a greater portion of carbonic acid, as well as a greater volume of the mixed gases. From hard, brisk pump water I obtained, by boiling, quantities which, in round numbers, and for the specific gravity, varied but little, in repeated trials, from sixteen cubic inches of carbonic acid, and the same quantity of a mixture of azote with a small quantity of oxygen. The water of the Aire, taken from the cut which supplies the waterworks, gave two inches and three-quarters of carbonic acid, and eleven inches and three-quarters of azote and oxygen, the total quantity of gases being less than half that from pump water. From that of a large fish-pond I obtained more gas than from the river, but less carbonic acid, viz., two cubic inches and a quarter of carbonic acid, fourteen inches azote, and two oxygen. It is only within these very few years that carburetted hydrogen has been recognised in water. Its presence was first noticed, I believe, by Dr. Scudamore and Mr. Garden, in Harrogate sulphur-water. It is found accompanying sulphuretted hydrogen in every water which I have tried in which that gas occurs, and is disengaged from many springs in much greater quantity than the water can absorb, so as to form large bubbles. This phenomenon has been observed in many parts of the world, and the inflammability of the gas disengaged in such situations had been often noticed, but its exact nature has been in most cases rather inferred than proved. This circumstance of an inflammable gas, great part of which is carburetted hydrogen, issuing spontaneously from water, may be seen in several places in our neighbourhood. At Harrogate large bubbles occasionally rise through the water. At Stanley there is a continual flow of small bubbles; the difference depends upon the figure of the well or boring, and that of the passages through which it is supplied. At Slaithwaite the disengagement of gas is still more abundant, so that there is a succession of large bubbles, and the gas may easily be collected in considerable quantities, or set fire to at the surface of the water. The nature and amount of gaseous impregnation, though often of moment in medicinal waters, is almost immaterial for domestic purposes, with the single exception of water used unmixed as a beverage. The gases do not appear to interfere with the solvent properties of water, at least while cold, and when heated they are quickly disengaged. The changes which take place as to the gases, when brisk pump water is exposed in open vessels, are rather curious. I found that water yielding twenty-six cubic inches, viz., ten carbonic acid, and sixteen azote, &c., when fresh drawn, gave, after standing five hours, twenty-five inches; the diminution was in the azote, the carbonic acid remaining the same. At the end of nine hours, the total gases, twenty-two inches, onefourth of the azote had escaped, but very little, not two per cent., of the carbonic acid. After three days, however, the case was different; no further escape of azote had taken place, the water yielded about fifteen inches per gallon, and of this only from one and a half to two consisted of carbonic acid. The quantity of this gas was smaller than in river or pond water. If these experiments, which I have not had time to repeat, are tolerably correct, they would shew that, on exposure to the atmosphere, the azote and oxygen contained in water very soon begin to separate from it; that after a time the carbonic acid partially escapes, the other gases remaining; and that this continues until very little carbonic acid is left. The power of water to retain gases in solution depends, the temperature and pressure remaining the same, on the affinity of water for the gas, and upon the proportion of that gas in the superincumbent atmosphere. Those having a great affinity for water fly off in some degree when the gases above the water are wholly different, and those least readily absorbed are retained under an atmosphere of the same gas. Now, these two are antagonist principles in the case before us, azote having little affinity for water, but constituting four-fifths of the surrounding common air; carbonic acid being much more abundantly absorbed, but having no atmosphere of its own dessciption to press on the water containing it. No calculation could enable us, I think, to ascertain beforehand the order and the degree in which these effects would take place; that is, to predict the result of these experiments. GENERAL REMARKS ON THE WEATHER IN MADAGASCAR, AND CHIEFLY AT ITS CAPITAL, TANANARIVOU, From the 27th of June, 1828, till the 1st of January, 1829; with a Meteorological Journal from the 1st of January to the 25th of March, 1829. BY ROBERT LYALL, Esq. British Resident-Agent, Member of many Foreign and British [Communicated by Mr. J. F. DANIELL.] E arrived at Tamatave on the 27th June, 1828. During WE our residence there till the 11th of July, and of four days at Ivondrou, the weather was very warm, and much resembled that we had experienced at Mauritius before our departure. On the journey to Tananarivou, it continued very warm, even during the