114 POWER DISTRIBUTION FOR ELECTRIC RAILROADS. 
4 
The net saving in cost of copper is then $60,500. In lay- 
ing out the feeders for an actual road these figures would 
doubtless be somewhat modified by the conditions of traffic, 
but the general condition is a saving of about $60,000 in 
first cost as against an extra yearly expenditure of over 
$14,000 in power. In the average case there would be a 
strong tendency to use the two stations. ‘There would be 
just so much less money to raise, the $14,000 would 
dwindle under the deft fingers of the bookkeeper and 
growth of the road would soon compel the use of two sta- 
tions anyway. On the other hand the skillful use of a 
boosting system might cut the extra expenditure in two 
and the single plant would be by far the more economical. 
The questions regarding probable growth involve very 
close judgments, and local conditions, such as cost of real 
estate and nearness to coal and water supply, may often 
properly turn the scale one way or the other. As between 
a single 1000 k. w. plant, and two 500 k. w. stations, there 
would be no doubt as to the propriety of installing the 
latter, while with a less aggregate than 500 k. w. capacity, 
the single station would very often be preferable. ‘The 
longer the line and the ‘greater the aggregate output the 
greater the advantage of using several generating stations. 
In most lines of twenty-five or thirty miles in length local 
or terminal demands for power will indicate the use of a 
pair of stations by raising the aggregate power or increas- 
ing the average distance to which power would have to be 
transmitted from a single station. Now and then how- 
ever exactly the sort of conditions set forth in the estimate 
Aare encountered and a single station is desirable. In 
strictly interurban work the suburban traffic near the ends 
of the line is almost certain to make the substation plan 
the cheaper. Such is naturally the case in the roads 
shown on the map (Fig. 62). 
On very long lines the question is still further com- 
plicated, for the distances may readily be too great to work 
easily even with two stations, and if the traffic is not unu- 
sually great the output of each station may be rather small