Prof. Magnus on the Motion of Fluids. 5 



to the orifice of the larger jet. For a greater ratio there is no 

 result given by Savart. 



\^1ien the difference of the pressures is sufficiently great, and 

 when the ratio of the diameters is as 1 to 4, or something greater, 

 the smaller jet penetrates quite into the wider orifice. 



10. In the exjDeriments described in § 7 this takes place. Two 

 jets, as it were, act against each other ; for tlie water in the cy- 

 lindrical tube may be regarded as a Avider jet under a small 

 pressure. Evidently a curved surface would in this case also be 

 formed, did not other circumstances come into plaj^, by which, 

 instead of one large surface, a number of smaller ones are formed ; 

 to these is due the foaming already alluded to. 



11. Proceeding further with these experiments, I permitted the 

 jet to project itself against a hemispherical concavity of metal. 

 Here also we obtain, by a certain pressure and a certain magni- 

 tude of the hemispherical vessel, an oval-shaped surface which 

 returns into itself*. If the hemispherical concavity be turned 

 to one side or the other, the axis of the surface of the water, or at 

 least the point where it unites again, falls on the same side of 

 the jet as represented in fig. 5. The coming jet goes then 

 right through the surface without derangingthe formation thereof. 

 The hemisphere, however, can be so turned that the water will 

 seek to unite at a point which lies in the direction of the jet. 

 Here again it will be met by the coming stream ; and in this way 

 is caused a peculiar foaming and scattering of the water par- 

 ticles, a representation of which is attempted in fig. 6. 



Supposing a similar action to occur within a tube the dia- 

 meter of which is not greater than that of the oval-shaped sur- 

 face, it is quite conceivable that the entire mass of water thrown 

 back against the narrow jet will be arrested by the latter, and, 

 forming a new surface, will be carried backwards. In this way 

 we can imagine how the outflow of a fluid from a vessel with a 

 wide orifice may be held back by a jet of much smaller diameter; 

 and we are also enabled to explain why it is that an entire large 

 surface cannot be formed, for it is broken up by the water which 

 meets it into a multitude of smaller surfaces of which the foam 

 is constituted. 



12. In order, howevei*, to obtain a more distinct idea of what 

 actually takes place, I have modified the experiment by furnish- 

 ing the tube de, fig. 7, with an addition, ??m, bent upwards at a 

 right angle, so that the entire tube presented the appearance of 

 a reversed T. The piece mn was so applied that the point k, 

 where the small jet met the water within the tube, lay between 



* I used ill these experiments a hemispherical concavity of 24 milhms. 

 diameter. The jet projected against it issued from an orifice .'} milhms. in 

 diameter, and under a pressure of 2".'i metres ; the distance of the hemi- 

 sphere from the orifice was about 05 of a metre. 



