_Stichococcus bacillaris_ and _Scenedesmus spp._, also occurring in soils, have been shown to be almost equally adaptable, though in these cases the organisms grow more slowly in the dark than on the corresponding medium in the light. Liquefaction of gelatine by the secretion of proteolytic enzymes has been shown to be a further property of certain species, resulting in the formation of amino acids such as glycocoll, phenylalanine, dipeptides, etc. This property is, however, possessed by only a limited number of species and in varying degree. Up to the present very little work of this kind has been done upon algæ actually taken from the soil, and our knowledge is therefore very scanty. Of the species so far examined all show considerable increase in growth on the addition to the medium of glucose and other sugars, and tend to be partially saprophytic; a few have been shown to liquefy gelatine to some extent. Servettaz, Von Ubisch, and Robbins have also demonstrated that the protonema of some mosses can make use of certain organic substances, especially the sugars, and grow vigorously in the dark. It has been shown, however, that light is essential for the development of the moss plant. It was thought at Rothamsted that some light might be thrown upon the activities of the soil-algæ by making counts of the numbers present in samples of soil taken periodically within a circumscribed area. A dilution method similar to that in use in the protozoological laboratory was adopted and applied to samples of arable soil taken from the surface, and at depths of 2, 4, 6 and 12 inches vertically beneath. A considerable number of samples were examined in this way from two plots on Broadbalk wheat-field, viz.: the unmanured plot and that receiving a heavy annual dressing of farmyard manure. The numbers in the unmanured soil were observed to fall far short of those in that containing a large amount of organic matter, while in both plots the numbers varied considerably at different times of the year. The chief species in both plots were identical, and their vertical distribution was fairly uniform, but it was observed that the numbers of individuals varied according to the depth of the sample. The 6th and 12th inch samples contained very few individuals of comparatively few species, but the 4th inch samples yielded numbers that were not significantly less than those in the top inch. The 2nd inch sample was usually much poorer in individuals than either the top or the 4th inch.