The results thus compared are probably as accordant as ought to be expected, and indicate the effect of smoothness of surface to be to lower the temperature of maximum vaporisation, but to increase the time required to vaporise at that temperature. Thus in the two extremes of high polish and considerable oxidation, the temperatures of maximum vaporisation are 2920 and 348°; and the times of vaporisation, 3 seconds and of a second. The nearness of the point of repulsion to the temperature of maximum vaporisation is shown in those cases where the point at which perfect repulsion took place, was noted, nearly; the temperature exceeds that of maximum vaporisation by about 21°. Vaporisation of Drops of Water by Iron.

5. Experiments were also made to deter mine the temperature of maximum vaporisation of water by iron with different states of surface, and as they preceded those made with the copper, the number of series was more considerable, that care might in a measure supply the place of experience. It will be wholly unnecessary to give the details of each series, since the mode of experimenting has already been stated, and the results can alone be of interest. At the same time, the

temperature at which perfect repulsion of the drops took place was observed. A portion of the experiments were made in an oil bath, others by communicating the heat through tin.

In the following table are the results for a bowl of wrought-iron (No. III.), three-sixteenths of an inch thick; the surface was cleaned with acid and alkali after each series: it was not very different in smoothness in the different series, until the closing one, which is marked in the table. The oil bath was used in these experiments. The drops of water were let fall from a dropping-tube, and 128 were required to make one-eighth of a fluid ounce; each drop, therefore, weighed about 45 of a grain.

The column of remarks in the following table, is intended to contain principally the temperatures at which the repulsion was observed not to be perfect, and gives an idea of the approximation to the true point of repulsion which each individual observation affords. These numbers obviously differ from those for the temperatures of perfect repulsion less than these latter among themselves, and much less than might have been expected, from the uncertain nature of the effect of slight inequalities of surface.

The experiments which follow the remark, "ten drops in six seconds," were all made by dropping several drops, not enough, however, to cool the surface down; measuring the time for the whole number, and dividing by that number. The point of maximum vaporisation is placed, with probability, at 4190, the times had certainly increased in rising to 444° but in descending, the same certainty is to be found only on reaching 32140. This slightly rough but polished surface, as it may be considered, had its temperature of maximum vaporisation very certainly above that of the polished copper and of the smooth iron. The time of vaporisation of a drop at this temperature was less thah one-sixth of the corresponding time for the polished copper, and less than that for the clean copper surface; agreeing more nearly with that for the smooth iron, which was much its inferior in lustre. A correct induction could only best had by varying the number of metals, and by frequent repetition of the results; but so far as these experiments go, they indicate that this repulsion does not depend alone upon the relative polish of the different metallic surfaces, des

8. The conclusions which they fairly war rant are as follows:

1st. With the same metal, the temperature of maximum vaporisation of water is lower, as the smoothness of the surface is greater, and the amount of vaporisation in a given time at this temperature is much diminished. In copper, the effect of polish and of oxida tion, the two extremes, is shown by a difference in the temperature of maximum vaporisation of 56, that point being in the two cases, 2920 and 3480. Further, the ratio in the times of vaporisation at these two points

2d. The temperatures of maximuni va porisation for copper and iron, in similar states of surface, differ between 30 and 400, the iron having the higher point. The time of vaporisation at the maximum is less in the copper than in the iron, in the ratio, probably, of 2 to, or nearly in the ratio of their conducting powers for heat, which are as 2 to 1.

3d. The temperature of maximum vapori, sation for oxidated iron, or for highly oxidated copper, corresponds nearly to that at. which steam has an elastic force of nine atmospheres. But the vapour was formed under atmospheric pressure only.

4th. A repulsion between the metal and water is perfect at from 20 to 400 above the point of maximum vaporisation, following more closely upon the temperature of maximum vaporisation in copper than in iron. At these temperatures the water does not wet the metal. The drops of water are put in rotary motion in variable directions, and sometimes remain at rest, slowly vaporising. When very small, they sometimes leap verti cally from the surface of the metal. They seem to vaporise from the side next to the metal.

