some extent overcome by the osmotic pressure of substances in the soil solution, and a plant may be compelled to take up a larger amount of substances like soda or magnesia than it requires, and which may be decidedly harmful. Such a case is furnished in my own experience by the growth of sugar-cane on soil infiltrated with brackish water, on certain parts of the Clarence River, New South Wales, and on what is known as Tiri land in Fiji—low-lying land subject to inundation with sea-water. In such cases, quite excessive amounts of chlorides are found in the juice when the cane is crushed.
Another interesting point which requires bearing in mind in case of further investigation of the water from plants, is that brought out by Professor H. H. Dixon[1], as to the variation in concentration in different parts of the plant. Dixon finds that the concentration of solids is greatest in the sap from the roots, and becomes less in the upper parts of the plant. This was determined by the electrical conductivity and freezing-point methods in samples of sap obtained by centrifuging pieces of the roots and stems. The figures obtained by both methods are proportional to the amount of dissolved substance present, and are in close agreement. It was found in the case, for instance, of Acer pseudoplatamus, that the sap in the branches at a height of 30 feet had only two-thirds the amount of dissolved substances as that in the roots. In another plant, Colocasia antiquoriim, a similar distribution was observed, and it is interesting to note that the droplets of liquid transpired from the tips of the leaves consisted not of sap, but of practically pure water.
These results, as regards the greater concentration of the sap in the roots, are certainly contrary to what I would have expected, but Professor Dixon's experiments are most lucidly described, bear evidence of careful execution, and seem to me quite convincing. The Rev. W. B. Clarke,[2] without giving his authority, states his belief that it had been ascertained that the specific gravity of
