ON THE KEW MAGNETOGRAPHS. 32? 



The tabulation is a far more serious matter as regards the time consumed 

 in the operation. Where traces vary from day to day so much as magnetic 

 traces are liable to do, the time required in tabulation must also be liable to 

 vary considerably; but it is found on a general average by Mr. Welsh, to 

 whose able and careful management Mr. Ronalds has confided this depart- 

 ment of the experimental trial, that the trace of each instrument in twenty- 

 four hours takes about three-quarters of an hour to measure and tabulate, 

 both sides of the trace being measured, and a mean taken between them. 

 The extent to which tabulation is carried is so calculated as to meet the va- 

 rious inquiries for which it is likely to be required, and generally sj)eaking 

 to reproduce the trace, though not of course perfectly in all cases. When 

 the processes of tabulation and copying are completed, the plate is cleaned 

 and is ready to be used again. 



Taking into account the labour of the photographist and that of the person 

 charged with the operations that have been just described, the work of a 

 photographical magnetic observatory in the present state of the apparatus 

 can by no means be deemed light ; but it has the advantage, and it is a great 

 one, of producing a continuous record. 



The plates are prepared and changed once in every twenty-four hours ; 

 the length of the plate is twelve inches, and as this corresponds to the time- 

 scale, which is an inch an hour, each plate is inverted at the expiration of 

 twelve hours, and thus at the expiration of the twenty-four hours each plate 

 is charged with a double trace, each trace having also its attendant zero line. 

 The width of the plates employed at Kew is three inches, being deemed a 

 fair ordinary width. In parts of the globe where great disturbances prevail, 

 and where at the same time great precision is required in the variations of 

 sjnall amount, a larger field than that of three inches may be desirable, and 

 may be given, but for moderate variations three inches seems an ample field. 

 It has been found by the concurrent experience of several persons that the 

 traces on the plates can be read off with full confidence to the 500th of an 

 inch; consequently the three-inch plate gives a scale of 1500 distinctly re- 

 cognizable parts. Taking the Declinometer as an example which will be 

 most generally understood, if a single division of the scale is made equal to 

 six seconds, the range of the scale will extend to 1500X6 = 9000 seconds, 

 or 2° 30'. 



Mr. Ronalds's instruments have a great advantage in the general stability 

 and freedom from sliake of the apparatus ; all parts which have a reference 

 to each other are bound together with marble and metal, the only wood em- 

 ployed being used strictly for exterior casing only. They have also a great 

 advantage in the constancy of the zero line, which is assured by mechanical 

 contrivance. The zero line has thus a permanent absolute magnetic value, 

 wliich remains unchanged from day to day, as well as at all times of the 

 same day. 



The time of exposure to the light by which each part of the trace has 

 been formed is one minute and a half. It is probable that by improvements 

 in photographical preparations the trace may hereafter be formed by an ex- 

 posure of much less duration, and that the length of the magnets, which is 

 now twelve or fifteen inches, may also be greatly reduced. Both j;hese im- 

 provements appear to be required if we desire that the trace should indi- 

 vidualize perturbations succeeding each other with great rapidity. If Mr. 

 Ronalds should succeed in substituting for the silvered plates a less costly 

 material, on which the traces might be taken with equal sharpness, and with 

 equal or greater rapidity, the photographic part of his invention would 

 seem to need scarcely any further improvement,—- Jit??/ 1851. 



