But not merely did Davy summarise and systematise the various results arrived at by others, he also made many valuable contributions to the science himself. The conclusions he drew from the results he obtained were, no doubt, in many cases false, and in other cases exaggerated; still the results possess a permanent interest. He may be said to have worked out many of the most important _physical_ or _mechanical_ properties of a soil, although exaggerating the importance of the influence of these properties on the question of fertility.[10] These experiments had to do with the heat- and water-absorbing powers of a soil. He experimented on a brown fertile soil, and a cold barren clay, and found at what rate they lost heat. "Nothing," he says, "can be more evident than that the genial heat of the soil, particularly in spring, must be of the highest importance to the rising plant; ... so that the temperature of the surface, when bare and exposed to the rays of the sun, affords at least one indication of the degree of the fertility." Again he says: "The power of soils to absorb water from air is much connected with fertility.... I have compared the absorbent powers of many soils, with respect to atmospheric moisture, and I have always found it greatest in the most fertile soils; so that it affords one method of judging of the productiveness of land." Where he erred was in overestimating the functions of the mechanical properties of a soil, and in considering fertility to be due to them alone. During the next thirty years or so, little progress seems to have been made in the way of fresh experimentation. _Boussingault._ In 1834, Boussingault,[11] the most distinguished French agricultural chemist of the century, began that series of brilliant chemico-agricultural experiments on his estate at Bechelbronn, in Alsace, the results of which have added so much to agricultural science. It was the first instance of the combination of "science with practice," of the institution of a laboratory on a farm; a combination peculiarly fitted to promote the interests of agricultural science, and an example which has been since followed with such magnificent results in the case of Sir John Lawes's famous Rothamsted Experiment Station, and other less known research stations. Boussingault's first paper appeared in 1836, and was entitled, "The amount of nitrogen in different kinds of foods, and on the equal value of foods founded on these data." In the year following other papers were published on such subjects as the amount of gluten in different kinds of wheat; on the meteorological considerations of how far various agricultural operations--such as extensive clearings of wood, the draining of large swamps, &c.--influence of climate on a country; and on experiments on the culture of the vine.