112 JOINT DISCUSSION ON TWO PRECEDING PAPERS. 
ing wake values between these two positions dependent upon the location of the wheel rela- 
tive to the wave crest or wave hollow. 
A change in speed will cause some difference in the location of wave crests along the 
length of the model, and this probably accounts for the fact that in fine models the wake 
values usually fall off with increase in speed above the normal, while in full models there is a 
rapid increase in the wake value as the speed is approached where a hump appears upon the 
resistance curve. 
If I may be allowed to emphasize certain points which I have made in the paper, it may 
answer some of the questions which have been asked. I have tried to bring out the varia- 
tion in wake which accompanies variation in draught. The wake decreases quite rapidly as 
the propeller is placed nearer the keel, with a fixed draught; and decreases less rapidly as the 
draught is increased with the propeller at a fixed height above the keel. One of the conse- 
quences of this is that the increase in apparent slip which is usually noticed as a ship is loaded 
down to a deeper draught may not indicate any material increase in the true slip. 
Another point which I have tried to bring out is the effect of fullness of form upon 
wake. Fig. 26 shows that the relation of wake to form fullness cannot be represented by 
a straight line, but is a curve which becomes very steep in the region of the full forms if 
plotted upon a longitudinal coefficient, and less steep if plotted upon a vertical coefficient. 
In Luke’s 1910 paper, Fig. 1 shows a number of wake values obtained from models with 
a block coefficient of less than .70. A straight line can be used here to represent fairly closely 
the variation of wake with fullness of form, but this same line cannot be used for the wake 
of fuller models. Furthermore, the ratios of breadth of ship to draught, of diameter of pro- 
peller to draught, and of elevation of propeller to draught should be known, and it would 
be found, probably, that a series of lines or curves would have to be used to give the relation 
between wake and fulness of form. 
Fig. 27 shows that the wake is very turbulent with full forms and raises the question 
as to the accuracy of the wake values derived from self-propelled models upon the assump- 
tion that the efficiency of the propeller is the same in the “behind” condition as in the “open” 
condition. 
No general statement can be made regarding the effect of bossing upon the wake, as this 
effect varies with the diameter and immersion of the wheel. If we can set up some system 
which will give the wake on the center line fairly accurately, I believe that the wake off the 
center line, both in the naked and bossed condition, can be expressed as a percentage of the 
center-line wake. 
As pointed out in the body of the paper, the wake values off the center line were ob- 
tained with in-turning wheels, so that the results would be expected to check more nearly 
with those obtained from in-turning propellers. 
The curve for the wake of the 10-foot plane in Fig. 27 seems to differ in character from 
those obtained for the models. This is due to the fact that the flow around the plane is 
probably two-dimensional in character, while that around the models is three-dimensional. 
A study of Fig. 27 will lead to the conclusion that the curves in Fig. 26 may be consid- 
erably modified for full models as the wheel diameter is reduced. A cross-section of the 
curves in Fig. 27 for a wheel elevation of .4 would show the wake for model 6 to be less than 
the wakes for models 3 or 5, when the diameter of the wheel is 334 inches instead of from 
5 to 6 inches as in Table IV. 
Prof. Chapman and Mr. Smith refer to the prismatic coefficient of the after body as a 
