ZOOLOGY AND BOTANY, MICROSCOPY, ETC. 217 



epidermal cell, if not tliin and easily permeable, as is the case where 

 the guard-cells are weak in mechanical structure, has invariably a thin 

 spot which is readily permeable. Where the wall of the guard-cell is 

 otherwise entirely cuticularized, this spot consists of ordinary cellu- 

 lose. The opposite ventral wall of the guard-cell, bounding the 

 fissure itself, also has always a thin spot. When the cuticle covers 

 the ventral wall of the guard-cells up to the fissure, it is not inter- 

 rupted at the thin spot. These thin strips constitute a hinge. 



Elsewhere the walls of the guard-cells are thickened in a great 

 variety of ways ; the outer and inner parts of the ventral surface 

 usually prismatically. By these prismatic thickenings, when the 

 guard-cells become more turgid, a strong elongation of the thin-walled 

 dorsal side is brought about, by which the guard-cells are made to 

 curve, because the thickenings of the ventral wall give them increased 

 power to withstand traction. No curvature of the guard-cells can 

 take place here in consequence of any greater increase in length of the 

 dorsal wall. The motility of such guard-cells appears therefore to 

 be less when mature than when young, when the form of the cells 

 more nearly resembles those with prismatic thickenings. In many 

 plants it is only the thickening ridges which face the interior of the 

 leaf that are capable of curving. The thickenings gradually dis- 

 appear at their ends, and do not usually coalesce. The guard-cells 

 are often considerably higher at the two ends than at the middle. In 

 many thick phyllodes and evergreen leaves the very narrow cell- walls 

 are bounded above and below by strong thickening stripes, often fur- 

 nished with prominent cuticular ridges. 



Those stomata of which the guard-cells have no cell-cavities are 

 often comparatively immotile, in extreme cases, perhaps, absolutely so. 



The mode of motion may be made out by comparing the open and 

 closed conditions of the stomata. For this purpose both vertical and 

 transverse sections should be made ; and care must be taken to allow 

 for the increase of turgidity caused by glycerin. The general 

 results of a careful series of measurements is that the size of the 

 guard-cells is greater when the fissure is open than when it is closed. 



The movements are caused by increase and decrease of the hydro- 

 static pressure in the guard-cells. When the turgidity is increasing, 

 the increase of the thin dorsal wall of the guard-cell amounts to about 

 9 per cent., and the increase in volume of the entire guard-cell to 

 about 17 per cent. The hydrostatic pressure necessary to produce 

 this effect on a cell-wall 1 or 2 /a thick, 's respectively that of 

 5 or 10 atmospheres. It is only when the pressure in tlae guard- 

 cells exceeds that of the adjoining epidermal cells that the stoma can 

 open. This is effected by a curvature of the guard-cells, caused by 

 the difference in structure of the dorsal and ventral walls, as can be 

 show^n experimentally by a caoutchouc-tube. 



When there is no tension the stomata are open only in some water 

 plants. In some Monocotyledons (as Tradescantia discolor) there is 

 a difference from the normal structure as regards the changes in form, 

 the peculiar structure causing an expansion of the guard-cells in a 

 direction vertical to the surface of the leaf, which increases and 



