AUTOMATIC RECORD OF PROPELLER ACTION. 257 
Mr. Everett :—In that case I would like merely to call attention to the cali- 
bration curve. It is quite an interesting curve, for the reason that the part that 
deals with half speed and lower occupies about 96 per cent. of the plate. Of course, 
in general, the part of the speed curve that is of interest is from half-speed up. 
From this plate, if we assume that the points lie exactly on the curve, the width of 
line of the curve itself permits an inaccuracy of about 3 per cent., in the determina- 
tion of speed, and further if one admits a deviation of the points from the curve, 
as they seem to have, of very slight amounts, possibly half of one of the coordinate 
spaces, that would correspond to an inaccuracy of five per cent. on speed. ‘This 
would be roughly about 16 per cent. on power, assuming that the power is propor- 
tional to the cube of the speed, and I would question the statement made by Mr. 
Emmet in the early part of the paper where he says: “As this curve indicates, the 
speed of the vessel through the water can be accurately determined from the marks 
on the paper down to a speed of about 150 feet per minute.”’ 
THE PRESIDENT:—Is there any further discussion? Mr. Emmet, have you 
anything to say in reply? 
Mr. Emmet:—I have not quite followed that last criticism. My meaning in 
stating that the speed can be determined down to 150 feet per minute had no refer- 
ence particularly to the scale of this curve. It had reference to the speed at which 
this curve bends. As long as the curve keeps reasonably straight, the action of the 
log is reasonable in proportion, but the characteristic of its action is unstable. How- 
ever, when the log begins to slip, then we could not trust it, and while the curve may 
be drawn to a bad scale, we should not regard it at all, because the regular speed is 
by dots which are spaced on a moving paper, and the spacing of these dots hasa 
very definite regard for the number of holes of the log. If you note the general 
characteristic of the log with relation to its motion in the water, we can determine 
the speed with accuracy. 
THE PRESIDENT :—If there is no further discussion, we will pass to the next 
paper, after recording the thanks of the Society to Mr. Emmet for his paper. 
The next two papers will be the last papers of this meeting. The authors are 
not present, and they will be read by title, in order to afford an opportunity for 
such discussion as may be offered. 
The first paper to be presented is entitled, “‘Ship Calculations; Derivation and 
Analysis of Methods,”’ by Naval Constructor T. G. Roberts, Member. 
