the reading noted. The calculated reading which it ought to have shown on the above assumption is put down, and it will be seen how closely the two agree.

This instrument is especially adapted for surveying-work where lines of sounding in comparatively deep water have to be run for considerable distances, as in the English or Irish Channels, or the North Sea. The sounding line or wire is arranged to pay over the stern, and arrangements are made for promptly heaving it in on bottom being reached. The engines are kept going at a convenient and uniform rate, and the revolution counteracted; it is then easy with good organisation to take soundings every hundred, two hundred, five hundred, or thousand revolutions. In the want of a revolutioncounter, which, however, ought to be fitted to every marine engine, equally good results can be obtained if the engineer pays attention to keep his engine going uniformly, checking its rate at frequent intervals by counting the revolutions in a minute, by making the soundings at regular intervals of time. The number of revolutions made by the engines in a vessel of known capabilities is a very accurate means of ascertaining the distance run, and if this method were adopted the deep sounding of a survey could be worked off with great expedition and accuracy.

There is one property of this sounding-machine which must not be lost sight of, namely, that it registers the sum total of the increments of pressure. If it is to give the depth correctly it must both descend and ascend without interruption, and in the work for which it is designed this condition is always fulfilled. Suppose, for instance, it be sunk to 20 fathoms and be then drawn up to 10 fathoms, the corresponding quantity of air will be eliminated. If it is now again sunk to 20 fathoms, the place of the air which left while it was rising from 20 to 10 fathoms, will be taken by water, and if it now be brought to the surface it will of course register a depth greater than 20 fathoms.

It follows from the principle of this instrument that the value of its graduation scale will vary to a certain extent with the barometric pressure. For purposes of navigation the error so caused is negligeable, but if used for surveying purposes a correction must be applied.

The instrument is graduated for a barometric pressure of 30 inches.

As an inch of mercury exercises a pressure equal to 13 inches of water, we have for the corrected depth

where D is the depth read off from the scale, and H is the barometric height. D' the corrected depth is given in fathoms.

4. On the Compressibility of Glass. By J. Y. Buchanan.

(Abstract.)

The experiments related in this paper were undertaken with a view to determine, by actual observation, the effect produced on solids by hydraulic pressure. The instrument used consists of a hydraulic pump, which communicates with a steel receiver capable of holding instruments of considerable size, and also with a second receiver of peculiar form. This receiver consists essentially of a steel tube terminated at each end by thick glass tubes fitted tightly. It is tapped at the centre with two holes, the one to establish connection with the pump and the other to admit a pressure-gauge or manometer. The steel tube may be of any length, being limited only by the extent of laboratory accommodation at disposal. The tube which I am using at present has a length of a little over six feet and an internal diameter of about three-tenths of an inch. The solid to be experimented on must be in the form of rod or wire, and must, at the ends, at least, be sufficiently small to be able to enter the terminal glass tubes, which have a bore of 0·08 inch, and an external diameter of 0.42 inch. The length of the rod or wire is such that, when it rests in the steel tube, its ends are visible in the glass terminations.

The experiment is conducted as follows:-A microscope with micrometic eyepiece is brought to bear on each end of the rod or wire. These microscopes stand on substantial platforms, altogether independent of the hydraulic apparatus. The pressure is now raised to the desired height, as indicated by the manometer, and the ends of the rod are observed and their position with reference to the micrometer noted. The pressure is then carefully relieved, and a displacement of both ends is seen to take place and its amplitude noted. The sum of the displacements of the ends, regard being had to their

signs, gives the absolute expansion, in the direction of its length, of the glass rod, when the pressure at its surface is reduced by the observed amount, and consequently also of the compression when the process is reversed. As, in the case of non-crystalline bodies like glass, there is no reason why a given pressure should produce a greater effect in one direction than in another, we may, without sensible error, put the cubical compression at three times the linear contraction for the same pressure.

The rod experimented on was made of lead glass, drawn by Messrs Ford of Edinburgh, and was 75.05 inches long. The temperature of the water in the hydraulic machine varied from 12.5° to 13.5° C. The pressure varied from 1 to 240 atmospheres. Ninetyone separate observations were made, and the general result is, that the linear compressibility of the glass under experiment is 0-96, and its cubic compressibility 2.92 per million per atmosphere.

5. Suggestions on the Art of Signalling. By Alexander Macfarlane, M.A., D.Sc., F.R.S.E.

(Abstract.)