A general view of the facts just deduced is given numerically in the following table;

vaporisation, an effective force of repulsion between the heated metal a and the water has ist dud)-elttog sid usowded,ga 70)67891093 18bud 8 9d ifiw 9791) yo

begun to be developed. For we may assume that heat will tend to pass from the metal to the water the more rapidly as the temperature of the former exceeds that of the latter, which would tend to increase the vaporisa

tion merced,

the repulsive action had com

The temperatures of maximum vaporisation are reached in practice in the highpressure steam engines. The locomotives with flues of copper use steam of 60lbs pressure upon the safety-valve, corresponding to nearly 306° Fah.; a temperature which is but 150 below that found for the maximum vaporisation by oxidated copper. The iron boilers of our high-pressure engines use steam of from ten to eleven atmospheres, or from 354° to 360° Fah., the higher temperature being about 20° below the temperature found for the maximum vaporisation of water by an oxidated surface of thick iron.

'It is possible, and indeed probable, that pressure may modify these results, all of which were obtained under atmospheric pres

19. The results already presented, however interesting they may be in a practical or in a philosophical point of view, cannot be said to touch the question of the effect of the contact of water suddenly made with hot metal, in producing explosions. It is necessary to suppose so large a quantity of water, brought under the vaporising influence of the metal as, except where there is a violent repulsion by the heated metal, to reduce materially the temperature of the surface. To study the question in this point of view, we must ascertain, if possible, the law, according to which a variable quantity of water, thrown upon heated metal, is capable of reducing its temperature, so as to produce the maximum amount of vaporisation. That such a maximum may be found will be seen by considering the foregoing results. They show that an effective repulsion is developed between water and heated metal, increasing rapidly after a certain temperature, at which the vaporisation is a maximum. Now, water thrown upon a surface at its temperature of maximum vaporisation, would cool it down rapidly below this temperature. Again, if thrown upon it at a temperature when the repulsive effect was very strong, it would not be able to cool it down as low as the tem-. perature of maximum vaporisation. Some where, then, between the points thus referred to, there will be an initial temperature, at which the vaporisation will be the greatest

possible, or a given quantity of water will be vaporised in the least time. It is obviously not as easy to solve this problem as the preceding one, nor can so satisfactory results be expected, nor results so constantly reproducible; its practical importance required 'that' its solution should be attempted; and the method adopted was as follows:-The same baths, viz. oil and tin, were used as in the foregoing series, to ascertain, generally, the effect of communicating heat through different media. Different metals, copper and iron, with different thicknesses of each, and different states of surface, were subjected to trial. The quantity of water was gradually increased, from small quantities, scarcely capable of reducing the temperature of the surface when the repulsive tendency was fully developed, to quantities as considerable as the bowls could contain. The study of each case was, of course, attended with much labour. In the greater quantities of water the temperature of the metal of the dishes was so much reduced as to affect that of the bath itself. Accordingly, a mean of the temperatures, observed at regular intervals, is taken as the temperature of the bath on which the water was thrown, and which, taking the entire mass into consideration, was supplying an amount of heat due to that temperature, to the parts adjacent to the bowls. The oil bath was stirred to produce, as nearly as practicable, a uniformity in temperature in the different parts.

1

Without knowing the temperature to which the parts of the heated metal, or of its bath, are reduced by the affusion of water, this kind of experiment supplies precisely the answer to the question in practice; at what temperature of a metal will water, thrown upon it in a limited quantity, be most rapidly: turned into steam? Making due allowance for the different modes of communicating heat in the experiments and in practice.

Copper Bowl, No. VII.

10. The same bowl used in a former series of experiments was again applied, the surface being smooth. This bowl was a portion of a spheroid, approaching nearly, in its inner surface, to a spherical surface of 3.09 inches radius: the versed sine of the segment, or depth of the bowl, was 16 inches; and its chord, or the breadth of the bowl, 539 inches. The thickness of the metal was '07 of an inch.

The quantity of water first introduced. was one-eighth of an ounce by weight (60 grains troy), the water being weighed in a small metallic dish, and thrown into the bowl placed in the bath. One experimenter observed the temperature of the bath, and gave notice to another of the instant of introducing the water; the other made a memorandum