ON DUST AND DISEASE. 159 



more or less distinct was always revealed by the powerfully 

 condensed electric beam. 



The floating motes resembled minute particles of liquid 

 which had been carried mechanically from the U-tubes into 

 the experimental tube. Precautions were therefore taken to 

 prevent any such transfer. They produced little or no miti- 

 gation. I did not imagine, at the time, that the dust of the 

 external air could find such free passage through the caustic 

 potash and sulphuric acid. This, however, was the case ; 

 the motes really came from without. They also passed with 

 freedom through a variety of aethers and alcohols. In fact, 

 it requires long-continued action on the part of an acid first 

 to wet the motes and afterwards to destroy them. By care- 

 fully passing the air through the flame of a spirit lamp, 

 or through a platinum tube heated to bright redness, the 

 floating matter was sensibly destroyed. It was therefore 

 combustible, in other words, organic, matter. I tried to 

 intercept it by a large respirator of cotton-wool. Close pres- 

 sure was necessary to render the wool effective. A plug 

 of the wool, rammed pretty tightly into the tube through 

 which the air passed, was finally found competent to hold 

 back the motes. They appeared from time to time after- 

 wards, and gave me much trouble ; but they were invariably 

 traced in the end to some defect in the purifying apparatus 

 to some crack or flaw in the sealing-wax employed to render 

 the tubes air-tight. Thus through proper care, but not with- 

 out a great deal of searching out of disturbances, the expe- 

 rimental tube, even when filled with air or vapour, contains 

 nothing competent to scatter the light. The space within it 

 has the aspect of an absolute vacuum. 



An experimental tube in this condition I call optically 

 empty. 



The simple apparatus employed in these experiments will 

 be at once understood by reference to a figure printed in 

 the last article (p. 98). s s' is the glass experimental 

 tube, which has varied in length from 1 to 5 feet, and 

 which may be from 2 to 3 inches in diameter. From 

 the end s, the pipe p p passes to an air-pump. Connected 



