266 W. R. HEBBLEWHITE. 



and required adjustment of the counter-weight for varia- 

 tions of the mean rate of flow, i.e., of the mean pressure. 

 The apparatus finally adopted (Fig. 5) is a modification of 

 the above and is automatic in its action. A spring replaces 

 the cord and counter-weight. Hence, the gasometer rises 

 under the air pressure until the tension of the spring has 

 relaxed sufficiently to give equilibrium. By choosing a 

 spring of very long range, a considerable variation in volume 

 may occur by rise or fall of the gasometer without any 

 appreciable change in the mean pressure of the air. The 

 efficiency of the apparatus is demonstrated by the sample 

 cards. 



The author desires to acknowledge his indebtedness to 

 Prof. S. H. Barraclough for his help in the construction of 

 the apparatus, and to Messrs. Parkinson and W. and B„ 

 Cowan for the use of their calibrating plant. 



