As a consequence of these counteracting influences, it appears that the share of the landowners has not increased disproportionately. The most comprehensive endeavour yet made to determine the growth and relative size of the different shares of the national product is embodied in Professor W. I. King's volume, "The Wealth and Income of the People of the United States," published in 1915. It estimates that the total annual income of the nation increased from a little less than two and one-fourth billions of dollars in 1850 to a little more than thirty and one-half billions in 1910, or slightly more than fifteen times. During the same period rent, the share of the landowners, advanced from $170,600,000 to $2,673,900,000, or about fifteen and three quarter times. In the year 1910, therefore, the landowners were receiving but a very small fraction more of the national product than their predecessors obtained sixty years earlier.[61] As to the relative size of the shares going to the different factors in 1910, the figures are even more remarkable. Wages and salaries absorbed 46.9 per cent.; profits, 27.5 per cent.; interest, 16.8 per cent.; and rent, only 8.8 per cent.[62] This was exactly the same per cent. that the landowners received in 1860. To be sure, these figures are only approximations, but they are probably the most reliable that can be obtained from our notoriously incomplete statistics, and they will deserve respectful consideration until they have been refuted by specific criticism and argument. In the opinion of their compiler: "The figures for wages and salaries are believed to be fairly accurate; those for rent are thought to have an error of not more than twenty per cent. The separation of the share of capital from that of the entrepreneur is very crudely done and no stress should be laid on the results. The total for all shares is thought to be more accurate than the mode of distribution, and for the last three census years should come within ten per cent. of the correct statement of the national income. For earlier years the error should not be over twenty per cent. at the outside."[63] If we make the maximum allowance for error in reference to the share of the landowner, and assume that the rent estimate is twenty per cent. too low, we find that it was still only ten and one-half per cent. of the total product in 1910, which represents an increase of less than three per cent. since 1850. It is significant that Dr. Howe, who has no bias toward belittling the share of the landowner, suggested as his minimum and maximum estimates of the land values of the country in 1910 figures which are respectively fifty per cent. below and only five per cent. above the amount taken by Professor King as the basis for his estimate of rent.[64] There is, consequently, a strong presumption that Professor King is right when he stigmatises as "absurd" the contention of the Single Taxer, "that