I believe that the hearings before the House committee contained more illustrations of this memory bias.

The CHAIRMAN. We will have those available.

Mr. BECKER. Yes. I do not believe it is necessary to take those up. I merely wanted to summarize our experience in a great many analyses that we have made, trying to eliminate this sort of bias, and to say that it is our feeling that the more nearly we can approximate the date that a particular event occurred, the more nearly can we get accurate figures; and in that connection, while it is true that corn is still partly unharvested and cotton is still not entirely picked on the 1st of November, that November 1 gets closer to a date when all of the information on production will be known to the farmer than any other date that we can think of or that our studies of final harvesting would indicate.

In connection with the livestock, on November 1, on the farms, with the exception of hogs, I think we come closer to a period of the year when the bulk of the animals are fairly mature, we come closer to the time of year when the number of animals to be used for breeding purposes is definitely known.

If we make an inquiry on cattle on the 1st of May, it is complicated by the fact that we have young stock coming on every day. Of course, calves are being born every day and lambs are being born every day and pigs are being born every day, and the farmer is a little bit confused in his mind in trying to separate these out for the enumerator when he is asked to classify his livestock by age groups, and November 1 comes closer to a date when the mature animals are being held on the farm as a minimum number, because, of course, the farmer must fit his numbers of the different animals on the farm to the supply of feed and grain, to carry them over the winter; and, of course, in the case of hogs he has a large number of hogs on the farm at that time that he is preparing for market; and it is true that during November and December a very large percentage of the hogs are being marketed, and that is one difficulty which must be considered in a census being made as of that date.

However, we have very satisfactory statistics of market receipts, and of publicly inspected slaughter, which are very helpful in determining the relationship, say, of the number of animals slaughtered during that period to the total number produced during the calendar year; which is, of course, the figure in which there is a great deal of interest, and the figure which is taken, year by year, gives us a measure of the importance of the industry as contributing to the farmer's income.

Senator BURTON. In that question of the proper percentages, what share of the calves and lambs are born after the 1st of May?

Mr. BECKER. Perhaps Mr. Harlan could answer that. Could you hazard a guess at that?

Mr. HARLAN. Ninety per cent of the lambs are born by the first or the middle of May, anyhow. The percentage of calves is probably not over 60 per cent. Sixty per cent of the calf crop at least is born by the first or the middle of May. Of course lambs are a much more seasonal crop than calves are.

Senator BURTON. How about the share of hogs and cattle sent away to the slaughtering houses for the market before the 1st of November?

Mr. BECKER. The spring crop of pigs from, say, Kansas, begins to move in October, and some of the spring crop from Iowa; just the very first start of the spring crop of pigs from Iowa. Of course the preceding fall's crop of pigs is moving, the tail end of that movement is on, in October, so that you have two crops of hogs. The spring crop, the pigs that were born in the spring, and the smaller crop in the fall. The spring crop begins to move, depending on the earliness or lateness of the season, some time early or late in October; and, as I pointed out, the big share of the spring crop is moved during the months of November and December.

The CHAIRMAN. Are there any other questions of Mr. Becker? Mr. BECKER. There is nothing else I have to bring up, unless there are some questions.

The CHAIRMAN. All right.

Mr. OLSEN. I think Doctor Gray would like to make a point or two, if he may.

Senator VANDENBERG. While we are hearing from the Agricultural Department, I would like to emphasize the fact that among the letters which I filed this morning with the committee is a letter indicating that the former Secretary of Agriculture, Mr. Jardine, is very heartily in favor of November 1 as the date, and that the same recommendation was made by Mr. Hoover as Secretary of Commerce.

Senator BURTON. Mr. Jardine's letter is in the report of the Secretary, and I must say that I had a bias in favor of the 1st of May before I read that. It makes a strong showing for November 1. That is in this report. Senator VANDENBERG. Yes. I was anxious to have that in our hearings.

STATEMENT OF L. C. GRAY, BUREAU OF AGRICULTURAL ECONOMICS

The CHAIRMAN. You may proceed, Mr. Gray.

