THE ASSIMILATION OF CARBON BY AUTOTROPHIC PLANTS. II 121 



If we create a number of artificial stomata on the upper side of a leaf whose 

 underside has been vaselined, by pricking it with a needle, or cutting it with 

 a knife, or by removing altogether a strip of cuticle or the whole epidermis, the 

 formation of starch takes place in the neighbourhood of such artificial aper- 

 tures. It is easily understood why such starch manufacture is entirely local, 

 seeing that the cells immediately underlying the gaps at once use up the small 

 supply of carbon-dioxide in the air. MOLL'S (1877) experiments, which have 

 been briefly referred to above, also show that starch formation is always limited 

 to those regions in the leaf which have direct access to the carbon-dioxide, and 

 it is on that account that the carbon-dioxide absorbed along with water from 

 the soil is of no importance so far as the leaves are concerned. 



The intercellular spaces, whose exits the stomata may be considered as 

 constituting, are of the utmost importance for the transference of carbon- 

 dioxide to the individual chlorophylliferous cells. Every one of these cells 

 abuts somewhere directly on an intercellular space and is thus in continuity 

 with the external air. Movement of carbon-dioxide in the intercellular spaces 

 is effected, at least in the first instance, by diffusion. The entry into the leaf 

 through the stoma depends on the partial pressure of carbon-dioxide in the 

 air ; on reaching the assimilating cells the gas is completely absorbed and the 

 pressure is reduced to nil, so that the necessary conditions for continuous 

 diffusion are complied with. It cannot be doubted, however, that, in addition 

 to diffusion, movements of the gases in the intercellular spaces in mass, resulting 

 in a rapid supply of carbon-dioxide to the mesophyll, also take place. Varia- 

 tions in temperature as well as bending of the plant due to wind, resulting in 

 changes in the shape of the intercellular spaces, must assist in producing such 

 air currents. 



Let us first of all consider stomata wide open and ask ourselves how it is 

 possible for carbon-dioxide, though present in such a small proportion in the air, 

 to enter the leaf through such minute pores in quantities so great that a sunflower 

 is able to manufacture i-8g. of carbohydrate per square metre of surface per hour. 

 A complete answer to this question is given in the recent researches of BROWN 

 and ESCOMBE (1900). These authors started from a consideration of purely 

 physical experiments. They allowed the carbon-dioxide of the air to diffuse 

 into a vessel through a narrow opening in a thin diaphragm, placing a solution 

 of potash at the bottom of the vessel. They found that the amount which 

 diffused through was proportional, not to the area of the opening, but to its 

 diameter. Thus, if an opening 4 mm. in diameter permitted two units of carbon- 

 dioxide to pass through in a unit of time, one unit of the gas passed through 

 an opening 2 mm. in diameter in the same time ; in other words, the amount of 

 carbon-dioxide stood in the proportion of 2 to I, while the surface extent of the 

 aperture bore the relation of 4 to I ; hence as the size of the opening decreases 

 the rate of diffusion must increase. When, therefore, numerous apertures are 

 made in the diaphragm, and when the activities of these are added together cases 

 must be conceivable where diffusion takes place through a partition pierced by 

 minute apertures as rapidly as if there were no wall there at all. BROWN discovered 

 that this result was reached if the individual openings lay so far distant from 

 each other that they were unable to influence each other's activity, and that 

 that was when their distance apart was at least ten times the diameter of the 

 opening. When we apply the results arrived at from a consideration of such 

 physical experiments to the inflow of carbon-dioxide into the leaf, the first thing 

 to notice is that the stomata possess not a circular but an elliptical form. If we 

 were now to consider the long diameter of the ellipse as the measurement on which 

 diffusion directly depends it might be said that the width of the opening plays 

 no part in the process, a view, as we shall see, directly opposed to observations. 

 According to the detailed statements of STEPHAN the facts point to an entirely 



