be this much too low. This error is, however, small, as will be seen from the following calculation.

Suppose that the distance of the piston were always kept 0.2 inches above the level of the acid, what would the error amount to at different heights and different gravities of acid if the instrument had the dimensions previously given? The volume of v in this instance was found to be 0.232566 cub. inches, easily deducted from the weight of water, filling 0-2 inches lineal of A B, and the whole of the capillary tube C D.

Supposing the acid in the chamber be 1.5 sp. gr., and its height 1 inch :

Now we know from the determination of the volume of v that

0.125226 cub. inches occupy 0·2 inch lineal of A B, hence 0.125226: 0·2 = 0·000831: x.

x = 0·00133 inch, representing the number which expresses in lineal inches the increase of v on the scale of our instrument, if it is withdrawn from a depth of 1 inch of acid at 15 sp. gr. This number, which represents the error, we will call e.

In like manner are obtained the values of e in the annexed table at the specified heights and gravities :—

We learn from this the ratio in which e increases within limits seldom exceeded in practice. Thus we have

being the average expansion-coefficient for v for every inch from 1 to 6 inches.

In the same way are found the expansion-coefficients of the following table :—

Supposing we had read off 5:4 inches as the height of the acid, which was found to have a sp. gr. of about 1·6, we should have to make the correction by adding 5.4 × 0.00154, or 0.008316, to the 5.4 inches observed, as being the real height of the acid inside the chamber. The error consequently is so small that for low levels it may be altogether neglected, as well as the slight error arising from the loss of the small portion of acid which influences the average specific gravity.

Having described the nature of the instrument, I may now briefly explain how to use it. After having wetted the piston, which is necessary for its easy motion, it is pushed into the tube to a position high enough not to touch the level of the acid of which a sample is about to be taken. The pinch-cock is now removed, to open C D, and the instrument slowly lowered into the acid till it touches the bottom. After C D is closed again with the pinch-cock, the instrument has to be raised, and the level reached by the acid read off on the scale. On again opening the pinch-cock the acid will run out. This sample has to be rejected, having merely been drawn to ascertain approximatively the height of the level, in order to adjust the piston to its proper place, i.e., 0.2 inches above the level. Before the next sample is drawn it will be well to blow through the piston-tube, in order to remove any portion of liquid which might have obstructed the passage. All the rest of the operation is then repeated as before, only with more care, especially in lowering the instrument with a slow, uniform, and perpendicular motion. After the acid adhering to the outside of AB has drained off, and the height of the level been read (the eye being carefully placed in a line with the surface), the pinch-cock is opened, the acid collected in a glass cylinder, and after mixture tested in the usual manner. When the acid in the chamber stands at a low level, it might be found troublesome to have to repeat the drawing of samples until a quantity of acid

is obtained, in which hydrometers of the usual size can float. This objection is easily met by employing smaller ones made on purpose.

Being now in possession of the means of ascertaining with accuracy the height and average specific gravity of the horizontal section of any liquid,* I may add a few words on the horizontal sections of the acid accumulating on the bottom of the chambers.

For reasons mentioned before, the sample has actually to be drawn from the inside of the chamber, which is generally effected by making a cavity of suitable dimensions in the side of the chamber, placed a foot or so above its bottom. A hole is provided in the bottom of this cavity for access to the acid underneath. If the bottom of the chamber be parallel with the surface of the acid, one determination is sufficient to fix the absolute weight of the whole. It will often happen, however, that, though the chambers were built level at first, they will in time become otherwise. If that is the case, one determination is insufficient, and the number and places of these determinations will depend upon the geometrical figure assumed by the vertical section of the acid. If the figure of this vertical section became too irregular, the task of fixing the accurate weight in this manner would have to be given up.

The heaviest acid will invariably gravitate to the lowest point in the chamber. Equal altitudes have equal gravities, and if samples drawn with the above instrument from the same chamber have different gravities, it is caused by irregularities in the level of the bottom.

