5'4 



NATURE 



[December i6, 1920 



is devoted. By measuring the distribution of pres- 

 sure on the surface of a cylinder with its axis at 

 right angles to the wind-stream in a tunnel, Mr. 

 Dryden shows that the decrease of pressure at the 

 back of cylinders of small diameter is greater in pro- 

 portion than the increase in front, and that, in con- 

 sequence, the distribution of velocity is not similar 

 about cylinders which are "dynamically similar" — 

 that is, for which the product of the diameter, the 

 speed, and the reciprocal of the dynamical viscosity 

 of the wind has the same value. After showing that 

 the size of the wind-tunnel, the effects of the guards, 

 and of the gap between them and the cylinder cannot 

 explain the difference, Mr. Dryden concludes that it is 

 due to some unknown property of the air in the 

 vortex motion behind the cylinder. He proposes to 

 test this by taking photographs of this portion of the 

 moving air. 



The second Report of the British Scientific Instru- 

 ment Research Association, covering the work of the 

 association from July i, 1919, to June 30, 1920, is 

 interesting reading. It is a striking example of the 

 importance of applying scientific research and know- 

 ledge to the comparatively small things in life. The 

 staff of the association, whilst preparing for a 

 number of important researches, has been carrying 

 out a series of investigations which, whilst small 

 in themselves, are of importance to the scientific 

 instrument industry. A polishing powder and some 

 abrasives have been developed which have desirable 

 properties, and which, it is hoped, will be manufac- 

 tured commercially in this country. A research into 

 tissue-papers has resulted in a specification which will 

 enable opticians to obtain paper in which optical glass 

 may be wrapped without tarnishing the glass. The 

 association has produced a solder, fusing at a tem- 

 f)erature of 195° C, capable of being used with 

 aluminium, and this in itself is no small achievement. 

 The report impresses the reader that every effort is 

 being made to get into close touch and co-operation 

 with other research organisations, such as the 

 National Physical Laboratory, and at the same time 

 the requirements of the users of scientific instruments 

 are being continually studied. The future work of 

 the association will be eagerly awaited by all makers 

 and users of scientific instruments. 



Three papers on notched-bar impact tests were 

 read at the Institution of Civil Engineers on Novem- 

 ber 30. The paper by Messrs. T. E. Stanton and 

 R. G. C. Batson gives particulars of a series of 

 tests made at the National Physical Laboratory with 

 the view of investigating (a) the effect on the work 

 of fracture and the consistency of the results obtained 

 of a variation in the angle and sharpness of the notch 

 in 10 mm. by 10 mm. specimens, and (b) the sensi- 

 tivity of the various types of V notch in revealing 

 faulty heat treatment. It was found that the shape 

 of the bottom of the notch has an important effect, 

 and the work of fracture has least value when the 

 angle at the bottom of the notch is as nearly zero 

 as can be obtained. Variations in the results of 

 individual tests are not inherent in the method of 

 test, but are due to a lack of homogeneity in the 

 NO. 2668, VOL. 106] 



material. M. Charpy has found that by taking 

 extraordinary care in heat treatment to ensure homo- 

 geneity it is possible to obtain a degree of uniformity 

 in the results of notched-bar tests which is higher 

 than any other mechanical test to which the material 

 can be subjected. The National Physical Laboratory 

 tests indicate that the 10 mm. by 10 mm. specimen 

 developed by the Aeronautical Insoection Directorate, 

 having a 45° notch with a radius of 025 mm. at the 

 bottom, is as effective in the detection of faulty 

 heat treatment as the Charpy specimen. Other 

 tests on the dimensional effect with specimens of 

 varying sizes indicate that the value of the impact 

 test, as at present understood, lies not in dis- 

 criminating between the impact resistances of different 

 materials, but as a means of ensuring that the impact 

 strength of any given material is at its highest. The 

 paper by Messrs. R. H. Greaves and H. Moore also 

 deals with notch radii, and that bv Messrs. R. M. 

 Jones and R. H. Greaves discusses tests made with 

 the view of investigating the effect of overstrain. 



The use of wire-rope conveyors of the class having 

 more than one rope has hitherto been restricted. 

 This is owing to the fact that it has been the practice 

 to attach the ends of the slats or their mechanical 

 equivalents to separate ropes. Since it is not possible 

 to splice separate endless ropes so that they shall all 

 be of exactly the same length, strains are set up in 

 the working which are fatal to success. This objec- 

 tion has been obviated in the Roe cable conveyor — 

 described by Mr. G. F. Zimmcr in Engineering for 

 November 19 — by fixing one end only of the slats 

 or transverse carrier-bars to one endless rope, and 

 the other end merely rests upon the other rope or 

 ropes without being attached thereto. Thus two or 

 more endless ropes of differing lengths m^ be used 

 to provide a satisfactory support for the conveyor 

 slats. In the conveyor illustrated in the article the 

 total length is 1800 ft., and the difference in altitude 

 between loading and unloading points is 15 ft. Ten 

 brake-horse-power is required to drive the conveyor 

 under full load at 150 ft. per minute. It is of interest 

 to note that the invention owes its existence to the 

 requirements of the War Office. 



Some particulars of the canal now under construc- 

 tion by the Hydro-Electric Power Commission of 

 Ontario, Canada, are given in Engineering for 

 December 3. The canal, which is said to be the 

 largest for water-power purposes in North America, 

 is about 85 miles long, and involves the removal of 

 about 19,000,000 cubic yards of earth and rock. 

 Water will be taken along the Welland River for 

 about 45 miles from Chippawa, and thence through 

 the canal proper to Queenston at the edge of the cliff 

 marking the limit of the Niagara River Gorge. The 

 power-house will be located at Queenston, where the 

 net head available will be 305 ft. The actual differ- 

 ence in level between Lakes Ontario and Erie is 

 326 ft., and the present power plants at Niagara 

 utilise heads of 150 ft. to 200 ft., which is the height 

 of the falls alone. The additional fall in the new 

 scheme is obtained by using a site beyond the rapids. 

 The canal is designed to carry sufficient water to 

 generate more than 400,000 h.p. 



