Ml 



QBODB8T. 



QBODK8T. 



line*, on a plate of metal, securely fixed in the ground. In order to 

 tnc the line of the hue, a tranoit-iusti uuient is adjusted over one of 

 the terminal points, and directed to a flag-staff or other signal erected 

 at the other, by which mean* the observer is enabled to direct an 

 eseistint to plant picket* in the ground, all ranging in the same vertical 

 plane. The measuring apparatus may be constructed and applied iu 

 various ways, but in all cases the following conditions must be ob- 

 served : 1, The successive measuring-rods (or chains) must be arranged 

 accurately in the vertical plane passing through the terminal points of. 

 the base ; 2, the temperature of the rods must be observed at the time 

 they are applied, and their rate of expansion determined by direct 

 experiment, in order that the apparent lengths may be reduced to the 

 length at a given temperature ; 3, they must be supported in such a 

 manner as to have no tendency to flexure; and 4, each rod, wh.-n 

 adjusted in the line of the base, must be exactly horizontal, or its 

 inclination exactly determined by levels, in order that the correspond- 

 ing horizontal distance may be computed. 



Previously to General Roy's measure of the base on Hounslow Heath 

 the measuring apparatus used in all similar operations (with the excep- 

 tion of one of the bases measured by Lacaille and Cassini in France) 

 consisted of deal rods, usually about 20 feet in length ; but deal rods, 

 however well seasoned, are found to bo considerably affected by the 

 hygrometrical state of the air, and liable to sudden and irregular 

 expansions and contractions, the effects of which cannot be accurately 

 estimated, even when they are compared from time to time with a 

 standard metallic bar. For this reason General Roy, after he had 

 measured the base with deal rods, measured it a second time with 

 hollow glass tubes. The tubes were 20 feet in length and about an 

 inch in diameter, and the extremities were defined by metal buttons, 

 ground perfectly flat, and perpendicular to the axis of the tube. At 

 one end this apparatus was fixed to the tube; at the other it was 

 attached to a sort of piston, or slider, capable of being pushed up a 

 certain way within the tube against a slender spring. A fine line 

 marked on the slider, and another oil the glass tube, were so adjusted, 

 that when the slider was pushed up until the lines came into coinci- 

 dence (which was observed through the glass), the distance betwi -n 

 the flat ends of the metal buttons was exactly 20 feet at the tempcra- 

 ture of 62 Fahr. In making the contacts, the flat end of one nl wa.< 

 pressed by a screw apparatus against the moreablc end of the other, 

 until the coincidence of the lines took place ; so that a constant pres- 

 sure was applied equal to the tension of the spring. When the Ord- 

 nance survey was begun, this base was again measured with a steel 

 chain, made by Ramsden, of 100 feet in length, and consisting of 40 

 links, terminated by a brass handle at each end. When in use the 

 chain was supported on wooden coffers, raised on trestles, and stretched 

 with a weight of fifty-six pounds. The extremities were denned by 

 marks on the brass handles ; and in order to bring the marks succes- 

 sively to the same point on the base-line the following contrivance was 

 adopted : a brass scale, firmly secured to a post driven into the ground, 

 but not connected with the chain or its supports, was placed contiguous 

 to the preceding end, A, and adjusted by means of a screw apparatus, 

 so that a division on the scale coincided exactly with the mark on the 

 handle. The chain was then carried forward one length ; and when 

 again placed on the coffers, was drawn back agaiust the stretching- 

 weight, until the mark on the handle at the following end, B, coincided 

 with a division of the scale which had remained fixed in its plan . In 

 this manner the marks on A and B either occupied the same point on 

 the line of the base, or the distance between them was known in terms 

 of the scale. The same method was followed in measuring several 

 other bases in England in the course of the survey, the chain being 

 compared before and after each operation with another similar chain 

 which was kept for this purpose. The advantage of the chain depends 

 solely on its great length, by reason of which the number of coinci- 

 dence* (in making which the chief difficulty consists) is considerably 

 reduced ; but this advantage is probably more than counterbalanced 

 by it< liability to alteration from wearing or corroding at the joints, 

 and the insufficient security afforded that, notwithstanding the - 

 ing, all its points are in the same straight line. The use of the chain 

 has been confined to the English and Indian surveys, and is now 

 abandoned in both. 



The apparatus used by Delambre and M ('chain in measuring the two 

 bones in France, on which the length of the meridian depends, was 

 constructed as follows : The measuring-rod was a thin bar of platinum, 

 two toisea in length, half an inch in breadth, and two lines in thickness. 

 Thiii was covered by another bar of copper, somewhat shorter, the two 

 being firmly connected by screws at one end, but free at every other 

 point, so that the expansions and contractions of each . 

 fleeted by the other. The relative expansions were thus indicated 

 by the distance between the free extremities of the two bars ; for 

 measuring which n finely divided scale was cut on a part i.f th. 

 num liar, and a vernier attached to the extremity of the copj.rr liar. 

