ment du faisceau est obtenu au moyen d'un prisme composé construit sur le principe de celui de M. Amici; ce prisnie composé est formé de deux prismes de flint à 90°, faisant corps avec trois prismes de crown. Le prisme de crown central est à 90°. Les deux prismes de crown terminaux, sont taillés sous des angles convenables pour procurer le redressement du faisceau. Ce système jouit d'un pouvoir dispersif considérable, et conserve au faisceau presque tout son pouvoir lumineux à cause de la faible valeur des réflexions intérieures. La Lunette qui sert à explorer le spectre porte deux objectifs placés à faible distance l'un de l'autre. Cette disposition qui augmente beaucoup le champ de la lunette, permet d'embrasser le spectre d'un coup d'oeil. Enfin une échelle gravée sur verre permet de mesurer la position des raies dans les spectres qu'on étudie. J'ai fait cet instrument en 1862. Il a été présenté à l'Académie des Sciences de Paris (Comptes Rendus, Oct. 1862) et à l'Académie de Rome (Déc. 1862). Il est employé par le P. Secchi dans ses recherches de physique céleste, mann, de Paris, l'a construit sur mes indications. M. Hof Sur le Spectre Atmosphérique Terrestre et celui de la vapeur d'eau. J'ai l'honneur de faire part à la section de la découverte d'une propriété optique nouvelle de la vapeur d'eau. Cette propriété consiste en ce que la vapeur d'eau agit spécifiquement sur la lumière de manière à produire un spectre d'absorption caractéristique. Ce spectre permettra de constater la présence de la vapeur d'eau, et, par suite, de l'eau, soit dans les hautes régions de l'atmosphère terrestre, soit dans les atmosphères planétaires, et d'une manière générale dans les corps célestes. En continuant les beaux travaux de MM. Brewster et Gladstone sur le spectre solaire, j'ai été amené à constater cette action de la vapeur d'eau par une longue suite d'observations sur la lumière solaire faites en diverses saisons de l'année. Pour des hauteurs égales du soleil, les raies telluriques du spectre solaire se montraient d'autant plus foncées que le point de rosée était plus élevé. (Voir les Comptes Rendus de l'Académie des Sciences de Paris ; 13 à 27 Juillet, 1863; 27 Juillet, 1864; 30 Janvier, 1867*.) Spectre de la vapeur d'eau.--Une expérience directe qui démontre définitivement cette propriété vient d'être faite à Paris dans l'usine de la Compagnie du Gaz. Un tube de fer de 37 mètres, fermé aux extrémités par deux glaces, a été rempli de vapeur à diverses pressions. Le tube de fer était placé dans une enveloppe de bois rempli de sciure de bois pour empêcher les déperdition de la chaleur. La lumière était donnée par une rampe de 16 becs de gaz. On sait que cette lumière donne un spectre continu et sans raies; or, quand la lumière eut traversé le tube plein de vapeur à 7 atmosphères, elle donna un spectre sillonné de raies et bandes obscures correspondant aux raies et bandes atmosphériques terrestres ou telluriques du spectre solaire. (Ce sont les bandes découvertes par l'illustre Sir David Brewster quand le soleil est à l'horizon.) L'expérience a été répétée dans des circonstances diverses. On a examiné les effets de la longueur du tube et ceux de la pression. Les raies se développent à mesure que la longueur augmente ou que la pression s'élève; elles s'affaiblissent dans les circonstances opposées. Quand le tube est vide de vapeur ou qu'il en contient fort peu, on ne voit aucune raie. Le resultat est donc parfaitement constaté. J'ai interrompu les expériences pour venir en faire part au Congrès, mais je compte en poursuivre les conséquences. En attendant je puis déja conclure: 1. Que les raies du spectre solaire dans les régions rouge, orangée, jaune, sont presque toutes dues à la vapeur d'eau de notre atmosphère. 2. Qu'il n'y a pas de vapeur d'eau autour de la photosphère solaire. 3. Que la découverte du spectre de la vapeur d'eau vient confirmer les résultats obtenus par M. Tyndall, touchant l'action absorbante de cette vapeur sur la chaleur rayonnante. * M. Cooke, en Amérique, vient d'annoncer des résultats semblables; je suis persuadé qu'il n'avait pas connaissance de ces publications où j'ai formulé les conclusions de son travail dix-huit mois avant lui. On a Fluid possessing Opposite Rotatory Powers for Rays at opposite ends of the Spectrum. By Professor JOHN H. JELLETT. The existence of this fluid was discovered in conducting a series of experiments with a new saccharometer which the author had formerly described to the Royal Irish Academy *, and which he now exhibited to the Section. In making use of this instrument, it became necessary to compare the rotatory powers of the two well-known species of oil of turpentine, namely:-1. The American oil of turpentine, which is obtained from the Pinus australis of North Carolina; and 2. the French oil of turpentine obtained from the Pinus maritima of Bordeaux. As these fluids, which are opposite in their rotatory powers, are chemically identical, and very slightly different in their refractive and dispersive powers, it was natural to expect that no difficulty would be found in determining