to take the census of the livestock at the time when it is at its minimum and when you can be sure as to what you are getting. We think that you can get the most accurate information at the time when the new stock is not coming in. In that way the farmers will not be confused as to the stock they have, as to whether they had certain stock before or after April 1.

Let me show you gentlemen what these charts develop in that connection. We will take this one.

Mr. THURSTON. Just a moment, before you leave the charts dealing with grains. I want to offer a practical observation which has been gone into by Mr. Lozier as to the percentage of the corn crop that has been picked.

Mr. OLSEN. Yes, sir.

Mr. THURSTON. By the 1st of November. I am sure that in the State of Iowa 30 per cent of the corn crop has not been picked by the 1st of November. Now it is a very common practice, among everyone who raises grain, to overestimate his production and what his corn will yield, and he makes a statement of the amount, and I believe that nine times out of ten I am safe in asserting that he will have a production of from 15 to 25 per cent or 30 per cent less corn than he has estimated.

So that figures based on November 1, in my judgment, would show an excess of production of from 15 to 30 per cent over the actual production. I know that is the situation and that condition to be actual.

Mr. MOORMAN. There never was a truer thing stated.

Mr. OLSEN. But, he is estimating for a very small percentage of his crop by December. There are very few cases where it is not out, or he has not gathered a sufficient amount of his crop to have a reasonably good idea as to what his yield will be.

Mr. THURSTON. I am not advocating November 1.

Mr. OLSEN. I am advocating December 1.

Mr. THURSTON. Even that same fallacy as to the matter of production would apply as to December 1. Then, there is another thing to be taken into consideration. Anybody that has had experience knows that before the crop matures the farmer is certain to exaggerate the number of acres he has. When the crop has matured and been binned and sold and the amount of yield ascertained he immediately decreases the number of acres and inflates the production of the acre. You know, that is human nature. It is char

acteristic.

Mr. THURSTON. I think that that applies in ninety cases out of a hundred.

Mr. MOORMAN. And, the result of this exaggeration would be to report overproduction, more than the farmers actually had raised. Mr. OLSEN. We have tried to make, and have made, as close a check as we could and the results seem to indicate to us the fact that April 1 would give us less accurate information than January 1. We are convinced that we could get a more accurate census on January 1. MI. THURSTON. Let me direct your attention to this fact: Of course, cattle feeding in the Corn Belt largely commences in the fall and is carried over until late in the spring so that on beef production the available statistics would be more reliable around the first of April than they would be when the feeding commences.

Mr. OLSEN. I think not. I think that that is very important. Take the census as of April 1 and what do we get? Your hog farrowing has begun. If it were taken on the 1st of January you would not be getting the spring pigs. Whereas on April 1 you get a figure which would not include all of the spring pigs. You are getting a figure then that is neither fish nor fowl-you do not know what it is. Mr. THURSTON. I am becoming confused here as to whether we are more interested in production or consumption. I would like to know what the attitude of the department is with reference to the gathering of these statistics. Is production or consumption the thing that you are primarily interested in?

Mr. OLSEN. We are interested in both.

Mr. THURSTON. Manifestly, a different rule applies to production than to consumption.

Mr. OLSEN. We are trying to get data on both. I do not quite get your point.

Mr. THURSTON. Well, the grain production is marketed during the latter part of the year, whereas a major portion of the livestock is marketed during the first half of the year, and you would appear rather closer to the production than to marketing.

Mr. OLSEN. I think that we want to get both and if you take the census during the latter part of the year you are going to get a pretty accurate report, whereas if you take it with breeding stock coming in during the first half of the year, you are going to get an incomplete inventory of the new stuff.

Mr. THURSTON. I am not attempting to say what is in the minds of the rest of the committee, but as far as I am concerned, I am confused as to the proposition that you are most interested in, whether it is the collection of production statistics, or marketing statistics.

Mr. OLSEN. Most of the census deals with the production statistics We are equally interested in marketing statistics and would be very glad if the census could provide more of such data.

The CHAIRMAN. Right here, I notice on April 1 that the cattle on hand on April 1 is shown as 101.1 as against

Mr. OLSEN. Where is that?

