2" well; sw—s/e" well; Ali—Auger hole; P—Piezometer . _ 54 - TABLE V (continued) Observ. Observ. Ground ElevationSITfilff Point Devicet Elev. ‘Ipril ‘May (ft.) 27 28 29 3O 3 4 5 6 35 P1 5.42 ---- ---- ---- ---- ---- ---- ---- ---- n P2 n ____ -_-_ __-_ ____ -_-_ -_-- ____ ____ 0 P3 " ---- ---- ---- ---- ---- ---- ---- ---- " P4 " 4.04 ---- ---- ---- 4.00 -~-- --—- ---- " P5 " 3.89 3.64 3.63 3.54 3.9513.68 3.57 ---- " P6 " 3.46 3.51 3.52 3.53 3.77 3.78 3.74 3.69 " LW " 4.08 3.74 3.62 3.58 4.05 3.75 3.62 3.50 " SW " 3.83 3.76 3.68 3.63 4.00 3.79 3.67 3.61 " AH " 3.88 3.61 3.58 3.55 3.91 3.67 3.56 3.55 S4 AH 5.47 4.10 3.76 3.69 3.61 4.1213.81 3.65 3.57 " SW " 4.12 3.78 3.69 3.61 4.13 3.83 3.67 3.52 " LW " 4.13 3.79 3.69 3.64 4.16 3.84 3.68 3.54 " P6 7 4.03 3.77 3.74 3.63 4.10 3.81 3.65 3.52 7 P5 2 3.75 3.77 3.76 3.73 3.96 3.95 3.90 3.83 " P4 " 4.05 ---- ---- ---- 4.12 3.84 ---- ---- " _P3 " ---- ---- ---- ---- 4.01)---- ---- ---- " P2 " 4.28 ---- ---- ---- 4.25 --—- ---- ---- n P1 u -_-_ ____ -___ ____ ____ ____ ____ ____ S5 P1 5.53 ---- ---- ---- ---- ---- ---- ---- ---- " P2 " ---- ---- ---- ---- 4.28 ---- ---- ---- a P3 7' 4034 4008 "“" ---- 4941 4015 ---- “--- " P4 " 4.24 3.92 3.88 ---- 4.30 3.98 ---- ---- " P5 " 3.88 3.90 3.89 3.85 4.17 4.15 4.06 3.93 " P6 " 4.27 3.92 3.90 3.72 4.30 3.98 3.77 3.59 " LW " 4.25 3.93 3.90 3.75 4.31 3.99 3.77 3.58 " SW " 4.34 3.99 3.96 3.79 4.37 4.04 3.80 3.62 " AH " 4.30 3.95 3.90 3.75 4.35 4.00 3.78 3.71 S6 AH 5.53 4.55 4.51 4.50 4.50 4.63 4.53 4.52 4.51 " SW n 4.49 4.23 4.18 4.13 4.61 4.25 4.11 4.07 " LW " 4.51 4.16 4.15 3.96 4.63 4.25 3.96 3.71 " P6 " 4.57 4.14 4.09 3.89 4.66 4.24 3.97 3.72 " P5 " 3.90 3.97 4.00 3.98 4.41 4.40 4.30 4.15 " P4 " 4.56 4.30 4.13 3.98 4.71 4.39 4.09 3.82 '9 P3 " 4058 4016 4014 """ 4065 4025 “" -"-“ " P2 7' 4057 4033 “--- ‘-""" 4.7]. 4035 '-"" ---- " P1 " ---- ---- ---- ---- 4.59 ---- ---- ---- ‘ LW—-2" well; SW——3/8" well; AH-—Auger hole; P-—Piezometer. (concluded) ,around ZlevationsIITT}) Observ. - fbservi- 318' . f April 7 may Pumice'ice (ft.)gz'r 23 29 50! 5 4 5f 6 - i 1. . , s7 21 5.5'7 4.62f4.51 ---- ---- 4.77 ---- -------- " P2 " 4.6l----I-—-- ---- 4.76 4.38 ‘---.---- " Pa " 4.5:.4.21:4.244 51 4.75 4.57 4.07 --—— ' P4 " 4.5; 4.21)4.24 4. 29 4.75 4.574 U.01 5. 78 7 ‘5 ' 4.6: 4.25-4.24l4. 20 4.75 4.57 4.00.5. 77 7 P6 ' 4.62-4.2214.25 4.07 4.77 4.57 4.03)3.78 ' L: 7 4.6;'4.22 4.24:4.25 4.75 4.57 4.19 4.10 " 5 ' 4.59 4.2654.22:4.24 4.72 4.55 4.00 5.75 " 13 " 4.54'4.;7)4.20§4.12 4.69 4.52 4.07 4.07 I 1 SS 13 5.55 4.5: 4.25 4.25‘4.15 4.76F4.40 4.04 5.94 ' s: 7 4.65 4.27 4.29 4.27 4.81 4.44 4.10:5.92 " 1: ' 4.67}4.31 4.52 4.11 4.85 4.47 4.16!3.87 ' P6 7 4.63.4.30 4.55s4.20 4.87 4.49 4.12!3.85 II 25 ” 4.74 403,-)4035140 144 4089 4050 4015i3066 ' 24 ' 4.59j4.40'4.55i 4.24: 4.80 4.71 4.49.4.21 " 35 ' 4.52 4.46:4.54 4.16 4.95 4.58 4.24i---- ' Pa 7 4.52 4. 47 4.58!4.29 4.96 4.69 4.57 ---- ' P; ' 4.7a.----§----§----“4.89 4.51 -------- 39 Pl 5.49 4.614.503----:---- 4.92,4.59 ~---,---- " 52 ' 4.27 4.29 4.29i---- 4. 52 4.55 4.45 ---- ' PE ' 4.25 4.11;4.12 4.12.4.55 4.57 4.36 4.51 " P4 7 4.2: 4.27 4.15:4.10 4. 42 4.46 4.45 4.54 ” 5 ' 5.75 5.85'5.68;5.914.24 4.29 4.29 4.25 " 25 ' 4.21 4.21;4.22;4. 