14 Prof. Magnus on the Motion of Fluids. 



orifice to a whirling motion of the water, which is very distinct 

 about ff. The tumult caused by the rebounding of the fluid in 

 the upjjer layer rendered a nearer observation of tlie action inside 

 impossible ; and to accomplish this, I have been compelled to 

 resort to another although somewhat more laborious method. 



31. Water which was perfectly free from salt and hydrochloric 

 acid was caused to stream into a large wide vessel containing water, 

 in which about 1 per cent, of salt was dissolved. To determine 

 exactly the quantity of salt in the mass, a measured volume of 

 it, generally about 10 cubic centimetres, was brought by means 

 of a pipette into a bottle, and a normal solution of nitrate of 

 silver was added until the precipitation ceased. From the volume 

 of the normal solution here necessary, the quantity of salt was 

 determined. This is a reversal of the method invented by Gay- 

 Lussac for the determination of silver. With a little practice, 

 the quantity of chloride of sodium can in this way be estimated 

 to O'Ol per cent, of accuracy. If now water, free as aforesaid 

 from salt or hydrochloric acid, stream into the solution, and if a 

 portion be taken from any point, with the precaution that the 

 motion of the jet shall not be disturbed, we can estimate from 

 the quantity of salt contained in the fluid so obtained the pro- 

 portion in which the streaming water has mixed with that already 

 present in the vessel. 



32. To take the fluid from any point of the jet, I used a glass 

 tube, abc, fig. 12, which was drawn out to a fine point at a. The 

 pointed portion was horizontal, and was brought into the line of 

 the horizontal jet. Behind the tapering point the tube was bent 

 upward, and reached above the surface of the water ; here an- 

 other tube was fastened to it by an india-rubber joint, which 

 permitted the tube to be bent downwards. Tlie tube abc was 

 firmly fastened on a strong piece of wood, ffh, which, being placed 

 upon another board, could be moved securely backwards and 

 forwards over the vessel ABCD. The point of the glass tube 

 had an inner diameter of 1 millim. and an outer one of 1'5 mil- 

 lim., and was gradually narrowed, so that the resistance offered 

 to the motion of the water was unimportant. When the distance 

 at which this point was placed from the orifice was not too con- 

 siderable, the motion was suflicient to drive the water into the 

 tube, and cause it to ascend therein until it flowed out through 

 cd, and was received in a glass held underneath. The first por- 

 tion of the fiuid thus obtained consisted of the solution which 

 was in the vessel before the commencement of the experiment, 

 and only the later portions could be regarded as the mixture 

 corresponding to the point a. For this reason the outflowing 

 fluid was received in three separate portions, the two last of M'hich 

 must show the same composition if the result was to be consi- 



