April ii, 1895] 



NA TURE 



569 



per square foot for deadwoud and keels moving at I foot per 

 second, and ihis was c infirmed by the behaviour of the Saltan. 

 Accepting Mr. Kroude's formula, the extinciive effect due to 

 bilge keels, such as have been added to the Rrjuige, was 

 calcula'ed, with the result that, in an extreme case, supposing 

 the ship to he rollinu; to 20^ an each side of the vertical, the 

 extinction value due to the bilge keels would appear to be not 

 quite 2'. Thi-, it will be seen, is far shoit of the observed 

 results obtained from the vessels themselves. Working on the 

 data that have been obtained in this way, new coefficients have 

 been calculated for bilge keel resistance, it being assumed that 

 the whole increase of work were credited to the bilge keel area. 

 For an angle of swing of about 10' instead of the coelTicient 

 being l '6 lbs., it woul I be about it lbs for a swing of 4' the 

 coethcient would reach as high a value as 15 or 16. It muit be 

 remembered that these coefficients are not put forward as truly 

 representative, they only hold good if the assumptions stated 

 are accurate. 



In any case the dilTerence is very great. Mr. R. E. Froude, 

 who conlributcJ a most interesting speech to the discussion, 

 confessed that the results took him, when he first had them 

 put before him, entirely by surprise, and, indeed, he did not 

 credit the statements made as to the improved behaviour of the 

 ships ; or, rather, he could not attribute this improvement to 

 the presence of the bilge keels. We judged, however, from 

 his remarks tliat he now accepts the observed data and the 

 truth of the recorded experimental conditions, but still con- 

 siders the phenomenon one for which he can offer no a-iequate 

 explanation. He hi nself had made tank experiments which 

 agreed /airly well with the results obtained by his father, and 

 was quite at a loss to account for the great difference between 

 these experiments and the results of the trials niw recorded. 

 The only explanation he could suggest was that bilge keels, on 

 a rolling ship, meet on the return roll with water set in motion 

 by the previous roll ; but this, he thought, was quite insufficient 

 to account for an increase in resistance of as much as, say, ten 

 times thai which would be calculated on the i'6 coefficient. 

 Naval architects will be glad to hear that the whole question is 

 to be madethe subject of exhaustive inquiry at Hislar. The prin- 

 cipal reasons that bilge keels are not fitted — putting aside expense 

 and difficulties as to docking — are that they add to the immersed 

 surface, and are thus likely to decrease speed. It is, therefore, 

 satisfactory to learn that "the practical tests of actual service 

 prove there is no sensible reduction in speed for power." As 

 it is also -tated that the keels have not sensibly reduced diameter 

 of circles made by the vessels, and, further, that additional 

 steadiness m sttering has been obtained, it is not hazarding 

 much to say that in future ships of this class in the Royal Navy 

 will all be fitted with bilge keels, unless exigencies of docking 

 forbid their application. 



The space which we have devoted to Sir William White's 

 paper will compel us to dismiss most of the other contributions 

 briefly. Mr. Milton's paper on steam-pipes was an excellent 

 practical contribution, and was followed by a no less excellent 

 discussion. The general conclusion arrived at appeared to he 

 that , with high piesure, steel steam-pipes are likely to take 

 the place of those of copper. Mr. Taylor's paper was read in 

 brief abstract, and as it was not in the hands of members until 

 a few minutes before the meeting, we must pass it by. The 

 same thing may be said of Fruf. Thurston's paper. It is very 

 gratifying to the members of an English institution to receive 

 papers from foreign members of such eminence as the two 

 American gentlemen just mentioned. We regret we have not 

 yet been able to devote the time to their contributions which 

 their merits doubtless demand. Mr. Yarrow's paper on the 

 aluminium torpedo boat he had built for the French Government, 

 was a very interesting contribution. The boat appears to have 

 been thoroughly successful, so much so that she is to be the 

 prototype uf a class. The discussion turned largely on the form 

 of test pieces for copper alloys, it being generally conceded that 

 there is a want of standard conditions for tests. The micro- 

 sections of various alloys thrown on the screen were also very 

 interesting. 



The last day of the meeting (Fiiday) was devoted chietly to 

 the vibration question, the sitting proving one of the most 

 instructive of the series. .\> will be seen, three papers were 

 contributed on this important and interesting suliject. 



