308 EXPEEIMENT STATION RECORD. 



While the results obtained by both methods are satisfactory the perchloric acid 

 method is deemed simpler and quicker. 



The determ.ination of potash as perchlorate in potash fertilizers, A. Strigel 

 and J. DoDT (Landw. Vers. Stat., 78 {1912), No. 3-4, pp. 179-188).— The authors 

 find that the best results are obtained with this method if the potassium 

 perchlorate solution is evaporated to sirupy consistency. While the figures 

 obtained agree well with those given when evaporated to complete dryness the 

 latter show a tendency to be a little higher, and still higher if the drying is 

 continued on a boiling water bath. Heating over 100° yields erroneous figures. 

 The presence of free hydrochloric acid, even in small amounts, was found to 

 influence the results markedly. In a series of tests it was soon noted that 

 when this acid was present in the perchlorate mixture, alcohol when added acts 

 upon some of the perchlorates (Ba(C104)2+2HCl=BaCl2-f2HC104), producmg 

 chlorid which is weighed and consequently produces erroneous results. 



The following procedure is recommended: Place 10 gm. of the potash fer- 

 tilizer in a * liter bottle with about 100 cc. of water and 2 cc. of hydrochloric 

 acid (specific gravity 1.125) ; bring to the boiling point, heat for 5 minutes, and 

 precipitate the sulphuric acid with barium chlorid, avoiding an excess of the 

 latter. After cooling fill to the 500 cc. mark and filter. Take 25 cc. of the 

 filtrate, which corresponds to 0.5 gm. of the original fertilizer, in a glass dish 

 with 10 cc. of perchloric acid (22 per cent), and evaporate to sirupy consistency. 

 Treat the residue in the usual manner, collect the potassium perchlorate in a 

 Gooch crucible, and dry at 130° C. 



Examination and judgement of soils, Hotteb et al. (Ztschr. Landw. Ver- 

 suchsw. Ostcrr., 15 (1912), No. 8, pp. 1002-1019).— After briefly reviewing the 

 factors involved in soil formation and soil fertility a description of the 

 mechanical and chemical methods for examining soils adopted by the Associa- 

 tion of Austrian Experiment Stations is given. The section on mechanical 

 analysis describes the elutriation and sifting methods. The chemical methods 

 deal with those for determining the plant nutrients, such as nitrogen, phos- 

 phoric acid, potash (by the perchloric acid method), and calcium. Special 

 methods are given for water, carbonates, sand, clay, humus, water capacity 

 (E. S. R., 26, p. 218), hygroscopicity, reaction, and the determination of sub- 

 stances injurious to plants. The latter methods include some for free acid, 

 sodium chlorid, sulphates, and iron in ferrous combination. Methods of sam- 

 pling are also presented. The remainder of the article deals with the in- 

 terpretation of the data obtained by the above mentioned methods, the re- 

 agents and apparatus necessary for examining soils, and forms for reporting 

 results. 



On the carbohydrate group in yam mucin, K. Oshima and T. Tadokoro 

 (Jonr. Col. Agr. TohoJcu Imp. Univ., If {1911), No. 6, pp. 2Ji3-249) .—As the 

 occurrence of mucin in the vegetable kingdom has not been fully determined, 

 an analysis was made of what is considered yam mucin. Cinehonin norisosac- 

 eharate was obtained, demonstrating positively the presence of the glucosamin 

 group. 



The occurrence of formaldehyde in plants, T. Curtius and H. Franzeis^ 

 (Ber. Deut. Chem. GeselL, 45 {1912), No. 9, pp. 1715-1718; abs. in ZontU. 

 Biochem. u. Biophys., 13 {1912), No. 16-17, p. 6.'/9).— Finding that the usual 

 method proposed for detecting formakiehyde in plant leaves reacttxl with alde- 

 hydes other than formaldehyde such as a, /3-hexylenaldehyde, the authors made 

 some experiments with dog-rose leaves for the purpose of finding an appropri- 

 ate method. It was discovered that if the volatile acids were first removed 

 from the dog-rose leaves the aldehydes could be converted into their respective 

 SLCids by shaking with sUver oxid. The acids so obtained were then estimated^ 