21 4.52 4. 54)4.m 4.52 ' L: ' 4.57 4.55 4.28 4.11; 4.84 4.50:4.21 3.95 ' s: ' 4.69 4.54 4.293L 4Q 4.87 4.595 .55 4.15 ' = ' 4.54 4.29 L 25)L 04% 4.85 4.49.4.19 5.91 L 1 1 j E ' 1 ~ . .. f}. 1 ‘0' g a. 4 Q ',,, O 1.1—2. I3--; ;5_3/;’ ‘g‘l; L;_‘-;€er {,Qle; r—Piezeae 962‘. 10. 11. 12.- LITERATURE CITED Aronovici, V. S., and W. W. Donnan. Soil-Permeability as a Criterion for Drainage-Design. Trans. Amer. Geophlg: Union. 27 (1946), pp. 95-101. Ayers, H. D. Ground water Studies as Related to Seepage and Drainage Problems on Irrigated Land. Unpublished M. S. thesis. State College of Washington, 1950, 67 pp. Baver, L. D. Soil Physics. John Wiley & Sons, Inc., No Ye. 2d ed.) 9 pp. Baver, L. D. Practical Values from.Physical Analyses of Soils. Soil Science. 68 (1949), pp. 1-14. Bethlahmy, Nedavia. A Method for Approximating the Water Content of Soils. Trans. Amer. Geophys. Union. 33 (1952), pp. 699-706. Bodman, G. B., and E. A. Colman. Moisture and Energy Conditions During Downward Entry of Water into Soils. Bouyoucos, G. J., and A. H. Mick. An Electrical Resistance Method for the Continuous measurement of Soil Moisture under Field Conditions. Mich. Agr. Expt. Sta., E. Lansing, Mich., Tech. Bull. 172, 1940. Bouyoucos, G. J. Nylon Electrical Resistance Unit for Continuous Measurement of Soil Moisture in the Field. Soil Science. 67 (1948), pp. 319-330. Briggs, L. J. The mechanics of Soil Moisture. U. 3. Dept. Agrio. Bur. Soils, Bull. 10, 1897. Seen in abstract only. Baver, L. D. Soil Phfisics. John Wiley & Sons, Inc., No Yo, 2d ado, 9 pp. Childs, E. C. The Transportation of Water Through Heavy Clay Soils: III. Jour. of Agric. Sci. 26 (1936), Childs, E. C. The Water Table in Drained Land. Soil Science. 59 (l945), pp. 313-327. Childs, E. C. The Water Table, Equipotentials and Stream- lines in Drained Land: V. The Moving Water Table. Soil Science. 63 (1947), pp. 361-376. l3. l4. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. - 57 - Childs, E. C., and N. C. George. Interaction of Water and Porous Materials, Soil Geometry and Soil-Water Equilibria. Discussions of the Faraday Society. No. 3, 1948. pp. 78-65. Darcy, H. P. G. Les Fontaines Publiques de la Ville de Dijon. Paris, 1856. Donnan, W. W., and J. E. Christiansen. Ground Water Determinations. Western Construction News. 19 (1944), Donnan, W. W. Model Tests of a Tile Spacing Formula. Soil Sci. Soc. Amer. Proc. 11 (1946), pp. 131-136. Donnan, W. W., and G. B. Bradshaw. Drainage Investigation Methods for Irrigated Areas in Western United States. U. S. Dept. of Agric., 8011 Gene. Serv., Washington, D. 0., Tech. Bull. 1065, 1952. Fair, G. M., and L. P. Hatch. Fundamental Factors Governing Streamline Flow of Water Through Sands. Jour. Amer. Water Works Assoc. 25 (1933),pp. 1551-1565. Haswell, J. R. Principles of Tile Drainage. Agric. Eng. 21 (1940), p. 310. Hazen, A. Some Physical Preperties of Sands and Gravels. Mass. State Board of Health, 24th Annual Rept., 1892, 1893. Israelson, 0. W. Application of Hydrodynamics to Irri- gation and Drainage Problems. Hilgardia (California Station). 