EXPERIMENT STATION BULLETINS. 565 



It will be seen from the above table that with few exceptions there is 

 quite a close relation between the degree of lowering of the freezing-point 

 and the weight of the solid matter in solution. On the other hand it will 

 also be observed that both the depression and the weight of the residue 

 var}"- considerably between the solutions of the various soils. Thus, for 

 instance, the depression varies from .002°C in some of the sands to .012°C 

 in some of the loams, while the weight of the residue ranges from .0043 

 grams in some of the sands to .0263 grams in some of the loams. 



At this point it should be mentioned that it is questionable whether 

 all the residue represents material which was in solution and affected 

 the depression of the freezing point. It appears very possible that some 

 of this residue consisted of solid particles of microscopic size. Although 

 the solutions were passed through the Chamberland-Pasteur filter and 

 appeared very clear, they still could contain solid particles of micro- 

 scopic size which had no influence upon the freezing-point depression. 

 In the light of our present knowledge of colloids the soil solutions may 

 be considered in part at least as colloidal solutions and may contain 

 solid particles of microscopic size which are able to go through the 

 various common filters. 



In spite of the error caused by the colloidal particles in the residue, 

 it is believed that a more true estimation of the concentration repre- 

 sented b}^ the freezing-point depression values can be obtained from the 

 soil water extracts than from pure chemical solutions. 



For performing the actual computation of the concentration of the 

 soil solution of the various soils as represented by the freezing-point 

 lowering, an average was taken of the depression and of the residue of 

 all the soils shown in Table 1 except the muck and peat. From these 

 averages the amount of depression that 100 p. p. m. produced was cal- 

 culated. The average depression amounts to .0041°C. For the calcula- 

 tion of the concentration of the solution of peat and muck, the depres- 

 sion and weight of residue obtained from the extracts of these soils, 

 were used. It would undoubtedly have been better to have used also the de- 

 pression and weight of residue of the extract of each mineral soil to 

 compute the concentration of its solution, rather than to employ the 

 above average. By using the average the concentration of the solution of 

 some soils is much higher than it should be, while that of others some- 

 what lower. 



The osmotic pressure was calculated from a table of osmotic pressures 



as worked out by Harris and Gortner (6) from a fonnula ^r = 12.06 a 



— 0.021 A 2^ where ^ is the osmotic pressure in atmospheres, and a 



the lowering of the freezing point in centigrade degrees. Fon conven- 



■ ience and immediate usefulness this table is reproduced herewith. 



(6) Am. Jour. Bot. 1, 75-78, 1914. 



