planation that next suggests itself depends on the acceleration of velocity during the fall, which should have as a necessary result a progressive diminution of the section-surface. It may be proved 
Fig. 8.—Form and Constitution of a Liquid Sheet falling. experimentally that this has nothing to do with the triangular form by giving an upward direction to the jet from a flattened tube; the stream takes the form of a tongue all the same (Fig. 10).
Is it because the lateral parts of the liquid ribbon, having to overcome in the canal or the tube a more considerable friction than in the central part of the current, have at their issue a less mean velocity than the latter, and are therefore attracted by them? The effect incontestably exists; but we shall demonstrate, when we come to speak of the liquid vein, that it is negligible in comparison with another cause. After trying these various causes and eliminating them one after the other, we are brought to the conclusion that the essential cause that draws toward the middle the edges of a liquid sheet falling freely is the force of cohesion, or capillary tension. It is the same force that causes a freely suspended liquid mass to take the spherical form, as in Plateau's, well-known experiment, and in the world of the stars. It is only in this state of least surface that the attractions between the molecules
| Fig. 9. | Fig. 10.—Tongue-like Form of a Jet of Water spurting from a Flattened Tube. |
are in equilibrium. A falling liquid sheet will, therefore, tend toward a cylindrical or contracting form. Does it really become a cylinder and keep the form? What takes place lower down? These questions require knowledge of other peculiarities of the primary tongue.
When water is subjected to a shock or to pressure, the pressure does not distribute itself insensibly through all the mass with
