276 Mr. Morgan's Experiments to a/certain the 



times before the charge has made a fmall perforation in- the. 

 bulb. If again a thermometer filled with mercury be inverted 

 into a ciftern, and the air exhaufted in the manner I have de- 

 scribed for making the experiment with the gage, a Torricellian! 

 vacuum will be produced ; and now the elettric light in the 

 bulb, as well as the fpark in the tube, will be of a vivid green; 

 but the bulb will not bear a frequent repetition of charges be- 

 fore it is perforated in like manner as when it has been ex- 

 haufted by an air-pump. It can hardly be neceflary to obferve,. 

 that in thefe cafes the ele£tric fluid affumes the appearance of a. 

 fpark *, from the narrownefs- of the pafiage through which it 

 forces its way» If a tube, 40 inches long, be fixed into a globe: 

 8 or 9 inches in diameter, and the whole be ex-haufted,. the eledtric 

 fluid, after pafling in the form of a brilliant fpark throughout 

 the length of the tube, will, when it gets into the infide of the 

 globe, expand itfelf in all diredlions, entirely filling it with a; 

 violet and purple light, and exhibiting a ftriking inftance of the. 

 vail elafticity of the eledric fluid. 



I cannot conclude this paper without acknowledging my obli^- 

 gations to the ingenious Mr. Brook, of Norwich, who, by com- 

 municating to me his method of boih.ng mercury, has been the 

 chief caufe of my fuccefs in thefe experiments t. I have lately 



learned' 



* By cementing the firing of a giiittar into, one end of a thermometer tube,, a 

 fpark may be obtained as well as if the tube had been fealed hermetically. 



t Mr. Brook's method of making mercurial gages is nearly as follows. Let a- 

 glafs tube L (fee fig. 2.), fealed hermetically at one end, be bent into a right- 

 angle within two orthree inches cf theother end. At the diflance of about an inch. 

 or lefs from the angle let a bulb (K), of about | of an inch in diameter, be blown 

 m the curved end, and let the reniaiadcr of this part of the tube be drawn out (I) 



