the ground only thirty-five minutes before. The temperature decreased rapidly at first, then moderately up to three miles, and above that it fell almost at the adiabatic rate. Even in this hot summer maximum of pressure and notwithstanding the maritime climate and south-westerly currents, a temperature about 29∞ below zero reigned at 27,000 feet, being only a few degrees warmer than Berson had observed in winter at the same height above Germany. Yet Glaisher, in all his ascents, two of which exceeded 26,000 feet, never recorded a temperature of less than 5∞ below zero. These relatively high temperatures, obtained also by Welsh, Tissandier, and Gay-Lussac, must be attributed to the insufficient protection of the thermometers against insolation, to the proximity of the instruments to the heated basket and its occupants, and lastly, to the sluggishness of the thermometers themselves, from lack of ventilation, during the rapid passage through air-strata of different temperatures. Plate VI. indicates the change of temperature with height observed during the four highest balloon ascents in Europe and in the United States. Dots indicate the observations while ascending, and crosses the observations while descending; these are connected by full and broken lines respectively, an inclination upward to the left showing a decrease of temperature with height and _vice vers‚_. The adiabatic lines, representing a fall of temperature of 1∞ Fahrenheit per 183 feet of ascent, serve for comparison. [Illustration: PLATE VI.--Temperatures observed in Four High Balloon Ascents.] This account of notable balloon ascents should not be closed without mentioning the most daring and unique of all, the voyage of Mr. S. A. AndrÈe towards the north pole in 1897. Although his was a voyage of geographical discovery, and not one for the exploration of the air, yet meteorological and other observations were to be made, and AndrÈe had familiarized himself with the instruments and the management of a balloon during several voyages in Sweden. The success of the polar voyage depended primarily upon the prevalence of southerly winds, and the ability of the balloon to keep afloat long enough to profit by them, even should they be light and variable at times. Therefore the impermeability of the balloon to hydrogen gas was of vital importance, and it was the conviction that the _Eagle_, of 140,000 cubic feet, was neither sufficiently large nor staunch to sustain itself for thirty days, the time which might be required to reach Behring Straits, that led Dr. Nils Ekholm, the meteorologist and physicist, to withdraw from the expedition. Unfortunately, his fears seem to have been well founded, and it is probable that we must now abandon hope of the safety of the brave AndrÈe and his two companions.