OF ARTS AND SCIENCES. 47 



These results are also represented graphically upon Plate No. 4, 

 where it will be seen they may be expressed with a good degree 

 of approximation in terms of the mean velocity of the stream, as 

 follows : — 



At Station No. 2. 



West side — 0.0063 U 2 



East side + 0.0086 U 2 



At Station No. 6. 



West side — 0.0033 U 2 



East side + 0.0074 U 2 



At Side Instrument near Station No. 6. 

 Down-stream Tube .... — 0.0025 U 2 

 Up-stream Tube + 0.0046 U 2 



In these results, we see that with the orifice inclined down stream 

 the lowering of the reservoir below the surface of the stream is less 

 in amount than the raising of the reservoir above the surface of the 

 stream where the orifice is inclined up stream at the same angle ; the 

 former being in the three cases seventy-three, forty-four, and fifty- 

 four per cent of the latter. 



Orifices parallel. Passages normal. Tubes projecting. 



At Station No. 3, holes were bored at right angles with the plane 

 of the side, and into these were fitted iron pipes. On the west side, 

 the pipe being 0.0G8 ft. in diameter outside, and 0.052 ft. in diameter 

 inside, had a well finished end parallel with the plane of the side, 

 with square edges. On the east side, the pipe was 0.109 ft. in diameter 

 outside and 0.08G ft. in diameter inside, with end designed to be 

 finished like that upon the west side ; but it was not done with care, 

 and the up-stream edge was, at the close of the experiments, found to 

 project 0.0009 ft. farther into the stream than the down-stream edge, 

 and the top edge to project 0.0003 ft. more than the bottom edge. 



At first, both pipes were kept flush with, or without any projection 

 beyond the plane of the side, and afterward were pushed out into the 

 stream, as indicated in the table giving a summary of results at 

 this station. 



With velocities from 0.64 ft. to 2.77 ft. per sec, the westerly pipe 

 being flush with the side, the average height in the reservoir is the 

 same as that at the middle of the stream, within the practicable limits 

 of measurement. 



The same result obtains upon the east side up to velocities exceed- 



