In the first place, the officer designated to take the census is, in every point of view, objectionable. That officer is the United States marshal, originally selected, probably, for no better reason than that, as there was such an officer in every State whose services could be made available, it was better to use him than to create a new office. But neither the legitimate duties of his office nor the department to which he belongs justify such a selection. His duties are chiefly connected with violations of law, and he is necessarily associated in public opinion with the criminal side of life. A police-officer is not a good census-taker. Moreover, many of the States are divided into several marshalships from considerations which do not at all enter into the taking of the census. Thus, New York has three districts, the largest of which contains more than two and a quarter millions of inhabitants, while Florida has two districts, the smaller of which, but by far the more important so far as the legitimate duties of the marshal are concerned, contains scarcely six thousand inhabitants. Massachusetts is a district with over a million and a third of people: so is Arizona, with less than ten thousand.

Then the methods of payment are unfair, irrational and cumbersome. They bear no relation to the amount of work performed, are irregular in their operation, are obscure in their manner of calculation, and impose needless labor alike on the officer to be paid and the census office. To say that the square root of an area multiplied by the square root of the number of horses indicates the number of miles travelled in taking a census is as absurd as to say that the square root of the yards of cloth in a suit multiplied by the square root of the number of stitches taken to make the suit will give the length of the thread used. In its practical working in 1860 the result was to give to one assistant marshal a per diem of $1.66 and to another $31.32 for the same labor. A proposition which works out such a result may serve for a joke in negro minstrelsy: it will hardly be accepted as honest figuring by the recipient of the minimum pay.

Lippincott's Magazine of Popular Literature and Science, Volume 22. July, 1878. — Part 127

But the greatest objection of all is to the schedules created by the law of 1850. The number of inquiries is limited by that law to one hundred, though why that number should be selected as the limit, except at haphazard, is a mystery. It is purely arbitrary, and in its practical working is mischievous. Statistical inquiries ought to be exhaustive, whether the questions asked are ten or ten thousand. To limit the number to one hundred requires the lumping together of incongruous facts or the entire omission of some of prime importance. Of what real value is the answer to the question, "Kind of motive-power?" in relation to manufactures unless other details are given? Yet only such questions can be asked where the margin is so narrow. In the census of Massachusetts for 1875, 304 inquiries were made, embracing 1337 topics; and so satisfactorily was the work done that out of a population of 1,651,912 only 43 persons were unaccounted for when the statistics of occupations were compiled; while in the United States census of 1870 the number thus unaccounted for exceeded 1,000,000. In Rhode Island no less than 561 inquiries were made in the census of 1875, and the result is the most complete census--not merely of persons, but of every kind of manufacture and production--yet taken in any State. The returns of cotton, woollen and iron manufactures show what can and ought to be done in that direction for the whole nation in 1880. They answer the requirements set forth by the superintendent of the census of 1870 by presenting "tables so full of technical information as to become the handbook of manufacturers."

Lippincott's Magazine of Popular Literature and Science, Volume 22. July, 1878. — Part 127