94 MR. GEOEGE C. SIMPSON ON THE 



I had my instrument was a little hut near to my host's large <' handlesmand's " 

 house. On the end of the large house was a flag staff, to the top of which I 

 attached an insulator and from it took a wire through a window into my room. 

 About a third of the way up the wire I attached two milligrams of radium bromide 

 which acted as a collector. On the accompanying photograph, the insulator, wire 

 and the position of the radium collector are shown. The height of the collector 

 above the ground was 5| metres. This arrangement acted extremely well and, as far 

 as I could judge, gave as good results as could be wished. 



The potential gradient was reduced to that over a level surface by making 

 simultaneous observations with a flame collector and leaf electroscope above the 

 most level piece of ground I could find. The country was so rough that a good and 

 accurate determination could not be made, but the error is certainly not 20 per cent. 

 During the year this reduction was several times repeated, no change being found. 

 Great attention was also paid to the insulations, which were never found defective. 

 As the collector was situated between two houses over a much frequented road, no 

 accumulation of snow took place under it, so corrections due to the height of the 

 snow were not necessary. 



Dissipation and lonization. In order to observe the ionization and dissipation 

 without being disturbed by the smoke of the village, two platforms (as shown in the 

 photograph) were built at different parts of the village, but as both were to the 

 north of a large part of the village, I could not observe when a south wind was 

 blowing ; with all other winds one of the platforms was on the windward side of the 

 houses. The platforms were about a metre above the ground and the instruments 

 on a shelf about a metre and half over the platform ; above all was a roof to protect 

 the instruments from rain and snow. By this arrangement the instruments were 

 exposed to the full force of the wind. In order to read the dissipation electroscope 

 in a high wind, a small screen was held to protect the instrument just at the moment 

 of observation. 



The usual method of observing the dissipation or ionization is to charge the 

 electroscope, take a reading, then return in 15 minutes and take another reading. 

 This method is open to great objections : first it is quite easy to make a false reading, 

 and secondly in open-air work the leaves are not steady enough to allow of one 

 reading being accurate. The method I adopted was to charge the two instruments, 

 then take a reading of the dissipation instrument, half a minute later a reading of 

 the ionization instrument, then at the minute take another reading of the dissipation 

 instrument, at the next half minute a second reading of the ionization instrument, 

 and so on for 5 minutes, when of course I had five readings on each instrument. 

 Ten minutes later I started reading again, and at minute intervals read each of the 

 instruments five times, then from a table found the value corresponding to each of 

 the readings, took the mean of the first five, then that of the second five, and used 

 these means as single values separated by an interval of 1 5 minutes. In this way 



