Colonel Sir H. James, R.E., in his _Instructions for taking Meteorological Observations_, says of the aneroid, "This is a most valuable instrument; it is extremely portable. I have had one in use for upwards of ten years, and find it to be the best form of barometer, as a "weather-glass," that has been made." One of the objects of Mr. Glaisher's experiments in balloons was "to compare the readings of an aneroid barometer with those of a mercurial barometer up to five miles." In the comparisons the readings of the mercurial barometer were corrected for index-error and temperature. The aneroid readings, says Mr. Glaisher, "prove all the observations made in the several ascents may be safely depended upon, and also that an aneroid barometer can be made to read correctly to pressures below twelve inches." As one of the general conclusions derived from his experiments he states, "that an aneroid barometer read correctly to the first place, and probably to the second place of decimals, to a pressure as low as seven inches." The two aneroids used by Mr. Glaisher were by Messrs. Negretti and Zambra. Aneroids are now manufactured almost perfectly compensated for temperature. Such an instrument therefore ought to show the same pressure in the external air at a temperature say of 40 deg., as it would in a room where the temperature at the same time may be 60 deg.; provided there is no difference of elevation. To test it thoroughly would require an examination and a comparison with barometer readings reduced to 32 deg. F., conducted through a long range of temperature and under artificially reduced pressure. A practical method appears to be to compare the aneroid daily, or more often, for a few weeks with the readings of a mercurial barometer reduced to 32 deg.; and if the error so found be constant, the object of the compensation may be assumed to be attained, particularly if the temperature during the period has varied greatly. _Directions for using the Aneroid._--Aneroids are generally suspended with the dial vertical; but if they be placed with the dial horizontal, the indications differ a few hundredths of an inch in the two positions. Hence, if their indications are registered, they should be kept in the same position.