124 CONSTRUCTION AND MANAGEMENT OF SMALL GASWORKS. 



day that is now the general experience, there is some 

 difficulty in arranging for systematic tests, section by 

 section, all over the district, and especially as regards 

 leading mains. If done at all, it is worth doing well, as 

 tests taken in a hurried and perfunctory manner are worse 

 than useless. Stopping off sections by the use of bags or 

 bladders is not reliable, and there is still room for a 

 dependable stopper that could be used for this and other 

 purposes. If the pipe is old and dirty, there is great 

 difficulty in getting a perfectly tight plug. I have used 

 Lyon's diaphragm syphon boxes on a small district with 

 satisfactory results, and find them most effective for 

 isolating sections of main in a few minutes and reinstating 

 as promptly, without any cutting or screwing other than 

 removing the plugs. With this apparatus, tests can be 

 taken early on summer mornings, or during an hour 

 or two in the afternoons. 



In country districts, the consumption between mid- 

 night and 6 a.m. is so small that it can be closely 

 estimated, and in that case it is possible to get a fair idea 

 of the loss by leakage by noting the quantity sent out 

 during the small hours of a summer morning, say 2 a.m. 

 to 4 a.m., as obtained by reading the height of the gas- 

 holder. The use of a thermometer is necessary, to prevent 

 error from variations of temperature during the test. 



Some good work can be done in the way of discovering 

 leakage by the use of an indicator and pointed bar. Find 

 a joint to start with, by excavation, and then measure off 

 9 feet, or whatever the length of pipe may be, and drive 

 the bar close down to each joint. 



The actual loss each quarter should be carefully esti- 

 mated as nearly as practicable, and recorded for future 

 reference. A comparison with the corresponding quarter 



