linder, and it is supplied with steam from the wrought iron-pipe B, and the air and condensed water are taken away by the pipe C. The admission of fresh air is to be regulated by the handle D. The steam occupies the space between the two cylinders at a, a, in the section. It is necessary that the cylinder should have an open top, as shown at E; and; as the height should not exceed about three feet, it is requisite to make the open work ornamental.

172. In other instances, I have used pipes joined in short lengths, nearly in the form of a distiller's worm, and placed an open trellis screen over them.

173. The proportion of pipe required to heat a given quantity of air per minute is easily calculated, by the following formula :•48 C (T-t)

the superficial content of sur

200-T face of steam vessel that will heat C cubic feet of air, from the temperature t to T, in one minute. Therefore, the quantity of ventilation and loss of heat per minute, being ascertained by the principles already given for ventilation, it is easy to provide the requisite supply of heat. It is assumed that the pipes are of cast iron, that being the fittest material, except for the small conveying pipe, which may be of wrought iron; but other surfaces will afford about the same quantity of heat, if of a dark color, and the surface a little rough and spongy a bronze color is very well adapted for the giving of heat. 174. In applying steam-heat, it should always

be made to warm that part of the air which is introduced for ventilation, before it enters the rooms; but it should only be warmed to a tem❤ perature a little lower than the mean temperature of the room; the proportion of pipe to be applied for that purpose our formula will show, and a register to regulate the quantity which enters, will put it in the power of any person to alter it at pleasure.

175. When steam-heat is employed in a dwelling-house, the distilled water is found useful for many domestic purposes; and, as the saving of fuel, by returning it to the boiler, is much less than it is imagined to be, not exceeding one-twelfth of the whole expenditure, the convenience of having this kind of water will be a sufficient compensation. For further information on this important subject, see Tredgold on Heating and Ventilating Buildings.

176. In the preceding account of the motion of air in conexion with ventilation we have but examined the tendency to a state of pneumatic equilibrium which is continually taking place on the large scale over every part of the terraquequs globe. It is by the aid of the winds that the earth's surface is ventilated, and the atmosphere fitted for respiration. If our globe were at rest, and the sun were always acting over the same part, the earth and air directly under it would become exceedingly heated, and the air would be constantly rising like oil in water, or like the smoke from a great fire; while there would be currents or winds in all directions below, towards the central spot. But the earth is constantly whirling under the sun, so that the whole middle region or equatorial belt may be called the sun's place; and therefore, according to the principle just laid down, there should be a constant rising of air over it, and constant currents from the two sides of it, or the north and south, to supply the ascent. Now this phenomenon is really going on, and has been going on since the beginning of the world, producing the steady winds of the northern and southern hemispheres, called trade winds, on which, on most places within thirty degrees of the equator, mariners reckon almost as confidently as on the rising of the sun himself.

177. According to the laws of statics, the air, which is less rarefied or expanded by heat, and consequently is more ponderous, must have a motion towards those parts of it which are more rarefied, and less ponderous, to bring it to an equilibrium; also, the presence of the sun continually shifting to the westward, that part towards which the air tends, by reason of the rarefaction made by his greatest meridian heat, is, with him, carried westward; and consequently the tendency of the whole body of the lower air is that way.

178. Thus a general easterly wind is formed, which being impressed upon the air of a vast ocean, the parts impel one the other, and so keep moving till the next return of the sun, by which so much of the motion as was lost is again restored; and thus the easterly wind is made perpetual.

179. From the same principle it follows that this easterly wind should, on the north side of the equator, be to the northward of the east, and in south latitudes to the southward of it; for,

near the line, the air is much more rarefied than at a greater distance from it; because the sun is twice in a year vertical there, and at no time distant above 231°; at which distance, the heat, being as the sine of the angle of incidence, is but little short of that of the perpendicular ray; whereas under the tropics, though the sun stay longer vertical, yet he is a long time 47° off, which is a kind of winter, in which the air so cools as that the summer heat cannot warm it to the same degree with that under the equator. Wherefore, the air towards the north and south being less rarefied than that in the middle, it follows, that from both sides it ought to tend towards the equator.

