46 VEGETABLE PHYSIOLOGY 



very young. Treatment with the reagent mentioned will 

 disintegrate the tissue of even the growing points of stems 

 and roots, and will cause their cells to become isolated. 

 A thin layer of this nature must therefore exist even in 

 the primary cell-wall. It is added to materially, however, 

 during the growth in thickness of the walls, and in many 

 cases it can be seen easily under a comparatively low 

 magnification. 



This middle lamella is composed of a material which 

 is very unlike that of the rest of the cell-wall. It dissolves 

 readily under the action of pofcassic chlorate and nitric acid 

 which do not affect the inner layers of the membrane. It 

 resists completely the action of sulphuric and other mineral 

 acids which cause the inner layers to swell and ultimately 

 to dissolve. 



Eecent investigations have led to the view that it is 

 composed of a calcium salt of pectic acid. 



Whether the primitive cell-wall is homogeneous or not 

 is uncertain. If it is, it must be regarded as being formed 

 of an intimate mixture or perhaps of a compound of cellu- 

 lose and pectose. At a very early period in its development 

 the middle lamella becomes differentiated, owing possibly to 

 the conversion of the pectose into pectic acid and the inter- 

 action of the latter with some salt of calcium derived from 

 the cell-sap which infiltrates the wall. The calcium pectate 

 becomes deposited in this way halfway between the con- 

 tiguous cells which are separated by the particular mem- 

 brane in which the change is taking place. 



If the cell-wall is not at first homogeneous, we must 

 suppose that the original thin membrane is composed of 

 three layers, a central one of calcium pectate, on each face 

 of which is a layer composed of the mixture or compound 

 of cellulose and pectose. We never find, even at the 

 moment of cell division, that the membrane is formed of 

 calcium pectate only. 



It is possible to explain the growth in thickness of the 

 middle lamella on either hypothesis. It is clear that the 