2 (1927), pp. 479-528. ‘Seenwin abetraOt only. ""T"‘A5r 0. Eng. 8 (1927).. p. 322. Kidder, E. H., and W. F. Lytle. Drainage Investigations in the Plastic Till Soils of Northern Illinois. Agric. {§§g. 30 (1949), pp. 384-386. King, F3 H. Principles and Conditions of the Movements of Ground Water. U. S. Geol. Survey 19th Annual Rept., Pt. 2, 1899. Pp. 59-2940 Kirkham, Don. Pressure and Streamline Distribution in Waterlogged Land Overlying an Impervious Layer. Soil Sci. Soc. Amer. Proc. 5 (1940), pp. 65-68. Klute, A. Relation of Pore Size Distribution to Perme- ability of Soils. Unpublished M. S. thesis. Michigan State College, 1948, 27 pp. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. -58.. Luthin, J. N. Percolation of Water Through Saturated Layered Soil Separates.- Unpublished M. S. thesis. Michigan State College, 1947, 29 pp. Marshall, T. J., and G. B. Stirk. Pressure Potential of Water Moving Downward into Soil. Soil Science. 68 (1949), pp. 359-370. Meinzer, 0. E. Outline of Ground Water Hydrology with Definitions. U. S. Geol. Survey W. S. Paper 494, 1923. Muskat, M. The Flow of Homogeneous Fluids Through Porous Media. McGraw-Hill Book Co., Inc., 1937, 763 pp. Neal, J. H. Proper Spacing and Depth of Tile Drains Determined by the Physical Properties of the Soil. Minn. Agric. EXpt. Sta., St. Paul, Minn., Tech. Bull. 101, 1934. Norling, Sven A. Subsoil as a Factor in Drainage Design. Agric. Eng. 8 (1927), pp. 311-319. Roe, H. B. Some Soil Changes Resulting from Drainage. Soil Sci. Soc. Amer. Proc. 4 (1939), pp. 402-409. Roe, H. B., and J. K. Park. A Study of the Centrifuge Moisture Equivalent as an Index of the Hydraulic Permeability of Saturated Soils. Agric. Eng. 25 (1944), pp. 381-385. Schlick, W. J. The Theory of'Underdrainage. Iowa Eng. EXpt. Sta., Ames, Iowa, Bull. 50, 1918. Schlick, W. J. The Spacing and Depth of Laterals in Iowa Underdrainage Systems and the Rate of Runoff from them with Data from Investigations. Iowa Eng. Expt. Sta., Ames, Iowa, Bull. 52, 1918. Schwab, G. 0. written communication. Slater, C. S. The Depth and Spacing of Tile Drains. Agric. Eng. 31 (1950), pp. 448-450. Slichter, C. S. The Motion of Underground Water.. U. S. Geol. Survey water Supply and Irrigation Paper 67, 1902. Terzaghi, K., and R. B. Peck. Soil Mechanics in Engineer- ing Practice. John Wiley & Sons,*Inc., 1948, p. 49. 40. 41. 42. 43. 44. 45. 46. -59- Terzaghi, K. Chapt. IXa. Soil Moisture and Capillary Phenomena in Soils. Meinzer, 0. E., Editor. H drolo (Physics of the Earth--IX). Dover Publications, Inc., N. Y., 1949, 712 pp. Trullinger, R. W. Research in Agricultural Engineering. Agric. Eng. 8 (1927), pp. 285-289. Uhland, R. E., and A. M. O'Neal. Soil Permeability Determinations for Use in Soil and Water Conservation. U. S. Dept. of Agric., Soil Cons. Serv., Washington, D. 0., Tech. Paper 101, 1951. Veatch, A. C. Fluctuations of the Water Levels in Wells. U. S. Geol. Survey W. S. Paper 155, 1906. Walker, J. P. Depth and Spacing for Drain Laterals as Computed from Core-Sample Permeability Measurements. Agric. Eng. 33 (1952), pp. 71-73. Walker, J. P. written communication. Whiteside, E. P. written communication. o V 4hr 25 '$ zoom USE omv "'TITI'I‘flflfllL )1! Mill] )Iflffllflilflflflifllflm'“