190 Professor Sir James Dewar [Jan. 19, 



bubbles. The lowering of the temperature will then reverse the 

 effect and the bubble is seen to thicken. The ordinary variations 

 from night to day are sufficient to cause many such repetitions in 

 the same bubble. Thus, when it was thought to increase the pro- 

 portion of water vapour by warming the water in the vessel, and 

 thereby increasing the thickness by a greater absorption, the rise of 

 temperature counteracted any increased absorption ; and once more 

 the bubble became thinner. The removal of the wet cloth allows the 

 condensed water to distil off more vapour, which is often sufficient to 

 thicken the bubble. This is best seen when a good condensed deposit 

 is accumulated before blowing the bubble, as was done in one case 

 (see fifth bubble, first vessel) by the application of bags of ice. The 

 bubble, which was then blown, remained almost entirely coloured 

 throughout its life. A small zone of black came and went, and the 

 thickness w^as reduced once or twice by a damp cloth. The w^eather 

 was rather warm at first, but became more temperate (the time was 

 August). Finally, the bubble was very much thickened, and prob- 

 ably burst from this cause ; for as the bubble became thick, it 

 overhung the ring unsymmetrically, and most probably became lop- 

 •^ided enough to swing round and touch the vessel. The supporting 

 ring was seen to be slightly out of the horizontal, and in vibration 

 due to the working machinery in the room below. 



The fourth bubble in the second vessel : during the first week 

 untouched ; maintained green and pink colour ; thick and overhanging 

 the ring ; then frequently treated by damp cloth above, thinned while 

 cloth was on and thickened when cloth was removed ; at third 

 operation, bubble half black : next time, the water below was warmed 

 by the arc lamp beam : result, 85 per cent, black obtained. 



The fifth bubble in the second vessel lasted eleven days only ; 

 short life in the hot weather ; disturl)ed at the end by dainp cloth, 

 also by vibration ; thinned to silver and black ; first three days 

 thickened rapidly by absorption from the deposit of water ; im- 

 mediately after being blown sagged over ring ; then steadily thinned 

 through several beautiful grades, until on the fifth day at 19^^ C. it 

 was amber and purple, with silver above. Damp cloth then put on ; 

 the silver extended all over, followed by a fairly uniform development 

 of black. Two grades of black were noted— the deepest black being 

 only a zone of a few centimetres at the top. The top of the bubble 

 was approximately 1 cm. higher wdien black than when thick and 

 heavy. 



The sixth bubble in the first vessel went quickly black in two days 

 (possibly from an unusual excess of ammonia in the soap solution), 

 and remained so until the end of the first week. It then thickened 

 to various colours for four weeks, and once again went quickly black. 

 This condition remained for three weeks, during which time a steady 

 diminution of diameter was measured. The blowing-tube having l)een 

 left attached to the bubble, the contraction had by this time narrowed 



