"It is in the spring, at the time of the first melting of the snows, that the ice must form. Then water at 0° pours over the surface, and penetrates by the fissures of the rock and by the large openings into the chilled cavern, which is also receiving the freezing air of the nights. The grotto then makes its annual provision of ice, which after this could only diminish little by little during the whole duration of the warm season." Professor Thury writes about the Grand Cave de Montarquis: "Here it must be when water and cold meet, that is autumn and especially spring, the time of the first melting of the snows." "During the winter * * * the colder, heavier air comes to freeze the water of the grotto, and chill the ice and the wall of rock." "During the summer, the radiation of the vaults and the proper heat of the ground only melt a small quantity of ice because this absorbs much heat to pass into a liquid state." "The heat of the air is entirely used to melt the ice; it does not therefore manifest itself as sensible heat." "The contact of the ice ready to melt, plays in a certain way, towards the air a little warmer than itself, the rôle of an extremely absorbing body, or one which has an excessive caloric conductibility." "Here the formation of the ice could not possibly be attributed to the cold caused by evaporation. The psychrometer indicated ninety-two per cent, of relative humidity: the atmosphere of the grotto was therefore almost saturated with evaporation of water, and the maximum of cold caused by evaporation was not over half a degree centigrade." About prismatic ice and a hollow pyramid, he says: "The prismatic (_aréolaire_) structure is produced later on in the ice, by a new and particular arrangement of the molecules of the already solidified water. Therefore the recent stalactites are never crystallized." "In the beginning of the hot season, the atmospheric temperature of the grotto rises slowly. Inferior to zero by some tenths of a degree, it produces first on the surface, in the stalactites, the prismatic structure. The temperature continues to rise, the central portions of the stalactites, still composed of ordinary ice, liquefy, and if the melting water finds some issue, either by accidental openings left between some prisms, or by the extremity of the stalactite or by some point of its surface which had escaped the action of the regular crystallization; by this opening the water escapes, and the tubular stalactite has been formed."