Mr. SEARS. In my State most of the oranges are moved from the groves to the railroad over vitrified brick roads, as you know, and the ultimate consumer pays the tax. Why put it all on gas? Why not put some of it on the 4-mule teams down in my friend's State that haul cotton to the depot instead of putting it all on gas?

Mr. MACDONALD. As a matter of equity I favor taxing the narrow-tired wagon and heavy freight wagons the same as I am in favor of taxes on motor vehicles.

Mr. LARSEN. I think a man who goes out in a horse-drawn vehicle is taxed enough by the time he gets to the end of his journey.

Mr. MACDONALD. Mr. Chairman, I have occupied a great deal of the time of the committee. I think you want to hear some other witnesses. I am anxious, however, to bring out a few facts on the matter of soils in subgrades, because whenever we talk prices, whenever we talk traffic, whenever we talk the kind of roads that are necessary, we have to go back to the character of the soils over which we are building roads. I brought some samples of actual subsoils. The thickness and to a large extent the cost of a road surface that is built to hold the traffic out of the mud is dependent on the subgrade underneath. This is a soil sample [indicating] from Hardin County, Ky., and this small sample shows the shrinkage of that soil; that is, the difference between the soil wet and dry. The sample originally filled the container and has been allowed to air-dry. You can see that the shrinkage even in this quantity of material is sufficient so that a concrete road supported on that soil, as the soil dried out, would be left without support under the edges; that is, the old grade itself would shrink away from the concrete to such an extent that you probably would be able to stick your hand right under the edge of the concrete slab on account of the shrinkage of the soil. That is an example of a very bad road-building soil. Then we have the adobe soils of California and the Western States which do exactly the same thing. When we made a study of those roads we discovered that the breaking down of many of the roads was on account of the shrinking away or pulling away of the soil from underneath the concrete slabs, so that there was nothing to support them, and the slabs were just sticking out in thin air, as it were.

The CHAIRMAN. Can you tell us something about the caliche soil that you find out in the mountainous region of the West. I found out something about that when I was in the West, and they said that that soil melts in very nicely with a moderate amount of cement and saves using an excessive amount of cement.

Mr. MACDONALD. I think that would be true. I have another soil here which is a very poor soil. It is a finely divided soil and was mixed with about 47 per cent of moisture two days ago and it has been air dried. I will pass this sample around, and you will notice that that sample, mixed with 47 per cent moistureit has been air drying for three or four days—would not support the pressure of the subgrade under a heavy load. That same soil with the same amount of moisture exactly mixed with a little cement is now very hard. It is the same soil, but there is just a small percentage of cement mixed with it.

The CHAIRMAN. There is a lot of calcium magnesia in that soil. Mr. MACDONALD. Yes; and you will see that it is very hard. Just the addition of a very small amount of cement has changed it from a very plastic material to a very hard material. This is a sample of the same soil heated. In the colloidal soils, the soil is so fine that if carried in water, it only gives a murky color; that is, holding it up to the light, it is impossible to see the individual grains or atoms or molecules or whatever they are.

The CHAIRMAN. I saw one road in Arizona made of that caliche soil that was mixed with a very small amount of cement, and it was a very good road. Mr. HUDSPETH. It looks very much like concrete, anyway.

The CHAIRMAN. It is a little rougher, I believe.

Mr. HUDSPETH. Yes.

Mr. MACDONALD. In contrast to this Kentucky sample which I showed you, here is a sample of soil taken from Alexandria County, Va. It has a great deal of sand in it and has no shrinkage. It was wet the same as the other sample, but when dry it has no shrinkage, so that it would support the subgrade in very good shape.

Mr. DOUGHTON. Mr. Director, I notice in connection with these two samples of soil you have in these containers, that one is harder or firmer than the other. Mr. MACDONALD. That is exactly the same soil, but one is mixed with a little cement.

Mr. DOUGHTON. Yes; this sample has been treated with a small amount of cement. Is it your idea that these soils need to be treated with cement before they can be used as a foundation for the main surface of the road?

Mr. MACDONALD. We are trying that out in an experimental way. We are trying in every way to improve subgrade material of that kind. That soil, when once wet, will hold its moisture for so long a time that any surface over it is likely to break down. It has little supporting power.

Mr. CABLE. How much investigation does the Federal Government make of subsoils before they approve the plans and specifications?

Mr. MACDONALD. We make investigations of that kind over all the roads. Mr. CABLE. How do you do that?

