278 



NATURE 



[November 13, 1919 



Surface-Tension. 



Owing to surface-tension, a surface of mercury 

 supports easily a sovereign placed flat upon it. Care 

 must, of course, he taken to avoid amalgamation. 



I shall be greatly obliged if one of your readers will 

 supply me with a 'formula for determining the size of 

 the largest sphere of gold that can just be supported 

 by mercury. As the numerical solution of the equa- 

 tion mav be troublesome, I venture to ask only for 

 the formula. C. T. Whitmell. 



Hyde Park, Leeds, November 3. 



Exceptional Dryness of October, 1919. 



Meteorologists have directed attention to the 

 exceptional dryness of the past October. It is also 

 interesting to note that the amount of drainage-water 

 percolating through 20 in., 40 in., and 60 in. of soil 

 in the open field for the month of October as recorded 

 bv the Rothamsted E.xperimental Station gauges is ml. 

 The three gauges, each measuring i/iooo acre, were 

 built in 1870, and in no previous year is October shown 

 quite dry, 1897 being the nearest with a reading of 

 0001 in.' The following are the figures for October : — 



2oiD. 4oin. 60 in. 



gauge gauge gauge 



1-848 1798 1-669 



5589 



nil 

 nil 

 The 50-year records show that October is one of the 

 four months when the ground is wettest. 



W. D. Christmas. 

 Lawes Agricultural Trust, Rothamsted 

 Experimental Station, Harpenden, 

 November 6. 



Average of 50 years 



Max. 1891 



Min. 1897 



1919 



40111. 



gauge 



1-798 

 5716 

 0001 



nil 



5479 



0001 



nil 



Rainfall 

 3233 

 6764 

 0960 



1073 



SOUND RANGING. 



SOUND ranging consists in the location of the 

 source of a sound, such as the report of a 

 gun, by means of measurements made on the 

 sound-wave which spreads from the source. 

 When it seemed probable, in the latter part of 

 1914, that the struggle in Prance was going to 

 develop into trench warfare, the possibility of 

 locating enemy batteries by this means was recog- 

 nised, and many experi-nents were started inde- 

 pendently to find a method of sound ranging 

 which could be used in the field. 



Suppose that there is a gun at the point S in 

 Fig. I. The report of the gun spreads as a spherical 

 sound-wave, with a uniform velocity, and is re- 

 ceived by stations at A, B, and C. If the time 

 intervals between the arrival of the sound at A, B, 

 and C are measured, a very simple construction 

 gives the position of the gun. For instance, if the 

 sound gets to B a time *i after it gets to A, 

 and to C a time t^ after it gets to A, circles are 

 described around B and C the radii of which are 

 equal to the distances travelled by sound in times 

 tj and tj respectively. If a circle is found which 

 passes through A, and touches the circles around 

 B and C, the gun position will be at its centre. 

 Therefore, by installing a series of observation 

 stations along the front at surveyed positions, 

 and recording the times at which the report arrives 

 at these stations, it is possible to plot the position 

 of the enemy battery on a map on which the 

 NO. 261 1, VOL. 104] 



observation stations are marked. This is the 

 essential idea underlying sound ranging. Three 

 stations only are necessary, but more may be 

 employed in order to confirm the location. 



There are other ways of plotting the gun posi- 

 tion, given the time intervals. For instance, if 

 the time interval between A and B is tj, the gun 

 must lie on a hyperbola with foci at A and B 

 which is such that the difference in the distances 

 from the foci of any point on the curve is Vt^, 

 where V is the velocity of sound. Another pair 

 of stations give another hyperbola, and by finding 

 where this intersects the first the gun position is 

 determined. This was the" method actually cm- 

 ployed on the plotting-boards used by the sound- 

 ranging sections. The hyperbola approximates 

 so closely to its asymptote near the gun position 

 that the asymptote can be used equally well, 

 which makes the method a very simple one in 

 practice. 



The French Army started experiments in sound 





 C 



ranging 

 showed 



n 1 91 4, and obtained results which 

 that the method was a promising one. 

 From the very beginning development took place 

 along two lines. Either observers were used, who 

 recorded the time of arrival of the sound by press- 

 ing a key, or the sound was registered automatic- 

 ally by some form of microphone. In both cases 

 the stations were connected electrically to a 

 central station, where the signals sent by the 

 observers or microphones were registered on a 

 chronograph of some form. It was soon found 

 that observers were not sufTiciently accurate. They 

 made errors amounting to one-tenth of a second, 

 whereas it is necessary to time the arrival of a 

 sound to 0005 second in order to make a satis- 

 factory location. This accuracy was attained by 

 the system in which the arrival of the sound was 

 registered by a microphone, and both in the 

 French .'Vrmy and ours a microphone system was 

 finally adopted. 



