554 
Analyses of Boohs. 
[September, 
Roorhee Hydraulic Experiments. By Capt. Allan Cunningham, 
R.E. Vol. I., Text; Vol. II., Tables; Vol. III., Plates. 
Roorkee : Thomason College Press. 
The rate at which water is discharged through channels, natural 
or artificial, is a matter of no small practical importance ; yet, 
strange to say, it has been so little the subject: of thorough- 
going investigation that engineers of well-known ability differ 
widely from each other if called in to determine the average flow 
of water passing along a stream or a drain. Instances of such 
discrepancy have frequently occurred to us when experimenting 
on the chemical treatment of sewage. Having ascertained the 
working formula for say ioo gallons, came the question what is 
the average fair-weather flow from the town concerned ; and here 
the results have varied widely. The flow from one town has 
been stated to us at every figure from 8 to 14 million gallons 
daily. 
The author finds that experiments on large rivers can seldom 
be executed with the precision necessary for discovering the 
general laws of fluid motion. Operations on a very small scale, 
on the other hand, give results not trustworthy if applied to 
larger bodies of water. The best opportunity for a fundamental 
study of the question is afforded by large canals, — such as the 
irrigation channels of India, — where large volumes of water are 
in uniform motion in regular channels. 
The chief objedls of the experiments conducted by Captain 
Cunningham and his assistants were the discovery of a good 
method of discharge measurement, an investigation of the appli- 
cability of known formulae for mean velocity, and the discovery 
of a good approximation to mean velocity. 
The chief results reached are that loaded tube-rods give a 
rapid and sufficiently close approach to mean velocity past a 
vertical. The author considers that for depths not exceeding 
15 feet they should supersede all other instruments. Rods made 
of tin tubing were found preferable to all others ; they require to 
be as thin as is consistent with rigidity, with a smooth surface. 
The centre of gravity should be as low as possible in the water, 
and the convex surface smooth throughout. 
Capt. Cunningham finds that discharge-measurements, when 
carefully executed on his system, do not vary respectively more 
than 3 per cent. This conclusion is based upon a careful com- 
parison of repeated measurements, and must be pronounced 
satisfactory. 
As regards the existing formulae for mean velocity, none of 
them seems generally applicable. That of Kutter, which is the 
most general, gives an approximation much below the results of 
direCt measurement. This conclusion the author admits as dis- 
appointing, especially as he is net able to propose a preferable 
