184 DISTRIBUTION OF GAS THROUGH MAINS. 



bend substituted for the quadrant ; the quantity delivered in the hour was 

 2824, a difference of -g^th of the first discharge. 



In Mr. Clegg's Journal I have the results of some experiments on the same 

 subject ; they serve to show that the obstruction to the flow of the gas is very 

 nearly as the number of bends, two semicircular bends making twice the dif- 

 ference, three bends three times, and so on. One semicircular and one right- 

 angle bend forms an obstruction equal to the sum of the two, or -fa + -fe ; 

 and this seems to be the same under all variation of circumstances. 



These results may be explained by the tendency of fluids to flow in right 

 lines and the obstructions formed, by the bends causing a reaction in the 

 contrary direction to the flow ; thus friction is increased, and the velocity 

 decreased in the same ratio. In pipes of large diameter the effects of bends 

 will not be so much felt, because the surface will bear a less proportion to the 

 area than in those of smaller diameter, and consequently the friction is not 

 so considerable. 



I have been favoured with the following, by Mr. G. A. Jermyn, whose 

 abilities in the surveying of towns for the purposes of distributing mains, have 

 tended in many instances greatly to facilitate the operations in the first 

 instance, and the regulation of the works afterwards : 



" In lighting towns with coal-gas, there is no branch more worthy consideration than 

 the arrangement of the mains, their ramifications, and the apparatus connected imme- 

 diately with them, in order that every part may be in perfect keeping with the size of 

 the magazine from which they draw their supply. To effect this a plan of the town 

 should be obtained; and not only is a plan giving the lengths and widths of streets 

 necessary, but their respective levels are indispensable, for upon this latter point many 

 circumstances will depend. Having ascertained the number of lights required in each 

 particular street, the main must be in accordance not only as to the value of ' diameter,' 

 but as to the density of supply likely to be obtained from its peculiar level, taking the 

 gas-holder curb always as a datum. Errors entailing a permanent expense and incon- 

 venience have been the result of the operations of most of the great companies, from 

 this simple point having been neglected, and many irregularities have come under my 

 own observation ; for instance, in a street of half a mile in length, where one, or at the 

 most two receivers, would have sufficed to drain the pipes, not less than twelve have 

 been discovered when taking up the main, and this was undoubtedly in consequence of 

 not having a proper map of reference as to where receivers had been laid down ; the 

 extra ten vessels would have covered the expense of making a map. 



" Experience has made manifest how necessary it is to have a perfect record of every 



