19 



olitaiii tlistinct indications we n'(|iiin' some incclianiial nuilti|i!ifati'in df tlio 

 displaei'iiiciit of (lie urouml iclativi'ly to tiic steady line nv steady imiiit. In the 

 iiorizontal ]H-iididnm seisnioineter tlie sini|ilest way tu I'tVect tiiis niniti|ilieatiiin is 

 t"i nialie a |ii>rtiiin nf tlie (Vame (§ 18) consist cC a iiuiit rod iirojeetinu to a enn- 

 >ideial)Ie distance iieyond //, and to use its end as a ])ointer wiiieli inseiilies the 

 record on a lixed or moving; plate of smoked fijlass. Tiieu wiien a small dis|)laee- 

 nu-nt of the axis of sup|)ort I'P takes place jiorjiendicnlar to the ])lane of FGP, 

 wiiile // remains at rest, the marking jwint moves in the o]i])iisite direction 

 thinULih a di>tance ei|iial to tiie disiilacenient of tlu' uronnd nudtipli<'d l)y 

 I )istance of ^üarkinij; Point from Instantaneous Axis 

 Distance of Instantaneous Axis from Axis of .-uppi.rt. 

 But at the sani(! time the jilate on wliicli the record is traced has l)eeii carried, 

 alon«; with otiier neighlionrini;- olijccts on the earth's surface, in the direction of 

 and i>y an amount e-jual to tiie ilisplaccmeut, and iienee tlie lentith of tiie line 

 traced on tlie smoked-t^la&s plate is the sum of these two movements, and i.- eijual 

 to the actual displaepmcnt nndtiplied hy 



Distance of ]Markin<r point from Axis of Support 

 Distance of Instantaneous Axis from Axis of Sup])ort. 

 ( )r, more simply, we may conceive a displacement o|i|'(isite to that df iIk ^ruinul 

 to l)e superposed on the actual motion of every point. Then the axis // is to he 

 taken as movinji while ohjects fixed to the earth remain at rest, and it is at oii<e 

 evident that the ahove is the ratio of multi[)licatinu. 



In the arrangement just described the jiointer is riijidly atlachcil to the 

 frame, and nnist therefore he inclii<]cd as fornuui;' j^art of the iramc in calculating 

 or finding experimentally the position oi' the steady line. 



§ 21. MiiUiplieadon by nn independent lever. 



Aithou<;h in the horizontal pendulum seismograph a mulli|ilicd rccdrd Is 

 \ery convem'ently got hy using as multiplying lever the |iivo((cd iiaMicofthc 

 pendiduni itself, it is fnipicntlv desirable in other instruments (and even in the 

 horizontal pcmluluni, if the length of the pcndidum is eonsideralile) to magnilV 

 the motion by the use of an imlependent lever, pivotlcil to the ground, aiul having 

 its short arm i-onneeted to .<ome point in the sns|)ended mas<, wiiile its long arm 

 carries the tracing jKiiiit. 



In order that suili a lever may not aH'ect the steadini-ss of any .seismometer 

 iw \>hii'h it is applied, it should in strictnr-.vs conform to the following condi- 

 tions, the lir.-^t static and the .'•econd kinetic: — ( 1 ) together with the suspended 

 m.-iss til which it is applie<l, it must foim a svstem whose iv|uilibriiiui isiniilral; 

 i'l) its point of «•onncetion with the principal mass nmst be relat<'d to its own 

 rliniensions and point or axis of support in such a way that, whiii a flislnrbance of 

 till- ground takes place, no .-Iri-ssdne to the inertia of the p.arts will be inlrodiiced 

 iwtween the nndliplying levi-r and the princ-ipal ma.ss at their point of conncc linn. 

 Tiic simplest way of .-t'ciiring the .«iccond < lition is to make the pnini nf 