The CHAIRMAN. I am speaking of 1924.

Mr. OLSEN. 1924, here [indicating on chart].
The CHAIRMAN. Yes. On May 1 it was 103.5.

Mr. OLSEN. And on June 1 it is 105.4.

The CHAIRMAN. Yes; and it runs comparatively level to August 1. There is only a difference of about four-tenths of 1 per cent between April 1 and the highest figure, which I think is August. There are at least four months when the figure is about the same.

Mr. OLSEN. And I might say that represents a high peak in cattle numbers.

The CHAIRMAN. Let us take December.

The CHAIRMAN. According to that, December, 1924, there is nothing on hand, nothing shown on that chart for December.

it in December than it is at

Linthera boieren according to 1923. hart there, 1923.

The CHAIRMAN. And, it is higher in 1922.

Mr. OLSEN. Yes, sir.

The CHAIRMAN. I wish that you would tell us what material difference there is between December 1 and those other figures that you have quoted.

Mr. OLSEN. We have here April 1 a lot of new stuff that is coming in only part of which would be included in the census and that does not give you an accurate figure. You have got a miscellaneous lot of stuff included in the census if taken April 1, and we do not know what

it is.

Mr. THURSTON. Could you not inquire of the farmer, Mr. Olsen, as to what he had on hand last year and not what he had on hand at the time of the taking of the census, or has on hand this year?

Mr. OLSEN. The point is that you are getting the calves and pigs that have come in since January, and you are getting those confused with the stock born last year.

Mr. THURSTON. The enumerator will ask the farmer what he had on hand during the preceding year, or last year, rather.

Mr. OLSEN. Certainly; he will ask him that.

Mr. THURSTON. And, you think, that he will get them confused? Mr. OLSEN. Without a doubt the farmers will be confused as to those two sets of data.

Doctor HILL. I think that I ought to correct one error.

The CHAIRMAN. You think that the farmer is going to be confused as to what he had on hand on the first of January as compared with what he has on hand when the census is taken?

Mr. OLSEN. Yes.

The CHAIRMAN. Now, you have introduced some extraneous figures here that you have not told us about yet, as to why those figures correspond so closely. If the figures correspond so closely what difference does it make whether it is the latter part of the year or April 1?

Mr. OLSEN. Then there is the problem of classifying livestock by ages. Such data are exceedingly important in determining the potential production of livestock. The enumeration of young animals is very difficult. It is easiest to make when the births are at their lowest numbers and when there is least confusion in the mind of the farmer regarding his livestock.

The CHAIRMAN. Do you not think if we took the farm census and the population census and the distribution census all at the same time, at one period, that we would be able to get a more accurate census, and it would be more convenient to the farmer to give his acreage, population, his stock and all of the other information in the return altogether than it would be to confuse him with a crop report and a population report at a different period? Is not the farmer to be considered, his pleasure, or his adaptability and would it not be better to get all of the information at the same time than it would be to interrogate him at two different times? In other words, can not a man give his agriculture, his crops, number of tenants, mortgages on his farm and all of the details which the Bureau of the Census require on one date rather than to have him mixed up with two dates?

I want to be convinced that separate dates are better than the system of having it all taken at one time. I am speaking now from the standpoint of the convenience of the farmer.

Mr. WHITE. We understand that this is a calendar year census, do we not?

The CHAIRMAN. Yes; this is a calendar year census.

Mr. OLSEN. You understand that I am not necessarily arguing for two censuses.

The CHAIRMAN. If we are to take an agricultural census for 1929, no matter what time of the year in 1929 it may be, we are obliged to take a population census in 1930, too. That is required by the Constitution. We can not get away from that in any way.

I am not insisting on April 1.

Mr. OLSEN. January 1 would meet your suggestion?

The CHAIRMAN. What we would like to get is a date that would be acceptable for the two censuses, agriculture and the population, so that they could be taken at the same time, both for convenience and economy. It seems to me that that is the thing that ought to be considered.

January 1 has been referred to in previous sessions of the Congress as a date for the taking of the census and the statement has been made that January 1 was not a good date for the taking of a population census throughout the country because of weather conditions,

etc.

