212 Mr. G. F. Rodwell on the Effects 



bably, from the whirling motion of the water around a conical 

 cavity, which, as we shall see hereafter, is produced in the trough 

 of machines similar to that mentioned by Francois, and in all 

 machines with inclined tubes : in both these respects there is a 

 resemblance to a waterspout. 



Several modifications of the trompe have been constructed 

 since its first invention, the main difference consisting in the 

 way in which air is allowed to enter the tube : thus there is 

 the trompe without orifices for the admission of air, as in that de- 

 scribed by Francois, and in the trompe mentioned by Mariotte 

 (in his Traite du Mouvement des Eaux), in which latter modifi- 

 cation a blast is produced by a stream of water falling from a 

 height of 10 or 12 feet into a funnel which communicates by 

 a vertical tube with a vessel for collecting the air carried down ; 

 or, again, there may be orifices in the tube, as in the trompe 

 described by Mr. Stirling, and in that now used ; or air may be 

 admitted by two small tapering tubes which enter the trompe 

 tube just below its juncture with the water- cistern, and pass 

 upwards through the water therein till they reach the air : this 

 latter arrangement was much used formerly. 



The modern trompe consists of a large cistern in which there 

 is a constant depth of from 4 to 6 feet of water ; from the bot- 

 tom of the cistern proceed two tubes from 20 to 30 feet long, 

 the lower extremities of which pass into a wooden wind-chest 

 furnished with an arrangement for keeping the water at a certain 

 level, so that no air can escape except by a blast pipe in the 

 upper part of the chest : beneath the lower extremity of each tube 

 there is a flat iron plate to break the fall of the descending water. 

 The upper part of each tube is contracted at the point where it 

 joins the cistern, and immediately beneath the contracted part 

 four holes are made in the circumference of the tube. When 

 water is allowed to flow from the cistern into the air-chest, a 

 quantity of air is carried down with it, and a perfectly regular 

 and constant current of air issues from the blast pipe. 



II. 



I have endeavoured in this paper to ascertain the most favour- 

 able conditions under which air is carried down by a stream of 

 water, and to arrive at a satisfactory explanation of the cause 

 of the descent of air in the different modifications of the trompe. 



In a former paper * I have given an account of some experi- 

 ments made with a view of determining the quantity of air car- 

 ried down by a known amount of water ; but as the method there 

 employed was scarcely so satisfactory as could be desired, I 



* "On some Effects produced by a Fluid in Motion," No. I., Philo- 

 sophical Magazine for January 1864. 



