THE SMITHSONIAN INSTITUTION. 



267 



If, as in Fig. 8, we find 29 inclies no tenths 5 hundredths, and on 

 the vernier the second line above that marked 2, is found to coincide 

 with the scale, each of these short lines indicates 2 thousandths — con- 

 sequently are so counted ; the reading is therefore 29.074. 



Fig. 8. 



Or it may be, as in Fig. 9, where we have 30 inches 1 tenth, and 

 Ihe line on the vernier mark 3 coinciding nearly, but not perfectly, 

 with a line on the scale, it "is a little too high ; the 2 thousandth short 

 line next above is, however, a like quantity too low ; so the true 

 reading must be the number between them — that is, 1 thousandth, 

 making together 30.131. 



These examples include all the combinations the scale allows. A 

 little practice with the barometer with reference to the examples will 

 soon enable the learner to read off the scale with facility. At first it 

 will be best to write down the inches and parts in full, as in the dia- 

 grams, not trusting the memory with the whole until experience shall 

 have given confidence. 



Be careful never to lower the mercury in the cistern much below 

 the necessary quantity, as it increases the risk of air entering the 

 tube. 



When the barometer is to be removed for transportation or change 

 of position, before taking it down, the mercury is to be screwed up 

 17 



