II.] STARCH FORMATION IN THE LEAF 27 



the box and pick off one or two leaves, which must be 

 put away in a box until they can be brought indoors for 

 testing ; to one or two other leaves still attached to the 

 plant fasten opaque designs covering part of the leaf, the 

 simplest plan being to cut two discs of cork and pin them 

 together with the leaf between. Leave the leaves thus 

 protected in places from the light until evening, and then 

 gather them. Follow the same routine as before in 

 testing the leaves for starch ; it will be found that the 

 leaves which have been kept in the dark for thirty to 

 forty hours are quite devoid of starch, from which we 

 may conclude that any starch present in the leaf when 

 the plant was covered up has been removed in the dark. 

 The leaves, however, which were afterwards partially 

 exposed to the light show a pattern ; the protected 

 portions still contain no starch, whereas the rest of the 

 leaf on which the light has fallen has gained starch, and 

 so will colour up when the iodine solution is applied. 

 Further experiments depending on the test for starch 

 can also be tried ; for example, a variegated leaf, white 

 and green, will show starch in the green parts and not 

 in the white. Again, a small plant may be kept for a 

 day or two under a bell-jar over water with some strong 

 cautic soda solution standing beside it ; the caustic soda 

 will remove all the carbon dioxide from the air with 

 which the plant is in contact, and the leaves after a day 

 or two cease to contain any starch, though they may 

 have been in full sunlight for some hours before they are 

 gathered. In this way it is possible to show that the 

 formation and presence of starch in the leaf is dependent 

 upon (i) exposure to light ; (2) upon the greenness of the 

 leaf; and (3) upon the presence of carbon dioxide in the 

 air with which the leaf is in contact ; all of which 

 conclusions agree with the views of assimilation that 

 we had before obtained by other methods. 



