Page:Popular Science Monthly Volume 48.djvu/258

246 they increased it is hard to say, for between 1803 and the time the census was taken in 1810 a large number of our people moved into the new territory, so that the population of the new territory for 1810 gives more than the number we ought to deduct. It is certain we ought to deduct something, and equally certain that we shall never know the exact amount. The best authorities, however, seem to indicate 20,000 as about the proper number. For Florida it is about 12,000, and for Texas, New Mexico, and California, which came to us by the Mexican War, it is about 90,000, to be deducted from the total whites as given by the census of 1850. We shall also have to make a deduction from the whites in the census of 1860, because part of the returns for California in the census of 1850 were burned, and the natives of that Commonwealth were not all given in the census of 1850, but appeared in the census of 1860. A deduction of about 70,000 will probably account for all of them.

The census of 1870 is now generally believed to have been an underestimate, owing principally to the difficulty of obtaining returns from the South so soon after the war. The rate of increase for that decade ought therefore to be a little more than 23·37, probably about 25·37; and this would lower the percentage of the next decade to about 29·05, instead of 31·05.

Following down the column of native increase, we find that from 1750 the rate remains at a little over 33 per cent for twenty years, until reduced by the Revolution to 28·81. But after the Revolution it returns again to 33·33 in the next decade, then rises to 34·14, and then to 34·79. In the next decade, 1810 to 1820, it fails suddenly about one per cent, and in the next falls one per cent again; and in the next, which is 1830 to 1840, falls more than two per cent to 30·64, which is much lower than it had been at any time in the previous eighty years, except during the decade