6 M. J. Plateau on Jets of Liquid under the 
The new hypothesis, as well as the old one, therefore establishes 
the close approximation to uniformity in the motion of the con- 
tractions through the small space in question, and consequently 
the remaining part of the paragraph may be accepted as correct. 
§ 3. These rectifications being made, we proceed to the subject 
in hand by recalling briefly the modifications which, according to 
the researches of Savart, every jet receives under the influence 
of vibrations. The first fourteen of the following numbers refer 
to jets descending vertically :— 
No. 1. The continuous part becomes shortened. 
No. 2. The thickness of the limpid portion appearsaugmented.. 
No. 3. Each of the masses which detach themselves at the 
lower extremity of the continuous part becomes flattened first 
in a yertical direction, that is to say, its horizontal diameter 
becomes greater than that of the sphere it tends to constitute. 
No. 4. These flattened masses, being abandoned to themselves, 
tend to assume the spherical form, about which they only oscil- 
late, however, in consequence of inertia; thus they become 
alternately flattened and elongated in a vertical direction, so that 
their horizontal diameter, which was at first greater than that of 
a sphere of the same volume, becomes afterwards less, then 
greater again, and so on. 
These periodic variations in the horizontal diameter of the 
masses taking place simultaneously with their translatory motion, 
the impression left upon the eye by the rapid passage of any one 
of these masses would be that of a figure presenting a regularly 
disposed series of maxima and minima of thickness; the first 
corresponding to the places passed by the mass in its phases of 
greatest horizontal development, and the second to those by 
which it has passed during its phases of greatest horizontal con- 
traction. Now as the successive masses pass (either exactly or 
nearly so) the same places in the same phases of their oscilla- 
tions of form, the impressions which they would individually 
produce become superposed more or less completely, and the 
agitated part of the jet presents in a permanent manner the dif- 
ferences in thickness in question; in other words, this agitated 
part appears composed of a regular series of ventral segments 
whose nodes occupy fixed positions. 
When the above superposition is imperfect, each ventral seg- 
ment has the appearance of an assemblage of waves, each of 
which constitutes a kind of cone having for axis that of the jet 
itself. About the half of the first ventral segment is formed by 
the passage of the expansions of the bottom of the continuous 
part, so that the latter terminates near the middle of this ventral 
segment. 
No. 5. The length and diameter of the ventral segments are 
