chasing clause of the Sherman Act; the discontinuance of government purchases was made up for by an increase of the exports of silver. The ratio of industrial consumption to the total domestic consumption may in consequence appear enormously increased, viz., from 9 per cent. to 44 per cent., whereas in reality it was stationary, averaging $6,000,000 annually throughout each period. If we compare the amount consumed in the arts with the amount coined from new material or from the stock of bullion accumulated through purchase, we obtain the following

This table shows that since the demonetization of silver the ratio between industrial consumption and normal coinage has increased approximately from 13 to 12, owing to the reduction of the amount annually coined. Still, now, as before, the demand for silver wanted in the monetary circulation of the United States exceeds the demand for industrial purposes.

IX.

The Bureau of the Mint has made repeated efforts to ascertain the consumption of the precious metals by direct inquiries addressed to the consumers. The subject has been referred to in the previous article on gold. The most complete data were those collected in 1899. As will presently be noticed, these data are greatly at variance with the preceding calculations. Of course, figures relating to one year are not comparable with annual averages for a number of years; their variance therefore in no way impeaches their reliability. Still less satisfactory are the results of a comparison of the aggregate returns from consumers for the year 1899 with the bureau's estimate for the same year. If the preceding calculation is made for the year 1899,

there is hardly anything left for consumption in the arts. It must be borne in mind, however, that both the consumption and the exports of one year may contain large quantities of bullion smelted or imported during the previous year.

With these qualifications the data on consumption in 1899 are given below:

If it is assumed that the mint bars sold by private refineries to the trade in 1899 were all bought by them during the same year, and are consequently included in the official total of the mints and assay offices, then the small shortage of 7 per cent. is easily explained by the fact that out of a total number of 43,050 inquiries sent and actually delivered to the parties addressed, 13,102, or 30 per cent., remained unanswered.

The excess of 9 per cent. in the reported amount of purchases of private refinery bars over the amount of sales may likewise be due to the fact that out of eighty-one firms addressed by the Bureau of the Mint twenty-one failed to respond. On the other hand, it is probable that a part of the reported value of mint bars sold by private refineries in 1899 was from stock previously purchased; it is further possible that through misinterpretation of the interrogatories some of the bars bought from private refineries, though bearing a government stamp, were reported by consumers as "private refinery bars."

The total industrial consumption reported by purchasers amounted to $7,309,596, exclusive of nuggets bought in the mining districts directly from miners, to the amout of $204,216; as this amount is, however, not included in the estimated silver production of the United States, it is not within our comparisons. The sales of mint and private refinery bars aggregated between

=

$7,229,291 and $8,248,260; the greater amount includes the value of mint bars sold by private refineries, and, in part at least, duplicated in the total of the sales from mints and assay offices. The discrepancy between total sales and total purchases is between $80,305 and +$938,664, i. e., between I per cent. and +13 per cent. Not all of this silver, however, was produced in 1899; of the total of $7,229,290 only $5,961,320 was new silver, i. e., 83 per cent. At this proportion the maximum consumed in the arts during the year amounted to $6,800,000. These are commercial values; at the average bullion value of the dollar 46.5 cents in 1899, they correspond to an industrial consumption of from $13,000,000 to $15,000,000, which is more than double the average annual consumption calculated above for 1889-93 and 1894-1900. The net exports of silver ore and bullion in 1899 amounted to $23,000,000 (commercial value) out of an annual production of $33,000,000 (commercial value), leaving $10,000,000 for domestic consumption. The ratio of industrial consumption, as reported by consumers, to this balance was not less than 59.6 per cent., whereas, according to the preceding calculations, it averaged at most 44 per cent. The variance, if real, would point to an underestimate of the silver production of 1899, amounting to from $7,000,000 to $9,000,000 (coining value), i. e., from 10 per cent. to 13 per cent. of the estimated amount. It is more probable, however, that the variance is due to a deviation of the figures for 1899 from the average computed for a number of years. It merely indicates that returns for one year, though accurate and laboriously collected, are valueless for statistical purposes.

Still, even if we take $6,800,000 as the value of silver consumed in the arts in 1899, which contains an obvious duplication, the maximum coefficient of industrial consumption of silver in the United States will be 68 per cent., which is far above the maximum coefficient of the world's consumption. This is another proof that in the United States, as well as in the rest of the world, over one-third, and probably as much as one-half, of the total consumption of silver is required to keep pace with the growing volume of circulation.

X.

As has been demonstrated, the coinage of silver still remains a very important, if not the principal, source of the demand for the white metal. Some idea of the relation between supply and demand in the silver market may be gathered from the following table, where the world's production is collated with the world's coinage for the twenty-eight-year period 1873-1900 and with the annual average prices represented by the bullion value of

$3,121,000,000 $2,864,153,902 $4,035,000,000 $3,703,148,492

We observe that up to and including 1879 the demand for silver from the mints exceeded the annual output of the mines; in 1880 the amount of the coinage is for the first time exceeded by the output of silver; the relative volumes of production and coinage oscillate in both directions until the year 1888, since

which the excess of the output over the coinage becomes a permanent feature. Throughout this period the value of silver exhibits a falling tendency.

Strong efforts were made in the past by the United States to change the relation between supply and demand by large purchases of silver. To study the effects of this policy annual figures are inadequate, being subject to violent fluctuations. For the purposes of comparison we eliminate the year 1873, when the free coinage of silver was discontinued by law, and divide the twenty-seven-year period, 1874-1900, into four parts: (1) 1874-77, from the demonetization of silver to the passage of the Bland Act; (2) 1878-90, i. e., to the passage of the Sherman Act; (3) 1891-93, from the passage of that act to the repeal of its purchasing clause; (4) 1894–1900, i. e., since the repeal of the purchasing clause and the closure of the mints in India. Only full years are considered. The figures of the Bureau of the Mint relate to fiscal years. They can, however, be easily reduced to calendar years. The government purchases of silver during the fiscal year 1878 fall entirely within the same calendar year, inasmuch as the Bland Act was passed in February. The purchases under the Bland Act for the fiscal year 1891 were all made from July 14 to August 13, 1890. From the date last mentioned to January 1, 1891, there were purchased by the government 21,300,141 ounces of silver. And, lastly, all purchases for the fiscal year 1894 were made prior to November 1, 1893. Reducing, accordingly, fiscal years to calendar years, and taking silver at its coining value ($1.2929 per ounce fine), we obtain the following table: