nent census office until March 19, 1896, when, by a joint resolution relating to the Federal census, the Commissioner of Labor was directed to report to the Congress, for its consideration, as soon as practicable, a plan for a permanent census service.

Following the passage of this joint resolution, a memorial (a) was presented to Congress by a joint committee appointed by the American Economic Association and the American Statistical Association to consider the question of the organization of the twelfth census, in which the attention of Congress was respectfully called to the importance of establishing at once a permanent and independent census office. The memorialists state that in so doing they are actuated by an earnest desire for the scientific development of statistics in the United States; that they represent learned societies, whose members come, through their professional duties, in constant, almost daily, contact with the work of the census; that they represent the point of view of those who use statistics, and that it is in the general interest that they plead for methods of census administration which would tend to increase its efficiency and heighten the value of its results. And in this connection the memorialists further

say:

In many departments of statistical work the publications of the United States Government occupy an honorable place. Through the liberal provision which has been made in the past for statistical inquiry, the United States has been able to contribute substantially to the development of statistical methods and to the extension of statistical research. But the work of the Government is uneven, and sometimes fails to reach the highest standard. This is true of the census, the largest statistical undertaking of the Government, upon which money and effort are so generously lavished. There can not be a moment's doubt that this work should be brought to the highest possible standard of statistical excellence.

It is no reproach to the census as a whole, nor to the gentlemen who have administered the census office, to say that in many respects the census reports are unsatisfactory to us as students of statistics and to the people of the United States.

We allude particularly to the attempt to cover too much ground, the enormous cost of the undertaking, and the delay in the publication of the completed tabular results. These criticisms all grow out of the legislation under which the census is taken. The defects of our present temporary organization may be summarized under the following heads: "

1. Accumulation of inquiries at the same period of time.

2. The lack of continuity in census work.

3. The haste with which the whole machinery of the census is placed in motion.

The permanent and independent census furnishes the best guaranties for improvement in statistical work, if established under proper conditions. But it would be a grave misfortune to postpone the organization of such a bureau until shortly before the time for taking

a Senate Doc., Fifty-fourth Congress, second session, No. 68.

the next census. We can not urge too strongly that consideration be given at an early date to this question. If, as it should, the census of 1900 is to be an advance over those of preceding years, thorough preparation for the work is indispensable. We are convinced that the gravest difficulty which has hitherto impeded the work of the census is the haste with which it has been planned and executed.

It is not alone in the interest of statistical science, but in the interests of the public, which has an undoubted right to the most accurate and prompt information which the census office can furnish, that we urge the adoption of a measure which will attain this end and mark distinct progress in the statistical work of the Government.

In accordance with the provisions of the joint resolution of March 19, 1896, the Commissioner of Labor, under date of December 7, 1896, submitted a report (a) on a plan for a permanent census service, in the form of a tentative bill providing simply for an organic administrative act, by which an independent census office was to be established, leaving the details of the twelfth and subsequent censuses to the officers, respectively, having them in charge. This report was referred to the Senate Committee on the Census, and on January 9, 1897, an informal hearing was had before the Senate Committee on the Census and members of the Committees on Appropriations of the Senate and House of Representatives, before which the Commissioner of Labor appeared, at the request of the chairman of the Census Committee. The bill practically as submitted by the Commissioner of Labor was introduced in the House January 14, 1897, by Mr. Sayers, of the Committee on Appropriations, to which committee it was referred, and in the Senate January 18, 1897, by Senator Chandler, the chairman of the Census Committee. Similar bills providing for a permanent census service, but charging the work upon the Department of Labor, were introduced later in the same month in the House and Senate by Mr. Sayers and Senator Chandler, respectively, and referred in the House to the Committee on Appropriations and in the Senate to the Committee on the Census. No result was reached, however, by either the House or Senate, and the matter of census legislation went over to the Fifty-fifth Congress.

TWELFTH CENSUS LEGISLATION.

At the second session of the Fifty-fifth Congress a general bill providing for taking the twelfth and subsequent censuses, under the direction of the Secretary of the Interior, as heretofore, but limiting the work of the decennial enumeration to four subjects, was passed June 16, 1898, by the Senate, and this bill was later presented in the House and referred to a select committee on the twelfth census. At the third session of the same Congress a substitute bill was passed by the House February 6, 1899, calling for the establishment of a census office

a Senate Doc., Fifty-fourth Congress, second session, No. 5.

