-130 PROF. A. E. H. LOVE ON THE TRANSMISSION OF 



AUSTIN'S earlier work. The values of received antenna current, obtained from the 

 "smoothed curve of observations," are recorded, and compared with the results that 

 would be obtained from the author's formula ( 20 above), and with those that would 

 be obtained from a formula described as the " Sommerfeld transmission formula," 

 which is effectively equivalent to (57). I find that, in this case also, the ratios of the 

 quantities recorded as observed currents at different distances are nearly the same 

 as the square roots of the ratios of the magnetic forces at those distances, as 

 calculated from MACDONALD'S formula.] 



23. From this critical discussion I draw the inference that there is nothing in the 

 experimental evidence, at present available, to compel us to adopt the view that the 

 diffraction theory fails to account for the facts. On the contrary, that evidence can 

 be interpreted in such a way as to support the view that the results of the diffraction 

 theory accord well with those of daylight observations. Until more complete experi- 

 mental data are available the question of the success or failure of the diffraction 

 theory must remain open. 



However this question may ultimately be, settled, the discussion shows that it is 

 impossible to accept the hypothesis that the law of decrease of the forces of the 

 electromagnetic field with increasing distance is a combination of the law of spherical 

 divergence with a law of absorption, expressed by an attenuation factor of exponential 

 type. However closely such a law may represent the facts it can have no value 

 except as an empirical formula. In particular, it is not admissible to draw from it 

 any inference as to the amount of absorption. Independently of any absorption which 

 may exist, there must be a law of diffraction expressing attenuation of the field on 

 account of the curvature of the earth's surface ; and the type of attenuation factor 

 required to express the effect of curvature Avoukl not differ much from the exponential 

 type in a moderate range of great distances. 



The investigations of SOMMERFELD and those of this paper throw some light on the 

 effect of the resistance of the medium, over the surface of which signals are trans- 

 mitted. It appears that, with such wave-lengths as are used in practice, a moderate 

 amount of resistance, such as that of sea-water, increases the strength of the signals at 

 great distances. A slightly higher resistance would increase it still more, but no 

 conclusion can safely be drawn as to the effect of so high a resistance as that of dry 

 ground. It seems to me, however, to be not unlikely that even a rather high 

 resistance may be favourable, and I am inclined to regard the known fact that 

 signals are in general appreciable at greater distances over sea than over land, as an 

 effect of the broken surface of ground covered with rocks, buildings, or trees. This 

 question also remains to be settled by quantitative experimental investigation. 



Another difficult question presents itself in the known fact that the signals are 

 generally stronger by night than by day, arid the related fact that the attenuation of 

 night signals by distance is sometimes less than it would be if they diminished simply 



