JOINT DISCUSSION ON TWO PRECEDING PAPERS. 111 
The propulsive coefficient at 5 knots was increased by 52 per cent when the propeller 
went back from the 1-inch position to the 6-inch position, and yet the shaft horse-power 
was only 9 per cent less; for a 52 per cent variation in the hull efficiency there was only a 
9 per cent variation in the power. At 6 knots a 35 per cent variation in the hull efficiency 
was accompanied by only a 9.5 per cent variation in the power. In going from the 6-inch 
position to the 30-inch position, the hull efficiency decreased about 10 per cent, but the shaft 
horse-power increased only about 3 per cent. 
I think that we need a little more light on the extent to which the wake fraction should 
enter into our expression for hull efficiency; i. e., whether it should be used to its full value 
or not. 
Also, referring to the question of the efficiency of the propeller in a disturbed stream, I 
notice that back in 1920, Prof. Abell, in the Engineer for September 24, wrote an article 
about some investigations which he made upon a propeller in a disturbed stream. He as- 
sumed a somewhat simple disturbance, but even so, he found from a purely theoretical con- 
sideration that the efficiency curve of a propeller in such a disturbed stream was somewhat 
different from the efficiency curve in the “open” condition. 
Perhaps some might ask just how these results of mine check up with those of Admiral 
Taylor, and I may say that I could not compare them very easily, because his propellers 
were tested at heights above the keel which corresponded to elevations of .2 and .6, while 
most of my results were grouped around an elevation of .4; so that it would require quite a 
bit of averaging in order to compare them. 
The fore and aft position which I used for my meter wheels was almost exactly half- 
way between position 2 and position 3 in his paper, considering position 1 as nearest the 
stern post and position 3 as the aftermost position. As far as I could compare them, they 
seemed to check in rather well. 
Now, replying to some of the questions on my own paper, first allow me to thank all of 
the gentlemen who have spoken so kindly of the results here given. 
I find that I have not stated anywhere in the paper just what I mean by the term 
“wake.” Since there are two definitions of this term, it may be well for me to state that 
I have used it in the sense that it is ordinarily used upon this side of the Atlantic, i. e., as 
the ratio of the speed of the following water to the speed of the ship. 
No attempt was made to explore the entire wake prism of the models but only that 
portion in which propellers ordinarily work. The wake prism varies in extent, dependent 
upon the fullness of the model. In the case of the fullest model, No. 6, the meter was car- 
ried below the surface a distance equal to the breadth of the model, and an appreciable wake 
of 2.5 per cent was registered. The meter was also carried out to one side a distance from 
the center line equal to the breadth of the model, and a wake of 1.5 per cent was registered. 
In the case of the 10-foot plane, however, the wake did not appear to extend below the 
bottom of the plane, as the wake became zero when the center of the meter wheel was below 
the bottom of the plane a distance equal to the radius of the meter wheel. 
A search of the wake prism in the fore and aft direction would probably reveal marked 
irregularities, as we would find different combinations of frictional wake, stream-line wake, 
and wave motion. In one case the meter protruded through the forward propeller post of a 
double-ended ferry, thus bringing the meter wheel into the position occupied by the forward 
propeller. In this case, a model which had a wake at the stern of about 33 per cent had a 
wake at the bow of 23 per cent. The meter wheel would probably have shown widely differ- 
