INTRODUCTION. XXXVTI 



ivory float is precisely halt' an inch in height, and, consequently, the distance between the 

 delicately cut zero line on the steel rod, and the divisions of the scale, is that amount lean than 

 the scale represents. For belli T reading at night it was found convenient to blacken the upper 

 eil^e of tlic ivory float, and fill the zero line with white paint. Adjustments for zero, rendered 

 necessary by flurtuntii.iu- .!' the level of tin- mercury in the cistern, are effected by means of the 

 tli 1 1 mil-screw at the top of the rod. 



A thermometer temporarily scoured to the tube, with its bulb immersed in the mercury of 

 the cistern, serves to indicate the temperature of the column. 



For transportation, the tube is furnished with a cylinder of wood that has an aperture lined 

 with velvet, and within which it fits very accurately. The packing plate and steel wedge remain 

 without the aperture. For additional security, the cylinder was placed on springs within an 

 oblong box which was suspended by a handle at one end the packed end of the tube upper- 

 most during the voyage (by steamer) from Washington to Baltimore, thence (by ship) to 

 Valparaiso, and finally (in a cart) to Santiago. In the last journey there were guide cords to 

 the bottom of the box to prevent it striking against the sides of the cart. On removing the 

 packing plate at Santiago, it could not be perceived that any mercury had been lost during the 

 several transportations; and the tube was not only full, but the column continued wholly free 

 from air. Nevertheless, to insure absolute repletion on reversal, a few drops of distilled mer- 

 cury were added, though only to be pressed out on replacing the packing plate and wedge. 



The bracket having been fastened to the wall in the office room at a suitable elevation for 

 reading of the barometer scale, and the shelf for the cistern having been placed at the proper 

 distance below it, the ferrule with the suspension loop was carefully put on whilst the tube was 

 still in an inverted position. The barometer was then reversed; its lower extremity was 

 immersed to the depth of an inch in mercury which had been poured into the earthenware 

 cistern ; and the tube and cup being lifted together to the places prepared for them, the former 

 was elevated by means of the screw until it swung freely above the bottom of the latter. The 

 packing wedge being next pushed out, the plate was removed, and the barometric column 

 descended to the height with which a column of tbe atmosphere of like diameter was in equi- 

 librium at the time. When the steel rod with its scale and ivory float were hung beside it, the 

 instrument was ready for use. 



At the termination of the observations in September, 1852, the barometer was taken down, 

 transported to a room of the INSTITUTO NACIONAL, and remounted there. To do this, the tube 

 was lowered to the bottom of the cistern, and being unhooked, it was slowly inclined until 

 entirely filled with mercury that flowed in from the latter. Whilst in this condition the 

 packing plate and wedge were inserted, and the upper ferrule being removed, the tube was 

 placed in its wooden cylinder, and conveyed by me to its new location. It was remounted with 

 all the care already detailed. 



As the mercury had been boiled in the tube, and the effect of capilarity for a diameter of 

 0.65 inch is less than 0.002 inch, no correction therefor has been applied in the reduction of the 

 observations. Nor is any correction required on account of the difference of the diameters of 

 the tube and cistern, because, by means of the ivory float and thumb-screw sustaining the steel 

 rod, the zero of the scale is adjusted to the surface of the mercury in the cistern, and the dis- 

 tance between it and the top of the column is directly measured. 



The correction for difference between the temperature at the time of observation and at 32 

 Fahrenheit consists of two parts, one of which depends on the change of specific gravity of the 

 mercury in the tube, and the other on alterations of the measuring scale. The dilatation of 

 mercury from the freezing to the boiling points of water being .018018, or .0001001 for each 

 degree of Fahrenheit, if we represent the height of the barometric column by fl, and its temper- 

 ature by t, the reduction to 32 will be 



.0001001 (t 32) A; 