the relative lengths of two columns of these fluids respectively, which should perfectly compensate each other. Two columns of fluid are said to compensate each other when a ray of polarized light, transmitted successively through these columns, emerges from the second column in the same state in which it entered the first. The actual result, however, was wholly different from this anticipation. When the relative lengths were so determined that the intensity of the light transmitted respectively by the two parts of the analyzer † was the same, the colours of these two spectra were wholly different. In reasoning on the difference of colour, the author was enabled to perceive that the American oil of turpentine was much more highly dispersive of the planes of polarization of the elementary rays than the French oil. It is plain, therefore, that if the lengths of the columns be so proportioned that the rotation may be the same for the mean ray, the more dispersive (in the sense just defined) fluid will turn the plane of polarization of the red ray through a less angle, and that of the violet ray through a greater angle than the less dispersive fluid. Hence, remembering that French oil of turpentine is left-handed, and American oil of turpentine right-handed, it is plain that if a red ray be transmitted through two columns, whose lengths are so proportioned, the total effect will be left-handed rotation; whereas, if a violet or a blue ray be so transmitted, the effect will be right-handed rotation. As these fluids, being identical in composition, could scarcely act chemically on each other, the same effects might be expected from a single fluid produced by mixing these two columns. This the author found to be, in fact, the case. The rotating fluid was formed by mixing the two oils in the following proportion ‡ :— American oil of turpentine. 583 67 33 When a column of this fluid, whose length was 4 inches, was traversed by a solar ray which had been previously transmitted through plates of red and blue glass, the rotation produced in the plane of polarization of this, which is the extreme red ray, was found to be -1° 35'. Again, when the same column was traversed by a ray which had been previously transmitted through a solution of ammoniacal sulphate of copper, the rotation was found to be+2°. This phenomenon is best shown with solar light, but it may be shown, though with less distinctness, with the electric or oxy-calcium lights. On Comets, and especially on the Comet of 1811. By CORNELIUS VARLEY. HEAT. Determination of the Mechanical Equivalent of the Thermal Unit by Experiments on the Heat evolved by Electric Currents §. By J. P. JOULE, F.R.S. + Ibid. p. 349. *Proceedings, vol. vii. p. 279. The proportion of oils, given above, must be understood to refer only to the particular specimens of the oils which were used in making these experiments. The rotatory power of commercial oil of turpentine, more especially that of the American oil, is very § Printed in extenso in the Reports. variable. ELECTRICITY. On the Electrical and Mechanical Properties of Hooper's India-rubber Insulated Wire. By W. HOOPER. The author at a previous Meeting described the method by which he secures the durability of india-rubber. Diagrams representing the effects of pressure and immersion were shown, from which it was seen that pressure improves the insulation of his wire in the same way as is observed with gutta percha. The result of carefully-conducted experiments, extending over three years, proves that the absorption of water is so small that the most refined electrical tests failed to discover it. On the Depolarization of Iron Ships, to prevent the Deviation of the Compass. By E. HOPKINS, C.E. Extract of a Letter from Senhor CAPELLO, of the Observatory, Lisbon, on Magnetic Disturbance, to BALFOUR STEWART, of the Kew Observatory. The author sent three Tables representing graphically the most important results deduced from the curves of our magnetographs for the year 1864. He had followed the plan of General Sabine in separating the greatest disturbances of the three elements. Thus he had considered as a disturbance of the declination every ordinate which differed from the monthly mean by 2'3 or upwards; while the separating value for the horizontal force was 0011 of the whole horizontal force, and that for the vertical force 00032 of the whole vertical force. The instruments were at work during the whole of the year 1864; and of the 8760 hourly observations of each instrument, the observers only failed in measuring 97 for the declination, 139 for the horizontal force, and 159 for the vertical-force instrument. The number of disturbances have been- From a diagram exhibited, giving the hourly variations yearly and half-yearly of the three elements, it was seen that the progress of the declination for each period is very regular. The mean daily range of declination during the six months from April to September, while the sun is north of the equator, is 9'20; while during the six months from October to March, when the sun is south of the equator, this range