The observations prove also that the inner temperatures vary less than the outer, that is that they range within narrower limits. They also show that the inner air is but slowly affected by the outer air when the latter is above freezing point, the inner temperature rising then only gradually. Per contra, when the outside temperature drops quickly much below freezing point, the inside temperature generally drops correspondingly at once, proving that the cold air has sunk by its weight into the cave. The observations also prove that the old idea that the temperature of caves is the same throughout, can no longer be considered correct. The observations also appear to show, that the temperature of a cave does not necessarily represent the mean annual temperature of a surrounding district. Observation is still entirely lacking on the mean annual temperature of glacières, so that one cannot speak definitely about the matter; but it seems likely that the mean annual temperature of a glacière cave is lower than the isotherm of its locality; and it seems more than probable that on the same isotherm different glacière caves may have different mean annual temperatures, varying with the elements of size, quantity of ice, position of body of cave and of entrance, water supply and other factors. _Ice near the Entrance of Caves and the Surface of the Soil._--An important proof that it is the cold air of winter which forms the ice is the fact that the latter is always found near the entrance of caves or near the surface of the soil. It never extends far within. To the best of my knowledge, ice has never been found two hundred meters from the entrance nor at any depth beyond one hundred and fifty meters. In all caves of great extent, the temperature far in is about the same as that of the surrounding rock, and in all deep borings the temperature increases with the depth and at great depths the temperature becomes high. This nearness of subterranean ice to the outside air is one of the best proofs, that, paradoxical as the whole phenomenon appears at first, yet in reality it is an extremely simple matter.