In 1893, Fugger writes: "The peculiar readings of temperature, which I made in August 1877, in the Kolowratshöhle, namely on the 13th at 12 M., 0.5°, on the 15th at 4 P. M., 0.35°, on the 23d at 10 A. M., 0.12°, on the 26th at 10 A. M., 0.17°, and on the 30th at 2.15 P. M., -0.10°, I think I can attribute to the workings of the winds. In the observations themselves there could scarce be an error. All five observations were made at the same place, with the same thermometer, after at least half an hour's exposure. In the time from the 13th to the 30th of August, the temperature minimum in the town of Salzburg, was 12°; before the 30th were several cloudless nights. During the whole of August scarcely any but southeast and northwest winds were blowing. The Kolowratshöhle opens in a rock wall to the east; the above named winds therefore affected during the entire month the entrance to the cave and may have produced a lively evaporation in the cave, through a sort of sucking up of the cave air, and thus have created the rather decided cooling off of 0.6° within seventeen days." This statement, coming from Professor Fugger, deserves particular attention, because it would go to show: first, that the air in the Kolowratshöhle, a _sackhöhle_ with only one entrance, is only apparently stagnant in summer and not really so; and second, that evaporation may act to a limited extent in a cavern where there is almost no running water. Captain Trouillet, in 1885, published a paper about Chaux-les-Passavant. He found that when it was colder inside than outside, the internal air was nearly cut off from the outside; when it was coldest outside there was a lively disturbance. He called these two classes _périodes fermées_ and _périodes ouvertes_. He says: "The duration of a _closed period_ is measured then on the curves [of a maximum and minimum thermometer] of the interior temperatures, between a minimum and the following maximum; that of an _open period_ is between a minimum and the preceding maximum. One can thus count from the 25th November to the 31st December 25 _open periods_ of a total duration of 200 hours or 8 times 24 hours: which gives for each a duration of 7½ hours. The shortest lasted 2 hours and the longest 16 hours. During the same interval, the _closed periods_ numbered 26, making a total duration of about 28 days; the longest, which lasted from the 3d to the 8th December, was 126 hours long." Trouillet also says: "From the 23d to the 30th December, the grotto was completely isolated from the external air, and yet during three consecutive nights, the interior had three marked chills. Such is the phenomenon whose cause can only lay, in our opinion, in the introduction of the dry air driven to the cave by the winds between north and east. This air on entering comes in contact with the ice and the humid roof of the cave; it saturates itself in producing a formation of vapors, and therefrom a consumption of heat which may be considerable."