508 - ii - 



interface was approximately normal to the light path between source and camera. In Figure 7a, 

 however, It was necessary to open the camera shutter before firing the charge so that the Image of the 

 detonated charge Is found as a very bright patch at the centra of the bubble. 



The bubble remains very nearly spherical, apert from sm^ll excrescences on its surface, until about 

 10 milliseconds before the first minimum (Figure yi). As the bubble collapses further It becomes 

 flattened, but In such a way that the lower half of the bubble remains very roughly hemispherical 

 (Figures 7e, f, ea, b, c), in Figure 8b the bubble Is nearly hemispherical with a flat top, while in 

 8c the flat top has become concave so that the bubble resemblesan inverted mushroom. During the 

 later stages of collapse the needle-like projections have become more accentuated relative to the 

 bubble as a whole, and appear to be slightly more marked on the lower surface. The shape of the 

 bubble at the minimum is somewhat obscure, perhaps owing to the relatively large duration of the 

 light flssh - about lOO microseconds. The f.symmpt rical lobe on the left of the bubble in Figures 8d 

 and e Is probably due to a perturbation of the bubble when large since in these cases the c'harge W3s fired 

 to the left of the centre of the framework. The framework in these was relatively stiff, and was made 

 of channel Iron; the bubble when large enveloped the left hand member of the framework. |n all 

 other photographs at the minimum no asynmetry was found. 



That tho bubble is moving downwards may be inftrred from the shape of the horizontal cross 

 wires nearest to the bubble seen In Figures 3d and e. These wires were terminated by stretched springs 

 df-slgned to keep them taut when the inwsrd rush of water pilled in the sides of the framework. In 

 practice the transverse pressure exerted en the wires by the moving water displaced them very 

 considerably. The actual shape of the wires is determined by the past histcry of the water velocity 

 In Its neighbourhood, and it is clear from their shape In Figures 8d and e that the motion has been 

 mainly radially Inward, but with a velocity vertically downwards superimposed on this flow in the 

 Immediate neighbourhood of the bubble. 



In all the photographs near the minimum there appears to be a region just above the bubble 

 where the water scatters the light rather than transmitting It In the normal manner. in Figure Be, for 

 example, the Intensely bright line of light produced by the detonated Cordtex appears to be Interrupted 

 by this region; the wire in Figure Be becomes almost, if not quite. Invisible when passing through this 

 region. it is suggested that this Is evidence of a "wake" In the rear of the moving bubble. perhaps 

 this region Is filled with many small eddies and fine bubbles. 



After the mlnlmum..th'e bubble expands somewhat irregularly - Figures 8f, 9a, b and c. Its 

 lower surface* shows small protuberances, but not the long needles which are such a prominent feature 

 of the lower surface before the minimum. The upper surface of the bubble, however, shows a very 

 marked array of approximately radial streamers or "whiskers" which appear to be long trials of bubbles. 

 The length of these "whiskers" Is of considerable interest; in Figure 9a they occupy water on either 

 side of the bubble which In Figure Be was quite clear and free from such bubbles. It seems somewhat 

 unlikely that these "whiskers" have actually been pushed out so far in advance of the main surface of 

 the bubble. These streamers (nay be composed of fins bubbles of gas, actually left behind In, the water 

 by the collapsing bubble which were compressed and rendered practically Invisible Oy the very high 

 pressure existing in the water round the bubble when near Its minimum. 



The curious radial streaks of light In Figure' 7e are thought to have nothing to do with the 

 bubble. It is thought that they are produced by small white hot particles - perhaps pieces of aluminium 

 from the detonator - projected outward at detonation. Although the solenoid which operated the camera 

 shutter was not energized until some 15 milliseconds after detonation, it Is probable that In this case 

 the impact of the Shockwave operated the rather delicately set shutter trigger. This would account 

 for the absence of the actual flash of the detonating charge, and for the absence of any streaks near 

 the centre of detonation. The Increased intensity of these streaks towards the end of their path would 

 be due to the increase of light-gathering power of the lens as the shutter opened. This same phenomenon 

 has been observed on several occasions when the timing of the main light flash and thus of the actual 

 bubble photograph differed very considerably. 



