214 



NA TURE 



[July 2, 1908 



given in full, as they form the most important part of 

 the paper. The loss for stalloy is somewhat lower than 

 that for the pure iron specimen above alluded to, while 

 that for lohys is slightly higher. 



The stalloy specimen requires careful attention in order 

 that a truly symmetrical hysteresis loop may be obtained, 

 more especially for values of B between 200 and 8000. 

 In an extreme case, after reducing the force H from 

 about 63 to 0-712 without subjecting the specimen to a 

 series of reversals of the magnetic force as it was reduced, 

 a complete hysteresis loop was obtained. This loop is 

 unsymmetrical in the sense that if the axis of H be so 

 placed that the coercive forces are equal, the positive and 

 negative values of the maximum induction B are not 

 equal, but the positive and negative values of the residual 

 magnetism are equal. The value of the permeability 

 defined as the ratio of half the total change of magnetic 

 induction to the maximum value of H is less than is the 

 case -when the loop is truly symmetrical. In the table the 

 figures for loops which are not quite symmetrical are 

 indicated by an asterisk. 



The Steinmetz coefficients have also been investigated, 

 the relation being ergs per cycle per cubic centimetre = 

 oB^. Both coefficients vary considerably. For stalloy the 

 coefficients are very nearly constant between values of B 

 of 600 and 11,000; over this range /8=i-7i and = 0000342. 

 For values of B from 0-937 to 825 the coefificient /3 is as 

 high as 2-69. In the case of lohys, between values of 

 500 and 8000 for B, the values /3=i.62 and = 0-00122 

 approximately hold. In this case also the coefficient 3 

 rises to a high value when B is small. 



Another matter investigated is the value of / HrfB/H(,Bo,„x , 



where H„ is the coercive force. Dr. Sumpner has pointed 

 out that this quantity is a linear function of B,,,,,^ over a 

 large range. For stalloy and lohys the relation only holds 

 apparently between values of B of 1000 and 9000. 



The specific resistance and temperature coeflicients were 

 obtained in the case of each of the materials. The follow- 

 ing figures are in each case the mean of the results of 

 three independent experiments : — 



Mean specific 



resistance at 



15* C. in io"6 ohms 



per c.cm. o^ to 50' C. o' to 100' C. 



Stalloy 49'63 ... 0-000975 •-- 000103 



Lohys I4'25 ... 0-00424 ... 0*00446 



It will be seen that stalloy has a high specific resist- 

 ance, which is important in connection with eddy current 

 loss, as this is thereby reduced. 



May 28.—" Effect of a Cross Wind on Rifled Projec- 

 tiles." By A. Mallock, F.R.S. 



The effect of wind on rifled projectiles is important for 

 practical reasons, especially in the case of small arms, but 

 the object of the present note is not so much to determine 



NO. 201S, VOL. 78] 



Mean temperature 

 coeilficienls 



the actual effect of wind as to show that accurate experi- 

 ments on the subject would afford valuable information 

 concerning the flight of projectiles in still air. 



It is easily shown that if the air resistance acts always 

 in the direction of the resultant of the velocities of the 

 wind and the projectile, the angle made by the resultant 

 velocity with the line of aim remains constant throughout 

 the range. 



In order, however, that the resistance may act in the 

 direction of the resultant velocity, the projectile must be 

 symmetrical about that direction. This, in the case of any 

 form except a sphere, means that the principal axis of the 

 projectile must take the direction of the resultant velocity. 



If this is assumed and we take x\ as the initial velocity 

 of the shot, tt) as the velocity of the wind (ui/ii„ being 

 small), and i\ as the coordinate of the shot perpendicular to 

 the line of aim, we have at the time t 



This result was first given by Captain Younghusband, 

 R.N., and would be correct if the axis of the projectile set 

 tself in the direction of the resultant velocity from the 

 ;ery beginning. 



kx first, however, the axis makes an angle it'/fo with 

 the velocity resultant, and the resistance has therefore a 

 horizontal component at right angles to that resultant, for 

 the same reason that a small angle between the axis of 

 the projectile and the tangent to the trajectory produces 

 an upward force on the former. 



The question, then, as to how far (1) may be looked on 

 as giving a true value for the effect of the wind turns on 

 the rate at which the projectile can set its axis in the 

 direction of the velocity resultant. 



It is shown, however, in a former paper,' that to pro- 

 duce a given angular velocity of the axis of a projectile 

 the couple must vary as the fourth power of the linear 

 dimension. 



For a given inclination of the axis to the direction of 

 motion the couple applied by action of the air will vary 

 as the cube of the linear dimension ; thus the angular 

 velocity of the axis will be inversely as the linear 

 dimension, or, in other words, the time for a given angle 

 will be as the linear dimensions. 



For a given inclination the lateral force will be as the 

 square of the linear dimension, and the distance to which 

 the lateral force will carrv the projectile while turning 

 through the angle U','z'„ will be proportional to the linear 

 dimension. 



Thus instead of the expression in (i) we should write 



71 = AL -I- n'(/ - R/i'o), (2) 



where L denotes the linear dimension and A some constant 

 depending on the form, weight, and initial velocity of the 

 projectile. 



If careful experiments were made on wind deflection, the 

 velocity of the wind being recorded at several positions 

 along the range at the instant that each shot was fired, 

 the value of A might be determined, and therefrom the 

 angle which the axis of a projectile fired in still air makes 

 with the tangent to the trajectory. 



Phvsical Society, Time 12. — Dr. Charles Chree, F.R.S., 

 president, in the chair.- — Experiments on a directive 

 system of wireless telegraphy : E. Bellini and A. Tosi. 



The authors describe the results obtained in the course 

 of their work upon a further development of their original 

 directive system. In the earlier method previously 

 described (Electrical Engineering, ii., p. 771, 1907, and 

 iii., p. 348, 1908) it was not possible to say from which 

 side of the receiving station the transmitted waves arrived, 

 for though the radiation was practically confined to the 

 plane of the aerial system, it was emitted equally in the 

 opposite direction to that desired. In the new unilateral 

 system the waves are sent in a single direction only, and 

 the problem of getting rid of the backwardly extending 

 radiation has thus been solved. The method adopted 

 consists in superposing a bilateral directive system, as 

 previously described, upon an ordinary or vertical antenna 

 system. The system of unilateral directive wireless tele- 



1 "The Behaviour of Rifled Projectiles in Air," Roy. Soc. Proc, vol. 

 Ixxix., p. 547. 



