EXTENSION-CURRENTS 69 



scattered in the surrounding fluid, such experiments do not conn- 

 off so clearly and distinctly as those performed with soap. Tin- 

 oil-drops rapidly become turbid by application of these solu- 

 tions, in consequence of numerous minute drops of fluid appear- 

 ing in them. These may partly arise in the way described 

 earlier, but may partly be thrust in from the surrounding fluid 

 by the violent streaming movements. Further, the fine granules 

 and droplets which appear round these oil-drops may be in part 

 very minute droplets of oil that have become split off. It was 

 shown that during the process (oil-drops + K CO.,, 2 '5 per cent) 

 the fine drops of fluid, which gradually made the oil-drop turbid, 

 collected in the posterior quiescent portion ./ (see above, 

 p. 62, Fig. 9), as a result of which the latter soon becomes quite 

 turbid. Without doubt the fine droplets were gradually carried 

 to this region by the current, since the superficial streaming zone 

 of the drop also became turbid and frothy to a small depth. 

 Finally, the clear axial streak m, which in this drop was 

 similarly well marked, also became bordered on each side by a 

 narrow, frothy, turbid line, which no doubt took its origin from 

 the posterior region . This behaviour of the fine droplets of 

 fluid in streaming olive oil-drops is the more peculiar since, as 

 has already been described, the fine particles of lamp-black mixed 

 with the drops .behave quite differently, and do not penetrate into 

 the region x. That this difference is not a peculiarity inherent in 

 the particles of lamp-black themselves, follows from the fact that 

 in streaming drops of paraffin oil they behave in just the same 

 way as the droplets of fluid in the olive oil. Thus if a drop of 

 paraffin oil impregnated with lamp-black is thrown into energetic 

 streaming movements, lasting for some time, by means of soap 

 solution or other fluids, after a relatively very short period all 

 the particles of black collect posteriorly in the resting region ,/', 

 which in consequence (see Fig. 11) appears as a black triangle 

 that extends forwards more or less far 

 into the similarly black axial streak. The 

 remaining part of the drop usually be- 

 comes completely clear and almost quite 

 free from black ; here and there only does a 

 particle of black emerge from the region 

 x into the streaming again. I have always 

 observed this phenomenon in the same Fi ,, 1L 



manner in the numerous experiments 



performed with paraffin oil. In the streaming drops of foam 

 also, that were prepared from olive oil containing lamp-black, 

 the soot particles, curiously enough, showed a behaviour more as 

 in paraffin oil. In just the same way they collected abundantly 



