154 JOHN M. SMAIL. 
On the first installation by the aid of thirteen induct and thirty- 
five exhaust shafts, on the average about 39,000 cubic feet of air 
per hour was thrown in and exhausted from the sewers. The 
results from the shafts being so satisfactory, the system has been 
extended to almost every borough which has a foul water system. 
The form of cowls, both induct and exhaust, has been very 
much improved since the inception of the works. Exhaustive 
tests were made at the cowl testing station at Crown Street. 
Reservoir, the competition being very keen between the various 
makers. The different designs are shown by the models exhibited, — 
representing the best forms. The system adopted in testing 
was at first by anemometers, but the results were so conflicting, 
that the efficiency of the cowl was afterwards tested by the 
work done in exhausting or forcing in air in proportion to the 
wind power. This was accomplished by sealing the bottom of 
the test tube and connecting a Thorp’s Automatic Pressure 
Register with it. The instrument was started and allowed to run 
twenty-four hours, the clock was surrounded by a diagram upon 
which was registered by a pen the amount of pressure or exhaust 
which the cowl exerted on the instrument in inches of water. The 
diagrams were then forwarded to the Government Astronomer, 
Mr. H. C. Russell, r.r.s., who kindly supplied the wind pressures. 
for each hour during the day. A further diagram was made from 
the register, and the pressures due to the cowl and those due to the 
wind were plotted on either side of a datum line, this represented 
two figures with twenty-four ordinates. As the area of the figure 
is clearly the base or datum line multiplied by the mean ordinate 
or mean pressure, it follows that the area of the figure representing 
the work of the cowl is proportional to the pressure exerted by 
the wind. The wind power was taken as the base, and the effici- 
ency of the cowl was determined by the ratio which the work done 
by the latter bore to the former. 
That the shafts are doing more work with the cowls than they 
would otherwise do, is shewn by the records of the tests made 
during the year. It was found that with a third class cowl on a 
ft 
