block of stone in which they occur should be unhesitatingly condemned. Want of attention to this obvious rule has led to the unsightly disfigurement of public buildings. III. GRANITES.-In Professor Pfaff's experiments, to which I have already referred, he employed plates of syenite and granite, both rough and polished. He found that they had all lost slightly in weight at the end of a year. The annual rate of loss was estimated by him as equal to 0.0076 mm. from the unpolished, and 0·0085 from the polished granite. That a polished surface of granite should weather more rapidly than a rough one is perhaps hardly what might have been expected. The same observer remarks, that though the polished surface of syenite was still bright at the end of not more that three years, it was less so than at first; and in particular, that some figures indicating the date, which he had written on it with a diamond, had become entirely effaced. Granite has been employed for too short a time as a monumental stone in our cemeteries to afford any ready means of measuring even appproximately its rate of weathering. Traces of decay in some of its felspar crystals may be detected, yet in no case that I have seen is the decay of a polished granite surface sensibly apparent after exposure for fifteen or twenty years. That the polish will disappear, and that the surface will gradually roughen as the individual component crystals are more or less easily attacked by the weather, is of course sufficiently evident. Even the most durable granite will probably be far surpassed in permanence by the best of our siliceous sandstones. But as yet the data do not exist for making any satisfactory comparison between them. [Note added 21st May 1880. Since the preceding paper was written, I have had an opportunity of examining the condition of the monumental stones in the graveyards of a number of towns and villages in the north-east of Scotland, where the population is sparse and where comparatively little coal-smoke passes into the atmosphere. The marble tablets last longer there than in Edinburgh, but show everywhere indications of decay. They appear to be quite free from the black or grey sulphate-crust. They suffer chiefly from superficial erosion, but I observed a few cases of curvature and and fracture. As a contrast to the universal decay of the marble tombstones, reference may be made to the remarkable durability of the clay-slate which has been employed for monumental purposes in Aberdeenshire. It is a fine-grained, rather soft rock, containing scattered cubes of pyrites, and capable of being readily dressed into thin smooth slabs. A tombstone of this material, erected in the old burying-ground at Peterhead, sometime between 1785 and 1790, retains its lettering as sharp and smooth as if only recently incised. Yet the stone is soft enough to be easily cut with the knife. The cubes of pyrites have resisted weathering so well, that a mere thin film of brown hydrous peroxide conceals the brassy undecomposed sulphide from view. The slate is slightly stained yellow round each cube or kernel of pyrites, but its general smooth surface is not affected. The lapse of nearly a century has produced scarcely any change upon this stone, while neighbouring tablets of white marble, 100 to 150 years old, present rough granular surfaces and half-effaced though still legible inscriptions.] 2. On a Realised Sulphurous Acid Steam-Pressure Thermometer, and on a Sulphurous Acid Steam-Pressure Differential Thermometer. By Sir William Thomson. A sulphurous acid steam-pressure thermometer, on the plan described in my communication on the subject to the Royal Society of March 1, has been actually constructed, with range up to 25° C., but not yet in a permanent form. The slight trials I have been able to make with it give promise that, in respect to sensibility and convenience for practical use, it will most satisfactorily fulfil all expectations, and have given some experience in respect to the overcoming of difficulties of construction, from which the following instructions are suggested as likely to be useful to any one who may desire to make such an instrument:— (1.) The sulphurous acid steam thermometer might more properly be called a cryometer than a thermometer, because it is not very convenient, except for measuring temperatures lower than the atmospheric temperature at the place and time of observation; for, it must be remarked, that the thermometric substance, that is to say, the infinitesimal layer of liquid and steam of sulphurous acid at the interface between the two in the bulb in the annexed drawing (fig. 1), must be at a lower temperature than any other part of the space of bulb and tube between it and the mercury 20° 15 surface in the shorter vertical column. It is satisfactory, however, that the instrument is really not needed for temperatures above +10° C., because for such the water steam-pressure thermometer, represented in the first of the three diagrams of my former communication, has ample sensibility for most practical purposes. Hence, instead of the range up to 25° C. in the instrument already realised, and the great length of tube (295 centimetres for the long vertical branch) which it requires, I propose in future to let +10° be the superior limit of the temperatures to be measured by an ordinary sulphurous acid steam thermometer. For this, the long vertical branch need not be more than 175 centimetres; thus the instrument is much more easily made, and when made, is much less cumbrous. (2.) The upper end of the long branch, being open to begin with, is to be securely cemented to a small and very perfectly air-tight iron stopcock L, communicating with an iron pipe, bent at right angles, as shown in the drawing (fig. 2). This iron pipe is, in the first place, to be put into communication temporarily by an india-rubber junction with the generator, and with an air-pump, by means of a metal branch tube, with two stopcocks R and S, as shown in the drawing. -30 -20 -10 10 15 20 150 cm. 246-cm: MERCURY 10 -5° -10° -15 -20° -25° -30° (3.) To begin, close R and open S and L; and exhaust moderately (down to half an inch of mercury will suffice). Warm the whole length of the bent tube moderately by a spirit lamp, or spirit lamps, to dry the inner surface sufficiently. Then, still maintaining the exhaustion by the air-pump, apply a freezing mixture to the bulb and shorter vertical tube, and all of the long vertical tube except a convenient length of a foot or two next its upper end, as shown in the drawing. Before joining the generator to Q, let enough of sulphurous acid gas be passed out through P to clear out fairly well the air from the generator and the purifying sulphuric acid wash Fig. 1. E S Fig. 2. temperature below 110 One of the sulphuric bottles and pumice tubes, &c. Then make the junction at Q, close the stop-cock S and open R very gently taking care not to let air be sucked in by the safety tube of the generating ap Sulphuric pumice paratus.* Continue until liquid sulphurous acid is seen in the long vertical branch, about the top of the freezing mixture, then close the iron stop-cock L, disconnect at E, and remove from the freezing mixture (but beware of Professor Guthrie's cryohydrates). (4.) Now, place the instrument with the long straight parts of the tube nearly horizontal, but sloping slightly upwards towards L, and by a hand or spirit lamp properly applied, boil the liquid, the stop-cock L being still closed. After boiling for some time cause the liquid to occupy the whole space from T to its free surface, then very carefully open the stop-cock L slightly, being ready to close it quickly and prevent the escape of any of the liquid in case of sudden ebullition. Repeat this process over and over again two or three dozen times, so as thoroughly to remove air or other gases more volatile than sulphurous acid from the liquid. To as much as possible remove water or any acid wash-bottles must be provided with a safety tube with overflow bulb. An ordinary pipette with its stem fitted into the india-rubber stopper of the bottle will serve for the purpose. fluid less volatile than sulphurous acid, proceed as follows:-Apply heat at T and in the bend next T until the liquid leaves that part of the enclosure and stands nearly at a level in the short and long vertical branch, the instrument being held with A down. Apply a freezing mixture to T, taking care not to cool it to quite as low a temperature as -11° C.; so that the pressure of the sulphurous acid liquid and steam may remain something above the external atmospheric pressure. Occasionally open the stop-cock L very slightly to prevent the liquid from being drawn up the short vertical branch through preponderance of temperature in the long vertical branch. Continue this until about a centimetre of liquid has been distilled over into the bulb T. Then open the stop-cock L very carefully until all the liquid in the two vertical branches is blown out, leaving that which has been distilled over into the bulb T, and then close L again. (6.) Then dip the end E under pure mercury, and by opening L very gently and warming the free surface of the liquid sulphurous acid, let gas escape bubbling up through the mercury. Close L again before or when the quantity of liquid in the bulb at T begins to be perceptibly diminished. Then apply a freezing mixture to T until mercury is drawn in. Incline the instrument with A up and L down, and watch until the mercury is drawn up to A, then incline with A down and let a little more mercury come in. Then close L. Lastly, keeping T still in the freezing mixture, melt the glass below L till it collapses and blows the mercury down, leaving Torricellian vacuum at the sealed end. The instrument is now complete and ready for use. Sulphurous Acid Steam-Pressure Differential Thermometer. This consists of a U tube, with its ends bent down, as shown in the drawing, containing mercury in the main bend and in the lower parts of the straight vertical branches, and sulphurous acid gas, steam, and liquid in the rest of the enclosure. Every other part of the enclosure must be kept somewhat warmer than the warmer of the two ends, T, T'. The infinitesimal quantities of matter in the transitional layers, between liquid and steam, at T and T', constitute the thermometric |