The establishment of soil erosion experiment stations in the various major land divisions of the country, especially susceptible to erosion, together with an aggressively prosecuted educational program, has done more to arouse the country to the extent of the damage and the need for protective measures than have the combined efforts of all the agencies of the past. For the first time research methods are being applied to this exceedingly diversified problem in accordance with a comprehensive scientifically coordinated plan, involving the cooperation of the Bureau of Chemistry and Soils, the Bureau of Agricultural Engineering, State experiment stations, and many other agencies.

Mr. SANDLIN. To what extent are you having the cooperation of the farmers in this work? Do they make visits to the stations?

Dr. McCALL. Yes; the farmers get the information through visiting the stations, and through attendance at these field meetings. The total number of visitors at the Bethany, Mo., station last year was approximately 2,000, the majority of whom were farmers, farm managers, and land appraisers. At the third annual field-day meeting, 255 farmers, extension agents, and teachers were in attendance. At the Statesville, N.C. station 1,200 visitors registered during the past year, and at the Zanesville, Ohio, station more than 800 visitors were shown over the farm. In southern Iowa, 720 people inspected the work of the Clarinda station during the season.

Mr. SANDLIN. You have a station at Tyler, Tex., do you not?
Dr. MCCALL. Yes, sir.

Mr. SANDLIN. We have the same type of land in north Louisiana that you have in the vicinity of Tyler, Tex.

Dr. McCALL. Yes, sir.

Mr. SANDLIN. How is that information made available to them, or of what benefit is that station to the farmers of north Louisiana? Do they make visits to this station to witness the demonstrations?

Dr. MCCALL. Yes, sir. Your extension people in the northern part of your State have been very keenly interested, and I think they have kept themselves pretty well informed as to what is going on.

Mr. SANDLIN. Do you know what other organizations have visited

this station?

Dr. McCALL. I could not say positively, but I am quite sure that certain groups have been there. Do you know, Mr. McCrory? Mr. MCCRORY. The agricultural engineers have visited the station, I know.

Dr. McCALL. As to just how many groups have visited this particular station we have no definite information.

Mr. SANDLIN. Do you have any data showing the number of farmers that are now using the methods that have been developed by your investigations?

Dr. McCALL. We have no statistics on that, except that Mr. McCrory, I think, will introduce some statistics as to the extent of the terracing program that has been developed.

Mr. MCCRORY. Yes.

Mr. SANDLIN. All right; we will be very glad to hear from you now, Mr. McCrory.

AGRICULTURAL ENGINEERING FEATURES OF SOIL-EROSION WORK

Mr. MCCRORY. The work under this appropriation has to do with the agricultural engineering features of the Department's soil erosion investigations and is financed by an allotment of funds made to the Bureau by the Secretary of Agriculture. In cooperation with the Bureau of Chemistry and Soils, the Forest Service, and State agricultural experiment stations and other agencies, 10 soil erosion farms in 9 States have been established. While much progress has been made, there is still much to learn.

The most important soil erosion control problem in the United States is the conservation of the fertile top soil of cultivated fields, which under common farm practices, is washing away at a rate of from 10 to 40 tons or more per acre annually. A great deal of land in the United States at present in cultivation should never have been cleared because due to steep slopes and erosive soil, each heavy rain causes serious erosion. In order to minimize erosion such lands should be replanted to forest or put into permanent pasture with grazing controlled so that a good vegetative covering can be maintained.

On cultivated land it is impossible to maintain a good vegetative covering at all times. During the period of plowing, preparation of seed bed, planting, and until the crop has attained a sufficient growth to furnish a suitable cover, cultivated land, because of the loosened and disturbed soil conditions, is especially susceptible to erosion. The practice of terracing or "hillside ditching" has been found to be a very effective method of controlling erosion on such lands.

Confidence in terracing as a means of controlling erosion on cultivated fields has been justified by the results obtained in our engineering experiments on our 10 cooperative soil erosion experiment stations. That this is recognized by farmers is shown by the fact that agricultural extension services in 22 States are actively engaged in terracing programs and that during the year 1932 approximately 2,000,000 acres were terraced on 50,000 farms. Records for 1933 have not yet been compiled. A summary of the county agricultural agents reports for the past 10 years shows a total of nearly 13,000,000 acres of terracing done principally in conformity with recommendations of the Department, and of over 18,000,000 acres on some 600,000 farms from 1915 to 1932, inclusive.

