EXPLORATION OF ATMOSPHERE AT SEA BY KITES. 247 



of the atmosphere could often be made from a steamship pursuing- its 

 regular course, and accordingl}^ such were attempted on a steamer 

 eastward bound across the North Atlantic. With the aid of my assist- 

 ant, Mr. Sweetland. and through the courtes}^ of Captain McAuley, 

 this was accomplished on board the Dominion steamship Cominonwealth^ 

 which left Boston for Liverpool on August 28, 1901. A view of the 

 stern of the ship, with the upper deck from which the kites were 

 flown is shown in Plate 11, figure 4. During- most of the voyage 

 we were within an area of high barometric pressure that was 

 drifting slowly southeastward and out of which light winds blew. 

 Although these were insuflicient to raise the kites, the ship's speed of 

 16 knots created a corresponding wind from an easterly direction that 

 sufficed to lift the kites on five of the eight days occupied by the voj^age 

 to Queenstown. On one of the three unfavorable days, a following 

 wind became too light on the ship for kiteflying, and on the two other 

 daj^s a fresh head wind, augmented bv the forward motion of the ship, 

 was so strong as to endanger the kites, but had it been possible to alter 

 the course of the vessel a favorable resultant wind might have been 

 produced every day. The maximum height attained was only about 

 2,000 feet, but with larger kites and longer wire this could have Ijeen 

 greatly exceeded. Automatic records^ Avere obtained of barometric 

 pressure, air temperature, relative humidity, and wind velocity, which 

 did not difl'er markedl}^ from records obtained in somewhat analogous 

 weather conditions over the land. The most striking feature was the 

 rapid decrease of the temperature with increasing height in all but one 

 of the flights. The fall of temperature was fastest in the first 800 feet, 

 where it exceeded the adiabatic rate of 1° F. in 183 feet, but in the 

 last-mentioned flight the temperature rose 6^^ in 450 feet, and during 

 the afternoon remained so much warmer than at sea level. The rela- 

 tive humidity varied inversely with the temperature, the direction of 

 the wind shifted aloft toward the right hand when facing it, and its 

 velocity generally increased with increase of altitude. The direction 

 and velocit}^ of the wind aloft were computed from the observed 

 position of the kite and the recorded velocitj^ of the wind at this level, 

 allowing for the speed at which the kite was being dragged through 

 the air by the vessel. Simultaneous records were obtained from a 

 meteorograph hung above the deck, with which the upper-air records 

 were compared. 



These are probably the first meteorological observations at a con- 

 siderable height in mid-Atlantic, and have a special importance, because 

 they indicate that at sea high-level observations may be obtained with 

 kites in all weather conditions, only excepting severe gales, provided 

 the steamer from which the kites are flown can be so maneuvered as 

 to bring the wind to a suitable velocity. It is evident that such obser- 

 vations as have been described, even if made like the preceding, only 



