[farmer] determination OF COEFFICTKNT OF DISCHARGE 49 



near!}' to the size required, and are then stam])ed Avitli a hardened steel 

 die. This stamp has been hardened and ground aecnrately to the proper 

 dimensions. The stamps used for making the ditîcrent orifices have been 

 carefully measured by means of a Brown and Sharp's micrometer, and 

 their dimensions recorded. In addition to this the orifices themselves 

 have been carefully measured up. This was done by means of a com- 

 parator Jiiade by the Waterville Iron Works of Waterville. Maine ; and a 

 standard scale. By this means. a discrepancy was detected between the 

 sizes of the punch and the orifice itself ; the latter l>eing the values used 

 in the calculations. 



The foregoing form the principal features of the tank itself, and are 

 those which enter into every class of experiment. For each particular 

 determination some further external apparatus is required, but this will 

 be described in counection Avith the particular series of experiments for 

 which it is iised. 



Determination of Coefficient of Discharcje. 



The first series of experiments made had for their object the deter- 

 mination of the coefficient of discharge, through various shaped orifices, 

 from a body of still water under various heads. 



This research was entered upon for two reasons. The first was that 

 it was hoped that with the a])paratus availal)le it would be possible to 

 obtain values on this important point which would be more consistent 

 and accurate than any hithertcj obtained. The second reason was that 

 the coefficient of discharge enters into almost every investigation on the 

 flow of water, and it was necessary to either make preliminary determin- 

 ations or to rely on previous determinations made under, possibly, difter- 

 ent circumstances ; and the first course was preferred. 



There have been a great number of determinations made of this 

 coefficient. Out of the numerous results given it was difficult to select 

 data which would be applicable as a basis of calculation in the experi- 

 ments it was desired to make, as the conditions of head and size of orifice 

 and the ratio between those two quantities in most cases differed widel}^ 

 from those under which the experiments were to be made. 



Of previous determinations the most applicable were undoubtedly 

 those actually made with the same apparatus by Prof. Bovey, and pub- 

 lished m his Treatise on Hydraulics. It was felt that those results might 

 be improved upon by alterations in some of the minor details of the 

 experiments, and thai it would be desirable to make a fresh series of 

 experiments, so as to obtain values for the coefficient of discharge, 

 through the large range of conditions allowed by the design of the appa- 

 ratus, even more accurately than had been done in the first instance. 



The chief alterations made in the present series of experiments were 

 the sul)stitution of an automatic time recorder for stop-watches, the alter- 



