48 REPORT OF THE SECRETARY. 



wliicli continua.lly flows througii the liighor regions of tlie atmospliere 

 over tlie wliole United States and wliich at intervals reaches tLe 

 surface of the earth, deposits its moisture, obtained from the Pacific, 

 on the Sierra Nevadii and other chains of the Rocky Mountain system, 

 and thus gives rise to the arid basins and woodless plains of the western 

 portion of our domain. A part, however, of this region may be rendered 

 arable by means of irrigation, or, in other words, a i)ortion of the pre- 

 cipitated vapor may be, as it were, reclaimed for agricultural purposes, 

 by artificial means ; and in determining the available resources of the 

 country it is important that the amount of water, either in the liquid 

 state or in the form of snow, which is precipitated in the mountains, 

 should at least be approximately determined, and hence we have directed 

 special inquiries to this point, supplying rain-gauges in all cases of 

 explorations and giving definite directions as to ascertaining the depth 

 of snow, the gauging of streams, &c. 



Much has been written on the subject of reclaiming arid wastes by 

 planting trees, but the facts which have been collected in regard to this 

 matter are frequently misinterj)reted, and the result of much research 

 in reference to it misapplied. It should be recollected that trees cannot 

 grow without moisture and that they have no power within themselves 

 to create this essential element of their existence. It is .true that in 

 some eases where moisture exists but is retained in an aeriform condi- 

 tion by the radiating character of the soil, or carried away by the wind, it 

 may be precipitated and rendered applicable to the uses of agriculture 

 by the judicious planting of trees; while it is equally true that there are 

 other localities in which the necessary water for vegetation can never be 

 l^rocured by artificial means. The necessary data for the investigation of 

 this question in relation to the western portion of the great valley of 

 the Mississippi can only be obtained by extensive series of observations 

 on the relative humidity and the direction of the moisture-bearing winds 

 of different portions of the country. 



During the past year a self-registering barometrical apijaratus, in- 

 vented by Professor George W. Hough, of the Dudley Observatory, has 

 been erected in the Institution, but owing to the position in which it was 

 first placed the series of records has not been continuous, though from 

 the records which have been obtained it would appear that the instru- 

 ment is capable of giving valuable results. This instrument consists of 

 an iron tube in the form of a siphon closed at the u^Ji^er end and filled 

 with mercury, and two cylinders or drums moved by clock work, to 

 the surface of which is attached the paper receiving the register. On 

 the surface of the mercury, in the shorter leg of the siphon, is an ivory 

 float, fastened by a line thread to one end of a lever, the other end of 

 which terminates between two points connecting the opposite poles of a 

 galvanic battery. As soon as the mercury begins to rise and while 

 it continues to ascend, the end of the lever is "pressed against the upper 

 point, thus completing the circuit of an electro-magnet, which unlocks 



