The presence of starch glucose in sugar-house molasses may be easily detected by the optical saccharometer when the sirup has the usual density of about 40 deg. B., and when starch sugar has been added in the usual quantities. For making the test the usual weight should be taken (16.35 grammes for Duboscq's saccharometer, and 26.048 grammes for Ventzke's instrument). The direct test should show a percentage of sugar not higher than the number of Baume degrees indicating the density, and it may be from 2 to 3 per cent. lower. To understand this, we must refer to the composition of cane-sugar molasses of 40 deg. B.: Sugar.......................................37.5 Insoluble impurities........................37.5 Water.......................................25 If the direct test should indicate 55 per cent. of sugar, and if the molasses were straight, the composition would be-- Sugar...........................................55 Soluble impurities..............................20 Water...........................................25 Now, a product of this composition would not be a clear sirup at 40 deg. B., but a mixture of sirup and crystals. Therefore, if the product is a clear sirup at 40 deg. B., and it tests 55 per cent., it cannot be _straight_. The presence of starch glucose in sugar-house molasses may also be detected by the copper test. The possibility of applying this test, as well as those already indicated, rests on the fact that starch glucose is always added in very large quantities for the purposes of adulteration. A very small addition could not be satisfactorily detected. The detection by the copper test rests on the observation that very nearly one-half of the soluble impurities in sugar-house molasses consists of glucose in the shape of inverted sugar. We have seen above that for a molasses of 40 deg. B. the soluble impurities amount to about 371/2 per cent. We may, then, lay down the rule: that the percentage of glucose shown by the copper test cannot, in a straight sugar-house molasses, be much greater than one-half of the number expressing the density in Baume degrees. The reason is obvious from what has been said of the test by the optical saccharometer. * * * * * FALSE VERMILION.--A curious case has been noticed in Germany, where a small cargo of vermilion was purchased, and, upon being analyzed, turned out to be red oxide of lead colored by eosine. This is an entirely novel sophistication. The eosine was separated from the oxide of lead by digesting the product for twenty-four hours in very strong alcohol. A much shorter time is sufficient to color the spirit enough to enable an expert chemist to detect the presence of this splendid organic coloring matter. Another kind of "vermilion" consists entirely of peroxide of iron, prepared especially to imitate the brilliant and costly sulphide of mercury, which it does very well, and is largely used in England, France, and America.