Ramsay and Young, in the February number of the Philosophical Magazine, on the results obtained by myself and others on the state of matter at the critical point.

In the first place, the authors observe that there are contradictions between the results arrived at by Galitzine, de Heen, and Zambiasi, and those obtained by myself; and these contradictions are reduced in the main to the two following:

(A) That according to my observations the meniscus disappears during the heating at a temperature higher than the true critical temperature, whilst according to the observations of Galitzine, de Heen, and Zambiasi, the dimness forms during the cooling at a lower temperature than the critical one.

(B) That Zambiasi and I found that the dimness (or cloudiness) during the cooling process is produced at a lower temperature when the substance contained in the experimenting tube is in a larger quantity, whilst de Heen found just the contrary, and Galitzine came to the conclusion that the temperature of the cloudiness is "practically independent of the relative quantity of liquid."

It will be well to examine more closely what are the foundation and value of these contradictions.

The first does not exist. In fact from the experiments both of Galitzine and myself, in the most favourable conditions the disappearance of the meniscus to the last trace of it occurs at a higher temperature than that which one perceives in the ordinary mode of observation, and the disappearance itself takes place at a higher temperature than that at which the cloudiness afterwards takes place. Whilst also it results from my experiments that the cloudiness takes place at a lower temperature than the true critical temperature, in conformity with the result obtained by the other three observers. That Zambiasi found the temperature of the disappearance of the meniscus coincident with that of the cloudiness, must certainly be attributed to his mode of observing the temperature.

In regard to the second point, on the contrary, the difference exists, and therefore merits fresh experiments in order to arrive at a definitive decision.

I would notice that, in the many observations made by me with several tubes containing various quantities of substance, I had each time a couple of these tubes in identical conditions, and the observation of these tubes was very clear and contemporaneous: therefore it may be that the difference depends on the methods of observation more or less adapted to the end in view, and thus it seems to me that I am right in insisting on my conclusions, which, however, I shall be ready to give up when decisive experiments have settled the question.

The reasons alleged by Professors Ramsay and Young for declaring incorrect the experiments in question, which lead to conclusions differing from their ideas on the point, are the following:

(1) That in our apparatus we could not have a perfectly uniform temperature;

(2) That we did not make certain of dealing with perfectly pure liquids and entirely deprived of air.

As to the uniformity of the temperature, Signor Galitzine has already replied, in the April number of the Philosophical Magazine, that the criticism regarding him was not well founded.

I can now repeat the same; for I must remind the reader that for heating I used petroleum distilled at determined temperatures, the vapours of which circulated in a stove with double sides, in the interior of which a closed glass globe contained the tubes. The length of these was a little more than the fourth part of the length of the entire globe, and they were placed in the middle of it, in correspondence with the height of the mouth of the tubes which conducted the petroleum-vapour into the stove.

I have already noted that an even temperature was preserved in the stove up to exactly degree for ten or fifteen minutes beforehand, and sometimes even for a longer period; thus it is absolutely impossible to suppose that there was not a uniform temperature in the globe, and especially in the small space occupied by the little tubes.

As regards the absolute purity of the liquids and the entire absence of air, one cannot perhaps give a categorical reply; from all the memoirs on the subject, however, it would appear the greatest care was taken to that end by every experimentalist.

But admitting that the apparent difference depends on the impurity of the substance or on the presence of air, still one could not altogether impugn the conclusion at which we have arrived.

In fact one could not explain merely by the hypothesis of Andrews the fact described in § 7 of my first memoir (Nuovo Cimento Genn. for Feb. 1893), namely, that in two bulbs placed in a stove at the same height, and connected by a small capillary tube in the form of a П, one of which contains liquid ether and not the other, when they become heated above the critical temperature, and then slowly cool, the characteristic cloudiness appears only in the bulb which originally contained the liquid. Nor, according to the same hypothesis, would the conduct of the isocores in regard to the critical temperature be as it is in reality, as I demonstrated in § 10 of the same memoir.

Finally, the fact observed by Galitzine (Wied. Ann. vol. 50. p. 541) is in contradiction to the ideas of Messrs. Ramsay and Young, namely, that if the two branches of a tube are separated at the U by a little column of mercury, and the one branch is filled entirely with ether and the other partially, when the tube becomes heated above the critical temperature the movement of the column of mercury from the first branch towards the second does not cease, even if the temperature is increased to 209°-5 C.

I was proving the same thing when Galitzine's paper appeared, using for this purpose a straight tube instead of a curved one, and

the observations were most carefully made, and some of them were repeated by Messrs. Cagnato and Strapazzon.

The result arrived at by me is considered incorrect by Messrs. Ramsay and Young-that "the vapour-pressure at a given temperature depends on the relative volumes of liquid and gas," because it is "absolutely opposed" to their experiments, and they thus maintain that my liquids must have contained some other liquid, or that they contained a permanent gas.

But I reply that the memoirs which describe my experiments detail all the minute care taken by me in order to obtain pure liquids; and they also give the calculation (in the case of ether and of carbonic sulphide) which shows that the increase of pressure could not be attributed to gas which might casually have remained in the experimental globe.

I would rather observe that, in order to observe such a phenomenon, an apparatus is necessary which enables us-as in my case-slowly to compress the vapour, and to maintain it for a time under constant pressure.

ON THE FORMATION OF FLOATING METAL LAMINE.