After considering the analogy which exists between the arts of writing and of signalling, the author proceeded to discuss what alphabet is the most suitable where the physical agent is not electricity. If we choose for elementary signals two qualities of the agent of communication A and B, which can be produced independently of one another, then the agent can be put into four states, viz., 1st, having the quality A, but not the quality B; 2d, having the quality B, but not the quality A; 3d, having both the qualities A and B; 4th, having neither of the qualities A and B. One of these states is required to separate letter from letter, and word from word; the fourth state where the agent is undifferentiated, is the one naturally adapted for the purpose. From the remaining three states we can get 3 permutations of one signal, 9 permutations of two signals, 27 permutations of three signals. Without going to a higher permutation than that of three signals, we get 39 symbols,*

Three of these would probably require to be omitted as repeating the same signal three times.

which are sufficient for the numerals and all the letters contained in the Morse alphabet, less by one. These symbols we suppose assigned to the letters according to their frequency of occurrence as given in that alphabet.

In the case of the Morse system we have only three states of the agent; the third of the above states is not, or cannot be, made use of. As one is required for the purpose of spacing, only two are left to form symbols. To form equivalents for the 39 symbols spoken of above, it requires 2 permutations of one signal, 4 permutations of two signals, 8 permutations of three, 15 of four, and 10 of five. Thus in the former case 102 signals are required to form the alphabet, in the latter 144.

If the elementary signals of the Morse system are made to depend on a difference in quantity, then the above qualitative system possesses other two advantages. Its signals, as they differ in quality, can each be made to occupy the minimum time necessary for a signal to be observed, whereas in the other case the longer signal occupies thrice the time of the shorter; secondly, elementary signals differing in quantity require, though belonging to the same letter, to be separated by an interval, whereas those differing in quality do not.

By assuming that each of the qualitative signals can be sent in the same period of time as the short signals, also that the time required in the Morse quantitative system for the space between the elements of a letter is equal to that period, and the time required for a space between the letters to thrice that period, I have been able to calculate the relative times required by the two systems to signal the words London, Edinburgh, Dublin. The respective ratios are 2-8, 2-6, and 2.9. We may therefore conclude, assuming that the other advantages and disadvantages neutralise one another, that a message can be signalled 2-6 times as quick by the qualitative alphabet described as by the Morse (quantitative) alphabet.

The four-state alphabet would allow one elementary sign to be invariably associated with the right hand, and the other elementary sign with the left hand, and the compound elementary sign with both together. The effect of this would be that a person who had learned to signal by means of any one agent, would have almost equal facility in signalling with any other.

The author then proceeded to offer some suggestions as to how this alphabet could be applied in the respective cases of signalling by means of the heliograph, the light of a lighthouse, steam-whistles, flags, and touch; and advocated the opinion first brought forward by Dr J. A. Russell in a paper read before the Royal Scottish Society of Arts in 1875, that signalling should be taught in the primary schools.

6. Note on the Wire Microphone. By R. M. Ferguson, Ph.D.

At our last meeting Professor Chrystal showed us that a fine platinum wire attached to a stretched disc of skin could act as an electric telephone receiver for the sounds of a violin. The wire was included in a galvanic circuit, and the variations of current were made by a microphone attached to the violin. The account he gave of this interesting experiment was that the receiving wire became extended by the heat of the current either as it was established or suddenly increased by the microphone, and correspondingly shortened on the current ceasing. These extensions and contractions were rendered audible by the disc. A similar demonstration with a like commentary was made by Mr Preece to the Royal Society of London, an account of which was published in "Nature" (June 10). Mr Preece got his wires to speak. At the first May meeting of this Society in 1878 I discussed the subject of the sounds emitted by fine wires, giving passage to intermittent currents. I found that the ordinary thread telephone gave us an easy means of hearing these sounds in non-magnetic metals. De la Rive had heard them in 1845, but since his time no one had been able to hear them, and they were almost looked on as apocryphal. I attached the thread of the skin or paper telephone transversely to the sounding wire, and not directly, as Professor Chrystal has done, for the simple reason that I found that the transverse method gave equally good results with very much less trouble. The cause in both cases seemed to me the same, viz., an internal molecular click which marked the setting in and stoppage of the current. In the kindly reference that Professor Chrystal made to my communication he considered it strange that his simple explanation should have been overlooked, that the sounds should be set down as having conditions the same as those of heat and yet the