These ttiree papers on vibration of steamers formed, with Sir 

 William White's paper on steadiness, thet.>o distinciive features 

 of the meeting. It is hardly necessary to insist on the import- 



NO. 1328, VOL. 51] 



ance of these two features ia steamship performance, both 

 of which affect alike the comfort of the passenger in 

 mercantile vessels, and fighting efficiency in a war vessel. 

 Since engine piwer has increased so greatly and speeds 

 have been raised, the vibration question has become one 

 of extreme importance in passenger steamers. Two of the 

 most recent largest and costliest of our ocean liners were almost 

 unfitted for carrying passengers — it any rate, they were fast 

 acquiring an uneviable reputation — on account of excessive 

 vibration. By the application of scientific principles the cause 

 of this defect was traced, and the evil cured ; a circumstance, 

 if measured by money value, now worth many thousands of 

 pounds to the owners. The extremely interesting paper and 

 series of experiments performed two or three years ago by Mr. 

 Yarrow, at a meeting of this Institution, will be remembered by 

 our readers ; and since then llerr Schlick has read two papers 

 on the su'iject. Records of these will be foun I in previous 

 volumes of N.\TtJRE. The seriousness of vibration in steam 

 vessels is largely dependent upon the period of the hull as a 

 structure synchronising with the beats of the , engines, 

 and thus it is that a vessel may vibrate excessively 

 at speeds less than the highest speed she can attain. 

 That is the elementary fact upon which a study of 

 the problem is based. Herr Schlick, in his previous 

 papers, has already considere i the case of vibrations of the first 

 order — that is to say, such oscillations of the longitudinal axis of 

 a ship in a vertical direction as have two nodular points. 

 Vibrations of this order claim most attention because they are 

 most common, and are more violent than those of higher orders. 

 It is in vessels with engines running at high speeds that vibra- 

 tions of a higher order are sometimes observed. It would, as 

 the author of the paper points out, be very advantageous if the 

 naval architect could ascertain beforehand the position of the 

 nodular points of a ship in getting out the design ; but this, he 

 is of opinion, cannot be done directly in a satisfactory manner. 

 Mr. Mallock also enters into this question, as will be seen when 

 we deal with h s paper later on. As the question cannot be 

 treated direcily in a satisfactory manner in the case of a ship, 

 Herr Schlick has recourse to the mathematical investigations 

 of the vibrations of an elastic, prismatical rod. Such investi- 

 gations have been made by sever il authorities, and the author 

 quotes at some length the formula; that have been constructed 

 for vibrations of the first and higher orders. These it is not 

 necessary to repeat. 



It is evident that in a complex structure like the hull of a 

 vessel, the vibrations will be of a very different nature to those 

 of a prismatical rod. Treating only of vibrations of the first 

 order — for the author has not yet succeeded in correctly ascer- 

 taining coefficients for the second order — Herr Schlick finds 

 that in a ship they are at a greater distance from the ends than in a 

 vibrating prismatic rod ; a circumstance which is explained by the 

 fact that a ship is less weighted in the ends than in the middle. 

 For ships of very fine lines, the only class investigated, for 

 vibrations of the first order the distance of the after nodular 

 point from the after perpendicular is 0231 to 0253 times the 

 length of the ship. The distance of the fore modular point 

 from the fore perpendicular varies from o'jio to 0365 times 

 the length. Tlie author had already shown that an ordi- 

 nary engine with three cylinders cannot produce vibrations of 

 the first order when the moving weights (pistons, &c. ) of each 

 cylinder are in such proportions to each other that the products 

 obtained by multiplying these weights by the distance between 

 the axis of the cylinder and the next nodular point are the 

 same lor all three cylinders. The same engine, therefore, will 

 produce vibrations of the second order when the number 

 of the revolutions increases accordingly. The new nodular 

 point, moving away from the engine, causes the moments 

 of the moving weights to be no longer equal to each other. 

 The author considers, therefore, that as the nodular points can 

 only be determined after the ship is completed, it is necessary 

 to alter the moving weights of the engines in such a manner 

 that their moments respecting the nodular point are made equal 

 to each other. The vibrations will thus be considerably reduced, 

 if not entirely avoided. The influence of the screw in pro- 

 ducing vibration, o*ing to the impulses it imparts at the 

 exticme end, is also discussed in this part of the pnper ; and 

 the author then proceeds to deal with the so-called " horizontal 

 vibrations," which he considers really consist of a twisting action 

 on the ship's axis, due to the turning. moment of the engines, 

 acting on a screw, in the case of a single propeller, .^n 



