Atkins — Factors affecting Hydrogen Ton Concentration of Soil. 411 



t,o the glass probably brought traces of alkali into solution. A duplicate 

 with half Plymouth tap-water and half Blachford water also reached pH 8-6. 

 A third vessel in good north light, and plentifully supplied with algae in 

 Blachford water, was at pH 8"5 ; and a fourth which had stood for over a 

 year, with added chalk, was at pH 8-95 when similarly illuminated. Since 

 pure calcite, when boiled with water, only gives a reaction slightly over 

 pH 9 0, this is near the hmit. The value pH 9-7 given by the Staddon 

 reservoir water, with its natural algal flora, is probably explicable on the 

 assumption that small amounts of magnesium were also present, as is highly 

 probable in a situation a quarter of a mile from the sea. Similar high values 

 are found in sea-water during insolation with algae. Such photosynthetie 

 changes have been studied in relation to their dynamics by Osterhout and 

 Haas (1918). 



Thus, both under its natural conditions and during intense insolation in 

 jars, soft upland water, such as the Blachford and Plymouth water, is less 

 alkaline than a water richer in the carbonates of the alkaline earths. The 

 difference in the algal and general flora is probably correspondingly marked, 

 though as yet it has not been examined in any detail. It is well known, however, 

 that certain species are found mainly or exclusively in these soft waters, and 

 the correlation of their occurrence and the pH value of the water will be 

 of interest. Skene (1915) showed a striking relationship between the 

 occurrence of various species of Sphagnum and their powers of growth in 

 solutions of different normality, from N/250 to N/5000 acid or alkali. A 

 study of the water in which they occur would probably demonstrate that 

 the hydrogen ion concentration is of importance here also. 



With natural waters of low bicarbonate content the buffer action is 

 extremely slight, so great care must be taken in the preparation of the 

 indicators, of which small quantities only should be used, so as to minimize 

 error from the reaction of the latter. With solutions which are boiled the 

 indicator should be added only after cooling again. Whenever possible, the 

 use of two indicators is recommended, one with the middle region of its 

 range above and one below the reaction of the liquid being tested. Eubber 

 stoppers should also be licpt away from water samples, since in contact with 

 distilled water a reaction of pH 5-4 may be developed. Prolonged boiling 

 of the rubber with changes of water minimizes this error. Special care 

 should be taken to avoid contact with the ground-glass portion of stoppered 

 bottles, especially those of white glass. Distilled water may become sensibly 

 alkaline in a bottle ; accordingly every bottle used should be tested. 



Pure toluene has been found serviceable in preventing the growth of 

 moulds in the Clark and Lubs standard buffer mixtures. 



