MECHANIC S. 
artifice will apply their lips to the other fide, and not meet 
with the fame difappointment. 
It is a known raft in Mechanics, that the vibrations of 
a pendulum are ifochrone, or of tire fame duration, though 
the arcs it defcribes are unequal. It is alfo acknow¬ 
ledged, that, in their duration, the vibrations of two un¬ 
equal pendulums are to each other as the fqnare-root of 
their refpeftive lengths. The motion of water vibrating 
in a fiphon follows the fame laws. To illustrate this, let 
us fuppofe Inom , fig. 6i, to be a fiphon confuting of three 
parts, or legs; two, In, mo, vertical, and one, no, horizon¬ 
tal ; and that it be of an equal diameter throughout its 
whole extent. Let us farther fuppofe, that the fluid, 
while at reft, occupies the fpace anod-, the two lurfaces, 
a b, cd, will be level. Now, if by any means the fluid be 
forced to defcend to gk in the leg me., it will rile to efin 
the leg In-, and, as ioon as this caule ceales to aft, and 
the fluid is left alone to the adtion of its gravity, the ex¬ 
cels in length of the column en over the column ho, will 
'force the fluid to defcend even below the level of the 
other, on account of the acceleration it acquires in de- 
fcending, which will raife the fluid in the other leg ; and 
it will thus continue rifing and falling alternately, form¬ 
ing ofcillations fimilar to thofe of a penduium ; and the 
duration of each vibration will be precifely the fame as 
the vibration of a pendulum whofe length is half the 
column pqr of the fluid. As the ofciilations of water 
follow the fame laws as thofe of a pendulum, if the length 
of the column of water is increafed ordiminifhed, the du¬ 
ration of the ofcillations will be alfo augmented and dimi- 
niflied, and_wil) be in a fubduplicate ratio of this length. 
We lhall conclude our notice of the fiphon with the ex¬ 
planation of a curious phenomenon obferved in many 
places. Certain fprings or fountains are obferved to have 
periods of repletion and fcantinefs, or feem to ebb and 
flow at regular intervals; and fome of thefe periods are 
of a complicated nature. Thus a well will have feveral 
returns of high and low water, the difference of which 
gradually increafes to a maximum, and then diminiflies, 
juft as we obferve in the ocean. A very ingenious and 
probable explanation of this has been given in N° 424. of 
the Philofophical Tranfaftions, by Mr. Atwell, as follows. 
Let A BCD (fig. 63.) reprefent a cavern, into which wa¬ 
ter is brought by the fubterraneous paflage OT. Let it 
have an outlet M N P, of a crooked form, with its higheft 
part N confiderably raifed above the bottom of the cavern, 
and thence (loping downwards into lower ground, and 
terminating into an open well at P. Let the diinenlions 
of this canal be fuch that it will difeharge much more 
water than is fupplied by TO. All this is very natural, 
and may be very common. The effect of this arrange¬ 
ment will be a remitting fpring at P; lor, when the ca¬ 
vern is tilled higher than the point N, the canal MNP 
will acLas a fiphon; and by the conditions alfumed, it 
will difeharge the water falter than TO fupplies it; it 
will therefore run it dry, and then the fpring at P w ill 
ceafe to furnifh water. After fome time the cavern will 
again be filled up to the height N, and the flow at P will 
re-commence. If, befides this lupply, the well P alfo re¬ 
ceive water from a conftant fource, we lhall have a reci¬ 
procating fpring. The fituation and diinenlions of this 
fphon-canal, and the fupply of the feeder, may be fuch, 
that the efflux at P will be conftant. If the lupply in- 
creafe in a certain degree, a reciprocation will be pro¬ 
duced at P with very lhort intervals; if the fupply dimi- 
nilhes confiderably, we lhall have another kind of recipro¬ 
cation with great intervals and great differenaes of water. 
