THE MATERIAL OUTGO OF PLANTS 327 
tion and entrance of the gases absolutely indispensable to life. To be 
sure, the outer walls of the surface cells are relatively dry, especially in 
plants of dry regions, where water loss is to be reduced to a minimum. 
Of the total water lost scarcely more than 20 per cent, and as little as 3 
per cent, escapes through the epidermis. This evaporation is sometimes 
distinguished as cuticular transpiration. The remaining 80-97 per cent 
diffuses through the stomata and constitutes stomatal transpiration. The 
efficiency of this arrangement in reducing transpiration and yet admit- 
ting gases freely is more obvious when one observes that the actual 
evaporation surface i.e. of the cells bounding the intercellular spaces 
is several times that of the leaf itself. 
The place of maximum cuticular evaporation has been shown in some leaves to 
be that part of the outer wall of the epidermis where the side walls abut. In these 
cases water of imbibition is more abundant there than elsewhere. 
It is impossible to determine the actual surface exposed in the very irregular air 
passages. If a leaf i mm. thick had an epidermis o.i mm. thick of coherent cubical 
cells on each face, and if the remaining cells were spheres each o.i mm. in diameter, 
tangent to each other, the internal surface would be about fifteen times the area of 
the corresponding outer faces of the leaf. This, of course, does not pretend to pic- 
ture the actual state of affairs; but it will give an idea of the relative magnitudes 
involved. 
Stomata. The guard cells of the stomata are different from the rest 
of the epidermal cells in form, in the peculiar unequal thickening of 
their walls, and generally in the possession of chloroplasts. These 
characters and the position of the guard cells with reference to the ad- 
jacent subsidiary cells determine simultaneous differences in turgor 
and make them behave differently from the others. In general when 
turgid, they become arcuate, and when flaccid, they straighten. The 
mechanics of these movements differs considerably with differences of 
form, structure, and position, and in none of the several types that have 
been distinguished is it fully understood. The chloroplasts are supposed, 
but on no very good grounds, to impart power to make osmotically 
active substances that do not exist in adjacent cells (or are present in 
smaller amount), so that these cells may be more turgid than the others 
with the same water supply. The longitudinal thickenings are elastic 
and are supposed to straighten the cells when they become flaccid. 
The auxiliary cells are supposed to offer proper bracing for the guard 
cells so that turgor will arch them. 
Regulation by stomata. Naturally the guard cells are most likely 
to be turgid when the water supply is good; then the opening of the slit 
