

HYGROMETRIC RESEARCHES 535 



Glaisher's tables should be so markedly different from those based on Regnault's 

 formulas, for the factors rapidly increase in value below the freezing-point, and the 

 effect of wind under such conditions becomes much more noticeable. 



The wet bulb thermometer in an ordinary screen is an unsatisfactory instrument in 

 calm or almost calm weather, and to obtain reliable results, some means of causing air 

 currents to pass it must be devised. 



There are two ways of doing this, (a) by moving the thermometer through the air, 

 as de Saussure and Espy did, or (b) by moving the air past the thermometer, follow- 

 ing the example set by Belli in 1830. 



American meteorologists have discarded the stationary for the sling-psychrometer, and 

 the tables published in the new edition of the Smithsonian Tables are intended for the 

 reduction of observations made by swung instruments.* 



The psychrometric tables in the Smithsonian collection are calculated from a formula 

 of the late Professor Ferrel : — t 



/=/- 0-000660 (t-t')p {1+0-00115 {t-t')}, 



the symbols having the same meaning as in Regnault's formulae on page 531. 



The sling-psychrometer should be continuously whirled ' for one minute at the rate 

 of 8 metres per second, or for two minutes at the rate of 3 metres per second, in order 

 to attain the true temperature of evaporation.' J Under such circumstances, differences 

 due to the shape and size of the thermometer bulb also vanish.§ But in these in- 

 vestigations the temperature observations are usually made every minute, and sometimes 

 oftener, and only rarely every two minutes. And this could not be done were a sling- 

 psychrometer used, except by employing another observer. Therefore the aspiration- 

 psychrometer was chosen, as the clockwork was sufficiently powerful to keep the 

 ventilating fan in action for nine minutes, and a few turns of a key wound it up again 

 when nearly run down. 



A description of the aspiration- psychrometer of Professor Assmann of Berlin used 

 in the following experiments may be given. 



Each thermometer near the bulb is surrounded by double tubes nickelled inside and 

 outside, the outer tubes being bent round to join each other, and then form a single 

 tube which passes parallel to and between the two thermometer stems to a chamber 

 with a ventilating fan which is worked by superposed clockwork. The fan draws the 

 air past the bulbs at a rate of 2^ metres per second, and the clockwork requires to be 

 wound every nine or ten minutes to keep the air-current constant. The nickelled tubes 



* Swinging need not be in a vertical plane, but may be in a horizontal one. A simple arrangement may 

 be used whereby the thermometers may be mechanically turned in a horizontal plane round a vertical axis. 



t W. Fehrel. Annual Report of Chief Signal Officer for 1886, Appendix 24. 



X Cleveland Abbe — Meteorological Apparatus and Methods, p. 375. Annual Report of the Chief Signal 

 Officer for 1887, Appendix 46. Washington, 1888. 



§ Assmann. Paper cited above, i. 114-270. 



