TRANSACTIONS OK SECTION B. 841 



of green, (6) the non-appearance of a precipitate, (c) tlie estimation of iodine 

 liberated. 



A potassium salt, formed by the action of hydrogen peroxide on cobaltous 

 oxalate dissolved in potassium oxalate, has been isolated and analj'sed ; the formula 

 appears to be K^Co.,(C._,0^).^4H^O. 



This body, unlike Kehrmann's K^Co.,(C204),6H30, exhibits no dicbroism. It 

 occurs in green, transparent, microscopic crystals, belonging in all probability to 

 the rhombohedral system. 



The salt is very stable ; its solution in water gives do precipitate with dilute 

 calcium chloride, while concentrated calcium chloride yields minute green, needle- 

 shaped crystals of the corresponding calcium salt. 



The properties of the potassium salt in regard to its behaviour when heated, 

 solubility, lowering of freezing-point, action of acids, alkalis, reducing and oxidising 

 agents, have been studied, from which it would appear that a structural formula 



KO.,C-C(OH), C(0H)2-C0..K 



o o 



\co = Co/ 

 O O 



/ \ 



KO,C-C(OH)„ C(OH)j-COjK 



is probable, the body having lost a previously acquired atom of oxygen between the 

 cobalt atoms. 



The absorption spectrum of the aqueous solution exhibits two bright bands — • 

 one in the red with centre near the C line ; the other iu the green with centre 

 approximately' X = fil50. The absorption spectra of other cobaltous and cobaltic 

 solutions have been studied ; that of the bicarbonate green exhibits no red band, 

 but gives a single bright band in the green with centre approximately X = 5365. 

 There appears to be a reversal of the band in the green when cobaltous solutions 

 are oxidised to form these green compounds. 



4. Analysis of Dorsetshire Soils. By C. M. Luxmoore, D.Sc, F.I.C. 



The investigation of the Dorsetshire soils, which has recently been commenced 

 at Reading College, is not forward enough to enable us to publish any results at 

 present, but an account of the methods that are adopted will perhaps be of interest. 



As the object of the work is to obtain a general knowledge cf the soils of the 

 county, the samples are taken from land lying on the various geological formations 

 that occur, and so selected as to be, in the opinion of local agriculturists, typical 

 of a more or less considerable area of land iu the neighbourhood. Three samples 

 of the soil of each selected field, usually along a diagonal line, are taken much in 

 the manner directed by the Royal Agricultural Society, using, however, boxes 

 G inches by 6 inches ia area and 1 8 inches deep, and made so that one side can be 

 readily removed, showing the soil in section. When the boxes are opened at the 

 laboratory the soil is separated from the subsoil, a division being made at the 

 arbitrary depth of 9 inches, and also at any other line where a definite change in 

 the character of the soil is manifest. The larger stones, if any, having been 

 removed, the sample is carefully divided so as to reduce it to about one kilo- 

 gramme ; the smaller stones and gravel are then separated by sifting under water, 

 and the fine earth passing a sieve ^-inch mesh is dried at a moderate temperature 

 in a copper oven. If the three samples taken from a field appear fairly similar, a 

 mixture of them is made for analysis, but the original samples are preserved for 

 reference. The chemical analysis of these mixed samples offers at present nothing 

 worthy of special notice ; it includes, of course, in addition to the usual complete 

 analysis of the portion soluble in hydrochloric acid, a determination of the available 

 potash and phosphoric acid, according to Dyer's method. A series of mechanical 



