WITHDRAWN FROM THE ACTION OF GRAVITY. 355 
to nearly 3 millimetres ; finally, in consequence of a new absorption, the figure 
disunites in the usual manner at the middle of the constricted portion. 
Second experiment.—Distance of the rings 49 millimetres. The bases pre- 
sent, in the end, a total loss of curvature, and then, as above, there is a sponta- 
neous transformation: the bases again become slightly convex, with a versed 
sine of about 1 millimetre. A new absorption brings on disunion. 
Third experiment.—Distance of the rings 47 millimetres. The bases again 
appear to become plane, and the figure continues in this state. Further absorp- 
tions seem, at first, to have no other effect than to deepen the constricted portion, 
while the bases still appear plane; then a slight convexity is re-established, 
but not now spontaneously ; it originates and increases in correspondence with 
te exhaustion; when the versed sine is about 1.5 millimetre, disunion takes 
place. 
Fourth experiment.—Distance of the rings 45 millimetres. The bases be- 
come first plane, then slightly concave. The versed sine of this concavity 
increases nearly to 2 millimetres, and again a spontaneous transformation is 
observed; the concavity is changed into a convexity, whose versed sine is 
nearly a millimetre. The action of the syringe then occasions disunion. 
Fifth experiment.—Distance of the rings 43 millimetres. The bases are ren- 
dered plane, then concave, and the versed sine of the concavity gradually at- 
tains 4 or 5 millimetres; the figure then disunites. 
§ 21. Let us consider what these experiments teach; first remarking that it 
is not easy to judge of the precise point at which the bases of the figures are 
rendered plane, for an exceedingly slight curvature eludes the sight. Hence 
arises some uncertainty in the determination of thé limit of height of the cate- 
noid; fortunately the particulars which we have noticed will furnish us a means 
of appreciation more exact. . 
In the fourth experiment we necessarily realize plane bases, from the cireum- 
stance that the curvature, from being convex, becomes gradually concave by 
the progressive absorption of the liquid; but is this the case likewise in the 
second and third, in which we seemed also to have realized planes? This, 
is a point. which we will attempt to elucidate. The first, second, and third ex- 
_periments have this in common, that a small spontaneous modification or trans- 
formation of the figure is produced therein, while in the third this phenomenon. 
does not occur; and this modification is observed decreasing from the first to. 
the second, disappearing in the third, and reappearing in the fourth. From this: 
we should infer that the third experiment forms a sort of transition, on one and 
the other side of which the spontaneous transformations are manifested ; but 
the effect was shown in the first experiment when the bases had still a visible 
curvature, and in the fourth when they had assumed one in an inverse diree- 
tion; it is highly probable, then, that in the second, at the moment when the 
spontaneous transformation was seen to occur, the bases still preserved a real 
curvature, though too feeble to be distinguished ; and that it was only in the third, 
where the distance of the rings was 47 millimetres, that bases entirely plane 
were attained. If, in this third experiment, the bases conceived to be plane 
seemed not to begin to lose this state until after the absorption of a very con- 
siderable quantity of liquid, that evidently results from the difficulty mentioned 
above of clearly distinguishing the point at which the curvature is annulled. 
Thus, for our rings of 71 millimetres diameter, we may admit that the dis- 
tance of 47 millimetres differs very little from that at which we begin to obtain 
bases strictly plane; and since 47 is obviously 3 of 71, we may conclude that 