Mr. GRAY. I just wanted to make one point in regard to the population matter which some of the Senators seemed to be concerned about. Mr. Olsen showed you the charts which seemed to indicate that probably as many people are on the farms on November 1 as on May 1. The fear that seems to exist in the minds of some, that there may be injustice done to rural communities from the standpoint of population and enumeration if the census is taken in November, I think is really met in a way by a provision in the bill as it is now drafted, and if I may read a few lines of that, reading from the top of page 6, the bill requires that the census enumerators shall visit personally each dwelling house, which of course will be, in the country, the farm house. In case he does not find the head of the family "or the member thereof deemed most competent and trustworthy," or "such individual living out of a family," he is supposed, "in case no person shall be found at the usual place of abode of such family, or individual living out of a family, competent to answer the inquiries, then it shall be lawful for the census employee to obtain the required information as nearly as may be practicable from the family or families or person or persons living nearest to such place of abode who may be competent to answer such inquiries."

47180-29

Now, evidently, the intent of the law is that a person shall be enumerated as of the place where he lives, and if the person has left, we will say, Iowa, and gone on a pleasure trip to Florida for a few weeks, presumably the enumerator gets the information from the neighbors and enumerates that person as of his place in Iowa.

It is of course probable that there are difficulties in that regard, and sometimes people are missed in the process; but certainly any resident of a vicinity who is absent from that vicinity for a few weeks, should be enumerated at the place where he lives.

The CHAIRMAN. If a boy, for instance, has left his family for the city, and expects to come back in the spring, he would be enumerated as of his home?

Mr. GRAY. Yes. The members of the family still remaining there, of course, would report for him.

The CHAIRMAN. Yes.

Mr. GRAY. Just one other point that was not perhaps brought out, with reference to the harvesting, and that is that in the Department of Agriculture we have not considered the yield data that we get from the census as extremely important. We lay very much more emphasis on acreage. Now, there are several crops, as was brought out by Mr. Olsen and others, that are not completely harvested by November 1, but for cotton we have the ginning reports annually, and they are probably more accurate for cotton than what you can get from an ordinary census enumeration, or at least they are as

accurate.

Furthermore, we get those ginning reports before the census figures will be available, and the census data on cotton production will really be old information by the time it comes out, so far as yields are concerned.

The acreage figures and the other economic information in the agricultural census are the lines of information which we really consider most important. I think that is all.

Senator FLETCHER. They are shipping citrus fruits all the time. I happen to know something about that, and have some reason for knowing a little about it. We do not begin shipping until November 1, and that is pretty early then, and we ship the last shipment along in February; but I suppose the census would want the crop for the previous year. They could not show the crop for the present year. Mr. STEUART. Do you want me to answer that? Senator FLETCHER. Yes.

Mr. STEUART. My tendency would be to postpone that until the latter part of November. Down in California they get the citrusfruit shipments still later.

Mr. GRAY. I am told that so far as the citrus fruits are concerned, the shipping reports are very accurate, and the department relies on those shipping records.

Senator FLETCHER. The trouble is you can not show the crop of 1929 by the 1st of November, because the crop has not begun to move yet. That is the point I am trying to get at; because if you say that the census is for the previous season's crop, that is all right; but you would have a pretty broad guess to make on the crop of 1929, beginning the 1st of November, because it has not started to move. I do not know how you would get it. Would you estimate what is on the

trees, or would you attempt to do that—to give the crop, for instance, of 1929 on the 1st of November?

Mr. STEUART. I say, in the first place, I would be inclined to postpone the canvassing until the latter part of November.

Senator FLETCHER. Then you would get an idea of what it was, but you still would not have an accurate estimate of what it was. Mr. STEUART. You would have to estimate what remains.

Senator FLETCHER. In Florida they begin the 1st of November and ship until June.

Senator TYSON. They do not have half of the cotton crop by the 1st of November.

Mr. STEUART. The ginning figures on cotton are much more accurate than the census figures.

Senator TYSON. But I say, you do not have half of it ginned then. The CHAIRMAN. Is there anything else? That will be all, then, so far as the Department of Agriculture is concerned. That will close the hearings on the census bill.