180. This motion, compounded with the former easterly wind, accounts for all the phenomena of the general trade-winds, which, if the whole surface of the globe was sea, would undoubtedly blow quite round the world, as they are found to do in the Atlantic and the Ethiopic oceans. But seeing that so great continents do interpose, and break the continuity of the ocean, regard must be had to the nature of the soil, and the position of the high mountains, which are the two principal causes of the variations of the wind from the former general rule; for if a country lying near the sun prove to be flat, sandy, and low land, such as the deserts of Libya are usually reported to be, the heat occasioned by the reflexion of the sun's beams, and the retention of it in the sand, is incredible to those who have not felt it; by which the air being exceedingly rarefied, it is necessary that the cooler and more dense air should run thitherwards, to restore the equilibrium. This is supposed to be the cause why, near the coast of Guinea, the wind always sets in upon the land, blowing westerly instead of easterly, there being sufficient reason to believe that the inland parts of Africa are prodigiously hot, since the northern borders of it were so very intemperate, as to give the ancients cause to conclude that all beyond the tropics was un inhabitable by excess of heat.

181. Mr. Clare, in his Motion of Fluids, p. 302, mentions a familiar experiment, that serves to illustrate this matter, as well as the alternate course of land and sea breezes. Fill a large dish with cold water, and in the middle of it place a water-plate, filled with warm water: the first will represent the ocean, the other an island, rarefying the air above it. Then holding a waxcandle over the cold water, blow it out, and the smoke will be seen, in a still place, to move toward the warm plate, and, rising over, it will point the course of the air (and also of vapor) from sea to land. And if the ambient water be warmed, and the plate filled with cold water, and the smoking wick of a candle held over the plate, the contrary will happen.

182. From the same cause it happens, that there are so constant calms in that same part of the ocean, called the rains; for this tract being placed in the middle, between the westerly winds blowing on the coast of Guinea, and the easterly trade winds blowing to the westward of it; the tendency of the air here is indifferent to either, and so stands in equilibrio between both;

and the weight of the incumbent atmosphere being diminished by the continual contrary winds blowing from hence, is the reason that the air here holds not the copious vapor it receives, but lets it fall in so frequent rains.

183. But, as the cold and dense air, by reason of its greater gravity, presses upon the hot and rarefied, it is demonstrable that this latter must ascend in a continued stream as fast as it rarefies; and that, being ascended, it must disperse itself to preserve the equilibrium; that is, by a contrary current, the upper air must move from those parts where the greatest heat is: so by a kind of circulation, the north-east trade-wind below will be attended with a south-westerly wind above; and the south-east with a northwest wind above.

184. That this is more than a bare conjecture, the almost instantaneous change of the wind to the opposite point, which is frequently found in passing the limits of the trade-winds, seems strongly to assure us; but that which above all confirms this hypothesis is the phenomenon of the monsoons, by this means most easily solved, and without it hardly explicable.

185. Supposing, therefore, such a circulation as above, it is to be considered, that to the northward of the Indian Ocean there is every where land, within the usual limits of the latitude of 30°; viz. Arabia, Persia, India, &c., which, for the same reason as the Mediterranean parts of Africa, are subject to insufferable heats when the sun is to the north, passing nearly vertical; but yet are temperate enough when the sun is removed towards the other tropic, because of a ridge of mountains at some distance within the land, said to be frequently in winter, covered with snow, over which the air, as it passes, must needs be much chilled. Hence it happens that the air coming, according to the general rule, out of the north-east, to the Indian Sea, is sometimes hotter, sometimes colder, than that which, by this circulation, is returned out of the southwest; and, by consequence, sometimes the under current, or wind, is from the north-east, sometimes from the south-west.

186. That this has no other cause is clear from the times in which these winds set, viz. in April; when the sun begins to warm these countries to the north, the south-west monsoons begin, and blow, during the heats, till October, when the sun being retired, and all things growing cooler northward, and the heat increasing to the south, the north-east winds enter, and blow all the winter till April again. And it is, undoubtedly, from the same principle, that to the southward of the equator, in part of the Indian Ocean, the north-west winds succeed the southeast, when the sun draws near the tropic of Capricorn. Philosophical Transactions, No. 183, or Abridgement, vol. ii. p. 139.

187. Some philosophers, dissatisfied with Dr. Halley's theory above recited, or not thinking it sufficient for explaining the various phenomena of the wind, have had recourse to another cause, viz. the gravitation of the earth and its atmosphere towards the sun and moon, to which the tides are confessedly owing.

188. From the laws of universal attraction it has been inferred that these celestial bodies must act upon the atmosphere, or that they must occasion a flux and reflux of the atmosphere, as well as of the ocean. Hence it has been alleged, that though we cannot discover aerial tides, of ebb or flow, by means of the barometer, because columns of air of unequal height, but different density, may have the same pressure or weight; yet the protuberance in the atmosphere, which is continually following the moon, must, they say, of course produce a motion in all parts, and so produce a wind more or less to every place, which, conspiring with or counteracted by the winds arising from other causes, makes them greater or less. Several dissertations to this purpose were published, on occasion of the subject proposed by the Academy of Sciences at Berlin, for the year 1746.

