272 
searches was the imitation of the phenomena of magnetic 
and electrical attraction and repulsion by analogous 
attractions and repulsions produced between pulsating or 
vibrating bodies immersed in liquid. The extreme im- 
portance of such experiments in. hydrodynamical theory 
was so well pointed out by Prof. George Forbes in the 
former article, that nothing need be said here in that 
respect. The present article is confined to a concise de- 
scription of the apparatus of M. Bjerknes, and of the 
results obtained by its means. 
Fig. 1 depicts the fundamental piece of apparatus for 
showing the action between two pulsating drums or tam- 
bours, Aand &, These tambours consist of metal cups 
covered with an elastic membrane. Each tambour com- 
municates by a tube with an apparatus by means of which 
Fic. «. 
the elastic membrane is set in regular pulsation. A 
pulley, D, driven by a band from a multiplying wheel, 
works, two small cranks whose rotations are converted 
into rectilinear movements by connecting-rods attached 
to two other tambours, C and Cc’; the latter serving as 
pumps to alternately compress and rarefy the air in the 
tubes which communicate with A and B. Fig. 2 shows 
the construction of a single pulsating tambour, When 
air is compressed into it the membrane is forced outwards 
as at C, when the pressure is withdrawn the membrane is 
drawn inas at D. Returning to the particular apparatus 
of Fig. 1 the result of setting the adjacent tambours in 
synchronous vibration when the whole apparatus is im- 
mersed in water, is as follows. If the tambours are 
arranged so that the movements are in similar phases 
Fic. 2. 
they attract one another; if in opposite phases they 
repel. This is exactly the inverse of what occurs for 
magnetic poles or electric charges, where similars repel 
one another and dissimilars attract one another. The 
tambour A is mounted upon a lever capable of turning 
upon a pivot at Pp, and balanced by a counterpoise at 4’. 
The connection with the pump C is made through a ver- 
tical tube of india-rubber which permits of limited rotation 
about a vertical axis. The tambour B is held in the hand 
at a convenient distance, just as one may hold in the 
hand a magnet to show its action upon a_ balanced 
compass needle. 
A second fundamental piece of apparatus is the oscil- 
lating sphere shown in Fig. 3. Toa pulsating tambour 
is fixed a small plate of metal, c, bearing a stiff wire 
NATURE 
[Fax 19, 1882 
terminated by a small ball, A, and supported by a metal 
guide. The oscillations executed by such a: ball when 
pulsations of air are directed into the tube D, are of 
course rectilinear displacements to and fro. Such an 
oscillating sphere presents at opposite sides opposite 
phases of displacement. Accordingly if the tube D be 
connected with the pumping apparatus, and a tambour, 
such as that just described, communicating with the other 
pump, be held near the sphere A, it is found that the 
effects are of two opposite kinds, according to the position 
of the tambour. The oscillating sphere resembles then a 
magnet in having two poles of opposite properties. These 
figures are from sketches furnished some months ago by 
M. Bjerknes. His apparatus, as shown in the Paris Ex- 
hibition, included a more powerful means of producing 
the pulsations. In Fig. 4, which shows the tank and the 
collection of small pieces of apparatus, the driving gear 
of the pumps is not shown; but the pumps themselves 
are drawn in the lower right-hand corner of the figure, 
and consist of two small metal cylinders fitted with 
pistons ; the connecting gear being so arranged that their 
movements can be made at will either in similar or in 
opposite directions. In the figure the fundamental expe- 
riment of the mutual action of two pulsating tambours is 
being shown. Two pulsating elastic spheres show similar 
results, but are less easy to manage. Some of the other 
portions of apparatus comprised in. the: collection are 
separately shown in Fig. 5. Of these the first is a double 
tambour whose two faces execute pulsations of similar 
phase. The second is a double tambour, the two cham- 
bers of which communicate separately with the two 
Fic. 3. 
pumps, and in which therefore the two membranes 
execute movements in opposite phases. Thirdly, we 
have two spheres, of which one (on the left) is mounted 
so as to execute horizontal oscillations ; the other (on the 
right) to oscillate vertically. It is possible to mount the 
oscillating spheres, either singly, or in pairs, upon hori- 
zontal axes, in which positions they act like mounted 
compass needles, following the action of another pulsating 
body ; but always with the inversion of phase alluded to 
previously, Ze phase producing attraction; wmdike, re- 
pulsion. 
In the foregoing cases the pulsating and oscillating 
bodies act on one another, producing (inversely) mutual 
actions and reactions as the poles of magnets do. M, 
Bjerknes has also succeeded in imitating the phenomena 
of diamagnetism and of magnetic induction. 
Diamagnetism is imitated by making the pulsating 
bodies act upon objects lighter than water suspended 
within the liquid by a thread attached to a weight. A 
small ball of cork thus suspended is repelled from both a 
pulsator and an oscillator in whatever phase the latter 
may move. Temporary induced magnetism is imitated 
by the behaviour of balls of some material heavier than 
water suspended from a float. Such a body is attracted 
by a vibrating or pulsating body. In the nearest corner 
of the tank of Fig. 4 are a pair of heavy bodies hung to 
threads for this very experiment. A little further to the 
right is shown a support from the top of which is sus- 
pended a little cylinder of heavy metal. When the pul- 
sating tambour is held near this it turns round and points 
towards the tambour as a suspended piece of soft iron 
does towards a magnet pole. The same support carries a 