That work is carried on by the extension service in cooperation with our extension engineer and the agricultural engineering departments at the colleges.

In a survey conducted jointly by the extension services of Oklahoma and Texas during 1932, 161 farm land appraisers and county agricultural agents estimated that in that region terracing increased the cash value of the land a minimum of $8 per acre.

Banks and banking associations have in various areas of the South required lands to be terraced before accepting them as securities. Some insurance agencies with large land holdings are now terracing these to protect them from erosion losses.

Custom terracing work is being done at present at costs ranging from $2 to $4 per acre. Where the county road machinery is available, the cost is running on an average from 75 cents to $1.50 an acre. Oklahoma, Texas, and possibly some of the other States have passed laws permitting the use of county road machinery for this purpose.

In Tallapoosa County, Ala., there is a very interesting development of that kind. The county as a group has decided that if they are to maintain a prosperous county the conservation of the soil is essential, and the county is buying the crawler type tractors and renting them to the farmers on a basis so that after a certain number of acres are terraced the tractor will be paid for. The last word I had from that county was that they had bought five such machines, and were not able to meet the demand for work of that character.

Costs with county crews and equipment have ranged between 75 cents and $1.50 per acre in the West for grading alone. Farmers may reduce cash expenditures by using their own farm labor and power in many cases, but very few of them acquire the technical skill for the surveying and grading, and ownership of the equipment is uneconomical on the average farm.

The objects of the engineering investigations on the erosion experiment farms are (1) to determine the water-carrying capacity of terraces for different soils and land slopes, and the required carrying capacity according to the fall, vertical spacing, and length of terraces; (2) to determine the effectiveness of terraces in preventing erosion and conserving moisture, by comparing water run-off and soil losses from terraced and unterraced land; (3) to develop the most economical methods of constructing and maintaining terraces under different farming conditions and practices; (4) to determine the best available machinery for building terraces, and to design better machinery for the purpose; (5) to study the operation of various types of farm machines over terraced land, with a view to recommending changes in design to improve their operation; (6) to improve the design and construction and to reduce the cost of various types of check and soilsaving dams for controlling erosion in gullies, terrace outlets, and road ditches; (7) to devise other methods involving the application of engineering principles to the control of erosion.

The most interesting results from the Guthrie, Okla., station were: During 1933 the soil losses from level terraces on land with an average slope of 5 feet per 100 was only 4 percent as much as that from a comparable unterraced area of 3.6 acres.

Yearly soil loss is about one fourth as much on a level terrace 1,500 feet long with open ends as on a similar terrace having a grade of 6 inches per 100 feet. This data indicates that the old practice of building terraces with steep grade caused the terraces to be inefficient in their purpose to conserve soil.

The terraces of the Guthrie farm were constructed during the spring of 1929. Data obtained to date indicates that the reduction of crop productivity on the soil disturbed by terracing operations practically disappears after the first 2 or 3 years of service for the terraces. In fact, after this period the disturbed area begins to show a higher yield than the undisturbed area.

The one-way disk plow, using disks 26 inches in diameter, is a necessary tool on all wheat farms in the Guthrie region. Experiments indicate that this machine will build terraces equivalent to those built with ordinary blade graders in approximately the same number of rounds. Only 1 man is required to operate the tractor and disk unit, while 2 men are required for the tractor and blade terracer. It will not function efficiently on wet or very damp ground that may be easily worked with a grader.

TEMPLE (TEX.) STATION

There has been a great deficiency in rainfall at the Temple station during the past 3 years, and as a result comparatively little information of a conclusive nature has been collected. It has been found that level terraces with closed ends, such as are used very successfully in conserving soil moisture and increasing crop yields in the western part of Texas, are not feasible in the blackland region of Texas as after rains causing run-off water stands above the terraces long enough to seriously damage the crops. Terraces having grades of 1 to 2 inches per 100 feet have worked satisfactorily so far at Temple and have not been overtopped; neither has water stood in the terrace channels. The results of one experiment demonstrated the effectiveness of grain crops in controlling erosion. The soil losses from three comparable areas planted to cotton, corn and winter oats during a year of light rainfall were 1.43 tons, 6.5 tons, and 0.05 ton of soil per acre. The loss from the area planted to oats was about one twentyeighth of the loss of that planted to cotton, and one thirteenth of that planted to corn. Similar results have obtained during 1931

and 1932.

HAYS, KANS.