is less, being barely 6'. For the dip, the corresponding curves are much disturbed from 6 P.M. to midnight, especially for the six months when the sun is north of the equator. The total force gives a well-pronounced minimum at 11 A.M. during the six summer months, and 11h 30m A.M. during the six winter months. The daily range is greatest for the six summer months, and least for the six winter months. The diagram of disturbances gives for the declination a maximum of the westerly disturbances at about 8 A.M., and a minimum about 10 in the evening. On the other hand, the maximum of easterly disturbances is about 10 in the evening, and the minimum about 6 in the morning. The curves for the horizontal-force disturbances are irregular. The maximum of disturbances tending to increase the horizontal force takes place about noon, while the minimum is about 1 A.M. But here one is much struck with the great disproportion between the disturbances tending to increase and those tending to diminish the horizontal force, the latter being both the most numerous and the greatest in amount. The maximum and minimum of these latter disturbances take place a little later than the maximum and minimum of the disturbances tending to increase the force. With respect to the vertical force, the curve of disturbances tending to increase this element resembles to some extent the curve of easterly disturbances, or disturbances tending to diminish the westerly declination. In this same diagram blue and red curves were made to represent the whole effects of the perturbations, or the quantities which it is necessary to apply to the line of no disturbance, reckoned a straight line, in order to reconstruct the curves with the perturbations. Thus the effect of disturbances upon the declination is to cause the needle to deviate towards the west during the hours of the day, but towards the east during the hours of the night. The effect of disturbances upon the vertical force is of a reverse kind, tending to diminish this element during the hours of the day, but to increase it during those of the night. With regard to the horizontal force, it appears that the disturbances tend to diminish this element almost during the whole of the twenty-four hours. A third diagram represented the mean hourly movements of the north pole of the freely suspended needle in a plane perpendicular to the direction of such a needle, both for the whole year, and also for the winter and summer seasons. On certain Phenomena which presented themselves in Connexion with the Atlantic Cable. By C. F. VARLEY. On a New Method of Testing Electric Resistance. By C. F. VARLEY. In 1860, Prof. Thomson and Fleeming Jenkin, F.R.S., invented a method of obtaining exact subdivisions of the potential of a voltaic battery. The apparatus consisted of a number of equal resistance-coils, say 100. These were connected one with one pole of the battery, and the other with the other pole. To the junction of each coil a piece of metal is attached, and a spring attached to a brass slide travelling along a square rod of the same metal traverses these different pieces, and so makes contact with whichever is desired. If the two poles of an electrometer be attached, the one to one pole of the battery, and the other to the brass bar on which the slide travels, it will be found that at the one end we have the full potential power, and at the other end nothing at all, and halfway half the potential; this is too self-evident to require further explanation, and is explained in Thomson and Jenkin's patent, 1860. Prof. Thomson has recently succeeded in making reflecting electrometers of such sensibility that they will give 200 scale-divisions for a variation of potential equal to one cell of Daniell's battery. In testing the Atlantic Cable this electrometer was used in the following way at Valentia, to get the potential of the ship's magnetism. The one pole of the electrometer was connected with the cable, and the other one with the slide, and by running it up and down the exact potential of the cable was measured. There were in the main slide 100 coils of 1000 units each, and it became necessary to subdivide these again 100 times to get sufficient accuracy. Some difficulty presented itself in getting a method for subdividing these coils, and the author was fortunate enough to hit upon the following very simple method of effecting this purpose. The slide consists of two square brass bars, over each of which travels a piece of brass, to the bottom of which is attached a spring, pressing upon the studs connected with the resistance-coils. Instead of using 100 coils in the main slide, the author uses 101, and makes the two springs to embrace two coils. Thus, then, the two bars of the slide have invariably a resistance between them of 2000 ohmads. The two bars are connected with a second set of 100 coils, each coil having 20 units resistance, and the 100 coils making up precisely the same resistance as that of two of the coils in the main slide. These two circuits of 2000 units each reduce the resistance to one-half, or to 1000 