Recent inquiries have introduced a new element into the problem of the influence of the forest on temperature, or rather into the question of the thermometrical effects of its destruction. I refer to the composition of the soil in respect to its hygroscopicity or aptitude to absorb humidity, whether in a liquid or a gaseous form, and to the conducting power of the particles of which it is composed. [Footnote: Composition, texture, and color of soil are important elements to be considered in estimating the effects of the removal of the forest upon its thermoscopic action. "Experience has proved," says Becquerel, "that when the soil is bared, it becomes more or less heated [by the rays of the sun] according to the nature and the color of the particles which compose it, and according to its humidity, and that, in the refrigeration resulting from radiation, we must take into the account the conducting power of those particles also. Other things being equal, siliceous and calcareous sands, compared in equal volumes with different argillaceous earths, with calcareous powder or dust, with humus, with arable and with garden earth, are the soils which least conduct heat. It is for this reason that sandy ground, in summer, maintains a high temperature even during the night. We may hence conclude that when a sandy soil is stripped of wood, the local temperature will be raised. After the sands follow successively argillaceous, arable, and garden ground, then humus, which occupies the lowest rank. "The retentive power of humus is but half as great as that of calcareous sand. We will add that the power or retaining heat is proportional to the density. It has also a relation to the magnitude of the particles. It is for this reason that ground covered with siliceous pebbles cools more slowly than siliceous sand, and that pebbly soils are best suited to the cultivation of the vine, because they advance the ripening of the grape more rapidly than chalky and clayey earths, which cool quickly. Hence we see that in examining the calorific effects of clearing forests, it is important to take into account the properties of the soil laid bare."--Becquerel, Des Climats et des Sols boises, p. 137.]