4i6 



NATURE 



[March "5, 1908 



side of the circle of Keswick. Tliese are doubtless 

 to be considered in relation to the direction of the 

 chambered cairn at Callernish. The rising- of the 

 Pleiades seems to have been in question. 



Still another stellar dolmen I measured in S. 

 Wales has already been referred to. 



NORM.iN LOCKYER. 



EXPERIMENTS ON SCREW PROPELLERS.' 

 'pHE screw-propeller was practically applied to 

 J- steamships by John Ericsson and Francis Petit 

 Smith about seventy years ago. It speedily became 

 ■1 formidable rival to the paddle-wheel. Long ago it 

 entn-ely superseded the latter for ocean navigation, 

 and in more recent years it has to a large extent 

 taken the place of the paddle, even in river steamers 

 of the shallowest draught. Accumulated experience 

 over this long period has proved of great advantage, 

 and has enabled naval architects and marine engineers 

 to meet new conditions in ships of much larger 

 dimensions and higher speed; but notwithstanding 

 this wealth of experience — largely based upon " pro- 

 gressive trials " of steamships and the analysis of the 

 results — it is still true that we are on the threshold of 

 exact know ledge in regard to the principles underlying 

 the efficiency of screw-propellers. 



Even in recent years, when the limits of experience 

 have had to be surpassed, thefe have been many proofs 

 of imperfect knowledge. On the whole, it is true 

 tliat success has been achieved, but not infrequently 

 as the result of numerous and sometimes costly ex- 

 periments on propellers of different forms. Perhaps 

 the most striking example of this general truth is to 

 be found in the case of torpedo vessels and motor 

 boats, driven at extraordinarily high speeds in propor- 

 tion to their dimensions; it is also true that, in vessels 

 of large size and of less speed in proportion to their 

 dimensions, remarkable results have been obtained 

 by a simple change of propellers. For instance, the 

 Drake class of cruiser in the Royal Navy, which are 

 the fastest cruisers afloat, had a guaranteed speed of 

 twenty-three knots on an eight hours' trial. The 

 guarantee was slightly exceeded in the first trials, 

 but there was evidence that the propellers became 

 relatively inefficient as the highest speeds were ap- 

 proached, and that the blade-area was insuflficient. 

 New propeller blades were made with greater blade 

 area, and with these the ship was driven at a speed 

 exceeding twenty-four knots, representing a gain of 

 about 25 per cent, in efficiency Obviouslv, incidents 

 of this nature point to the possibilitv of verv large 

 economies if our knowledge of screw-propeller action 

 and efficiency could be made more definite as well as 

 iiiore extensive. Trials in actual ships, especially 

 those of large size, are necessarilv costlv, and are 

 often impossible to make because the vessels are 

 required on service. Hence, at a verv earlv date, 

 attempts were made to introduce a system of experi- 

 ments with model screw-propellers, and from these 

 useful information was obtained. It was left for the 

 laie Mr. William Froude to perfect the method of 

 experiment in connection with his well-known system 

 of " tank " experiments on models of varving ship 

 forms; and his son, Mr. R. E. Froude, superin- 

 fendcnt of the :\dmiralty experimental tank at Haslar, 

 has carried on and developed the investigation so 

 far as the pressure of other and more urgent experi- 

 ments connected with the construction of ships for 

 the Royal Navy has permitted. 



The niodel propellers used by Prof. Durand were 

 forty-nine in number, of 12 inches diameter, with 



„} '' Reseir.-hes on the Performance of the Screw Propeller." By Prof. 

 W. F. Durand. Pp. 61. (Washington : Carnegie Institution, 1907.) 



NO. 2001, VOL. yy~\ 



bosses of uniform diameter (2'4 inches) ; all the models 

 had four blades, and all the blades were elliptical in 

 shape. Blade-areas and pitch-ratios were varied over 

 wide limits, going beyond the range of variation oc- 

 curring in actual practice. For example, the pitch- 

 ratios tried extended up to 2'i from o'g by differences 

 of o'2, and the blade-areas were carried down to 

 unusually small proportions of the disc area. Great 

 care was taken to shape the model screws truly and 

 to measure the pitch accurately. For each propeller 

 there was a determination of the power absorbed and 

 the thrust developed for a given number of revolutions 

 per minute, and a corresponding record of the speed of 

 advance in undisturbed water. Practically uniform 

 motion was ensured, and accurate measurements were 

 made of time, distance and force. From these experi- 

 mental data the actual and comparative efficiencies of 

 the model-screws were ascertained, and the percentages 

 of " slip " could be estimated. The facts are tabulated 

 and graphically illustrated in the memoir. They 

 require and deserve detailed study. In this brief 

 notice it is not possible even to mention the most 

 striking features. Prof. Durand briefly summarises 

 his conclusions in regard to the character of the 

 efficiencv-curves of the different model screws, and 

 supplements this section by a description of the 

 method he recommends for applying experimental 

 results to propeller design for actual ships. 



One cannot peruse this memoir without regretting 

 that, as vet, no British university, or public institution 

 primarily devoted to scientific work, possesses an 

 experimental tank such as is attached to Cornell 

 University, the University of Michigan, and to the 

 Technical High .School at Charlottenburg. Its value 

 for purposes of instruction is great; but its import- 

 ance as a means of research can hardly be over- 

 estimated. When tanks are closely associated with 

 the detail-work incidental to the design of actual 

 ships, the opportunities for research are less, and the 

 interruptions of research-work more numerous and 

 serious when undertaken in the intervals of ordinary 

 employment. In other words, research has to give 

 \\ay to urgent demands connected with ship-designs,, 

 and the special apparatus required for research has 

 to be removed or dismantled at short intervals. This 

 has been the experience at the .Admiralty tank, and 

 at the two tanks attached to the shipbuilding yards 

 at Dumbarton and Clydebank. .\- great need exists, 

 therefore, in this the greatest shipbuilding and ship- 

 owning country in the world, for an experimentaf 

 tank in which research work on ship-forms and pro- 

 pellers can be undertaken systematically and uninter- 

 ruptedly. This need has been recognised for a long 

 time. The Institution of Naval .Architects has made 

 efforts to interest ship-owners and ship-builders in the- 

 establishment of such a tank at the National Physical 

 Laborator)'. Considerable support has been obtained 

 from ship-builders and from a few ship-owners, but 

 hitherto it has not been possible to secure the whole 

 amount needed for the construction and equipment 

 of the tank, estimated at 15,000?., or for its mainten- 

 ance, estimated at 1500Z. a year. This failure is 

 greatly to be regretted, and is not creditable to the 

 community interested in shipping. It is certain that 

 the investigations made at such an establishment 

 would secure large economies and enable great ad- 

 vances to be made in the constructiori and propulsion 

 of ships. In connection with screw-propellers alone 

 there is a great opportunity for economies in coal- 

 consumption, the benefits of which would be secured 

 by ship-owners, and the amount of which in a single 

 year's operations of our immense mercantile marine- 

 would far exceed the cost of the research-tank. .Seeing 

 that the I'nited .States and Germany alread\- have ,'i' 



