ralogical composition of the granites of Donegal, and the minerals associated with them." In furtherance of this object, Mr. Haughton and Mr. Scott repaired, last Easter, to the northern part of the county, as they had visited the S.W. portion of the district in the summer of 1861. They were accompanied on their tour by Mr. Jukes, Local Director of the Geological Survey of Ireland, who gave them the valuable benefit of his experience and assistance throughout the tour. The exploration commenced at Moville, on the E. shore of Innishowen, whence a section was carried along the N. coast of that peninsula nearly as far as Malin Head. This section exhibited a great thickness of primary rocks, consisting of quartzite and mica-slate, accompanied by several beds of limestone, and a number of beds of igneous rocks, which appeared to be contemporaneous with the sedimentary rocks. These are best exhibited at a place called the Mintiaghs or Bar of Inch, where there are several alternations of quartz-rock and syenite exhibited in an escarpment of several hundred feet in height. This locality is situated about five miles N. of Buncrana. From Buncrana, the granite of Urrismenagh, near Dunaff Head, was visited.

From Milford an excursion was made to the extremity of the promontory of Fanad, lying between Lough Swilly and Sheep Haven, in order to visit the granite of this district. This patch of granite is not a continuation of that which traverses the country in a N.E. and S.W. direction, as it lies to the N. of that axis and exhibits a slight difference in composition from the granite of the central axis. From Milford the route lay to Dunfanaghy; and a section was made across the northern end of the granitic axis of the county at Glen, in which its gneissose character was very strongly exhibited. This was marked in a most decisive manner between Lackagh Bridge and Creeshlagh, where the rock might be observed changing from gneiss, by almost insensible gradations, on the one hand into granite, and on the other into hornblende slate and crystalline syenite. The latter is most highly crystalline at Horn Head, where it contains large quantities of titanic iron. On the return-journey from Dunfanaghy to Letterkenny, it was determined to make two sections across the granite; so that Mr. Haughton and Mr. Scott took the road from Creeshlagh through the Gap of Barnesbeg, while Mr. Jukes took that by Owencarrow Bridge, about four miles higher up the valley.

It having now been found necessary to compare the facts observed with those which were to be observed in other countries, Sir R. Griffith repaired to Scotland in the month of July. Mr. Haughton traversed the centre of Scotland, and paid a visit to Sweden, Finland, and Russia. Both these gentlemen discovered facts strongly confirming the views propounded at the Manchester Meeting, of the similarity of the geological structure of Donegal to that of the Scandinavian peninsula and of Scotland. For this latter fact the Committee had been prepared by the examination of a series of specimens of Scotch granites which had been furnished to them by Sir R. I. Murchison, in accordance with his kind promise made at the last Meeting.

While these tours were in progress, Mr. Scott repaired, for the third time, to Donegal, and spent the month of July in the re-examination of several points connected with the geology of the southern district. He visited the granite of Barnesmore, near the town of Donegal, which is essentially nongneissose, and is penetrated by numerous pitchstone dykes, some of which are amygdaloidal. Numerous minerals were discovered here, which were in some cases new to the district. In the neighbourhood of Glenties, a considerable quantity of andalusite was found in the mica-slate-a mineral which is replaced near Barnesmore by kyanite, and in the Rosses, near Dungloe, by a white variety of kyanite.

From Dungloe, as head-quarters, the structure of Crohy Head was carefully examined, and also the island of Arranmore, which differs materially in its structure from the mainland of Ireland, from which it is only distant three miles. The southern portion of this island is nearly entirely composed of white granite, penetrated by numerous dykes of syenite and of felspathic porphyry. The strike of these rocks is nearly E. and W., while that of the flaggy quartz-rocks on the northern shore of the island approaches N. and S. During the course of this tour, two more sections were made across the granite of the main axis, exhibiting the same facts which had been observed before, viz. numerous beds of limestone and of altered slate lying in the granite, stratified nearly conformably with it. These were observed in the centre of Glenveagh, close to Ballaghgeeha Gap, on the pass through the Poisoned Glen from Dunlewy. At Glenleheen, where the same occurrence of non-granitic rocks had been observed in the previous year, four beds of limestone and several beds of slate were discovered. Almost all these beds of limestone contained garnet, idocrase, and epidote in quantity; and at Glenleheen itself, scapolite, a mineral whose occurrence in the British Islands has escaped the notice of modern English mineralogists, was discovered. Inasmuch as the specimens brought home by the members of the Committee from their several tours are very numerous, it is not possible for them to present their complete report at this Meeting. They hope to embody in it some valuable information relating to the granitic rocks of Canada, which Dr. T. Sterry Hunt has kindly offered to supply to them. They have to express their thanks to him and to Mr. Harte, C.E., county surveyor of the western district of the county, who, with the Rev. Frederick Corfield, has afforded them most efficient assistance. They have succeeded in procuring some of the granite of Rockall, through the kindness of the officers of H.M.S. Porcupine, who furnished it to Mr. Harte, and will include its analysis in their paper.