 The relative expansion gave the means of computing the absolute ex- 

 pansion of the platinum bar at the time of the measurement. 1 

 these compound bars were used together ; and when all were adjusted 

 in the line of the base, and the requisite observations made and re- 

 corded, the last was carried forward and placed first, the others being 

 left in their places. The ends were not brought into contact ; a small 

 interval wae left between each and the next, which wu measured by a 



slider, or email scale of platinum, attached to that end of the platinum- 

 bar which was not covered by the copper bar. For the measure of the 

 Hanoverian base Gauss employed three bars of hammered iron, two 

 toises in length, and about an inch and a half in breadth and thickness, 

 enclosed in boxes so as to leave only the ends projecting, and supported 

 by counterpoises to prevent flexure. The ends were covered with steel 

 plates, one flat and the other spherical ; and when the bars were 

 placed in the line of the base, a small interval was left between the 

 flat end of one bar and the spherical end of the preceding one, which 

 was measured by dropping into it a thin wedge. Strove also employed 

 iron bars in measuring the Russia base. In this case, however, the 

 ends of the bars were brought into contact, and the contacts made 

 on a principle similar to that which had been adopted by General Roy 

 in using the glass rods. But the apparatus which formed the move- 

 able extremity of Struv.-'s r.xl consisted of a bent K-MT which turned 

 on an axis passing through the bar, near its extremity, the mo: 

 the short arm of the lever being in the direction of the length of the 

 bar. The short arm terminated in a hemisphere which projected a 

 little beyond the end of the bar ; and in making the contacts, the flat 

 and fixed end of the next bar was pressed against the hemisphere, 

 until an imlex on the extremity of the long arm of the lever stood at 

 Idle of a scale connected with the bar, in which poeitioirthc 

 distance between the plane in which the bar terminated at one end, 

 and the apex of the hemisphere at the other end, was exactly two toises 

 at the normal temperature. A steel spring acting on the long arm 

 of the lever regulated tho degree of pressure applied in making the 

 contacts. 



In measuring the Prussia base Beesel used four compound bars con- 

 structed on the same principle as those of Delambre. The apparatus, 

 which is simple and ingenious, will be readily understood from the 

 annexed figure, in which the two ends are shown, i i' is a liar of iron 



Fig. 1. 



<L 



two toisea in length, an inch 1 .nod, and a quarter of an inch thick 

 z z' is a bar of zinc of the same thickness and half the breadth. The 

 two bars are firmly connected by screws and soldering at the end c i. 

 The zin rminatcs in two knife-edges of steel, the edges being 



horixontal. X- /' in a irt of the iron bar, terminating in knife-edges, 

 which are placed vertically, or at right angles to the former. Tho 

 Icmrth of the measuring bar, or two toises at the normal temperature, 

 in defined I>y the knife-edges at z and L J ; while the interval h< 

 :' and k varies with the temperature, and indicates the relative 

 sion. 1 1 om which the absolute expansion of the iron bar becomes known. 

 This interval was measured by inserting a thin glass wedge b. ' 

 the knife-edges ; and when the bars were placed in the line of the base, 

 the interval between the knife-edge f of one bar and z of tho adjacent 

 one was measured in the same manner. By means of fine di-. 

 on the parallel sides of the glass wedges an interval in the 

 tion of the base so small as the 12,000th port of an inch was made 

 visible. 



For the measurement of the Irish base an apparatus was en. 

 by Colonel Colby in which the expansion of the measuring-rods was 

 compensated in such a manner that no reduction was required on 

 account of variations of temperature. The principle of tint- very in- 

 genious apparatus will be understood from the following description : 

 A B is a brass bar, 10 feet in length, c D a bar of iron, tirmly connected 



Fig 2. 



with tl i uigties A ]', n (j, arc con- 



with the extremities of tho liar* l.y double conical joints which 

 efficient play to the tongues to prevent interference with the 

 free expansions of the bars. At a certain normal temperature the 

 steel tongues arc perpendicular to the direction of the bars. Suppose 

 them in this position, anil that au increase of temperature takes place : 

 the brass bar will become longer than the iron bar in consequence of 

 its greater relative cxi>ansion, and a straight line A c p (on the middle 

 of the tongue) will come into the position a cr, intersecting its I 

 direction at a point P, so xituatcd that p c has tor tin' ratio of the 

 expansion of the iron bar to that of the lirass bar. But this ratio 

 instant, P ie a given point ; an. I a-' tin oorre ponding point o, at) 

 the other tongue is in precisely the same circumstances, it folio,, 

 the distance between r and o. will remain unaltered in all temperatures, 

 provided at least both bars have the same temperature. This in- 

 variability is, however, not absolute, for when the tongue comes into 

 the oblique position a f is longer than A r, and consequently the point 

 marked on the tongue will be between a and the invariable point. 

 Hence an increase of temperature must increase the distance between 



