INVESTIGATED l:V TIM. MKTIIOD OF JET VIBRATION. 355 



This reduction is thus seen to be double that originating from the capillary-pressure 

 for a water JH with u diameter of 1 mm. Incomes A, = circa 3'0 cm. 



This fact has also been the subject of but little attention, though A. DUPR* has 

 undertaken some interesting experiments alxmt the height to which a jet can rise. 



A closer examination of how the reduction of velocity corresponding to A, spreads 

 itself over the jet I am obliged to leave to another opportunity, only a particular 

 characteristic fact l)eing named here. If the pressure is reduced more and more so 

 that it comes near to the value h t + h, the jet can be observed to deviate more and 

 more from a paral>ola, the curvature of the jet just outside the hole becoming much 

 too great. If the pressure is reduced almost to Ai + A, the jet first runs a short 

 distance horizontally and then falls vertically down. If we determine the discharge 

 and cross-section of the jet, the velocity in the horizontal part of the jet can be 

 calculated, and it will be observed that this velocity is almost equal to that 

 corresponding to the pressure 1^. It is very difficult to maintain the jet in the above 

 position ; the slightest disturbance will cause the jet to cease. 



It is also possible to determine the velocity of a jet by measuring the pressure it 

 produces by normal impact on a sufficiently large plane surface. Measurements in 

 this manner have been made, for example, by BoFF.t who worked with a jet with 

 diameter from 5 to 7 mm. This method seems to give quite reliable results and may 

 most probably be available even for much smaller jet thicknesses, but the difficulties 

 connected with its use will be, in consequence, considerably greater. The balance 

 used must, in such a case, be made very sensitive, and it will probably then be 

 difficult to keep the sensitiveness constant. In the use of this method there must 

 also be taken into consideration a correction resulting from the surface-tension. The 

 pressure measured must be reduced by the capillary-pressure in the jet multiplied by 

 the area of its cross-section, in other words, by 



..=. 

 d 4 2 



At the same time the surface-tension along the jet's circumference, or TTK/, must 

 be added to the pressure measured. The final correction will accordingly be 4- ^TrTrf. 



The cross-section of a jet has hitherto, as a rule, been determined by direct 

 geometrical measurements, which ordinarily take place in such a manner that the 

 points of some micrometer screws are brought exactly to touch the surface of the jet. 

 In this manner a sufficient exactness can be reached, as a rule, with thick jets. The 

 condition is quite the reverse when the diameter of the jet is only about 1 millimetre. 

 In this case it will most probably be impossible to get even a moderately satisfactory 

 exactness. In this respect a great progress has been made by the elegant method 



* A. DtiPRfc, ' Throne m&anique do la chaleur,' p. 376, Paris, 1869. 

 t BUFF, 'Poou. Ann.,' 137, p. 497, 1869. 



2 Z 2 



