g4 NOTES ON THE 



and which extended over the entire length of the tube. In addition to the empty 

 tubes and the scales, I carried with me two small wooden vessels or cisterns each 

 of two inches diameter, and two inches deep, together with a quantity of mercury. 

 On arriving at the point to be measured, the two cisterns were filled to the depth 

 of an inch and a half, and the tubes entirely, with pure mercury; care being taken 

 to expel the small bubbles of air by slight sliaking. The aperture of each tube was 

 then closed with the finger, and the whole inverted with the open end below the 

 surface of the mercury in the cistern. After the mercury had settled in the tube, 

 the scale was read at the top, and also at the level of the quicksilver in the cistern. 

 The difference of these readings gave the length of the mercurial column. This 

 should be reduced for temperature, on account of the expansion of mercury and 

 the wood, but the correction, arising from the latter, was neglected, being very 

 small. At the trigonometrical station on the Ixtaccihuatl, at a temperature of 

 + 9°. 6 C, the longer of the tubes was found to agree perfectly with the syphon- 

 barometer. In Tlamacas the difference scarcely exceeded 0"'°.2, showing that the 

 result, obtained by this simple barometer, has claims to a considerable degree of 

 accuracy. 



The instrument used for measuring the vertical angles, and for the triangulation, 

 was made by Ertel, of Munich. It had a repeating circle, which could be mounted 

 vertically for measuring altitudes, or horizontally for azimuths; and by means of four 

 verniers it read to 10". When the circle was in a horizontal position it carried a 

 telescope with broken axis; for observations of altitudes an extra telescope had to 

 be attached as well as a level, and so many adjustments were required, that 

 much time was consumed in mounting it vertically. The instrument belonged to 

 the Mining School, in Mexico, and was kindly lent to me by that institution. It 

 had been treated rather roughly before, and, owing to this as well as to its compli- 

 cated construction, it required great care and much patience in using it. The 

 readings when the telescope had been moved from the left to the right, or from 

 above downwards, were always different from the readings when it had been moved 

 from the right to the left, or from below upwards. This difference, sometimes 

 amounting to more than two minutes, was the principal cause of the want of agree- 

 ment among the single measurements, and I have on this account given them, 

 in full detail. The observations were, however, so arranged, that all constant 

 errors, which miglit arise from torsion, or from tension, and collimation, were 

 eliminated. 



The standard used in measuring the base-line was a brass scale, the length of 

 which was determined in Mexico, and found to be 622 millimetres, at a tempe- 

 rature of +15° C. AYith the aid of this standard the length of a rod used in 

 measuring the base-line was accurately determined. The direction of the line 

 was carefully marked by a thin rope, stretched, and at several places supported, 

 so as to form very nearly a straight line. Besides the wooden rod, a steel-tape of 

 ten metres length was used, and the two measurements of the base, obtained with 

 these instruments, agreed in both cases within 0.06 of a metre. 



Corresponding barometrical observations were made in Mexico by Mr. Diaz- 

 Covarrubias, at a temporary observatory near the gate "San Lazaro," and by Mv. 



