[ 225 ] . 



XXX. On Scheutz's Calculating Machine. 

 , By G. B. Airy, Esq.,, Astronomer Royal. 



To the Editors of the Philosophical Magazine and Journal. 



Royal Observatory, Greenwichj 

 Gentlemen^ August 25, 1856. 



I RE MARK that in your Supplementary Number for July 

 you have transcribed the Report of the Royal Society's Com- 

 mittee on M. Scheutz's Calculating Machine. I had the pleasure 

 of examining that machine shortly after its arrival in this country, 

 and I cannot refrain from expressing my general admiration of 

 the beauty of its arrangements, and my assurance that such a 

 machine is not likely to get out of order, and that it can be con- 

 structed at comparatively small expense. I am also impressed 

 with the ability and accuracy of the Committee's Report. But 

 I wish to guard myself from giving an opinion on the utility of 

 the machine ; remarking only that, as I believe, the demand for 

 such machines has arisen on the side, not of computers, but of 

 mechanists. 



Permit me, however, to point out to you a course of computa- 

 tions, of a most mechanical and monotonous kind, which is going 

 on frequently in every place of extensive and systematic calcu- 

 lations, and to which I think the attention of the constructors 

 of mechanical computing machines might advantageously be 

 directed. 



An immense number of computations consist of parts follow- 

 ing each other in the following order : — 



1st. Independent calculations are made (for instance, from 

 the lunar tables) for equal intervals (for instance, for every twelve 

 hours) . 



2nd. The differences of the computed numbers are taken, 

 probably to the fourth order, as a check on th« accuracy of the 

 computed places. 



3rd. Differences of the first and second order are formed, ap- 

 plicable to the interpolation of places for every hour. 



4th. These differences are appKed to form the places for every 

 hour. 



The most tedious of these operations is the second (I assume 

 that the first can never be dispensed with) . 



This second operation appears to be well adapted to mechanical 

 action. Speaking (for the present) without mature consideralfton, 

 I should conceive, that, by reversing the order of the figures oU 

 some of the wheels, and with some other instrumental changes, 

 and by feeding the proper wheels continually with new computed 

 numbers instead of new differences, the additive operations of 

 M. Scheutz's engine might be made subtractive ; and the new 

 Phil Mag. S. 4. Vol. 12. No. 78. 8ept, 1856. Q 



