The following text may have been generated by Optical Character Recognition, with varying degrees of accuracy. Reader beware!

horsepower should be 4,300. That al- ters the water consumption." This is absolutely wrong in every particular. Four hundred and fifty was used in the computation for maximum horsepower and not 400 as he states. Moreover, it is evident that Mr. Anderson either does not figure out his 4,300 horsepowen or else he made a grievous error in his arith- metical work. The discrepancy is his, and the rate of water consumption would not be altered anyway because it is definitely stated that this rate is figured on average (3,747) and not on _ the (4,100) as he supposed. Thus, these three discuss'ons fall absolutely flat, the last two contain- ing not a single sound suggestion. It is no wonder 'that these gentlemen were misled. Many of the critcisms offered are worthy of careful. explanation, but these unfortunately are not the ones that attracted the most attention at the meeting. To the charge of incompleteness we must, in a measure, plead guilty for we are aware that very little data is given concerning the vessel, and none at all of the propeller or tur- bines, but after reading Mr. Fletcher's discussion the reason for these omis- sions ought to be apparent to every one, but these omissions do not in any way affect the accuracy of the test or of the results reported, nor does it make our observations any the less complete. He criticizes us se- verely for not giving the draught and trim of the vessel, but evidently he did not read the paper very carefully for the draught ts given on page four. As to preparations forthe test there is little to be said: The test was run on a regular trip of the boat, under service conditions, and it would hardly occur to any one of experi- ence that this bore any direct rela- tion to trial trip conditions. The boat had just been overhauled for the summer's work and the test was run on her second trip thereafter. Mr. Fletcher says, "Enginee-s in general will think it her best per- formance." Perhaps they will but if they do it is mo fault of ours, and Wwe cannot be responsible for . the Opinions of supposedly intelligent men who msconstrue plain English or who jump at erroneous conclusions. Engi- neers are supposed to be familiar with service conditions. Mr. Fletcher questions the evap- oration figures per pound of coal and endeavors to show that if these are horsepower maximum in error then the steam consumption "TAE MarINE REVIEW is in error. This is certainly an un- expected conclusion and one that is absolutely groundless and misleading. On reflection we are inclined to be- lieve that Mr. Fletcher is too good an engineer to be misled by his own conclusions, but for the benefit of others I want to say that the evapor- ation of those boilers is correct. I read the meter myself at the begin- 'ning and at the end of the test-and checked the readings at frequent in- tervals. The meter, moreover, was calibrated under my personal direc- tion and I know whereof I speak. Those boilers on the Governor Cobb evaporated 183,376 pounds of water in four hours. If one chooses to question the coal consumption he is at liberty to do so. I did not count the buckets myself but I made frequent trips to the boiler room and coal bunker. The method of getting coal is definitely stated, the result, therefore, is not misleading and the worst any fair minded man can say in regard to it is that the meth-- od is not one of precision, but even so it is 'the boiler test and not the turbine test that is affected by the coal. It is true that we could not twist the shaft to determine its modulus of elasticity but from a large number of tests it has been found that 11,600,000 is a fair mean, the variation either way being usually less than 3% per cent. Assumptions, as Mr, Fletcher says, may sometimes be wrong but he has no objection to accepting the records of the ordinary steam engine indicator which we know is always wrong and frequently subject to an error as serious as that. mentioned above. The log was not towed over a measured mile at a speed of 18 knots, it is true, but a progressive speed trial at lower speeds showed practi- cally a constant error. The log was towed from the end of a 20-ft. boom both times. It is inconceivable that the rate should hold, up to say 16 knots, and then suddenly fall off and we do not believe that it did, but right here let me say ithat the de- terminations of.speed and coal were not considered the essential features of the test. What we wanted to find out was the steam consumption of the turbine under- service conditions and we found it. Mr. Fletcher makes a point of the fact that our results do not agree with his own. He says that at 459.3. revo- lutions he obtained a speed of 18.12 knots, but he does not say how he. obtained the speed, nor does he give Sill any information by which one may judge of the probable accuracy of his result, and he fails to give the draught and trim. It is, therefore, difficult to compare the results, but it is hardly to be expected that the boat carried © cargo on hen trip from the builders' yard to Boston. At 459 revolutions our curve shows -a speed of about half a knot less than Mr. Fletcher gives, all of which could readily be accounted for by a difference of 1 ft. in the draught, and the boat at that time was fresh from the builders' hands. Mn, Fletcher has made a considerable demonstra- tion only to help prove in the end that our results are just what might have been expected under the existing conditions and that they check ad- m'rably with his own. Both Mr. Taylor and Mr. Fletcher object to the form of speed curve, but as the curve itself is no part of the original data every one is free to plot for himself. It is too much to expect that every point will fal on a fair curve, or that everybody would plot exactly the same curve, through those points. If the speed at 443 revolutions were half of one per cent less and at 463, one-half of one per cent greater, a fa'r curve could readily be put through these points of a shape doubtless more satisfactory to all concerned--but its departure from the original curve would be very slight and half of one per cent is a small matter any way and might exist even on a measured mile trial. If a straight line should be drawn from 'the origin of the curve on Plate II at. 45° to the base line, it would cut the 450 revolutions ordinate at 18 knots, and would pass fair through the three lower points of the curve. If there were no loss of propeller efficiency whatever this straight line would ishow 18.33. knots at 459 revolu- tions so Mr. that Fletcher's own figure, 18.12 knots, shows that the propeller begins to lose efficiency somewhere in the vicinity of 17 to 17% knots. If it is true that all our points are too low, then his curve would be more precipitous than the 45° line, and the falling 'off at 459 revolutions would be relatively greater and the loss of efficiency. would bégin earlier than 17 to 171%4 knots, perhaps as early as 16% knots and yet he criticises our curve severely because as he says it begins to fall off at about 17 knots. It is evident to us and must be to any one tthat this really is a very close agreement between