During the calendar year 1933 but 16.26 inches of precipitation occurred at the Hays station as compared with a total of 31.66 inches for 1932, and a normal of 23 inches. The 1933 wheat crop received 8.65 inches of rain during its growing period (Sept. 30, 1932, to June 30, 1933). During this period terraces have increased crop yields at the station. In one experiment, the wheat yield on an area with level terraces was 16.5 bushels per acre compared with a yield of 7.8 bushels per acre on an unterraced check plot. In another, the yield was 14.2 bushels on the terraced area and 13.5 bushels on the unterraced check plot. In every experiment the terraced area yielded more grain per acre than the adjacent or comparable unterraced plot.

STATESVILLE, N.C.

Experiments with terraces with various grades indicate that in this region also terraces with variable grades of 0 to 3 inches per 100 feet are the most effective in controlling erosion. It has been found that the length of terraces can be increased very materially over what was formerly thought feasible

TYLER, TEX.

An area with level terraces 700 feet long lost 2.35 tons of soil per acre during the year 1933 as compared with a loss of 41.02 tons per acre from a similar unterraced check plot. Both areas were in cotton and cultivated as nearly alike as possible. At this station also, terraces with variable grade (0-3 inches per 100 feet) were found more effective in controlling erosion than terraces with uniform grades. Level terraces with closed ends proved unsatisfactory under the rainfall and soil conditions existing on the Tyler project.

A strip cropped area of 1.73 acres in corn and sorghum lost 12.33 tons of soil, and 15.77 percent of the rainfall ran off. An area of 1.24 acres above a long level terrace planted to corn lost 1.24 tons per

acre and 4.68 percent of the rainfall ran off. In another experiment short terraces lost less soil than long terraces.

The effect of timber cover on run-off and erosion was shown during December 1932 when 10.35 inches of rain fell. During this month only 1.08 inches of rain ran off from a wooded area and the soil loss was only .06 ton per acre as compared with 4 inches of runoff and a soil loss of 7.02 tons per acre from an unterraced cultivated watershed.

BETHANY, MISSOURI

Rainfall was nearly normal; 13 major storms caused run-off. On September 26, 4.8 inches of rain fell in a period of 9 hours. During 1933 measurements of run-off and erosion were obtained from 22 terraced areas and 4 unterraced areas. No terraces broke or were overtopped. The data obtained during the year demonstrates the benefits from terraces on Shelby soils regardless of the crop grown. A 5-acre unterraced corn field lost 27.6 tons of soil per acre during the year. On similar terraced fields in corn the soil losses were 3.2 and 4.4 tons per acre.

The measurements of run-off and soil loss from an unterraced alfalfa area shows that during the first year after planting alfalfa the soil loss may be considerable The unterraced alfalfa area lost 10.7 tons per acre as compared with a loss of 1.5 tons per acre from a similar terraced area.

During the last half of the year a natural unterraced watershed planted to strip cropping had an erosion loss of 20.1 tons per acre for the half year period, as compared with a loss of 3.1 tons from similar, though not identical, terraced areas. One half of the strip cropped area was in oats followed by clover; one fourth in corn, and one fourth in soybeans followed by wheat.

The measurement of run-off and erosion from an established blue grass sod about 40 years in age shows that it is fallacious to assume that the run-off and erosion from blue grass pasture are negligible. This pasture area had a soil loss of 1.27 tons per acre. This loss is small in comparison to the unterraced fields in cultivation, but is sufficiently large to justify some thought to the control of erosion in pastures. The loss of 1.27 tons per acre a year represents pasture erosion with gullies controlled. Thousands of acres of gullied pasture land in northern Missouri indicate that if pastures are not terraced, the gullies should be controlled by the use of water-fall flumes, soil saving dams and seeding.

A large gully 1,400 feet long varying in depth from 10 to 25 feet and in width from 75 to 150 feet is being reclaimed. A fill of 10 feet has been obtained immediately above a soil saving dam at the lower end. The sides of the gully have been plowed in and a tile drain laid in the bottom. The banks are being cultivated and the gully, together with the adjacent slopes, is rapidly taking the form of a new tillable valley. At this station also, experience indicates that terraces with little fall are most effective in reducing erosion.

The Bethany method of controlling erosion at terrace outlet ditches with the use of broad, flat, shallow ditches protected with hidden dams and surface seeding is working satisfactorily in the outlet ditches of 5 different fields with drainage areas varying from 6 to 18 acres each.