Mr. STEUART. The condition that has been spoken of, of people moving about, is greater now than ever before in our history, and we are giving that a special study down at the office, trying to devise a method to avoid duplicating people, and at the same time to get all the people who are not at their usual places of abode. Now, I can not say whether that work will result in a larger duplication or a less duplication. I think, if anything, they will about offset each other. That is what we are going to try to do; but you can see that it is getting worse and worse.

Senator JOHNSON. There is a great moving around of the population.

Mr. STEUART. Yes. One of the Members of Congress, in the Nassau district, just outside of New York, says: "There are numbers of people in my district that have fine residences there and who pay taxes and live there a part of the year but who go into Manhattan and live in an apartment house in November, and I do not want you to enumerate them in Manhattan when they belong in my district." He is perfectly right about that, and I am going to try to devise some method of getting them all back in the district where they belong.

Senator JOHNSON. In the first place, you ought to have a law to make everybody go into the places where they live and stay there during the taking of the census. [Laughter.] That is the way it would be done in Turkey, but I am afraid it would not work here. The CHAIRMAN. That will close the hearings on the census bill. Senator Vandenberg has introduced his apportionment bill and I think it would be well if he would make a statement now in regard to that bill.

STATEMENT OF HON. ARTHUR H. VANDENBERG, A SENATOR FROM THE STATE OF MICHIGAN

Senator VANDENBERG. There are copies here, Mr. Chairman, of the committee print of the bill I introduced.

The CHAIRMAN. It is quite a different bill in terms, and one that harmonizes very largely all the experts.

I think Senator Vandenberg has done a wonderful work in getting these experts together.

Senator VANDENBERG. Thanks to Doctor Hill and Doctor Steuart. The CHAIRMAN. The bill will be found printed at the beginning of this hearing. It has no number now, but it will be printed regularly to-morrow.

Senator VANDENBERG. This bill is in the precise language which the conference committee reported favorably last January, with three exceptions. The first two exceptions are minor. The first exception is in the title. The former bill claimed for itself that it was a bill for the reapportionment of Representatives in Congress. I think it was justly criticized on the ground that it did not actually provide an apportionment, but merely provided the ultimate lines for an apportionment. Therefore, in order to more accurately identify this measure, it is called "A bill to provide for apportionments of Representatives in Congress."

The second change is in line 5, page 1. The former bill called upon the Secretary of Commerce to transmit such census arithmetic and apportionment deductions, to the Congress. This bill calls for the substitution of the President as the reporting officer, the idea being that since this undertakes to be a permanent enabling act rather than a specific apportionment, that its purposes would be better served by identifying a constitutional officer instead of a mere statutory officer, for the purpose of this ministerial act.

The third and last change is in lines 3 to 7 on page 2. The former bill called for this census arithmetic on the basis of 435 Representatives, apportioned according to the method of major fractions. This bill, instead, identifies the existing number of Representatives at any given time, instead of a specific number, and it identifies the method used in the last preceding apportionment, instead of the method of major fractions.

It is quite true that the last apportionment-if you can remember back that far-was on the basis of fractions.

Senator LA FOLLETTE. I have a slight memory bias on that. Senator VANDENBERG. We will say, as a minority of the Senate. The 1910 apportionment was on the basis of fractions, and that' called for a House of 435 members. Therefore, if Congress failed to make an independent apportionment in 1930, a thing which it is invited to do and expected to do under the terms of this bill, if it sees fit-if it does not make that actual approtionment in 1930, within the automatic terms of this bill they revert to a House of 435 Members, and the method of major fractions, which would be precisely the result which would have been obtained under the so-called Fenn bill, which we recommended to the Senate in January. But if Congress in its wisdom, in 1930, in making a new apportionment, should decide that it wanted to make a House of 440 Members, or if it decided that it favored the method of equal proportions, then this bill would accommodate itself to that decision in 1940. In other words, we have taken all the identification of controversial detail, out of this bill, and we have written it now so that it can accommodate itself to the decision of Congress in relation to these particular things, so that all that is left is a general enabling act which takes nothing from Congress except its right of anti-Constitution inertia. That is all it takes from it.