So, if we could arrive at some date that is a most convenient date, not only for the Department of Agriculture or the Bureau of the Census, or as far as we are concerned our convenience, but to select a date so that we can do it all at one time, so that all of the censuses will be taken at the same time for the convenience of the country and the convenience of those who furnish the information, file their returns, etc., and also for economy in taking the census, that is what we should do. I think that we can arrive at such a date.

Mr. THURSTON. I am sure that all of this discussion leads us out into different fields.

The CHAIRMAN. Surely.

Mr. THURSTON. And we are merely approaching the problem with a view to solving it.

The CHAIRMAN. I think that we should bear in mind that we are attempting to select a date that is suitable to the taking of the agricultural and a date that is suitable to the taking of the population

census.

Mr. THURSTON. I want to observe here that I am not nearly so solicitous of the inability of the man on the farm.

The CHAIRMAN. I am not either.

Mr. THURSTON. To make some fair statement of his business. In fact, we all know that the farmer is able to do that. I do not feel apprehensive along that line. I feel that the average men, average farmer, being a man with a small business, as probably the average farmer is, yet I think that he has the main facts in his mind, and that they can be elicited and that he can keep them in his mind over a period of three or four months.

The CHAIRMAN. In other words, the farmer knows what he is doing?

I think that Doctor Hill wanted to make the statement
in that connection. The Census Bureau has been
on livestock at the time the census is taken.
The farmer's inventory will apply to the date on
is to be taken. If the date of the taking of the

census is as of April 1 he is required to give the situation as it existed on that date. If it does not relate back.

Mr. THURSTON. Of course, it would be better to take a calendar year some time prior to the taking of the census.

Doctor HILL. That has been the practice. The practice has been to ask what he had on hand. I do not lay so much stress upon precedent, but I should question the wisdom of asking him to give the situation as it existed three months before, as far as the inventory is concerned.

Mr. THURSTON. I believe that we are assessed for taxes on the property that we possessed the prior year, not what we have at the time the assessor comes around. Is not that true?

Doctor HILL. I think that is true, but the assessor usually comes around within a week or two after that date.

Mr. OLSEN. I have a number of other charts that illustrate what would be the situation on livestock. Since we are taking an agricultural census, and getting the various classifications of livestock and finding the existing conditions, I think it is important to get the facts that will enable us to determine what the potential livestock production will be. Suppose that we asked for heifers 1 year and under at April 1. What are we going to get? You are going to get the stuff that was born between January 1 and April 1, and a lot of stuff that has come over from the preceding year. You will have to take that into consideration.

Suppose we ask for milk cows 3 years old or under. You are going to have exactly the same situation. Suppose that you ask for the number of yearlings. You will find that by April 1 there are a great many of them that are nearing 2 years, but they are still going to be classed as yearlings.

Mr. MOORMAN. You will have that difficulty at any time during the year with some kinds of stock.

Mr. OLSEN. I think that we will have far less difficulty along that line if we take the census as of January 1 than we would if it is taken as of April 1. Take the situation with reference to range cattle and sheep. They are branded and tallied, in a very large majority of cases, by the end of the year, by December 1, or by January 1 in the western part of the country, and they are pretty well settled down for winter.

By April 1 they have gone out on the range and in many instances are in transit. That is something that you will have to figure on. Mr. LozIER. You think that the range men would not have a record of the number of sheep they have?

Mr. OLSEN. Probably they will, but how are you going to maintain the comparability of the 1930 census with those taken in 1925 and 1920 as of January 1, when by April 1 the sheep bands are on the move from winter to summer ranges?

The CHAIRMAN. Let me ask you this: We have talked about crops, cereals, cattle, and all of that. Your department, I presume, gives

considerable attention to fruits?

Mr. OLSEN. Quite right.

The CHAIRMAN. When do you think would be the proper time to take the fruit census? That is a changing condition in all parts of the country. They have ripe fruits in Florida to-day, while the trees are without leaves in the North. To-day, the fruit crop of the United States is getting to be a very, very, important crop.