Although the importance which is attached to these determinations lies less in the knowledge of the absolute weight of acid present at the time in the chamber, than in that of the exact quantity of acid which has been formed during a certain period, there is no possibility of knowing the one without the other. By fixing the level to-day and at the same point to-morrow, we learn from the altered height the volume formed in the interval, but not its weight, or at least only conditionally, that is, only under the assumption that the determinable part of acid, x, has

The intimate mixture of large quantities of liquids demands considerable time and labour. As the average specific gravity after mixture is generally taken as the guide for fixing the desired proportion between two liquids to be mixed, numerous experimental mixtures and tests of the whole bulk may be avoided by the taking of samples with the above instrument.

the same specific gravity as the indeterminable part, y, and such might seldom be the case. If therefore y be of a higher specific gravity than x, x will become heavier at the expense of y, and we assume a greater yield of acid than we have actually made during the period in question. If y is weaker, the reverse takes place. The intervals between these determinations are generally so long, that the acid has to be removed from the chambers. In its further treatment it undergoes such a change that it becomes a homogeneous liquid, or a well-defined sulphate, in which state it is weighed. By this means all the errors arising from the heterogeneity of the acid are the more diminished the larger the proportion of homogeneous to heterogeneous acid. In other words, the longer those periods are, the nearer the calculation will come to the truth; on the other hand, the shorter they are, the more particular have we to be in all details which fix the weight.

XLIII. On some Products derived from Indigo-Blue.

By EDWARD SCHUNCK, Ph.D., F.R.S.

[Abstract, from the third volume of the third series of Memoirs of the Literary and Philosophical Society of Manchester. Session 1864-5.]

My experiments on the formation of indigo-blue, an account of which I had the honour of presenting to this Society several years ago, led me to make some inquiries regarding the processes employed in tropical countries for the production of indigo from the various plants yielding that dye-stuff. I found that all the authors who have written on the subject agree in affirming that the process of fermentation, which is the one usually adopted for the purpose of extracting the colour from the plant, requires to be conducted with the greatest care, in order to yield a successful result. Unless certain precautions are adopted, a product of very inferior quality will be obtained; in some cases, indeed, the colouring matter is entirely lost. This will not be surprising to any one who considers that though indigo-blue, when once formed, is a very stable compound, the substance existing in the cells of the

plant from which it originates, and which I have named indican, is decomposed with the greatest facility in various ways; that indigo-blue is only one of its products of decomposition, and may be formed or not, according to the nature of the process to which it is submitted. With this sufficiently obvious explanation I should have been inclined to rest contented, had I not acquired a knowledge of some other facts relating to indigo-blue, to which the same explanation cannot be applied, but which evidently belong to the same class.

It is well known to those dyers who employ the so-called woadvat, in which the reduction of the indigo-blue is effected by the action of various organic matters, such as woad, madder, and bran, together with lime, that if the process be not carefully managed it may change its character entirely, the contents of the vat entering into a state of complete putrefaction-a change which results in the total destruction, or at least disappearance, of the colouring matter. Now this phenomenon, the reality of which cannot be doubted, though its nature has never been subjected to scientific scrutiny, cannot be explained in accordance with what is at present known regarding indigo-blue, which is considered by chemists to be a body of such a stable character as not to be decomposed by any except very potent agents, such as chlorine, bromine, and nitric acid. In no work on scientific chemistry is it stated that indigo-blue may be decomposed by any process of fermentation or putrefaction, in the same way as sugar or albumen.

In my experiments on indigo-blue I have generally employed for its reduction and purification the process of Fritzsche, which consists in acting on it with a mixture of alcohol, grape-sugar, and caustic soda. The colouring matter dissolves when the mixture is heated, and is again deposited on exposure to the atmosphere in crystalline needles. Now in performing this operation with very small quantities of indigo-blue and an excess of alcohol and grapesugar, I found that the colouring matter did not make its appearance again on agitating the solution with air. The yellow colour of the liquid passed as usual through red to green; but, instead of the indigo-blue being precipitated, the whole became yellow or brownish-yellow, and the colouring matter disappeared entirely. In this way I had the mortification of losing a quantity of indigoblue, which I had prepared with much labour from human urine, though the loss resulted, as it afterwards turned out, in some gain of information.