14 M. J. Plateau on Jets of Liquid under the 
extension and this vertical flattening must render. the capillary 
pressure at the contour of the masses greater than that at points 
near the axis, whence there will arise an increasing resistance 
which will ultimately destroy the transversal velocity. 
The differences of pressure will then act freely, and the mass 
will return towards its spherical figure of equilibrium; but the 
phenomenon taking place with an accelerated velocity, this latter 
figure will not be a permanent one; on the contrary, the mass 
will become contracted in a horizontal, and elongated in a ver- 
tical direction, until the increasing resistance which results from 
the new inequalities between the pressures shall have destroyed 
the velocity acquired; afterwards the mass, being solicited by 
the differences of pressure which produced this resistance, will 
again return towards the spherical figure, and again go past it 
in order to become extended a second time horizontally and flat- 
tened vertically, after which it will reeommence the same series 
of modifications, and will continue its oscillations of form as long 
as it continues to descend. 
Thus, in the case of unison with the note which the shock of 
the discontinuous part would produce, the phenomena recorded in 
the first four numbers of paragraph 8 are very simply explained. 
Since, however, the continuous part of the jet reaches almost 
to the middle of the first ventral segment, and consequently 
almost to the point of the agitated part corresponding to the 
first of its maxima of thickness, we must admit that each mass 
attains its first phase of greatest horizontal development a little 
before it becomes completely detached, and at the moment, no 
doubt, when it is connected with the mass which follows it only 
by a mere thread. 
As to the appearance of systems of waves presented by the 
ventral segments when the phenomena are not altogether regular, 
it is evidently, as recognized by Savart, the result of the inexact 
superposition of many ventral segments produced individually by 
the successive masses: these ventral segments are seen simulta- 
neously, and appear as it were through one another, in conse- 
quence of the persistence of their impressions upon the retina. 
§ 9. It is evident that the interval of time between two phases 
of greatest horizontal contraction, or, in other words, the time 
employed by each mass in executing a complete oscillation of 
form, is independent of the velocity of translation ; consequently 
the space described by a mass during the time in question in- 
creases with the velocity of translation ; but this space is evi- 
dently the distance between the central points of two nodes or 
the length of a ventral segment*; this length, therefore, ought 
to increase with the charge. 
* It is thus that Savart appears to regard ventral segments whenever he 
