436 



Sir James Dewar 



[Jan. 28, 



shallow vessel of slightly tepid water. At first single drops of liquid 

 air were allowed to fall from a height of 35 feet from a dropping 

 pipette.'" Each drop produced a little whirling cloud as it wandered 

 about in the spheroidal state on the surface of the water. The rate 

 of dropping was then increased to a rain, which soon obscured most 

 of the water surface under a misty mantle. A dense rolling cloud 

 mass was then obtained by tossing spoonfuls of the liquid air about 

 all over the vessel. With a localised spot-light illumination all the 

 miniature alpine valley effects were seen very distinctly. 



A more widely distributed cloud was produced by a whirling 

 inverted metal cup fixed on the vertical spindle of a small motor, 

 and supported a few inches above the surface of liquid air in a squat 

 metal cylindrical vacuum vessel. The cup was divided by radial 

 plates into eight sectors, but with a small centre space left clear. 



jUN (half minute) 



Fig. 21. 



About a litre of liquid air was poured into the vacuum cylinder, and 

 the motor started. As the speed was continuously increased, a 

 waterspout of liquid air was steadily lifted and finally projected out 

 into the room to form a thick cloud mantle several feet across. 



Solar and Lunar Measurements. 



A series of measurements were made during the partial solar 

 eclipse of April 8, 1921. A silvered-surface heliostat was employed 

 with a platinised mirror above the portable thermoscope, all arranged 

 on the laboratory roof some 80 feet above the grouud. A well- 

 defined decrease and increase in the solar radiation was recorded, 

 using 1/100 of the full aperture employed for zenith measurements. 



* Proc. Roy. Inst, xx. pp. 586-7. 