If the cavern have another fimple outlet R, new varie¬ 
ties will be produced in the fpring P, and R will afford a 
copious lpring. Let the mouth ot R, by w hich the water 
enters into it from tiie cavern, be lower than N, and let 
the fupply of the feeding-fpring be no greater than R can 
difeharge ; we lhall have a conikint fpring from R, and P 
will give no water. But fuppofe that the main feeder in- 
747 
creafes in winter or in rainy feafons, hut not fo much as 
will lupply both Pnnd R ; the cavern will fill till the wa¬ 
ter gets over N, and R will be running all the while; but 
foon after P has begun to flow, and the water in the ca¬ 
vern finks below R, the ftream from R will (top. The 
cavern will be emptied by the fiphon-canal MNP, and 
then P will ftop. The cavern will then begin to fill, 
and when near lull R will give a little water, and foon 
after P will run and R flop as before. 
Various Improvements and Modifications in Pumps for raifing 
Water. 
To deferibe or even to enumerate, the immenfe variety 
of combinations of the three fimple pumps would fill a 
volume. We lhall (cleft a few, which are more deferv- 
ing of notice. 
The common fucking-pump may, by a fma -11 addition, 
be converted into a lifting-pump, fitted for propelling, 
the water to any diftance, and w ith any velocity. Fig. 64. 
Plate XXIII. is a fucking-pump, whofe working-barrel 
A C D B has a lateral pipe AEG HF connected with it 
clofe to the top. This terminates in a main, or rifing pipe, 
I K, furnilhed or not with a valve L. The top of the 
barrel is flmt up by a ftrong plate MN, having a hollow 
neck terminating into a fimall flanch. The pilton-rod 
Q R pafles through this neck, and is nicely turned and 
polifned. A number of rings of leather are put over tlie 
rod, and ftrongly comprelfed round it by another flanch 
and feveral lcrewed bolts, as is reprefented at O P. By 
this contrivance the rod is clofely grafped by the leathers, 
hut may be eafily drawn up and down, while all paflage 
of air or water is effectually prevented. The pifton R is 
perforated, and furnifhed with a valve opening upwards. 
There is alfo a valve T on the top of the fuftion-pipe' 
VX; and it will be of advantage, though not abfolutely 
lieceflary, to put a valve L at the bottom of the rifing 
pipe. Now, fuppofe the pifton at the bottom of the work¬ 
ing-barrel. When it is drawn up, it tends to comprefs 
the air above it, becaufe the valve in the pifton remains 
Unit by its own weight. The air therefore is driven 
through the valve L into the riling pipe, and efcapes. In 
the mean time, the air which occupied the fmall fpace be-v 
tween the pilton and the valve T expands into the upper 
part of the working-barrel; and its elafticity is fo much 
diminifhed thereby, that the atmofphere prelfes the water 
of the ciftern into the fuftion-pipe, where it will rife till 
an equilibrium is again produced. The next downward 
ftroke of the pifton allows the air, which had come from 
the fuction-pipe into the barrel during the afeent of the 
pifton, to get through its valve. Upon drawing up the 
pifton, this air is alio drawn off through the rifing pipe,. 
Repeating this procefs brings the water at laft into the 
working-barrel; and it is then driven along the rifing-pipe 
by the pifton. 
This is one of the heft forms of a pump. The rarefac¬ 
tion may be very perfect, becaufe the pifton can be brought 
fo near to the bottom of the working-barrel; and, tot- 
forcing water in oppolition to great preffures, it appears 
preferable to the common forcing-pump ; becaufe, in that, 
the pilton-rods are comprelfed and expofed to bending, 
which greatly hurts the pump by wearing the pifton and 
barrel on one fide. This foon renders it lets tight, and 
much water fquirts out by the tides of the pilton. But in 
this pump the pifton-rod is always drawn or pulled, which 
keeps it tiraight; and rods exert a much greater force in 
oppolition to a pull than in oppolition to comprellion. 
T iie collar of leather round the pilton-rods is found by- 
experience to need very little repairs, and is very i nper¬ 
vious to water. The whole is very acceliible for repairs , 
and in this refpeft much preferable to the common pump 
in deep mines, where every fault of the pilion obliges us 
to draw up lome hundred feet of pilton-rods. By this 
addition, too, any common pump tor the lervice of a 
houfe is converted into an engine for extinguifhing fire, 
i/St 
