1 66 SUBMERGED LEAVES [CH. 



form of the epidermal cells may be correlated with the feeble- 

 ness and uniformity of the transpiration stream. He supposed 

 that the active and variable flow of sap in land plants might 

 bring about* the accumulation of nutriment at certain points 

 of the epidermisj thus favouring localised cell-multiplication 

 and the production of hairs and stomates. It seems possible to 

 the present writer that this suggestion contains an element of 

 truth. But on the other hand it must be remembered that 

 stomates have been observed in a large number of submerged 

 leaves, such as those of Lobelia Dortmanna 1 , Fillarsia ovata 2 and 

 Pontederia cordata 2 ^ and on the lower surfaces of certain floating 

 leaves, such as Limnocharis Humboldtii* and Hydrocharis Mor- 

 sus-ranae*. Porsch 5 , who has considered the subject compre- 

 hensively, concludes that the stomatal apparatus must have been 

 gradually evolved over a long period of time, so that its charac- 

 ters have become fixed with great tenacity; for, in cases where 

 its existence must be not only superfluous, but attended by a 

 certain danger to the plant, instead of being discarded, it is 

 often modified secondarily in such a way as to render it func- 

 tionless. He shows that, in the case of submerged plants which 

 retain their stomates, four different modifications are found, 

 each of which must have the result of preventing water entering 

 the tissues through the aperture between the guard cells : 



(1) The guard cells may close on submergence, even in full 

 illumination, e.g. Callitriche verna and Hippuris vulgaris. 



(2) The aperture may be permanently closed, as in the case 

 of Potamogeton natans (Fig. 107 B\ in which the whole stomatal 

 apparatus remains roofed in with cuticle. 



(3) The development of each stomate may actually cease at 

 an early stage. This is rare, but such abortive stomates are found 

 in the submerged parts of a species of Oenanthe*. 



1 Armand, L. (1912). 2 Costantin, J. (I885 1 ). 



3 Duchartre in discussion following Chatin, A. (I855 1 ). 



4 Goebel, K. (1891-1893). 5 Porsch, O. (1903) and (1905). 

 6 Porsch uses the specific name "Oenanthe aquatilts, L."; he is pro- 

 bably referring to Oe. Phellandrium, Lamk. var. fluviatilis, Colem. 



