Il6 PHILOSOPHICAL TRANSACTIONS. [aNNO 1745. 



more than it receded the day before on the same piece of coach-glass, before it 

 was broke into this triangular form. 



He was naturally led now to make use of 2 supporting tubes, instead of the 

 triangular glass plane. These were about 18 inches long each, and -j-'^ of an 

 inch in diameter, and placed parallel to each other at the distance of about 2 

 inches, so as to support the moving tube near the middle of it. When very 

 nearly horizontal by the level, the supported tube moved from the fire about its 

 axis to the distance of 13 inches : when the supporters were a little raised at their 

 remote ends, so as manifestly by the level to descend towards the fire, it receded 

 to the distance of 10 inches, moving as before about its axis; but in this latter 

 case the fire had declined a good deal ; otherwise probably the tube would have 

 receded farther though up-hill. 



The next day, the same tube, when the same supporting tubes were 8-I- inches 

 distant from each other, receded nearly as before : when 1 2-^ inches from each 

 other, it stood still; and when removed to the distance of 16-L inches, the sup- 

 ported tube very manifestly changed its motion, and went towards the fire; as it 

 did afterwards, when the inclination of the supporting tubes was altered, so as to 

 ascend towards the fire. 



When the tube had 4 others under it, all supporting, one near each extremity, 

 and one on each side of its centre, no motion at all was perceived ; and when 2 of 

 them on the same side of the centre were taken away, the supported tube moved 

 into an oblique situation with regard to the fire, the unsupported half receding 

 from the fire. So that on the whole, it appears sufficiently plain, that the stream 

 of air up the chimney is not the cause of the rotation. 



Mr. W. next made an experiment or two, to show that the motion is not owing 

 to any attraction or repulsion in the tubes. He suspended two fragments of small 

 tubes, 8 inches long, and about ^V of an inch in diameter, near the fire, from 2 

 pins, by blue silk lines, which had each a loop at one end, were tied at the other 

 to the top of the tubes, and hindered from slipping off by a little sealing-wax. 

 The tubes came together at the upper end, and receded manifestly from each 

 other at the lower, appearing to be in a state of attraction above, and a state of 

 repulsion below : but suspecting this to be owing to the sealing-wax, which soon 

 began to melt, he scraped it off both, leaving only as little as was possible, to 

 hinder the silks from slipping. The consequence then was, they came together 

 at the lower ends, and very near so at the upper; and when suspended from one 

 pin, so that the loops of the silks touched each other, the tubes seemed equally 

 close all the way down, without any appearance either of attraction or repulsion. 

 But imagining still that a repulsive power in the heated supporting tubes, when 

 placed nearly together, might possibly be the occasion of the receding of the upper 

 tube at contact with them. To put the matter out of all doubt, he wet the 3 