189. Although the atmospherical air is much more variable than water, and the action of the sun and moon upon it becomes much less apparent to us, because they must frequently concur with or be counteracted by the much more powerful effects of heat and cold, of dryness and inoisture, of winds, &c., so that their action upon the barometer has been long disputed and even denied, yet, that the moon in particular, as well as the sun, has such an action has been for a considerable time surmised; and of late years it has been in a degree observed and rendered sensible by means of very accurate and longcontinued barometrical observations, and perceived only by taking a mean of the observations of many years.

190. Toaldo, the learned astronomer of Padua, after a variety of observations made in the course of several years, found reason to assert that, ceteris paribus, at the time of the moon's apogeum, the mercury in the barometer rises the 0-105 of an inch higher than at the perigeum; that at the time of the quadratures the mercury stands 0.008 of an inch higher than at the time of the syzygies; and that it stands 0.022 of an inch higher when the moon in each lunation comes nearest to our zenith (meaning the zenith of Padua, where the observations were made), than when it goes farthest from it. Journal des Sciences Utiles.

191. In the seventh volume of the Philosophical Magazine there is a paper of L. Howard, esq., which contains several curious observations relative to this subject. This gentleman found, both from his own observations, and from an examination of the Meteorological Journal of the Royal Society, which is published annually in the Philosophical Transactions, that the moon had a manifest action upon the barometer. It appears,' he says, ' to me evident, that the atmosphere is subject to a periodical change of gravity, by which the barometer, on a mean of ten years, is depressed at least one-tenth of an inch while the moon is passing from the quarters to the full and new; and elevated, in the same proportion, during the return to the quarter. A great fall of the barometer generally takes place before high tides, especially at the time of new or full moon.

192. The causes, it is said, which render the

diurnal tide of the atmosphere insensible to us, may be the elasticity of the air, and the interference of the much more powerful effects of heat, cold, vapors, &c.

193. It has been calculated by D'Alembert, from the general theory of gravitation, that the influence of the sun and moon in their daily motions is sufficient to produce a continual east wind about the equator. So that, upon the whole, we may reckon three principal daily tides, viz. two arising from the attractions of the sun and moon, and the third from the heat of the sun alone: all which sometimes combine together, and form a prodigious tide.

194. In corroboration of the opinion of the influence of the sun, and principally of the moon, in the production of wind, we must likewise mention the observations of Bacon, Gassendi, Dampier, Halley, &c. ; namely, that the periods of the year most likely to have high winds are the two equinoxes; that storms are more frequent at the time of new and full moon, especially those new and full moons which happen about the equinoxes; that, at periods otherwise calm, a small breeze takes place at the time of high water; and that a small movement in the atmosphere is generally perceived a short time after the noon and the midnight of each day.

195. M. Muschenbroeck, however, will not allow that the attraction of the moon is the cause of the general wind; because the east wind does not follow the motion of the moon about the earth; for in that case there would be more than twenty-four changes, to which it would be subject in the course of a year instead of two. Introd. ad. Phil. Nat. vol. ii. p. 1102.

196. Some action in the production of wind may also be derived from volcanoes, fermentations, evaporations, and especially from the condensation of vapors; for we find that, in rainy weather, a considerable wind frequently precedes the approach of every single cloud, and that the wind subsides as soon as the cloud has passed over our zenith.

197. Wherever any of the above-mentioned causes are constantly more predominant, as the heat of the sun within the tropics, there a certain direction of the wind is more constant; and where different causes interfere at different and irregular periods, as in those places which are considerably distant from the torrid zone, there the winds are more changeable and uncertain.

198. In short, whatever disturbs the equilibrium of the atmosphere, viz. the equal density or quantity of air at equal distances from the surface of the earth; whatever accumulates the air in one place, and diminishes it in other places, must occasion a wind both in disturbing and in restoring that equilibrium, as above stated.

199. Mr. Henry Eeles, apprehending that the sun's rarefying of the air cannot simply be the cause of all the regular and irregular motions which we find in the atmosphere, ascribes them to another cause, viz. the ascent and descent of vapor and exhalation, attended by the electrical fire or fluid; and on this principle he has endeavoured to explain at large the general phenomena of the weather and barometer. Philosophical Transactions, vol. xlix. art. 25, p. 124.