limit 01 plant food in reserve; no direct experiments to test the validity of 

 this theory have been made, but the substitution of the ordinary tempera- 

 ture for that of the steam bath in the hydrochloric acid extraction method 

 seems to fit in exactly with the conditions of this Colony, and the result* 

 thereby obtained tally closely with the practical experiences of farmers. 



It must again be observed that although every soil which may yield 

 large quantities of soluble plant food to this method of extraction is not 

 thereby proved to be fertile, yet, as the method represents a maximum, 

 we may definitely consider all soils that show small proportions of soluble 

 constituents under its treatment, to be unmi&takeably poor, and in need of 

 replenishment. To a certain extent, also, we may assume the fertility of 

 a soil which yields good results not only by that method, but also when 

 treated according to the citric acid method of extraction outlined below. 



Two other extraction methods by Hydrochloric acid have been used in. 

 several of the analyses; they are the second and third of the methods de- 

 scribed below, the former being that of the German Experiment Stations, 

 and the latter one stated to have been devised by Professor Maercker, of the 

 Halle Experiment Station. 



METHOD I. STANDARD METHOD OF SOIL EXTRACTION WITH HYDROCHLORIC 



ACID. 



Two hundred grammes of the air-dried " fine earth "* are placed in 

 a large rubber-stoppered flask, and treated with 400 c.c. of hydrochloric 

 acid of specific gravity I 1 115 (plus any needed for neutralising carbonates 

 in the soil) ; allowed to remain for five days at the ordinary temperature, 

 shaking thoroughly from time to time. After the prescribed period of 

 digestion has expired, the extract is filtered, through a. dry pleated filter, 

 into a, dry flask. 250 c.c. of the clear filtrate, representing 125 grammes 

 of soil, are evaporated to dryness in a shallow porcelain dish, at first over 

 a small open flame, then on the water bath, and finally on a sand bath, 

 or in an air oven, at 120C, until perfectly dry. During the evaporation 

 a few cubic centimetres of strong nitric acid are added to the extract. 

 The dry residue is moistened with strong nitric acid, and again evaporated 

 to dryness; to expel the nitric acid, the residue is moistened with hydro- 

 chloric acid, and evaporated on the water bath to as near dryness as pos- 

 sible, taking care to stir towards the end of the evaporation, so as to 

 prevent the formation of crusts. This final residue, after warming in 

 the air bath for an hour at 105C to 110C, is treated with warm water 

 and a 20 per cent, solution of hydrochloric acid, and is then washed over 

 into a 250 c.c. flask, boiled for 15 minutes, allowed to cool, filled up to 

 the mark with distilled water, and filtered into a suitable bottle. This 

 filtered soil extract is then employed for the actual estimations of lime 

 and potash. 



Determination of Lime. 50 c.c. of the extract (equivalent to 25 

 grammes of " fine earth "), obtained as described in the preceding para- 

 graph, are removed, by means of a pipette, into a 250 c.c. boiling flask: 1 



* As more than once indicated, only soil grains below a certain size are included in the 

 material submitted to extraction. Loughridge (Proceedings of the American Association 

 for the Advancement of Science, Vol. 22, p. 81) has found that, of all the grades into which 

 soils are usually separated by mechanical analysis, the " clay," i.e. the finest grade, is by 

 far the richest in mineral constituents, and that the quantity of acid-extractible matter 

 in the grades of soil particles of over '04 mm. diameter was practically negligible. Hence 

 it follows that soil sifted through a \ mm. sieve will contain all the mineral constituents 

 available for plants. The first requisite in selecting a standard method is to utilise for 

 extraction only the soil passing the mm. sieve. Hence, too, the method t.g. of sifting 

 the soil through a 3 mm. sieve and pounding the sifted portion for extraction must be 

 rejected as furnishing misleading results. 



