CHAPTER XIII 

 UNDER-REGISTRATION OF METERS 



A thoroughly serviceable meter should have great sensitiveness, but need 

 not have a very high degree of accuracy. Emil Kuichling, Trans. Am. 

 Soc., C.E., Vol. XXV, p. 68, 1891. 



The physical difficulties of making a water meter that will 

 record all the water that passes it are apparently very much 

 greater than the corresponding difficulties in getting an accurate 

 gas meter or an accurate electric meter. 



There has been much discussion as to the probable average 

 amount of under-registration of meters in actual service. Actual 

 data are almost impossible to get. It would take a line of 

 specially designed experiments to throw light adequately upon 

 this matter, and as far as the writer knows, such experiments 

 have never been undertaken in any adequate way. 



Without being able to prove it, it is the writer's idea that 

 probably the under-registration of meters is the most important 

 of the reasons why the whole output of water cannot be accounted 

 for. The statement of Mr. Kuichling in regard to the conditions 

 of twenty years ago is still in the main directly applicable.* 



The average amount of water not accounted for as shown by 

 returns to the Committee on Meter Rates for completely metered 

 systems amounted, on an average, to 130 gallons per service 

 daily, as stated in Chapter XI, and of this only 32 gallons, 

 more or less, is fairly to be assigned to probable leakage from 

 street mains. 



Kinds of Water Meters. An illuminating statement as 

 to water meters was made by John Thomson,! a meter inventor 



* Chapter II, p. 8. 



t Trans. Am. Soc. C.E., Vol. XXV, p. 40, 1891. 

 140 



