METHODS. 23 



paper test, that on the white parts of leaves the stomata do not function, 

 illustrates the invalidity of arguing from the behavior of such indications. 

 In many plants with non-chlorophyllous areas, the stomata are known to 

 function, i. e., to open and shut, and their guard-cells are often provided with 

 chlorophyll. 



Francis Darwin very properly admitted the inadequacy of the hygrometer 

 methods devised by him, and classified them with other methods for demon- 

 strating stomatal as contrasted with cuticular transpiration. Among these, 

 however, we find regarded by Darwin as the least accurate, "microscopic exam- 

 inations of the uninjured leaf," a method clearly to be used for the determina- 

 tion of the condition of the stomata, unless we assume what must rather be 

 proved. It seems perfectly clear that Darwin's hygroscopes are available 

 for demonstrating sharp contrasts in transpiration, as occurs, e. g., when the 

 hygroscope is pushed from a white to a green portion of the leaf. But it 

 would seem improper to draw conclusions as to the condition of the stomata, 

 except when a rough contrast of the closed and open conditions is to be deter- 

 mined. Weiss (1878) may be perfectly right in asserting that the stomata 

 on non-green parts behave as normal ones, because, as I have satisfied myself 

 by personal examination, the stomata on white, and on colored but non- 

 chlorophyllous, parts of many variegated leaves are normal (Zebrina, Poin- 

 settia, certain grasses, etc.), while Kohl's (1886, p. 39) non-chlorophyllous 

 stomata, I am equally persuaded, are not incapable of movement. Instruct- 

 ive and suggestive as Darwin's work with stomata is, and in spite of the fact 

 that I have found his the most suggestive of essays on the general subject, 

 I nevertheless feel urged to insist that the only way to know what stomata 

 are doing, and to know the relation between their condition and the diffusion 

 of matters into and out of the leaf, is to determine these quite independently, 

 and above all to know exactly what the condition of the stomata is. To do 

 this we must see the stomata, not one or a few, but many, and on different 

 leaves. On this point Copeland (1902, p. 330) is correct. 



All of the methods above referred to may properly be used for certain 

 purposes, as, e. g.,to determine roughly the ratio of transpiration of the upper 

 and lower surfaces of the same leaf, but for exact work on stomata some 

 method of direct observation was needed. 



Immediate microscopic study of the stomata on a leaf in situ, as has been 

 carried on by Kohl (1895 and earlier) and others, was practically, though 

 not wholly, impossible in both of the plants studied. To study the living 

 stomata in epidermis removed alone, or in thick tangential sections of the 

 leaf, and mounted in water, offered difficulties and sources of error which 

 could not be overcome. 



Air bubbles, which cling so persistently, prevent accurate measurement; 

 and it would have defeated my purposes to have used the method of mounting 

 the epidermis or sections in boiling water, which, upon standing, will take up 



