TEN THOUSAND DEADWEIGHT-TON TANKER. 321 



The wind thus had approximately four times as much retarding effect as aiding 

 effect. 



The M. E. P. curves feel only tlie wind effect, as the tidal currents move the whole body 

 of water in which the ship is floating. From the above, therefore, we may approximate the 

 location of the corrected M. E. P. curves at runs 2 and 3 as about one-fifth the distance be- 

 tween the curves drawn with and against the wind. At higher speeds the wind dropped and 

 the ratio of wind effects was less ; at lower speeds the wind was stronger and the ratio was 

 larger. In this manner the curve of still air M. E. Ps. was drawn. 



The next step was to make corresponding corrections on the two speed curves. The 

 simplest explanation of the method is an example: The M. E. P. at 60 R. P. M. against 

 the wind is 18.5 pounds; in still air this value would occur at slightly over 62 R. P. M. 

 Therefore, from the "against" curve, the speed at 62 + R. P. M. can be read and plotted as 

 corrected for wind on 60 R. P. M. The corrections for the "with" curve were made in a 

 similar manner but are decreases instead of increases. The two speed cur\res corrected for 

 wind can now be drawn. 



As the runs were made at fairly imiform intervals, the most accurate way of correct- 

 ing for the tide is by halving the differences in speed between the "with" and "against" curves 

 corrected for wind. 



The curves w^ere then replotted, using speeds as the abscissae. From the corrected curves 

 of speed on R. P. M. and M. E. P. on R. P. M. we can read for any speed, R. P. M. and 

 M. E. P. : thus the I. H. P. at V knots can be computed. 



FINAL CURVES. 



The final power and standardization curves are given on Plate 65, together with curves 

 of apparent slip and Admiralty coefficient derived therefrom. Table I, page 322, gives 

 the obsen'ed and resultant data. 



Plate 63, already referred to, shows the curve of effective horse-power derived by towing 

 the model at the U. S. Experimental Model Basin at Washington. The curve is for a slightly 

 larger displacement (salt water) than the trial (fresh water) and was corrected to trial dis- 

 placement. The difference was small, and the reduction was made directly proportional to 



E H P 

 the displacements. The ratio ' ' " gives the curve of propulsive coefficients also shown 



i . ±1. P. 



on Plate 65. 



CHECK ON WIND CORRECTION. 



The curve of M. E. P. on R. P. M. is independent of tidal or current effects and is af- 

 fected only by the force of the wind. This being so, it should be possible to obtain from the 

 curves the actual magnitude of the wind force on the ship. The no-wind curve is (subject 

 only to variations in slip and propulsive efficiency) a direct measure of resistance of the ship, 

 so that at the same speed the difference between the curves for "still air" and "against the 

 wind" will be the M. E. P. used to overcome the force of the wind. As the M. E. P. curves 

 are plotted on R. P. Ms., however, it is necesary to obtain the effect first from the speed 

 curves in terms of R. P. Ms. (for constant speed), then for these R. P. Ms. to read the 

 corresponding M. E. Ps. For example, we already know the corrections to the speed curves 

 to allow for wind effect, and by adding and subtracting these to the final or no-wind curve, we 

 can derive curves of speed on R. P. M. "with" and "against" the wind but without tidal effect. 



