378 Dr. Balth. van der Pol on the Propagation of 



The experimental data published up till now are very few 

 in number. Practically all the work in this field has been 

 carried out by Austin *, who measured the received signal in 

 a telephone- receiver by shunting the latter down to such an 

 extent that dots and dashes could just be discriminated, and 

 then the value of: this shunt gave an indication of the 

 amplitude of the wave at the receiving spot. 



Now the numerical relations entering in the action of 

 crystal detectors and thermionic valves are not yet com- 

 pletely known, and a certain variety of explanations of their 

 working has been given, while a definite numerical theory of 

 their action has not yet been generally adopted. 



What was therefore actually determined in these experi- 

 ments is the strength of the sound in the telephone-receiver, 

 representing a quantity of energy which was ultimately 

 drawn (wholly or partially) from the energy of (he radiation 

 field. But it only appears in the form of sound waves after 

 two transformations : (i.) in the detector, and (ii.) in the tele- 

 phone-receiver, and neither the one nor the other trans- 

 formation can completely be calculated. Consequently the 

 calibrations had to be carried out under conditions approach- 

 ing as near as possible those of the actual measurements. 

 Unfortunately the description of the experiments and 

 methods is occasionally very short and not quite clear, so 

 that, while reading the accounts, one often regrets that 

 several important particulars have been omitted or not dealt 

 with. What makes a full appreciation of these experiments 

 and complicated calibrations further difficult is the fact that 

 it is not always clear from the descriptions whether certain 

 values or constants were obtained experimentally, by calcu- 

 lation, or by mere estimation. This at once precludes before- 

 hand any suggestion for a different interpretation of the 

 results than that given by Austin himself. 



The smallest energy in the receiving antenna producing a 

 barely audible sound is given as 



1-225 x 10~ 15 watt, 



It is obvious that the measurement of these amazingly small 

 alternating energies is a matter of considerable difficulty 

 and that variations in the observations of a few hundred per 

 cent., as actually obtained, are to be regarded as relatively 

 small, especially in view of the atmospheric disturbances 

 which occasionally prevent the observations altogether. 



* See Bull. Bureau of Standards, vol. vii. p. 315 (1911) ; Bull. Bureau 

 of Standards, vol. xi. no. 1 (1914) ; Proc. Inst, of Radio Engineers, vol. iv. 

 no. 3 (1916) (New York). 



