the size of the drop is partially recovered. There is a stage of dilution when the specific gravity is 1.0680, where the drop-size is a minimum. Further, it is seen from column 5 that the quantity of nitre in a drop increases continually as the strength of the solution increases, although both the weight and the volume of the drop vary.

Inversely, the regularity of the variation of drop-size, in the case of nitre, points to the absence of hydrates of that body.

It would be delusive to endeavour to construct a formula connecting the specific gravity with the drop-size or drop-weight of the solution; but, as before, a graphic representation serves to show the connexion between the variables. In curve D, fig. 2, the abscissæ represent the quantity of nitrate of potash in solution, the ordinates show the corresponding dropsizes. As with chloride of calcium, it is seen that the drop-size of water is larger than that of any solution of nitre. Curve E, fig. 2, having the

same abscissæ as D, has ordinates which represent the drop-weights.

It is confessedly a matter of great interest, and still greater difficulty, to determine exactly the relation which exists between a dissolved solid and its solvent-that is, to find out whether or when a solid should be viewed as being in combination with a portion of the liquid in which it is dissolved. Such questions may perhaps receive additional light from experiments similar to the above, but more extensive, and performed with this special object in view. Comparing the curves C and D, for instance, there can be little doubt that the secondary maxima and minima of C are owing to the existence of hydrates of chloride of calcium in solution. The only known hydrates of chloride of calcium are Ca Cl, 2 HO and Ca Cl, 6 HO, the latter of which contains 50.7 per cent. of Ca Cl. Solution S contains about 42.5 per cent. It is noteworthy that, while the six-water chloride in the solid state absorbs heat on solution, the solution S evolves heat on dilution, as already mentioned. In the case of nitre we have in the drop-sizes evidence only of the opposite efforts of two cohesions, that of the water and that of the nitre. By pursuing this direction of experimental inquiry, evidence may probably be got concerning the truth of Berthollet's hypothesis of reciprocal recomposition in the case of the mixture of the solutions of two salts, AX and BY, where AY and BX are also soluble in water.

III. "On Drops."-Part II. By FREDERICK GUTHRIE, Esq., Professor of Chemistry and Physics at the Royal College, Mauritius. Communicated by Professor STOKES, Sec. R.S. Received October 17, 1864.

We have next to consider the influence which variation in the chemical nature of the drop-forming liquid may exercise upon the drop-size in the case SLG.

The liquids which were selected for this purpose were chosen as being

VOL. XIII.

2 M

typical of extensive classes, rather than as being connected with one another in immediate chemical relation. They were

These several liquids were allowed to drop under the same conditions, from the bottom of a hemispherical platinum cup. The arrangement of the apparatus was quite similar to that described in Part I., the ivory ball being replaced by the platinum cup, and the overflow of the cup being deter mined by strips of paper bent over its edge. The case of mercury is the only one which requires some explanation. A few years ago I noticed the fact that mercury which holds even a very little sodium in solution has the power of "wetting" platinum in a very remarkable manner. The appearance of the platinum is quite similar to that presented by amalgamable metals in contact with mercury. But the platinum is in no wise attacked. Further, the amalgam may be washed off by clean mercury, and the latter will also continue to adhere equally closely to the platinum. All the phenomena of capillarity are presented between the two. surface of the mercury in a platinum cup so prepared is quite concave; and a basin of mercury may be emptied if a few strips of similarly prepared platinum foil be laid over its edge-just as a basin of water may be emptied by strips of paper or cloth, and under the same condition, namely that the external limb of such capillary siphon be longer than the internal one.

The

I generally use this curious property of sodium-amalgam for cleaning platinum vessels. It enables us now to examine the size of drops of mer cury under conditions similar to those which obtain in the case of other liquids*. After the cup had been used for the other liquids, its surface

In regard to the above-mentioned property of sodium, the following observations may be of interest. At first the explanation naturally suggests itself, that the effect wrought by the sodium may be due to an absorption of oxygen, in consequence of the oxidation of the sodium, the consequent diminution of the gaseous film between the two metals, and the resulting excess in the superior pressure of the air. This, however, cannot be the true explanation, because it is found that the perfect contact between the two, or "wetting," takes place equally well in an atmosphere of nitrogen, carbonic acid, or in vacuo. Hence, if I may venture upon a guess, unsupported by experimental evidence, I should be rather disposed to assign the phenomenon to the reducing action of nascent hydrogen derived from the contact of sodium with traces of water. Perhaps even the least oxidizable metals are covered with a thin film of oxide, which is reduced by the nascent hydrogen at the same moment that the mercury is presented to the reduced metal. It is found that iron, copper, bismuth, and antimony are also wetted by mercury if their surfaces are first touched with sodium amalgam. Not only do the latter metals lose this power on being heated (as we might expect, in consequence of their superficial oxidation), but platinum, from which the adhering mercury film has been wiped by the cleanest cloth, or from which it has been driven by heat, also loses the power. It is true that the surface of clean platinum is supposed to condense a

A

was rubbed with sodium-amalgam and washed with clean mercury. few strips of similarly prepared platinum foil being bent over the edge and pressed close to the sides of the cup, the mercury could be handled similarly to the other liquids.