entirely independent of any department, but with the same limitation of the decennial work as provided for by the bill which passed the Senate at the close of the second session. In the bill which finally passed both Houses of Congress, as agreed upon in conference, and which was approved by the President March 3, 1899, a census office is established in the Department of the Interior, the chief officer of which is denominated the Director of the Census (instead of Superintendent of Census, as heretofore), to be appointed by the President and confirmed by the Senate. Provision is also made for the appointment by the President of an Assistant Director of the Census, who shall be a practical, experienced statistician, and for the appointment by the Director of the Census of five chief statisticians, who shall be persons of known and tried experience in statistical work; of a geographer, chief clerk, disbursing clerk, five expert chiefs of division, and the various clerks and employees essential to the prosecution of the census work. Under the provisions of this act, the census office was organized by the appointment by the President of Hon. William R. Merriam as Director of the Census, March 3, 1899, and of Dr. Frederick H. Wines as Assistant Director of the Census, March 5, 1899. The statistical organization of the office was later completed by the appointment by the Director of the following-named persons: William C. Hunt, chief statistician, population; William A. King, chief statistician, vital statistics; S. N. D. North, chief statistician, manufactures; Le Grand Powers, chief statistician, agriculture; Walter F. Willcox, chief statistician, methods and results; and Henry Gannett, geographer. By the act of March 3, 1899, the Director of the Census is given the entire direction and control of the work, including the appointment of the statisticians, clerks, and other employees of the census office, as above stated, but the act provides that nothing therein contained shall be construed to establish a census bureau permanent beyond the twelfth census. The decennial work is limited to inquiries relating to population, mortality, agriculture, and manufactures, but provision is made after the completion of the decennial work for the collection of statistics relating to various special subjects, such as the insane, feeble-minded, deaf, dumb, and blind; crime, pauperism, and benevolence; deaths and births in registration areas; social statistics of cities; public indebtedness, valuation, taxation, and expenditures; religious bodies; electric light and power, telegraph and telephone business; transportation by water, express business, and street railways; and mines, mining, and minerals.

GROWTH OF CENSUS INQUIRIES.

In the historical review of the Federal census it has been shown that there was constant inquiry concerning the population at the first six enumerations and that this inquiry was supplemented by attempts to

secure additional data concerning manufactures in 1810 and 1820, and concerning mining, commerce, agriculture, manufactures, fisheries, and schools in 1840. The inquiries relating to these subjects were made, however, on schedules prescribed, in most cases, by the several census acts, in which the detail required was expressly stipulated and beyond which no tabulation or combinations of facts were possible, owing, as already explained, to the manner in which the returns were required to be made. But, beginning with the census of 1850, an individual return was required, that is, the detailed enumeration was made with respect to each living inhabitant, each decedent, each farm, and each establishment of productive industry, etc., and in presenting a summary of the census inquiries, showing their inception and growth from the beginning to the present time, the details called for at the enumerations prior to 1850 have been brought to the basis of the modern idea of census enumeration, as will be more fully explained in connection with the analysis of the inquiries relating to each of the various subjects considered.

Although a material modification and extension of the census inquiries was had at the censuses of 1850, 1860, and 1870, taken under the same general provisions of law, the great, and what is fair to be termed the extraordinary, increase in the scope of the census did not come until 1880 and 1890, when, instead of the census being limited to but five or six schedules, comprising only about 150 details, there were at each of the last two enumerations more than 200 general and special schedules, relating to very many subjects and comprehending several thousand inquiries or details.

At the first nine enumerations, those from 1790 to 1870, inclusive, there were, in all, 26 schedules of inquiry, the number of schedules used at each census, classified according to the subjects to which they related, being shown by the following statement:

By the terms of the acts of March 3, 1879, and of March 1, 1889, which governed, respectively, the work of the censuses of 1880 and

1890, the number of subjects of investigation was greatly increased, and in addition to the five general schedules prescribed by section 2206 of the Revised Statutes of 1878, there were very many special schedules prepared and used at each of these two censuses, as shown by the following summary:

a Includes insanity, feeble-mindedness, deafness, blindness, physical disabilities, etc.
b Includes newspapers and periodicals.

The very large number of schedules used at these two censuses, as compared with the preceding enumerations, was due, of course, to the great increase in the scope of the census, calling for an investigation at the decennial period of very many subjects or special topics which had not previously been touched by the census inquiries or for which very limited data had been secured. Following the more comprehensive plan of census investigation in 1880 and 1890, supplemental and special schedules were used, wherever deemed necessary, by which the inquiries were specialized, but in which there were, as a matter of course, very many inquiries or details that were common to some or all of the schedules relating to each particular subject. In attempting to arrive at, approximately, the number of different inquiries or details asked concerning each subject of inquiry, or the total number for all the subjects combined, every effort was made to conform the work, as far as possible, to a fixed rule or plan, although realizing fully the difficulties arising in such a work, covering so many different subjects, for which the schedules of inquiry have been prepared by very many persons, and in accordance with various ideas or bases. In making the effort to determine, however, for the purposes of this article, the number of different inquiries or details called for concerning each topic of investigation at each of the various censuses from 1790 to 1890,