BY F. MYLIUS AND O. FROMM.

The following are the results of this investigation:

1. Oxidizable metals such as zinc, iron, cobalt, cadmium, copper, silver, and antimony, have the property that when separated by electrolysis they float on the surface of solutions of their salts in coherent laminæ.

2. This diffusion depends on two circumstances-firstly, the presence of an impurity which does not mix with water; and, secondly, the chemical action of the oxygen present. The same effect is produced by sulphur on the halogens.

3. For the spreading of the metals on the boundary of two media, the thickness of the oily layer is not of essential importance. 4. The form of the bounding surface has no appreciable influence on the phenomenon: hence the spreading occurs even when one medium is in the form of drops.

5. Oxides and sulphides which conduct the current have the property of spreading out on the bounding surface; thus, for instance, the lower degrees of oxidation of silver and of cadmium, peroxide of lead, subsulphide of copper.

6. The growth of the floating lamina is influenced by the capillary attractions which those parts of the liquid experience from which the precipitate is deposited.

7. During the passage of the current a tension is often observed which disappears when the current is broken, and apparently depends on the difference of potential, like the surface-tension of mercury when it is polarized.-Wiedemann's Annalen, No. 4, 1894.

THE

LONDON, EDINBURGH, AND DUBLIN

PHILOSOPHICAL MAGAZINE

AND

JOURNAL OF SCIENCE.

[FIFTH SERIES.]

SEPTEMBER 1894.

XXVIII. On the Velocity of Sound in Air, Gases, and Vapours for Pure Notes of different Pitch. By J. WEBSTER Low, Ph.D., B.A.*

§ 1. Introduction.

Y the publication of Regnault's† great work and the immediate corroboration of his results by Le Rouxt, the general confidence in the previously accepted value of the velocity of sound was severely shaken. Since then several experimenters have sought, by measuring the wave-lengths of notes of different pitch, to arrive indirectly at the velocity of sound. With this object Kundt§ and Kayser|| have utilized the former's dust-figures, Schneebeli ¶ and Seebeck ** Quincke's interference-tubes ††; and all have agreed in finding a greater velocity for the higher notes than for the lower ones, a result the reverse of that found by Regnault and König ‡‡.

From theoretical considerations, Helmholtz §§ and Kirchhoff || have shown that friction and the conduction of heat *Communicated by the Author.

+ Compt. Rend. lxvi. pp. 209-220; Mém. de l'Inst. xxv. (1867). Ann. de Chem. 4 série, xii. pp. 345-418.

$ Pogg. Ann. cxxvii. p. 497 (1866); and cxxxv. pp. 337-372 and

527-561 (1868).

|| Wied. Ann. ii. pp. 218-241 (1877); and vi. P.

¶ Pogg. Ann. cxxxvi. p. 296 (1869).

** Pogg. Ann. cxxxix. p. 104 (1870).

++ Quincke, Pogg. Ann. cxxviii. p. 177 (1856).

1 König, Mém. de l'Inst. xxxvii. p. 435.

465 (1879).

$$ Verhandlungen des natur.-histor, medicin. Vereins zu Heidelberg vom Jahre 1863, iii. p. 16.

Il Pogg. Ann. cxxxiv. p. 177 (1868).

Phil. Mag. S. 5. Vol. 38. No. 232. Sept. 1894.

S

must greatly affect the wave-length and the velocity of sound in narrow tubes. Their theory agrees only imperfectly with the results of Kundt, who employed mixed notes; with those of Schneebeli, Seebeck, and Kayser, however, all of whom used pure musical tones, the accord is somewhat better. The methods of the last named inquirers, though correct in principle, are, however, in detail liable to various objections; I have therefore subjected the whole question of the indirect determination of the velocity of sound to a fresh investigation. The questions I set myself for answer were:

1. How does the velocity of sound vary in air and gases for pure notes of different pitch in tubes of different diameter? 2. How can the true velocity of sound in unlimited space be determined from that found in tubes?

§ 2. Method of the Inquiry.

At the suggestion of Prof. Quincke I measured the wavelengths for tuning-forks of known vibration-frequency by means of his interference-tubes. I observed, not one minimum of vibration-intensity, as Seebeck had done, but successive maxima, by shortening the tubes by one, two, or more half wave-lengths. My apparatus (fig. 1) consisted of a wide glass tube O U, closed at the bottom with a cork and a stopcock H. From H a long piece of guttapercha tubing led to a water-bottle F; a second piece of narrower tubing connected the side-tube A, distant about 5 centimetres from the upper end of the main tube, with the ear of the observer at C, ending in a glass pipe coated with sealing-wax, so as to fit exactly into the outer passage of the ear. By raising and lowering the bottle F a swimmer B could be brought to any desired point of the interference-tube, and the exact position of its upper smooth surface could be read off on a millimetrescale fixed behind the tube. The swimmer consisted of a cork 4 centim. long, loaded at the lower end with lead and coated with stiff paper and paraffin. The cork had almost the same diameter as the tube.

The theory of vibrating air-columns, as developed by Kirchhoff, postulates a regular motion of the air-particles parallel to the axis of the tube. In the course of my experiments, however, whether the bare prong or the resonancebox of the tuning-fork was held over the opening of the tube, or the fork-handle pressed firmly against any point of its sides, or the fork, with box attached, removed to any part of the room, I failed to observe any change in the positions of the maxima. These positions could, however, be most easily