The following Table VII. shows,

1. The liquids examined.

2. The number of drops which were weighed.

3. The weights found.

4. The mean weights of single drops.

5. The observed specific gravity at the given temperature.
6. The relative sizes of single drops.

TABLE VII.

T=26° C.

gt=2".

Radius of curvature of platinum cup=11.4 millims.

Relative size of

gravity.

single drop.

Name and formula Number Weight Mean weight of Specific of drops. of drops. single drop.

of liquid.

film of oxygen; and the removal of this might alter the adhesion between the mercury and platinum; but such a film could scarcely exist in vacuo or in another gas.

The experimental numbers obtained are given without omission. The liquids are arranged in the order of magnitude of their drop-sizes. It appears from column 5 (of the specific gravities) that some of the liquids employed were not perfectly pure. This, however, is quite immaterial in the present direction of examination, provided that in all cases where the liquids named are in future employed and compared with those of Table VII., identically the same liquids are meant.

The numbers of column 6, with which we are now exclusively concerned, present several points of great interest. In the first place, it appears that the specific gravity of a liquid is not by any means the most powerful determinant of the drop-size. Thus butyric acid, which has sensibly the same specific gravity as water, gives rise to a drop less than half the size of the water-drop; while mercury, of singular specific gravity, has no exceptional drop-size. Lastly, it may be observed how that remarkable body water asserts here again its preeminence. The first impression which these numbers make is, that there are three groups of magnitude, n, 2 n, 3n. But it is possible that a change in the nature of the solid might throw these drop-sizes into a different order of magnitude; and certainly until a very much greater number of bodies is examined in this sense, it would be premature to attempt to establish anything like a law.

It is sufficient for the present to point out that the drop-size is not directly dependent upon either the specific gravity or boiling-point; nor does it stand in any obvious relation to what is sometimes called the liquidity, mobility, or thinness of a liquid. For we find that glycerine and (from former experiments) cocoa-nut oil both form smaller drops than water, the one being heavier and the other lighter than that body, and both being viscid or sluggish. On the other hand, alcohol and acetic acid, both perfectly mobile liquids, give rise to drops about half as large as those of glycerine*.

Hence it is clear that we are still ignorant of that property of a liquid upon which its drop-size mainly depends. We are not yet in a position to connect the drop-size with any of the known physical or chemical properties of liquids. We approach the solution of the problem by studying the effects of change in some others of the variables.

The adhesion between the liquid which drops and the solid from which it drops is also affected by the curvature and general geometric distribution of the solid at and about its lowest point. And the variation in the adhesion between the solid and liquid, caused by the variation in the geometric distribution of the solid, may and does in its turn affect the size of the drop.

From this aspect, one of the simplest kinds of variation is that offered

The evaporation of the more volatile of these liquids is a source of slight error; not so much on account of the direct loss in weight of the drop in falling, as by reason of the cooling which it causes, and the consequent variation in density and adhesion. Such source of variation we shall examine in the sequel, and find insignificant.

t

by a system of spheres of various radii, but made of the same material. And this case is an important one, because it undoubtedly offers the key to all drop-size variation arising from a similar cause. To study this point we may make use of any one convenient liquid, such as water, and cause it to drop at a fixed rate from spheres of various radii, including the extreme case of a horizontal plane. This extreme case, however, presents certain practical difficulties. From a plane it is almost impossible to get a series of drops uniform in growth-time and in position. A ripe drop hanging from a horizontal plane will seek the edge thereof. Several drops may form upon and fall from the same plate at the same time and independently of one another. It is only by employing a plate not absolutely flat, that an approximation to the required conditions can be made. Taking r for the radius of curvature, the first numbers for r∞ can therefore be considered only as an approximation. The arrangements for the other cases were quite similar to that described in Part I., fig. 3.

No. 1. A glass plate, fastened to and held by a vertical rod.
Nos. 2, 3, 4. Selected globular glass flasks.
Nos. 5, 6, &c. Perfectly spherical glass spheres.

It appears, therefore, that the drop increases in size according as the radius