RESTRICTION OF TRANSPIRATION BY EPIDERMIS 105 



testing the extent to which the epidermis as a whole, and its outer 

 wall in particular, are capable of checking transpiration. 48 But 

 the majority of the objects that have been employed for this purpose 

 (such as apples, plums or pieces of Cactus-stem.) are ill fitted to afford 

 any exact information as to the restrictive effect exercised by the 

 epidermal layers of foliage-leaves and other vegetative organs. A 

 transpiration experiment of this type carried out by the author may 

 be reproduced at length. Two similar pieces were cut from leaves 

 of Aloe vulgaris and coated all over with tallow, except for an area 

 of 10 sq. cm. in each case. The epidermis was then carefully removed, 

 with the help of a scalpel, from the uncoated area of one of the pieces. 

 A couple of apples were treated in precisely the same way. The 

 loss of water by transpiration (estimated by weighing at stated intervals) 

 worked out, in grams per sq. cm., as follows : 



Leaf of Aloe vulgaris. Apple. 



In Aloe, therefore, the loss by evaporation during the first three 

 hours subsequent to the removal of the protective epidermis was 23*5 

 times as great as that suffered by the intact leaf ; after twenty-four 

 hours it was still 15*6 times as great. In the case of the apple the 

 corresponding ratios were very similar, namely, 25*6 : 1 and 15 : 1. 

 Evidently the epidermal layer of Aloe vulgaris and that of the 

 particular sort of apple selected happen to possess a similar restrictive 

 power. 



In experiments like that which has just been described the 

 controlling action of the thickened and cutinised outer walls cannot 

 be fully effective, owing to the presence of more or less numerous 

 stomata. As a matter of fact, the stomata probably close very soon 

 under the conditions of the experiment; nevertheless entirely satisfactory 

 measurements of cuticular transpiration can only be obtained by 

 employing epidermal layers which are completely devoid of stomata 

 and thereby eliminating the effects of stomatic transpiration altogether. 

 There is yet another source of error which must be taken into account. 

 It was known to Nageli that potatoes and apples suffer a greater loss 

 by evaporation, under otherwise similar conditions, if they are first 

 killed by freezing, than if they are observed in the living state. The 

 difference is probably due to a restrictive influence exerted upon 

 evaporation by the plasmatic membranes of the living cells, an effect 

 whic i, of course, disappears when the protoplasts are killed. This 