On the Vertical Movements of the Atmosphere considered in connexion with Storms and Changes of Weather. By HENRY HENNESSY, F.R.S., M.R.I.A., &c., Professor of Natural Philosophy in the Catholic University of Ireland.

THE labours of the Committee, consisting of Admiral FitzRoy, Mr. Glaisher, and myself, who were appointed, at Manchester, for the purpose of studying the vertical disturbances of the atmosphere with the aid of instruments, have, for the present, been restricted to the work of a single observer. This has arisen from the circumstance that the money-grant appropriated to the Committee has sufficed only to defray the cost of erecting a single instrument. As this instrument is likely to afford opportunities for observing the vertical motions of the atmosphere more completely than has been hitherto possible, it is to be hoped that similar apparatus will before long be in the hands of the other members of the Committee. The fact that all the preliminary work has thus necessarily devolved on the writer of the present Report will sufficiently account also for its provisional nature.

Hitherto the only kind of atmospherical currents which have formed the subjects of definite observation by instruments are those whose existence is manifested by the movements of ordinary wind-vanes and anemometers. But as these instruments indicate horizontal movements exclusively, ordinary

winds as well as storms are almost always conceived as currents flowing in perfect parallelism to the earth's surface. It is true that no physical theory of the motions of the atmosphere can be attempted without some considerations which involve the necessity of vertical and oblique motions among the masses of air, as well as horizontal motions; but while direct comparisons of the latter among themselves have continued for many years to be made in different parts of the world, we possess scarcely any such data relative to non-horizontal movements as would enable us to make them subjects of exact inquiry.

The only writer who, as far as I am aware, has hitherto endeavoured to deduce any well-defined results from observation relative to the vertical movements of the atmosphere is M. Fournet, and his studies were almost exclusively directed to the elucidation of the phenomena of some remarkable local winds that frequently prevail among the Alps and in the valley of the Rhone*. A local phenomenon in Ireland + induced me to study the vertical motions of the air in a more general way than was necessary for the explanation of this phenomenon itself; and my first step was an attempt at devising a vane capable of showing the existence and direction of non-horizontal currents. This was a non-registering instrument, and the results obtained were therefore somewhat unconnected; but they seemed to establish some important relations between vertical currents and other atmospherical disturbances. Among these, I may be permitted to notice the phenomena which preceded the disastrous gale of February 9, 1861. For many days, at the close of January and beginning of February, the weather was remarkably fine, and no vertical currents were observed; but on the 7th very distinct evidences of vertical disturbance came under my notice, while the air had as yet no remarkable horizontal motion. On the 8th, at 2 P.M., my attention was called to the vane by its shifting round through N. towards N.E., with decided and frequent downward plunges of the disk exposed to the vertical action of the air. It appeared as if showers of cold air were descending; for the thermometer showed at the same time a rapidly falling temperature. While vertical convection had become already highly developed, the horizontal motion of the air was not as yet greater than that of an ordinary brisk breeze.

Next day, during the storm, although the disk of the vane was in constant oscillation from. the undulatory motion which my observations had already shown to be a necessary accompaniment of all high winds passing over terrestrial obstacles, no marked prevalence of upward or downward motions could be observed corresponding to the plunges of the disk noticed on the preceding day. The mercury in the barometer had been falling with great regularity during four days before that on which I had noticed the first decided indications of vertical disturbance. On that and the next day, as well as on the very day of the storm, the barometric column was rising, while the temperature was steadily falling. Here the rise in the barometer was accompanied by north-easterly winds, and the air at the earth's surface was thus rapidly mingled with cooler masses descending from above, as shown by the vane; so that the increased pressure was due to the increased density of the entire aërial column above the barometer,

* See Annales de Chimie et de Physique, tome lxxiv. p. 337; and a résumé of his results in a note to M. Martin's translation of Kaemtz's Meteorologie, p. 35.

