Some interesting observations were made by Guthrie and Cohen (1910) on certain Australian soils. A bowling green that was initially covered with a healthy mat of couch grass developed a number of small patches after about three years growth, on which the grass died off. No reason was apparent for this phenomenon, as the cultural conditions were uniform and to all appearances the soil over the whole area was similar in character. Analyses of soil samples from the dead patches and from the neighbouring healthy parts of the green showed that the chemical composition in both cases was practically the same, except that while no manganese occurred in the soil from the unharmed part, as much as ·254% Mn₂O₃ was found in that from the dead patches. As no other differences were found it was argued that the manganese, present in such large quantities, acted as a toxic agent and killed off the grass. Other instances of manganese poisoning in which wheat and barley were affected are quoted by these authors, the analytical results indicating that possibly barley is able to withstand without injury a greater quantity of manganese compounds in the soil than is wheat. 2. _Effect of manganese compounds on germination._ Nazari (1910) rolled wheat grains in a paste of manganese dioxide, iron sesquioxide (both with and without organic matter), and in what he terms “artificial oxydases.” The seeds rolled in the last-named showed the greatest energy in germination, while those with manganese gave an appreciable acceleration. The presence of organic matter decreased the action of manganese. The plants from the manganese seedlings gave an increased yield in both straw and grain, while those treated with sesquioxide of iron showed no gain over the check plants. The Hills Experiments yielded some information as to the differing effects of various compounds of manganese on germination. With wheat plants in pot experiments manganese oxide (MnO₂) distinctly retarded germination when applied at the rate of 2 cwt. per acre. With barley MnO₂, manganese carbonate and sulphate all retarded germination, while with the iodide 50% of the seeds were entirely prevented from germinating. 3. _Does manganese stimulate higher plants?_ With manganese the evidence in favour of stimulation is more weighty than with such poisons as copper, zinc and arsenic, and the literature on the subject is correspondingly plentiful. (_a_) _Stimulation in water cultures._