Mr. MACDONALD. Our engineer goes over the road mile by mile. For example, in Ohio the question came up of building a broken-stone or crushed-stone base under brick roads. We investigated very carefully the soil, and determined because of the soil conditions we would have to use a greater thickness of broken stone over the clay soils than we would in other localities. Here is a sample of a typical agricultural soil from Polk County, Iowa. It contains a great deal of the finely divided material, and you can see the amount of shrinkage as it dries out. That shrinkage is sufficient to pull that subgrade away from the road. This soil will hold its moisture almost indefinitely.

This is a sample of the top soil that we use a great deal in the South for building roads. This is a good top soil. This is the material as we find it— naturally.

Mr. WARD. It looks like it is granular.

Mr. MACDONALD. Yes; it is 44 per cent of coarse sand, 20 per cent of clay— that is all the clay there is in it, only 20 per cent-and silt, 9 per cent; and fine sand, 27 per cent; that is, you have 27 per cent of fine sand and 44 per cent of coarse sand.

Mr. SANDLIN. What States are those samples from?

The CHAIRMAN. Would you approve the building of cement roads on that pasty soil you have there? Have you been approving cement roads on that kind of soil?

Mr. MACDONALD. We have built some on that soil, ordinarily trying to treat them in some way. That soil treated with sand or cement or gravel will make a very much better base; that is, when mixed with some other material of that kind. We are conducting a large number of experiments in which we are using cement and lime and oil and sand and all the different materials we think may solve the problem of providing a better support for these subgrades. The CHAIRMAN. You have an artificial base of some kind under all concrete roads, have you not?

Mr. MACDONALD. Not under all kinds. For example, when we get into soils of this kind [indicating] it would be entirely unnecessary. I might qualify also what I said the other day about building the road grade a year before the surface is placed. It would not be necessary where we have soils of this character. That is done in the adobes, the clays, and all the soils which contain a very large amount of finely divided material.

Mr. HUDSPETH. Mr. Sandlin asked you where those soils were taken from. Have you that information? You stated they were taken from the Southern States.

Mr. SANDLIN. Is it about the same in all the Southern States? Mr. MACDONALD. No; it varies. Here is what we call a poor top soil. It has 45 per cent of clay. This you can see would be a better agricultural soil than the other sample. This sample has only 20 per cent of clay while this soil has 45 per cent. That would be a poorer road material than the other but very much better agricultural material.

Mr. ROBSION. Generally, a good road soil is not such a good agricultural soil. Mr. MACDONALD. Exactly. Just to show the difference, Mr. Chairman, this diagram shows the bearing value of the different kinds of soil. This is California clay thoroughly saturated and this is a sand clay with the moisture at 16. Now, you can see that gravel and cinders, even though thoroughly saturated with moisture, have a very good bearing value. Here is a California clay thoroughly saturated and here is a California clay dry, or practically dry, with a minimum percentage of moisture. In one case we figure, when dry, about 45 pounds per square inch, and in the other case probably about 1 pound or one-half a pound per square inch. This table shows the difference in the bearing value of these soils when dry and when wet. I should like to submit for the record another table, and, as I remarked the other

day, there are blue prints for any of the members of the committee who wish to have one. This table shows the cost of all surfacing which has been placed and paid for up to the 31st of December. There are about 7,500 miles, and this table gives the different classes of construction and the cost by the geographic sections of our country. We find that the costs vary between the different geographic sections but are reasonably uniform within each geographic section. I think this should be incorporated in the record.

I should also like to prepare a table, as requested, showing the traffic and the kind of traffic we have on the roads, and I should like the very great privilege or rearranging this testimony so that it will come in consecutive order under different subjects.

Mr. Chairman, I wish to incorporate in the testimony a table which shows by States the different classes of work and the cost of it and the amount of Federal aid.

The CHAIRMAN. As applied to those roads?

Mr. MACDONALD. By States, the amount of graded roads, sand and gravel roads, and macadam. It is divided up by types of construction and by States, so that the committee can see for each State the amount of roads, and this table is made up on the basis of all of the roads that have been paid for up to the first of the present year. These projects include all the Federal aid projects that have been completed and paid for.

The CHAIRMAN. Does it include the mileage cost?

Mr. MACDONALD. The mileage is given under the amount of Federal aid. We give the estimated cost from both funds and the amount of it that is Federal aid and the mileage.

The CHAIRMAN. Is the forest mileage paid partly by the States?

Mr. MACDONALD. No; I am not talking about forest roads.

The CHAIRMAN. I thought you were talking about forest roads.

Mr. MACDONALD. No; I am talking about all the post roads that were completed and paid for up to the first of this year. Every mile is included in this table.