14 DR BREWSTER on the Existence of Two New Fluids 



with heat, and never mixes with the new fluid NN. By a little 

 management, the vacuity V may be made to come in contact 

 with the bounding lines mn,mn, &c. ; but it never affects its 

 curvature, and seldom enters the fluid W. When the vacuity 

 V has been made to vanish by heat, these bounding lines remain 

 exactly the same. 



Having at first observed this second fluid only in the angles 

 of cavities, as in Fig. 8., I experienced considerable difficulty in 

 establishing its fluidity. The improbability of two fluids exist- 

 ing in a transparent state, in absolute contact, without mixing 

 in the slightest degree, induced some of my scientific friends 

 to refer it to an optical illusion, and to consider the line which 

 separated it from the new fluid as a septum or partition in the. 

 cavity. The beautiful curvature of the bounding line, and its 

 perfect similarity to that of two contiguous fluids, rendered 

 this conjecture untenable. It was next supposed to be a va- 

 cuity into which the new fluid could not expand itself; but 

 though this idea explained the curvature of the bounding line, 

 it was inconsistent with other facts, and especially with the im- 

 portant one, that the second fluid acted upon light neither like 

 topaz nor a vacuum, but like water. 



These difficulties were gradually overcome by more numerous 

 observations. 



Although the cavities were generally like those in Fig. 8., 

 where V is the vacuity, NN the new fluid, and WW the sup- 

 posed second fluid; yet I found several in which the second 

 fluid filled a great part of the cavity, as in Fig. 9., where NN is 

 the new fluid, and W the second fluid, or as in Fig. 10., where a 

 vacuity V also appeared within the globule N of the new fluid. 



This great enlargement in the quantity of the second fluid, 

 removed most of the difficulties which had formerly presented 

 themselves ; but something was still wanting to prove its fluidity. 

 This desideratum was fortunately obtained in a specimen of to- 



