O. N. Purvis 341 



just appearing, until a few days before the hay was cut, so that all 

 stages of growth were investigated. 



A large number of measurements of cell-walls and lumina were 

 taken at these various stages; but they do not show that potassium has 

 a well-marked eSect. 



In the presence of added potassium salts, especially in the early 

 stages of growth, the thickening of the sclerenchyma cell-walls was 

 less pronounced, and not more so, as would be expected if changed 

 wall thickness were the cause of the additional strength. As the summer 

 progressed, the difference became smaller, and in the case of plants 

 which had in addition received nitrogenous manures, the cell-walls were 

 as thick as those of plants receiving no potassic fertiliser. 



The addition of potash, however, caused considerable increase in 

 the rate of growth, and nearly doubled the crop. It is possible, therefore, 

 that in the early stages material was used for growth that might other- 

 wise have been used for wall thickening. The circumstances that the 

 walls ultimately had the same thickness as in plants which had not been 

 supplied with potash shows that the rate of thickening was greater with 

 added potassium salts, and may, therefore, indicate that these substances 

 do actually help in thickening the walls. 



Eepetitions of the observations in other seasons, comparison being 

 made with the rate of growth, would be necessary to decide this question. 

 This possibility receives some support from the fact that on May 31st 

 the plants which had been supplied with potash had the thicker walls 

 (see Figs. 6 and 18). 



In the xylem potassium salts had no effect on the wall thickness, 

 and on the diameter of the lumen the effect was irregular and incon- 

 clusive. 



Further evidence was obtained by studying the ratio of lumen to 

 wall in the same tissues. The ratio of the lumen to the wall in each 

 individual cell was taken as a numerical expression of its mechanical 

 strength. While the material of the wall remains the same, a cell where 

 the lumen is small in proportion to the thickness of the cell-wall would be 

 mechanically stronger than one where it is large. It is not safe to lay 

 too much stress on this, because it is possible that the cell-wall material 

 would differ in plants which have received dissimilar foods. If, however, 

 it were true that the mechanical effect of potassium salts were due to 

 their influence on the anatomical character of the plant, then this ratio 

 would be inversely proportional to the mechanical strength of the cells. 

 The results lead to the conclusion that the strengthening effect of potas- 



Journ. of Agric. Sei. ix 23 