passage of the great forest, and until we crossed the river Mangoor: it then became gradually cooler; and, as it was cloudy and windy on traversing the mountain called Augave, (the height of which, above the level of the sea, may be five thousand feet,) it was even cold. Indeed, the change of climate was very remarkable; and the weather continued cold, not only on the road to the capital, but after our arrival in it. I entered Tananarivou on the 31st of August, on a beautiful morning, with a splendid sun. The weather continued very fine, and in the middle part of each day it was warm for a considerable period; but, as there was no rain, the mountains had a very barren and bleak appearance. East and south-east winds blew hard, almost every evening, and rendered it so cold, that, in slender houses, we were necessitated to have recourse to woollen clothes, to a small fire both morning and evening, and to blankets in the night. Excepting a few days, on which it was warm, (as the 12th of August, when his late majesty, Radama, was interred, and a few hours before and after mid-day,) the thermometer ranged from 50° to 60° of Fahrenheit for a considerable time. On the 17th of August it was windy, and so cold, that we put on our cloaks to go to church. The sympiesometer and the barometer were very little affected, and the medium altitude of the latter may be reckoned 25.32 inches. We had neither rain, nor storms, nor even any high winds. 6 The weather, about the end of August, became considerably warmer, and continued fine; the evenings and the mornings were beautiful and highly salubrious, being clear, dry, and cool. Afterwards the thermometrical range became higher, the temperature being generally between 60° and 70° Fahrenheit now and then, however, for a few hours after the middle of the day, it rose as high as 75° and 80°. Again, when the heat had been less intense in the day, it descended in the evening to 65° and 60° of Fahrenheit. The sympiesometer and the barometer seemed nearly stationary during September and October; and the wind was regular, at least every evening, and blew from the east and south-east. From about the beginning to the 22nd of November we had, now and then, a heavy shower, but no great quantity of rain fell; so that the Malagash government and people began to fear the loss of their rice crops, in consequence of long-continued drought. They had recourse to their idols, or gods, for assistance, as recorded in my journal of this period. On the afternoon of the 22nd, however, a day sooner than the gods predicted,' rain fell very copiously, and continued to do so all night, and even during a part of the 23rd. This date (though afterwards we had some fine days) may be reckoned the commencement of the rainy season in 1828. This remarkable epoch has therefore been late; but I have been told that it has occurred, though very rarely, that the rainy season has not commenced till January, which must always be a serious misfortune. It has been remarked, that the periodical rain has not followed the usual course-of falling in heavy showers, some time between the hours of two and six o'clock, P. M. On the contrary, it has frequently commenced earlier, and, more frequently, at a later hour; indeed, it has sometimes rained the whole night, and even in the morning and forenoon we have had heavy showers. Again, after heavy rain-falls, there have been periods of one day, of two days, and even of three days, very fine weather, and without a drop of rain; but with such heavy dews during the night, as, in the morning, led me to suppose that it had rained. Thunder-showers, which have been fre quent, have generally fallen in the afternoon. The lightning was vivid, and the thunder loud and near, so that a number of lives were lost by the former, in the capital and in its vicinity. Very often in the evening, and especially after thunder-storms, as in Russia, a great part, and even nearly the whole of the hemisphere was illuminated by that kind of lightning (called zara by the Russians) which flashes from cloud to cloud, but never approaches the earth, and by which lives, I believe, are never lost. About the end of November, or the beginning of December, at four o'clock, P. M., a very heavy shower, mixed with large hail, fell, to the astonishment of Mr. Chenard, the tutor of my children, who had often heard of, but had never seen such a ' phenomenon.' Ever since the rainy season has set in, with the exception of a few hours before and especially after noon, the heat has been very moderate. The barometer, comparatively speaking, has varied little; nor has the sympiesometer been greatly affected by the changes of weather. The wind, since the 22nd of November, has been more variable, and frequently from the north, north-west, and west. The quantity of rain which fell previous to the 22nd of November may be estimated at two inches, and that since the 22nd of November at about twelve inches-total, fourteen inches-till the commencement of the report for the month of January, 1829, which accompanies these observations. JOURNAL VOL. I. Ост. 1830. E |