45 

 V. GENERiVL Instructions 



(A) FOR CARE AND USE OF BAROMETERS 



89. Exposure of harometers. — The two important considerations in 

 selecting a proper location for a barometer are (1) that the cistern 

 and top of the mercurial column may be in a good light, and (2) that 

 the temperature may be as constant as possible. The best conditions 

 for light are obtained when the barometer can be placed between the 

 observer and a window, preferably a north one, covered either with 

 tissue paper or fitted with ground glass. Very nearly as good results 

 are obtained by a light from one side reflected from clean white 

 paper or white glass immediately back of the barometer. The top 

 of the column should be about the height of the observer's eye. The 

 barometer should not be exposed either to the direct rays ot the sun 

 or to the air currents that are always found in the vicinity of cracks 

 and crevices in windows. 



In establishing stations officials will use special care in selecting 

 the exposure of the barometers and satisfy the conditions stated above 

 as nearly as possible. In general, it will be necessary to avoid expo- 

 sures near windows, as proper temperature conditions can not be 

 found in such locations. 



As houses, no matter how tightly built, always permit the free flow 

 of air in and out through crevices, ventilators, chimneys, etc., it 

 results that the air pressure within is exactly the same as without, 

 except possibly for very slight differences of very short duration. If 

 such were not the case it would be necessary to expose barometers 

 out of doors to obtain the real air pressure. 



89a. Pumping of 'barometers. — Notwithstanding what has just 

 been said about the pressure indoors and out being the same during 

 very windy, gusty, weather barometers within doors are often sub- 

 jected to very rapid and irregular oscillations of pressure, caused by 

 gusts of wind blowing into doorways, windows, or chimneys, and 

 momentarily increasing the pressure, or by blowing across chimney 

 tops and otherwise, so as to produce a sort of suction that momen- 

 tarily diminishes the general pressure. In consequence of these 

 effects the mercurial column of a barometer may be observed on. 

 such occasions to rise and fall irregularly within narrow limits, the 

 motion in many cases being little more than changes in the curvature 

 of the meniscus. This action is called the " pumping " of barometers, 

 and, of course, interferes with accurate pressure observations. 



The term " pumping " is also applied to much more violent oscil- 

 lations of the mercurial column, such, for instance, as will occur 

 when an ordinary barometer is exposed on a vessel at sea, or when 

 carried in an upright position in the hand. In barometers for use on 

 shipboard this action is prevented by making the lower portion of 

 the glass tube of very fine bore, so that the movement of the mer- 

 cury is necessarily too slow to follow sudden and irregular oscilla- 

 tions of pressure. 



90. Verticality of harmneters. — For accurate results it is necessary 

 that barometers should be exactly vertical when the adjustments for 

 reading are made. For this purpose the better forms of barometers 

 are arranged to be suspended from rings at the top, so that the in- 



