204 



Professor Sir James Dewar 



[Jan. II 



were thus left practically in the same relative proportions as in the 

 original l)ul)ble, water and ammonia bein^- driven off. If necessary, 

 the ammonia was separately estimated by Nesslerising. This is one 

 advantage attending the use of ammonium oleate in such bul)bles, as 

 compared .with the oleates of a non-volatile base like soda or potash. 

 The arrangements employed are shown in Fig. 11. The l)lo wing- 

 tube A, with its nozzle N to carry the bubble, is supported in an 

 india-rubber cork fitting the neck of a 10 to 12-litre tabulated 

 " aspirator " vessel. D is a bulbed vent tube, and is packed uniformly, 



30 DAYS 40 



•.ra ' Diameter 



Fig. 12. — Aie" Bubble ovek Water remains Coloured. 

 Contraction and condensation. 



but not tightly, with cotton-wool moistened by glycerine. This 

 allows any variations of atmospheric pressure to be equalised with- 

 out endangering the purity of tlie internal atmosphere. B is the 

 graduated collection tube fitted in the lower tubulation ; its nozzle, 

 inside, is directly under the drops whicli accumulate on the l»ul)ble. 

 Before opening the stop-cock on B, to draw off' the accumulated 

 liquid, the stopper in I) is removed to allow the equalisation of 

 barometric pressure, otlierwise the small quantities of liquid could 

 not be smoothly withdrawn ; alternatively, D may remain open. 

 Before commencing observations any excess liquid is removed from 

 the bul)b]e by drainage along the rod 0. This is also done very 

 effectively by a small bundle of very thin irlass rods, about ^ mm. 



