Prof. Backers Report 



161 





The measurement is now pursued with an apparatus devised 

 by Professor Bache, founded on the known system of compensa- 

 ting bars, bat perfected by new precautions and nice contrivances 

 of mechanical adjustment. Heretofore, no further approximation 

 to accuracy of compensation had been attained, than resulted 

 from joining together bars, of about the same size and surface, 

 of different metals. The masses of the bars, have been propor- 

 tioned by Prof. Bache upon. the principle that the absorption va- 

 ries directly with the conducting power and inversely with the 

 specific heat, of the material. Hence masses proportioned to the 

 ratios of those constants, and having equal surfaces, from which 

 conditions their linear dimensions may be deduced, would theo- 

 retically bear to each other the due compensating relation. Two 

 sources of error however remain — the inexactness of the values 

 assumed for those constants, and the different rapidity of absorp- 

 tion of the surfaces, during changes of temperature. These are 

 obviated by proportioning the bars by actual experiment, subse- 

 quent to the calculated approximation, and by coating the mate- 

 rials respectively with varnishes suitable to each ; so that the 

 absorption of both will be equal in a given time and the compen- 

 sating relation continue during changing temperatures. The 

 bars while in use are constantly compared with the standard bar. 

 They have b^en exposed beforehand to a higher temperature 

 than they can be while in use, in order to give them a set, and 

 are always protected by a covering, throughout the greater part 

 of their length. These precautions, with arrangements for per- 

 fect contact, of which our limits forbid description, have ensured 

 all the accuracy desirable. During the season embraced in this re- 

 port, a base line of 6| miles was measured, in the whole of which 

 the probable accidental error did not exceed one-tenth of an inch. 



The Chesapeake triangulation will furnish the length of an arc 

 of the meridian of over 2i degrees, measured, with but moderate 

 deviation, on each side of the central meridian. 



The local attraction at the astronomical stations having been 

 found to modify essentially the determinations, those stations 

 bave been multiplied, and with the astronomical have been united 

 magnetic and meteorological observations. The method of azi- 

 niuth by observations at equal intervals before and after the cul- 

 mination of a circumpolar star has given excellent results^ requir- 

 ing but brief computations. 



By telegraphic connection with the coast stations, the obser- 

 vations of moon culminations and occultations at Cincinnati have 

 been made available for differences of longitude from Euro|)e. 

 In general, it is to be remarked that the observations of scientific 

 J)^en, where they could be made available, have been procured 

 for the Survey, and various distinguished professors have been 

 eniployed to compute results from the materials accruing to the 



Second Skries, VoL XII, No. 35.— Sept., 1851. 21 



