All this brilliant work had been accomplished in the short space of the ten years 1880 to 1890. The inspiration had in each instance come from France, and is traceable direct to Pasteur, although coming long after his own work on bacteriology. It is impossible for us now to realise the thrill of wonder and astonishment with which students, teachers, and writers of those days learned that the nutrition of plants, and therefore the growth of crops and the feeding of themselves, was largely the result of the activity of bacteria in the dark recesses of the soil. It is not surprising that the ideas were pushed somewhat too far, that the soil population was regarded as solely bacterial, and that important chemical and physical changes were sometimes overlooked. Gradually there came the inevitable reaction and a somewhat changed outlook. Continued examination showed the presence in soil of almost every kind of bacteria for which search was made. Some of them were almost certainly in the resting condition as spores, and the new generation of workers had an uneasy feeling that the case for the overwhelming importance of bacteria in the economy of the soil was not too well founded. It was shown that the decomposition of nitrogen compounds to form ammonia would take place without micro-organisms if, as presumably would happen, the plant enzymes continued to act after they got into the soil. Even the oxidation of ammonia to nitrate--the great stronghold of the biological school--was accomplished by chemical agents. The fixation of nitrogen in soil conditions was beyond the power of chemists to achieve, and here it was universally agreed that bacteria were the active agents. And finally, chemists were themselves bringing into prominence a set of bodies--the colloids--whose remarkable properties seemed indefinitely expansible, and were in addition sufficiently incomprehensible to the ordinary student to attain much of the magnificence of the unknown. All the time, however, a faithful body of workers was busy exploring the ground already won, improving the technique, making counts of the numbers of bacteria in the soil, and trying to measure the amount of bacterial activity. Much of the value of this work was limited by the circumstance that the bacteria were regarded as more or less constant in numbers and activities, so that a single determination was supposed to characterise the position in a given soil.