68 



NATURE 



[_March 20, 191; 



mechanical gearing with gear-wheels suitably de- 

 signed. Careful investigations have been made "of the 

 causes producing noise in the gearing, and show that 

 the noise is due to slight inaccuracies in the teeth ; 

 it should be noted that the noise is an engine-room 

 noise only, and is not perceptible elsewhere. This 

 has led to a method of cutting- the gear-wheels, which 

 greatly reduces the errors involved in reproducing 

 the parent gear. Two rotating tables are used in the 

 new machine; the wheel to be cut is fixed to the 

 upper one, and is given a creep in advance of 1 per 

 cent, in relation to the motion of the lower table ; the 

 lower table is driven by worm-gearing at 1 per cent, 

 less speed than would be the case if a single table 

 were employed ; hence the wheel on which the teeth 

 are being cut has a motion compounded of the motion 

 of both tables, and equal to that required for the given 

 number of teeth to be cut. This device has the effect 

 of causing the errors in the teeth to lie in very oblique 

 spirals around the wheel, and also reduces the errors 

 themselves. In the actual machine, the errors are 

 reduced to about one-fifth of the original magnitude. 



Mr. W. Reavell contributed a paper on the use of 

 compressed air for working auxiliaries in ships pro- 

 pelled by internal-combustion engines. It is of in- 

 terest to note, in the operation of deck winches in 

 cargo steamers, that although steam at a pressure 

 of 90 lb. per sq. in. may be supplied, the 

 actual pressure demanded by the winches in 

 working did not exceed 16 lb. per sq. in. Earlier 

 attempts to deal with such cargo-hoisting problems 

 with high-pressure compressed air have been waste- 

 ful ; modern installations in which air at low pressure 

 is used for operating the winches have been success- 

 ful and economical. 



Baron A. Roenne contrasted the advantages and 

 disadvantages of airships and aeroplanes, and gave 

 a suggested design for an airship S53 ft. in length 

 and 72 ft. 3 in. in diameter, having a displacement 

 of 104 tons at o° C, and 760 mm. of mercury. A 

 speed of fifty-two miles per hour could be obtained 

 with 2000 h.p., and it should be possible to carry a 

 regular passenger service and to master the air on 

 almost every day of the year. 



In a paper on the longitudinal stability of skimmers 

 and hydro-aeroplanes, Mr. J. E. Steele states that the 

 most notable machine in the aeroplane show at Paris 

 this year from the point of view of inherent longi- 

 tudinal stability was one designed by M. Drzewiecki. 

 The principle embodied in this design is that of differ- 

 ence in pressure intensity on the forward and the 

 after curved planes, due to the different cross sections. 

 On the involuntary rising of the front part of the 

 machine, the increase in the angle of attack has quite 

 a different effect on the fore to what it has on the 

 rear plane. The pressure per sq. ft. on the front 

 plane is but very gradually increased for changes of 

 the angle of attack between the limits of 5 and 18°, 

 -whereas that on the after plane increases very rapidly 

 with the angle at which the wind meets it. The 

 result is an excess of lift aft, which restores the 

 machine to its original position. The converse holds 

 if the front of the machine is involuntarily depressed. 

 The reduction in the angle of attack leaves the pres- 

 sure on the front plane but slightlv altered, but 

 reduces quickly that on the rear plane, resulting in 

 a drop of that part to the normal position. 



Mr. G. S. Baker gives the first published account 

 of systematic research work carried out at the William 

 Froude tank at the National Physical Laboratory - . 

 The experiments had for their object the testing of 

 the effect upon the resistance of the ship of varying 

 the relative lengths of the entrance to run (i.e. those 

 portions of the bow and stern respectively which are 

 clear of the perfectly parallel midship bodv), main- 

 XO. 2264, VOL. 91] 



taining the same general form, water-line, and prin- 

 cipal dimensions. Five parent models have been 

 chosen, and with each of these, four or five propor- 

 tions of entrance and run have been tried. Another 

 set of experiments has been carried out with the view 

 of testing the effect upon model resistance of various 

 possible terminations to the lines, both in fore and 

 after body. The alterations tried have affected both 

 the area curve and the water-line, and, in addition, 

 the effect of the presence of the rudder has been tested 

 in one case. 



Mr. C. E. Inglis contributed a mathematical paper 

 dealing with the stresses in a plate due to the pre- 

 sence of cracks and sharp corners. Exact results are 

 obtained for the distribution of the stresses around 

 a hole in a plate, the hole being elliptic in form. 

 If the axes of the ellipse are equal, a circular hole 

 is obtained ; by making one axis very small the 

 stresses due to the existence of a fine straight crack 

 can be investigated. One of the several results ob- 

 tained may be quoted. A strip of plate of indefinitely 

 great width is pulled in the direction of its length, 

 the tensile stress intensity being R. There is an 

 elliptic hole in the plate having major and minor 

 axes, ia and 2b respectively, and arranged so that 

 the major axis is at right angles to the pulls. At the 

 edge of the hole situated at the extremity of the 

 major axis, a tensile stress occurs having an intensity 

 R(i + 2a/b). This stress decreases rapidly as w 7 e pro- 

 ceed along the section of the plate made by produc- 

 ing the major axis, and, at a short distance from the 

 edge of the hole, attains the normal value R. It will 

 be seen that the maximum value becomes very large 

 if b is made small; if u & = iooo, the maximum tensile 

 stress has a value of 2001 times the intensity of the 

 mean stress. In this case the ellipse would appear 

 as a fine straight crack, and a very small pull applied 

 to the plate across the crack would set up a tension 

 at the ends sufficient to start a tear in the material. 

 The increase in the length due to the tear exaggerates 

 the stress yet further, and the crack continues to 

 spread in the manner characteristic of cracks. 



A paper on the distribution of stress due to a rivet 

 in a plate, by Prof. E. G. Coker and W. A. Scoble, 

 is also of considerable interest. In a former paper 

 measurements have been described of the differences 

 of principal stresses at points in plates having notches 

 and holes of various kinds. In the majority of the 

 former cases, the stress distributions w-ere such that 

 the minor principal stresses vanished or were of little 

 importance. In many practical problems, both prin- 

 ripal stresses are of considerable magnitude, and it 

 is then important to obtain each stress separately. 

 The present paper describes a general method for 

 determining both the sum and the difference of the 

 principal stresses at a point in a plate, considered as 

 averages taken over the normal at the point, and 

 bounded bv the two faces of the plate. The stress 

 difference may be measured directlv by mechanical 

 or optical means, advantage being taken in the latter 

 method of the fact that plates of glass, celluloid, and 

 like transparent bodies, become temporarily doublv 

 refractive when stressed, and that in polarised light 

 there is, in consequence, a relative retardation, R, 

 between the ordinary and extraordinary ravs, which 

 is proportional to the stress difference, and to the 

 thickness T of the plate. If /,. and />, are the mag- 

 nitudes of the principal stresses, the law is given 

 verv approximately by R = (-'/,. /V)T, where c is an 

 optical constant. The sum of the principal stresses 

 may be determined by taking advantage of the fact 

 that a plate, when subjected to stresses in its own 

 plane, alters in thickness. Thus, if both stresses /,. 

 and p, are pulls, there is a lateral contraction of 

 amount (p +/>,)T'mE. where in is Poisson's ratio 



