6 



coincidence than the seventh. We, therefore, estimate that the true 

 reading is about 30.277. We might, probably, with as great accuracy 

 have selected 30.278. 



If the scale and vernier are accurately graduated, such readings 

 by a practiced observer will rarely be in error by more than 0.002 

 inch. It is important in estimating the fractions that the eye be 

 exactly in front of the lines being studied. 



14. In Figures 7 and 8 are shown verniers applied to a barometer 

 scale having 20 parts to the inch. In this case 24 parts on the scale 

 are divided into 25 parts on the vernier. By the principle already 

 explained in paragraph 12, the value of the subdivisions effected by 

 such a vernier, or, as it is most frequently expressed, the least count 

 of the vernier, will be -^ of ^^=5^ of an inch. In reading the 

 vernier, therefore, each line will represent 0.002 inch, so that the 

 fifth, tenth, fifteenth, twentieth, and twenty-fifth lines will repre- 

 sent one, two, three, four, and five hundredths of an inch, respec- 

 tively, and are so numbered. 



As described in paragraph 13, the lines in this kind of vernier also 

 may not be exactly in coincidence; but in such a case, owing to the 

 smallness of the spaces, it is not of any special advantage in making 

 our estimate to consider whether coincidence is nearer one line than 

 the other. In ordinary practice we simply take midway between. 

 Thus in Figure 8 the reading is between 30.176 and 30.178 ; we there- 

 fore adopt 30.177 as the proper reading. 



15. Caution agakist error. — ^When the zero line of this style of, 

 vernier is next above one of the shortest lines on the scale, as was the 

 case in the example above, some attention is necessary in order to 

 take off the correct reading. For example, in Figure 8 we find that 

 coincidence on the vernier is between lines designated 26 and 28, 

 which corresponds to a reading of 0.026 or 0.028, or, taking midway 

 between, 0.027. On the scale itself, however, we see the graduation 

 next below the first line of the vernier is 30,150. The complete read- 

 ing is found by adding the parts thus: 30.150+0.027=30.177. It 

 frequently happens with beginners that the 0.050 represented by the 

 short line on the scale is overlooked and omitted entirely — that is, 

 the above reading might be called 30.127. Whenever readings are 

 made with a scale and vernier of this character, special pains must 

 be taken not to omit adding 0.050 to the vernier reading when the 

 first line below the zero of the vernier is a short one. 



16. Strain on the cistern of a Jyarometer. — ^When the mercury is sent 

 up to the top of the tube of a barometer by screwing up the cistern, 

 an internal hydrostatic pressure is produced proportional to the 

 amount by which the length of the column has been increased. 

 This pressure tends to force the mercury through the joints of the 

 cistern or the joints and pores of the leather bag. This is more par- 

 ticularly the case with a barometer at an elevated station, where, 

 owing to the diminished air pressure, the column may need to be 

 raised 10 or 15 inches in filling the tube, greatly endangering the 

 cistern. For this reason the Fortin barometer cistern is not wholly 

 satisfactory, as it is difficult to make and keep the joints so tight that 

 the mercury will not be able to find its way through some very small 

 crevices ; such leakage soon impairs the barometer. 



