of this period, the thermometer shows 212 degrees C. at the first aludels.  This lasts for 18 hours, and then the third or “cooling period” begins, which takes from 24 to 26 hours, and during the beginning of which the temperature in the furnaces still rises.  It is then opened and cooled down.  A very elaborate series of observations made on the temperatures of various parts of the condensing apparatus of the Almaden furnaces has shown that at the aludels nearest to them the heat increases steadily until it reaches 249 degrees C., 44 hours after the beginning of the operation; that in the middle of the line, at the depression, the maximum is 50 degrees 50 hours after starting the fires; and that at the end it does not surpass 39 degrees.  In the final condensing chamber, the temperature varied, running downward from 40 degrees during the heating period to 14 degrees, rising again to 29 degrees toward the close.

The loss of the quicksilver during the operation has been vary variously estimated, some stating that it is 50 per cent. and more, while others place it at 30 per cent.  Escosura, in his work, gives the details of an operation checked by a royal commission in 1872, according to which the loss in working ore running 9.55 per cent. was only 4.41 per cent.—­a loss which he considered inevitable.  In 1806, two Idria furnaces were put up at Almaden, but the engineers are not favorably impressed with them.  The first cost is stated to be more than ten times greater than that of an aludel furnace, while the capacity is only 50 per cent. greater.  One pair of Idria furnaces in five years produced 120,000 kilogrammes of quicksilver, against 843,000 kilogrammes made by eight sets of the Bustamente furnaces, the cost per kilogramme of quicksilver being respectively 0.121 and 0.056 peseta.

* * * * *

THE BALLOON IN AERONAUTICS.

While it is undoubtedly true that the discovery of the balloon has very greatly retarded the science of aerostation, yet, in my opinion, its field of usefulness as a vehicle for pleasure excursions, for explorations, and for scientific investigations, has not been fully developed for the want of certain improvements, the nature of which it is the object of this paper to point out.  The improvement of which I am about to speak relates to the regulation of the buoyancy of the balloon.  This is now done by throwing out ballast or by allowing some of the gas to escape—­a method which necessitates the carrying of an unwieldy amount of sand and the expenditure of an unnecessary amount of gas.