O. N. Purvis 347 



in the diameter of the lumen (Fig. 10), a steady increase occurred 

 from May 16th until June 15th. After this it decreased slightly until 

 the end of the investigation. The change during the first week in June 

 was again similar to that seen in the sclerenchyma, but cannot here be 

 attributed to increased turgor. 



The ratio, like the mean lumen diameter, increased until June 8th, 

 showing in the early stages a gradual loss of mechanical strength in the 

 individual cells. From that date onwards, however, it diminished, and 

 finally on June 26th reached its original value of 13-4 /la (Fig. 11). 



These changes in the dimensions of the xylem vessels cannot be con- 

 nected with wall deposition or turgor changes, such as occur in the living 

 cells of the sclerenchyma. 



Comparison of Development in Presence and Absence of Potas- 

 sium Salts where no Nitrogen Fertiliser has been applied. 



Plot 8 is supplied with a yearly dressing similar to that given to 

 Plot 7 except that potassium salts are omitted; differences between 

 plants from the two plots can therefore be attributed to the influence 

 of potassium salts. 



A. Sclerenchyma. 



Plot 8. Minerals without potassium salts. 



Date May 16th May 31st June 8th June 15th 



Walls 3-3 ±09^ 207±02m 304±07/ii 5-3 ±12/^ 



Lumina 15-1 ±-29^ 9-4 ±-20m 9-3 ±-22m 5-6 ±12^ 



Ratio 3-51 4-60 3-21 112 



Plot 7. Complete minerals. 



WaUs 1-88±-09ai 2-81 ±-11/* 2-39+06^ 506 ±- 10m 



Lumina 15-6 ±-27m 8-1 ±-23m 101 ±-32/i 75 ±-21^ 



Ratio 9-37 3-22 4-3 1-57 



In the case of the earliest material collected, the stem sections of 

 plants from Plots 7 and 8 showed a striking difference in the structure 

 of their mechanical tissue. Those from Plot 8, where no potash manure 

 was added, possessed a well-marked sclerenchymatous zone, whose cell- 

 walls readily took a basic ptain, and which were twice and sometimes 

 three times as thick as those in Plot 7 stems, which had received 

 potassium salts. In the latter the sclerenchyma was hardly differentiated ; 

 it could only be distinguished from ground tissue by the characteristic 

 refractive properties of its walls and would not take sucli stains as 

 aniline safranin (Figs. 4 and 5). 



