Prof. Ketteler on the Boundary-Conditions of Reflection and Refraction for the Principal Section of Media in motion . . 411 Mr. O. Heaviside on Telegraphic Signalling with Condensers. 426 Mr. J. Croll on the Physical Cause of Ocean-currents Prof. J. Thomson on certain Relations between the Dr. T. Wright on a new genus of Silurian Asteriadæ I. Illustrative of Mr. J. Y. Buchanan's Paper on an Apparatus for Gas- II. Illustrative of Mr. J. Croll's Paper on the Physical Cause of Ocean- III. Illustrative of Prof. J. P. Cooke's Paper on the Vermiculites, and IV. Illustrative of M. F. Kohlrausch's Papers :-on the Determination of the Absolute Value of the Siemens Mercury Unit of Electrical 3 и LONDON, EDINBURGH, AND DUBLIN PHILOSOPHICAL MAGAZINE AND JOURNAL OF SCIENCE. [FOURTH SERIES.] JANUARY 1874. I. On an Apparatus for Gas-analysis. By J. Y. BUCHANAN, Chemist on board H.M.S.' Challenger '*. A [With a Plate.] VERY important subject of investigation in the chemistry of the ocean is the nature and quantity of the atmospheric gases dissolved in it. These are extracted by boiling in vacuo in an apparatus recently described by Jacobsent, and at the end of the operation are obtained hermetically sealed in a glass tube, in which they are capable of indefinite preservation. Although it would be absurd to waste time in making necessarily imperfect analyses of valuable specimens at sea, which are capable of being kept and analyzed with the greatest accuracy on shore, still cases might occur during a three years' voyage where it would be very desirable to make an approximate analysis of a specimen of gas, not necessarily proceeding from sea-water, but, for instance, from hot springs or volcanoes on shore. The first desideratum for such an apparatus for use on shipboard appeared to be freedom from a mercurial trough. All existing apparatus, besides having this source of inconvenience, were far too large and cumbrous to be at all suited to the modest allowance of laboratory space available on board ship. What seemed to give most promise of success was in some way or other to adapt the original Ure's eudiometer to the purpose. The form of apparatus finally adopted is that represented in Plate I. fig. 1. It was constructed, according to my drawings Communicated by the Author. † Ann. der Chem. und Pharm. vol. clxvii. p. 1. Phil. Mag. S. 4. Vol. 47. No. 309. Jan. 1874. B and instructions, by Dr. Geissler, of Bonn; and it is needless to add that as a piece of glass workmanship it is a chef-d'œuvre. Before describing it, it may be as well to state the objects sought to be attained, when the motives in the design of the various parts will be more apparent, and the measure in which they fulfil their end more correctly appreciated. First of all, the size of the instrument must be reduced to a minimum; nor must its shape be so eccentric as to interfere with its being packed into a reasonable space of symmetrical form. Again, when packed it must be safe against rolling, and be easily unpacked and mounted, and as easily dismounted and packed away again. I wished further to be able to use eudiometrical as well as absorptiometrical methods, using in the latter liquid reagents. The necessity for a special gas-analysis room had to be dispensed with; and, as before mentioned, the use of a mcrcurial trough was to be avoided. The advantages secured by the peculiar way of packing, so that the case in which it is packed when out of use forms its support, and a working tray for saving spilled mercury when in use, were not contemplated from the beginning, but suggested themselves only after the glass work was finished, as until then the form and size of the case could not be determined on. It was made for me by Messrs. Kemp and Co., of Edinburgh; and the workmanship is in every way satisfactory. The apparatus consists essentially of two U-tubes. The one, which according to precedent we may call the "laboratory tube,' is wholly of glass; the other (the eudiometer) has the legs of glass united by an india-rubber tube of suitable length. These are affixed to mahogany boards A and B, which fit into the wings of the box C, where they are secured each by two bolts (2, 2) passing into the sides of the box. D is a strong mahogany box for holding mercury. The glass U-tube, which for convenience we shall call A, as well as the board to which it is attached, is 0.42 metre high; the shorter leg is 0.18 metre from the bend to where the capillary tube is joined on, and its diameter is 0.02 metre. The diameter of the capillary tubes is 0.003 metre. In the U-tube B the eudiometer (g) is 0.17 metre from capillary to india-rubber; the movable leg (p) is 0·34 metre long; and the diameter of both is the same as in A. The length of the box C is 0.45 metre, and the width and depth of each wing 0·195 metre and 0.105 metre respectively. The mercury-box D measures 0·12 metre by 0.10 metre by 0.08 metre, and when in use fits into the place where it is represented in the figure. When dismounted it is not packed in the case with the rest of the instrument. When in use the parts A and B are screwed to the back of one wing of the box, A with one and B with two screws (x, x, x). The back of the box carries four nuts, so that A may be fixed to it close up to the end of the box and with the stopcock y projecting over the side, leaving a space between A and B, the object of which will be explained further on. The different parts of the apparatus will be best explained by describing the manipulations occurring during the analysis of a gas as well as those necessary for introducing the gas into the apparatus. But first we must describe the capillary part of the apparatus between the eudiometer g and the laboratory tube m. The part belonging to A is shown separately in two sections, figs. 2 and 3. The stopcock a has two tubes-the one affording direct communication upwards between the two portions of the capillary tube shown in fig. 3, the other communicating through the prolongation of the stopcock with the air, shown in fig. 2. b, c, and d are ordinary stopcocks pierced to the same bore as the tubes they connect. The capillaries of A and B are connected by a piece of india-rubber vacuum tubing. This tube is 0.003 metre diameter in the bore; and the thickness of its walls is likewise 3 millimetres. The stopcock d communicates with the air through the cup e. When the instrument has been set up as in fig. 1, it is filled with mercury by pouring in at the open legs of A and B, all the stopcocks being open, so that finally the mercury rises a little way up in the cup e. The apparatus being full of mercury, the gas to be analyzed is introduced into it either through the opening, y, of the stopcock a, or, if it is contained in a sealed tube, by interposing it at f between the two parts A and B, A being for that purpose shifted to the further end of the box, and the two ends of the tube in which the gas is collected connected by vacuum-tubing, filled with mercury, with the ends of the capillaries of A and B. The pressure in one side of the apparatus is now reduced, either by running mercury from A by means of the stopcock y, or by lowering the movable leg p of B. When the points of the tube are broken the gas rushes into the parts of the apparatus where the pressure is least, its place being supplied by mercury from the other side. The stopcocks are then closed, the tube full of mercury removed, the parts A and B reunited by f, and the analysis proceeded with. When the gas is to be admitted through a, an india-rubber tube is connected at y and filled with mercury by opening a, as in fig. 2, and running mercury through it. The point of the collecting-tube, having previously been touched with the file, is pushed into the india-rubber tube; the other end having also received a stroke of the file, is immersed in a cylinder full of mercury and the lower point broken off against the |