Positive Ionization from Hot Salts. 695 



neighbourhood of 900° C. The reduction to equal areas is a 

 still less definite matter. It is not yet certain whether the 

 emission is clue to the salt directly or to the interaction with 

 the platinum of vapours which it gives off. 1 think there is 

 a considerable amount of evidence that the latter is an 

 important factor. I have therefore reduced to equal areas 

 of platinum surface and have taken the effective area of the 

 platinum tube to be 60 sq. cm. In the case of the strips the 

 area of both sides has been counted. The other observers do 

 not seem to have recorded the area of 1he strip covered by 

 the salt, but it is probable that if equal areas of salt were 

 compared the values in column 10 would be altered relatively 

 in favour of higher values for the platinum tube. 



One of the most striking features of the numbers in column 

 10 is the comparatively small difference in the greatest 

 currents given by the different salts when te-ted by the tube 

 method. With the exception of pure aluminium phosphate, 

 which will be considered below, the smallest maximum 

 emission is given by BeS0 4 (or BeO) with 11 x 10~ 10 amp. 

 and the largest by BaS0 4 (barite, but probably due to BaS) 

 with 190 X 10~ 10 amp. If the emission after heating in air 

 at atmospheric pressure is taken to be that characteristic of 

 BaS0 4 then this substance would be the least efficient with 

 an emission of only 7x 10~ 10 at 800° G. I have not found 

 the phosphates to be especially efficient. 



A comparison of the maximum currents obtained by 

 the two methods indicates that the tube causes a greater 

 emission than the strip when equal areas are compared. 

 Horton's measurements with sodium phospluite are an 

 exception in this respect, but the salt he used was, of course, 

 a different specimen from that tested in the tube experiments. 

 My experiments with trisodium phosphate gave much smaller 

 currents at the same temperatures than tliose recorded by 

 Horton for " sodium phosphate." 



One advantage of testing the emission by two widely 

 different methods such as those which have been described, 

 and at the same time making tests on a number of different 

 kinds of salts, is that it offers some chance of a separation of 

 the physical from the chemical effects. One would expect 

 that effects which were physical in their origin would depend 

 upon the method of testing employed, but would be inde- 

 pendent of the nature of the salt. Conversely effects which 

 arose from chemical changes might be expected to vary con- 

 siderably from one salt to another, but not to depend so much 

 upon the method of testing employed. 



