IX Leaves 1 79 



the discussion. Here, as everywhere, the aim has been to 

 interest the reader in it, or at least to make it clear that 

 there is one, and that scientific studies involve primarily, 

 beside wide observation, the critical reason, not the 

 credulous memory. Hence before leaving this subject it 

 may be profitable to cite yet a third writer. Thus Schrenk 

 finds cases in which chlorophyll grains do occur upon 

 those very cross -partitions to which Haberlandt denies 

 them, and not unnaturally urges that he cannot see 

 how the presence of a few granules at these walls could 

 materially obstruct the passage of the sap-current. 



For him then "neither Stahl's nor Haberlandt's theories 

 are necessary to explain the structure of the palisade cells." 

 He reminds us of Frank's observations, that the chloro- 

 phyll grains are especially abundant in the neighbourhood 

 of air spaces. There are of course vertical chinks in the 

 palisade parenchyma, hence also the absence (or com- 

 parative scarcity) of granules at the bottom of cells, on 

 which Haberlandt has so much insisted, becomes intelli- 

 gible for quite another reason, that these generally abut 

 upon another cell, and not upon an air space I 



Discussion of this kind may seem to some unprofitable, 

 as if there could be no certainty in physiology. Yet we 

 should be losing one of the best lessons in biology if we 

 failed to see the variety and complexity of its problems, 

 and to see too how each new theorist may be doing good 

 service by bringing some new factor or aspect of the case 

 into view. Thus Schrenk goes on towards a compromise : 

 " How can we imagine a plan better adapted to fulfil these 

 two conditions {i.e. of access to air and to sunlight) than 

 that on which the palisade tissue is constructed ? Into a 

 given volume of this tissue the greatest possible number of 

 cells is packed in such a manner that each presents its 

 upper surface to the incident rays, permitting them to 