units, so that the resistance of the 101 coils of 1000 each is reduced to that of 100 coils. By passing the traveller along the 20 unit coils in the second slide an exact subdivision of the potential between these points is obtained; and in this way the potential of the battery is accurately and quickly subdivided to 10,000 parts. By these means Prof. Thomson has been able to introduce a method of testing, on the Wheatstone balance system, so extremely simple that it should be made known as soon as possible. The battery is connected permanently to the main slide, so that a current is always passing through it. Its resistance, 100,000 ohmads, is such that no sensible elevation of temperature is produced. The current is also passed into the cable through a definite re sistance, R. At the junction between the end of the cable and the resistance R a key is attached, which is connected by either the reflecting electrometer or a reflecting galvanometer with the slides. That position is sought upon the slide which has precisely the same potential as that of the cable at the point where it joins the resistance R. If now the potential of the battery be represented by p, and the resistance of the junction of the cable with R be represented by p', and if the two portions of the coil necessary to balance this potential be n and m, it will be evident, on the principle of the Wheatstone Balance, that n: m :: R: cable x (the cable resistance). Thus, then, the resistance R being known, p and p1 being known, and the resistance or position on the slide noted, the resistance of the cable is accurately obtained. sea. METEOROLOGY. On the Climate of Aldershot Camp. By Sergeant ARNOLD, F.M.S. The military station of Aldershot is in the county of Hampshire, bordering on Surrey, and is situated on an elevated site, about 320 feet above the level of the It is distant about 40 miles from London, and about 50 from Portsmouth and Southampton, being in lat. 51° 15' 25'' N., long. 45° 36′ W. The extensive area of ground occupied by the North and South Camps was formerly a barren heath, the soil consisting mostly of sand and gravel, covered by about seven inches of peat. On the north and south the Camp is much exposed: on the east it is slightly sheltered by hills that run from the eastern boundary of the North Camp to the South. On the south-east is a range of hill, called the "Hogsback;" these are the highest in the neighbourhood, affording great protection to the cultivation of hops, which is carried on so successfully that their growth is rapidly extending. The north-west and west are bounded by land under cultivation. Small woods or copses are numerous in the locality, consisting principally of stunted fir-trees and brushwood. A small river named Blackwater is the only one in the vicinity. as Meteorological observations during the past eight years yield the following results: The mean height of the barometer at 325 feet above the mean sea-level is 29-610 inches; this, however, is only an average of seven years, a standard instrument not having been used for the whole of 1858. The highest observed reading of the barometer was 30-452 inches on January 9th, 1859; the lowest, 28.269 inches on January 14, 1865 (these observations being reduced to 32° Fahrenheit). The adopted mean temperature of the air is as follows:-January, 384; February, 39°; March, 42°1'; April, 484; May, 53°2; June, 580.8; July, 599; August, 599; September, 566; October, 51-5; November, 418; December, 403. The mean for the past eight years is 49°2. The mean of all highest readings is 576, and the mean of all the lowest 41°.9; the mean daily range of temperature being 158. The highest temperature was 93° on July 12, 1859, and the lowest 8 on December 29, 1860, so that the extreme range of temperature is 85°. The mean degree of humidity (saturation=100) is follows:-January, 89°; February, 82°; March, 84°; April, 76°; May, 78° June, 78°; July; 78°; August, 78°; September, 85°; October, 86°; November, 90°; December, 89°. The yearly mean is 83°. The amount of cloud is estimated on the usual scale, O being a clear sky, and 10 an overcast sky. The mean amount is 6.1. The month in which the largest amount of cloud occurred was December (7·2), and the least (51) in September. The individual monthly averages show that the most cloud state (85) occurred in December 1865, and the least (3-2) in June 1859. Rain falls on an average of 143 days of the year. The greatest number of days was 183 in 1860, and the least 113 in 186-4. The average yearly rainfall is 25-24 inches. This is less than at any other station in Hampshire. The greatest yearly total was 33.89 inches in 1860, the least 17-13 inches in 1858. The greatest monthly fall was (5.80) in October 1865, and the least (0.16 inch) in February 1858. Taking the average of eight years, the wettest month is October, the mean amount being 2.96 inches, and the driest February, being 1.23 inch. The mean monthly amount of ozone is 17; the largest quantity occurred in May, the mean of which is 22, and the least 11 in December. The yearly relative propor |