INSTRUCTIONS TO MARINE METEOROLOGICAIi OBSERVERS 25 



with the temperature of the air. The "temperature of evaporation" 

 is given by the reading of a tliermo meter wliose bulb is covered with 

 thin muslin and moistened with water. If two thermometers, one 

 with its bulb moistened, the other with its bulb completely dry, are 

 exposed to a rapidly moving current of air, the thermometer with 

 moistened bulb (the so-called wet-bulb thermometer) will show a 

 lower temperature than the dry bulb, or ordinary unmoistened ther- 

 mometer. The difference between the two bears a known relationship 

 to the humidity, the latter being obtained with the aid of appropriate 

 tables. 



Thermometers are also employed in obtaining the temperature of 

 the ocean surface. This is an important element in all meteorological 

 observations made at sea. 



Exposure of thennometers. — The free air is nearly always in mo- 

 tion and therefore is so thoroughly intermixed locally that within 

 a small area no appreciable temperature differences are likely to 

 exist. Wlien any portion of the air is confined, however, so that it 

 cannot intermingle freely with the general air currents, its tempera- 

 ture will be influenced to a marked extent by the local surround- 

 ings and wdll not be a free-air temperature. 



These ideas show us at once that if we aim to make accurate ob- 

 servations of the air temperatures our instruments must, if possible, 

 be placed in an open space where the circulation of the air is entirely 

 unobstructed. It will not do, however, to place the thermometers 

 simply in the open air, exposed freely to the sky and the direct rays 

 of the sun. The sunshine would cause the thermometers to register 

 too liigh, and even if the instrument were not exposed directly to 

 the sun it could not be depended upon to indicate the true air tem- 

 perature. To overcome these difficulties, it is customary to emplov 

 some sort of shelter. The form adopted by the Weather Bureau for 

 the use of its observers on land is a box with louvered sides and a 

 double roof, made in such a way that the air can move through it 

 with the greatest possible freedom. To minimize the effects of in- 

 solation and conduction of heat, shelters should be made of wood and 

 painted white. Tliis is an essential condition for proper exposure. 

 The object of the shelter is simply to screen off the direct and re- 

 flected sunshine and the radiation to and from the sky and to keep 

 the instruments dry. 



Although a thermometer shelter can easily be set up in a suitable 

 permanent position on land, experience has shown that on shipboard 

 it is virtually impossible to mount a slielter in a convenient, fixed 

 location in such a way that its exposure is not appreciably affected 

 at times (particularly in the case of a following wind) by "^radiation 

 from the deck or some nearby portion of the ship's superstructure 

 and by currents of warm air issuing from the interior of the vessel. 

 These undesirable effects may be eliminated to a large extent if use 

 is made of a portable shelter, which can be suspended on the weather 

 side before the observation is taken, in a position where it is freely 

 exposed to the wind and unaffected by artificial sources of heat. 



However,_ even a portable shelter has serious draw-backs. In the 

 first place, it possesses the undesirable feature common to all types 

 of shelters, namely, that it affords little, if any, protection from 

 flying spray. If it is allowed to stand continuously out of doors the 