+ Proceedings of the Royal Irish Academy, vol. iv. p. 279.

Atlantis, vol. iii, p. 166; Phil. Mag. for May 1860; and Proceedings R. I. A. for May 1861, p. 232,

Among the phenomena attending the more tranquil conditions of the air, I had noticed in my earlier observations, during the summer of 1857, that upward currents generally prevailed by day, while downward currents became more prominent at night. This alternation was manifestly connected, as shown by the horizontal vane, with the action of land and sea breezes; for at this time the observations were made at a point situated about two miles from the sea-shore. By day, the convection due to the heating of the lower stratum of air in contact with the ground could not take place by equal upward and downward exchanges of masses of air, because the place of the ascending warm air was partly supplied by the lateral influx of colder sea air, which, in its turn, would become sufficiently heated to ascend and give place to a fresh lateral influx. By night, the colder air from the land flowed towards the sea, and its place was filled by descending currents from above. At the same time the warmer air from the sea probably tended to occupy the place of these currents, and thus to equalize the temperature of the upper and lower strata of air so as to lessen the energy of the convective movement over the land.

Before the termination of the Meeting of the Association at Manchester, I had resolved, with the concurrence of Mr. Glaisher, the only other member of the Committee then present, to cause a registering instrument to be constructed which would record the existence of non-horizontal atmospheric motions. The following is a description of the anemoscope which I ultimately decided upon as most suitable in its construction for the purposes we have in view. Fig. 1 is a vertical section of the portion of the apparatus which is exposed to the wind, and fig. 3 an elevation of the same portion. A is a cast-iron pillar which supports a cup, h, containing friction-balls made of gunmetal; on these a disk, g, rests, and this is firmly attached to a box from which an arm projects at one side, and is terminated by the cone, P, which acts as a counterpoise for the opposite and working arm of the anemoscope. A short arm, n, shown in fig. 3, supports a wheel, d, in one side of which teeth are cut; the other side is firmly attached to a hollow light copper box, B, which forms the tail. This box is a truncated pyramid, and while its vertical sides are exposed to the horizontal action of the wind, its upper and lower surfaces are exposed to its vertical action. This tail is balanced by a counterpoise, i, which is connected by a bent arm with the axle of the wheel, d. The teeth of this wheel catch those of the pinion, e (fig. 1), and this catches in the rack, f. The rack is attached to a shaft, c, which descends through the hollow supporting pillar and communicates with the registering apparatus. In fig. 2 the most essential part of the arrangements for registering the indications of the upper part of the instrument are shown. The shaft, c, passes through brass guides, and carries a small circular projecting piece, s, which catches in a notch made in the bit, v, attached to the pencil-carrier, p. This pencil-carrier is capable of upward and downward motions only, and the rod to which it is attached passes through guides. The carrier is, moreover, supported by an ivory friction-wheel, t, which turns when the piece, s, revolves beneath it.

From this brief description, it is apparent that the cone, P, will always indicate the direction of the wind in azimuth, like ordinary vanes. At the same time the vertical component (if any) of the wind will raise or depress the tail, B. In the former case it is manifest that the wheel, d, will cause e to turn, so as to raise the rack, f, and in the latter case the effect will be to lower the rack. It follows, therefore, that the shaft, c, and consequently the pencilcarrier which it moves, must rise or fall according as the vertical motion of

the air is upward or downward. A spring within the pencil-carrier constantly presses the pencil against a sheet of paper placed in front of it. This paper is for the present carried on a flat board, which is moved by a clock. The registering sheets are ruled with vertical hour lines and with horizontal

lines which assist in estimating the angle of inclination to the horizon made by the disk during the action of an upward or downward impulse from the air. This follows because the tail and the wheel, d, revolve on the same centre, and each tooth in d describes an arc similar to that described by the axis of the tail. An equal number of teeth in e are raised or lowered, and thus the rack and the shaft, c, move through spaces proportional to arcs described by the teeth of the wheel, d, and the axis of the tail, B. The board