Water-pipes and Si/phonsjrom disengaged Air, <S*c. 105 



limb with a force more and more retarded. This double action, 

 therefore, of the gravitating force in the syphon tends to weaken, 

 and at last to overcome, the corpuscular attraction in the middle 

 part or summit of the double column, as we have seen actually 

 to take place as soon as the weight of the column of water came 

 into equihbrium with that of the atmospherical column. 



A proof of the important share which the diminution of the 

 corpuscular attraction has in causing the evolution of the com- 

 bined air, is to be found in what takes place in the case of water 

 flowing through conduit pipes along a declivity of considerable 

 but unequal inclination. In this case the water, although it 

 enters the pipe with only the force corresponding to its depth 

 beneath the surface, that is, to the pressure of a hydrostatic 

 column of the diameter of the pipe, and of a height equal to its 

 depth (aB^ fig. 3) below the surface of the water, yet tends ♦ to 

 issue from the lower end with the velocity due to the total height 

 of its fall (or in the proportion of the square root of the height 

 c D, to the square root of the height a B), modified by the fric- 

 tion of the pipe, which, however, according to the experiments 

 of Bossut, is neutralized when the inclination of the pipe amounts 

 to 6° SV, or about 1 in 8. It would appear, therefore, that 

 the intermediate part of the moving column must be, as it were, 

 in a high state of tension. As its velocity increases it exerts less 

 and less pressure on the sides of the pipe, and receives less pres- 

 sure in return, — the increasing momentum of the descending 

 column, particularly of the interior filaments, which are not af- 

 fected by the friction of the pipe, tends to elongate the mass, to 



• I say tends to issue, because the affirmation of some writers that in long 

 horizontal pipes the water enters the pipe with much greater velocity than it 

 leaves it, must be taken with some qualification. For, if this were strictly 

 true, it is evident that an accumulation and condensation must take place in 

 the interior of the pipe, which would quickly attain a maximum, and prevent 

 the water from entering materially quicker than it was discharged. And, on 

 the other hand, its tendency in pipes of great inclination to issue with greater 

 velocity than it enters is controlled by the increased friction of the pipe, and 

 by the corpuscular attraction of the water, which, together with the pressure 

 of the atmosphere at the lower extremity of the pipe, prevent the continuity 

 of the column from being broken, as it would be, did the water actually issue 

 with much greater velocity than it enters, and reduce its final to very nearly 

 its initial velocity. 



