the water is so agitated by the wash that when we are rising at a speed 

 of 3 ft. per second we see, floating in the water, small particles which 

 move up and down without our being able to distinguish whether 

 the bathyscaphe is rising or descending. To avoid these eddies, we 

 mounted our anemometer at the top of the tower and 3 ft. away from 



Fig. 12, Vane anemometer for measuring the vertical speed of the bathyscaphe 



the side. In these circumstances it was impossible to observe the vane 

 anemometer directly, and so it had to transmit its signal electrically. 

 This was quite simple: at the base of the anemofneter shaft (see 

 Fig. 12) we fitted a cylinder of insulating material upon which were 

 placed two metallic segments s s (one short and one long) so that a 



[98] 



