49 



The aiMitioi) wliicli ^Fr. finiv lias inadc, willi tlic virw nf makinji; tlif 

 iMluililiriiini .■^till ni.tii' nearly neutral, is a troiicrh or tiiUe contaiiiin»- a li(|uiil, 

 which is eonneeted to the horizontal har in sneh a manner that when the liar is 

 depressed the liijuid moves in the tuiie so as virtually to inerease the load on tiie 

 liar, and when it rises t!;e lii|nid moves so as virtually to decrease the load. The 

 apparatus isshown in sectional elevation in 1 'late A' IT, tiii-.n.'j,\vhieii, with the follow- 

 ing description, is taken from the Piiiiosopiiieal ^^a<razine for Sept. 1 SSI , p. •2011 :— 



"A vei-tieal sprinji; S is fixed at its njiper end hy means of a luit //, which 

 re.<ts on the top of the frame F, and serves to raise or lower the sprin«; through a 

 short distance as a last adjustment fur the ])osition of the cross-arm A. The arm 

 A rests at one end on two sharp ])oints, ji, one n'sting in a conical hole and tiie 

 other in a V-sIot; it is sup])ort<'d at Ji hy the sprini;- S, and is weighted at C 

 with a lead ring /.'. (^ver a pin at the point C a stirrup of thread is placed 

 which supports a small trough f. The trough f is pivotteil at a, has attached to 

 it the index i (which is hinged hy means of a strip of tough paper at /(, and 

 re.-ts through a fine pin on the gla.ss plate fj), and is jiaitly filled with iiierciny.... 

 AVhen the plane earrying the .«])ring A' is raised and lowered, the \wint a rises and 

 falls, hut in eonstKjuence of the inertia and slow ])eriod of li the point C remains 

 liehind. In coasc<|nenec of this the end of the trough t, falls and rises relatively 

 to rt ; and the mercury, running 1 ackwards and forwaids, pni-^ more or less force 

 on the point ( \ and hence tends to keep this ))oint stationary." 



An ohjection to this arrangement is that horizontal oscillations of the ground 

 produi-e a Ix'aping up of the liipiiil at one and the other end of the tuhe alter- 

 natelv, and .-o throw the har into a ,-tat" of forced vertical oscillation. 



?! -47. A'txllf ffiiri-.i'iilii/ Il'ir, irilliinii llijiiiil. 



A .-im|)ler and e<|iially clVci-livc meiliod nl ri ducing a loaded horizontal har 

 to apjiroximate astaticism, tlie invention of the picseui writer,* is as follows: — 

 Instead of changing the load upon the har, when the har ri.-^'s or falls, we may 

 change the leverage at which the spring acts. The moment of the weight is 

 lialriiieed, al'out the fulernm, hy (he moment of the iijiward pull of the spring. 

 In any >mall displacement the moment of the weight is sensihiy con>lant. For 

 neutral r'(|uilil)riinn we mir-t therefore make the moment of thi' pull of the spring 

 constant. To do this we must cau.^e the point at which the spring is attached to 

 till' l.ar to niove towatds the fideium when the spring lengthens, and away from 

 (he fidcrnin when the; s|)ring shortens. This i- ea.-ilv done hy jiutting the point 

 of attachment // (in the ski tch on |i, |H| not in the hm i/onial line joining .1 with 

 (', iilit at a di>-tanee r vertically lielow that line. Let // le the horizontal 

 di>laiie<- of the spring's line of action from the fulenini .1. Tlicii if we suppose 

 the har to he displaced downwards thiongh any very small angle ilO from its 

 normal hmizontal position, the point of altachment of the spring goes vcitically 



• 'I'miiK, ufllie Sciniiioliigioal Soi-ici} uf Jn|.iiii, Vol. Ill, |i. I in. 



