Positive Ionization from Hot Salts. 699 



materials all o£ which had undergone distillation. This was 

 done by the action o£ ammonia on a mixture of dry aluminium 

 chloride with excess of orthophosphoric acid. The dry 

 aluminium chloride and phosphorus pentoxide were supplied 

 by Eimer and Amend and had been made in Germany. The 

 orthophosphoric acid was prepared by boiling the phophorus 

 pentoxide in distilled water. The ammonia solution was 

 analysed and contained 28 per cent, of NH 3 and 0'0004 per 

 cent, of non-volatile matter. It was added cold to the 

 mixture of the solutions of aluminium chloride and ortho- 

 phosphoric acid, which had a temperature of about 40° C. 

 The mixture was stirred constantly and the ammonia was 

 added until the precipitation was almost but not quite 

 complete. The precipitate was thoroughly washed on the 

 filter with boiling distilled water. Jt was divided into two 

 parts, one of which was allowed to remain pasty and the 

 other dried in an air oven at such a temperature that a piece 

 of filter-paper placed over it was slightly charred at the 

 edo-es after about two hours' drying. 



0*2 gram of the diy phosphate, about the same amount as 

 had been used in working with Kahlbaum's phosphate, was 

 tested in the platinum tube apparatus in the usual way. It 

 was found that it did not give rise to thermionic currents of 

 the order of 10~ 8 amp. until temperatures above 1000° C. 

 were reached. The mean current-pressure curve at 1050° C. 

 is given in fig. 8 (curve 4). K educed to a temperature 

 of 800° 0. this would correspond to about 1/150 of the 

 emission given by Kahlbaum's aluminium phosphate, which 

 in the light of the work of other observers appears to have 

 been small rather than the reverse. It is not even certain 

 that this small emission was due to the aluminium phosphate 

 itself as it was less than what would have been given by the 

 platinum tube alone at the beginning of the experiments. 

 No doubt the emission from the platinum had fallen, but 

 there is no evidence to show how much. 



The small emission from the pure aluminium phosphate is 

 not due to its having been dried, for the paste was tested by 

 the strip method and found to give a small emission. There 

 was a measurable current during the early stages of the 

 heating, but it soon died away to a very small value. A 

 temporary emission of this character would have escaped 

 notice in the tube experiments. 



There are an enormous number of slightly different 

 aluminium phosphates, according to the chemical handbooks, 

 and the differences between the different samples might 

 conceivably be due to differences in chemical constitution. 



