Some fair idea of the conditions prevailing in the upper air may have been gathered from the many and various observations already recorded. Stating the case broadly, we may assert that the same atmospheric changes with which we are familiar at the level of the earth are to be found also at all accessible heights, equally extensive and equally sudden.

Standing on an open heath on a gusty day, we may often note the rhythmic buffeting of the wind, resembling the assault of rolling billows of air. The evidence of these billows has been actually traced far aloft in balloon travel, when aeronauts, looking down on a wind-swept surface of cloud, have observed this surface to be thrown into a series of rolls of vapour, which were but vast and veritable waves of air. The interval between successive crests of these waves has on one occasion been estimated at approximately half a mile. We have seen how these air streams sometimes hold wide and independent sway at different levels. We have seen, too, how they sometimes meet and mingle, not infrequently attended with electrical disturbance

Through broad drifts of air minor air streams would seem often literally to "thread" their way, breakng up into filaments or wandering rills of air. In the voyage across Salisbury Plain lately described, while the balloon was being carried with the more sluggish current, a number of small parachutes were dropped out at frequent intervals and carefully watched. These would commonly attend the balloon for a little while, until, getting into some minor air stream, they would suddenly and rapidly diverge at such wide angles as to suggest that crossing our actual course there were side paths, down which the smaller bodies became wafted.

On another occasion the writer met with strongly marked and altogether exceptional evidence of the vehemence and persistence of these minor aerial streamlets. It was on an occasion in April weather, when a heavy overcast sky blotted out the upper heavens. In the cloud levels the wind was somewhat sluggish, and for an hour we travelled at an average speed of a little over twenty miles an hour, never higher than 3,000 feet. At this point, while flying over Hertfordshire, we threw out sufficient ballast to cause the balloon to rise clear of the hazy lower air, and coming under the full influence of the sun, then in the meridian, we shot upwards at considerable speed, and soon attained an altitude of three miles. But for a considerable portion of this climb - while, in fact, we were ascending through little less than a mile of our upward course - we were assailed by impetuous cross currents, which whistled through car and rigging and smote us fairly on the cheek. It was altogether a novel experience, and the more remarkable from the fact that our main onward course was not appreciably diverted.

Then we got above these currents, and remained at our maximum level, while we floated, still at only a moderate speed, the length of a county. The descent then began, and once again, while we dropped through the same disturbed region, the same far-reaching and obtrusive cross-current assailed us. It was quite obvious that the vehement currents were too slender to tell largely upon the huge surface of the balloon, as it was being swept steadily onwards by the main wind, which never varied in direction from ground levels up to the greatest height attained.

This experience is but confirmation of the story of the wind told by the wind gauges on the Forth Bridge. Here the maximum pressure measured on the large gauge of 300 square feet is commonly considerably less than that on the smaller gauge, suggesting that the latter must be due to threads of air of limited area and high velocity.

Further and very valuable light is thrown on the peculiar ways of the wind, now being considered, by Professor Langley in the special researches of his to which reference has already been made. This eminent observer and mathematician, suspecting that the old-fashioned instruments, which only told what the wind had been doing every hour, or at best every minute, gave but a most imperfect record, constructed delicate gauges, which would respond to every impulse and give readings from second to second.

In this way he established the fact that the wind, far from being a body of even approximate uniformity, is under most ordinary conditions irregular almost beyond conception. Further, that the greater the speed the greater the fluctuations, so that a high wind has to be regarded as "air moving in a tumultuous mass," the velocity at one moment perhaps forty miles an hour, then diminishing to an almost instantaneous calm, and then resuming." In fact, in the very nature of the case, wind is not the result of one simple cause, but of an infinite number of impulses and changes, perhaps long passed, which are preserved in it, and which die only slowly away."

When we come to take observations of temperature we find the conditions in the atmosphere above us to be at first sight not a little complex, and altogether different in day and night hours. From observations already recorded in this volume - notably those of Gay Lussac, Welsh, and Glaisher - it has been made to appear that, in ascending into the sky in daytime, the temperature usually falls according to a general law; but there are found regions where the fall of temperature becomes arrested, such regions being commonly, though by no means invariably, associated with visible cloud. It is probable, however, that it would be more correct not to interpret the presence of cloud as causing manifestation of cold, but rather to regard the meeting of warm and cold currents as the cause of cloud.