FOR DETACHED OBSERVATORIES. 525 



the arbor on which the minute hand turns is in the prolongation of the 

 axis on which the index of the barometer turns. Now, designating the 

 index by A and the minute hand by B, it is clear that B will at some 

 point of each revolution be directly over A. 



The planes in which the two hands move being supposed to lie a few 

 hundrcthsof an inch apart, let us attach a "point"* to tiie extremity of B, 

 which, passing over the extremity of A, will establish metallic connection 

 between the two without however producing sufficient friction to move 

 A out of its position. Let us also connect A by a conducting wire with 

 tin' positive pole of a battery placed at the central station, and B with 

 the negative pole of the battery ; then, at each revolution of B there 

 will be an instant when the circuit will be closed. 



At some point on the positive wire, near the battery, let us place an 

 apparatus consisting of a vertical metallic cylinder moved by clockwork, 

 so as to make one revolution in the same time as the minute hand B, 

 and also descend a little at each revolution. Let this metallic cylinder 

 be covered with a sheet of dampened paper impregnated with some salt 

 which will be changed in color under the action of a galvanic current 

 (tor instance, iodide of potassium and starch). If then a "point" con- 

 nected with the negative pole rest on the dampened paper, it is evident 

 that at each revolution of the cylinder a colored spot will be made on 

 the paper at the instant when the tw r o hands A and B come in contact. 

 As the hand B and the cylinder have an isochronous movement, it will 

 now only be necessary to have divisions on the paper corresponding to 

 those on the aneroid barometer in order that the indications of the lat- 

 ter be recorded on the paper. 



Since, however, the index of the aneroid barometer in its greatest 

 oscillations describes at most an arc of but 120° it will be evident that 

 this apparatus may serve to record more than barometric indications 

 alone. For instance, let us place opposite the aneroid and at a short 

 distance from it a circular metallic thermometer, the index of which C 

 shall have the same length as that of the aneroid, and have its axis in the 

 prolongation of the axis of the aneroid. Let this thermometer be so 

 placed that the middle point on its graduated arc (00°) will be diametri- 

 cally opposite to the middle position of the index A of the aneroid ba- 

 rometer (30 ins.). Let the clock be as before described, except that the 

 minute hand P> revolves between the planes of the two indexes A and 

 C. Nov.-. if an S-shaped "point" be attached to the extremity of the index 

 B, instead of the single point before described, then, if the instruments 

 are properly adjusted, at each revolution of B it will come once into con- 

 tact with A and once with 0. The latter being connected like A with 

 the wire from the positive pole, two records will be produced on the 

 sensitive paper at each revolution, which will never be confounded on 

 account of their different positions. 



'[This is the technical term for a metallic .strip or spring, serving t<> make and 

 break the electrical connection according as it rests on a conducting or an insulating 

 substance. — Tr. ] 



