352 On a Remarkable Effect produced by the 



then wrapped in a piece of muslin, and with loose packing of 

 cotton waste it was enclosed between the two copper hemi- 

 spheres, which were then soldered together. The holes at tlu- 

 poles of the copper sphere gave free communication with the 

 sea-water. The copper ball was then attached to the dredging 

 cable, which took it to a depth of 3000 metres. When it came 

 up no external effect was visible. I could not believe that even 

 a small flask of the kind could support a pressure of 300 atmo- 

 spheres, and I concluded that it had collapsed shortly after 

 leaving the surface. Still, as the line was going to make a 

 second excursion, and this time to 6000 metres, I re-attached 

 the ball along with a larger one to it. 



On returning to the surface the ball had the appearance 

 which you see. If the soldered welt represent the equator, it 

 will be seen that both polar areas are as they were. Perpen- 

 dicularly to the equator a system of folds or creases runs 

 northwards and southwards and extends very little beyond the 

 tropics. The creasing is most accentuated at a part of the 

 equator where there is a slight flattening. It is evident that 

 the glass flask when it collapsed was relatively near this part 

 of the ball. I did not open the ball, as I thought it would be 

 more instructive to keep it as it is. The debris of the glass 

 flask with the cotton waste is still inside it. 



The effect of the sudden relief of pressure on the copper ball 

 is distributed much more uniformly over it than is the case in 

 the brass tube. In the latter the effect is very powerful and 

 very local. In both cases the effects which we see have been 

 produced in a moment of time, and are properly speaking, the 

 effects of violent shock. It is remarkable that in the ball the 

 equatorial zone which has the welt to stiffen it should be the 

 field of all the disfigurement, while the polar areas which have 

 no strengthening have not been exposed, or at least have not 

 yielded to strain. 



If we examine the brass tube, Figs. I, 2, we see that, with 

 the exception of the portion nearly in the middle which held 

 the sealed glass tube, the case has perfectly preserved its 

 cylindrical form. The distortion or crumpling affects only 

 the part where the tube collapsed, and it is evident that it did 



