iv] THE FROGBIT 45 



surface of the leaf, have slightly prominent, external, cuticular 

 ridges (Fig. 26); it is probable that here, as in Trianea and 

 in certain other plants with floating leaves, the closure of the 

 stomates is brought about by the approximation of these ridges, 

 rather than by the bulging of the ventral walls^. Haberlandt 

 has suggested that this form of stomate is adapted to diminish 

 the risk of capillary occlusion of the aperture by water. 



The palisade parenchyma, which lies beneath the upper 

 epidermis, is, in normal leaves of Hydrocharis^ extremely well 

 differentiated (Fig. 25). On one occasion, however, in the 

 latter part of May, the present writer found a number of 

 plants which were entirely submerged, the winter buds having 



Fig. 26. Hydvocharis Morsus-ranae, L. 



T.S. upper epidermis passing through a 



stomate. (X318.) [A. A.] 



Fig. 27. Hydrocharis Morsus-ranae, L. 



T.S. leaf of young plant growing entirely 



submerged at the bottom of a ditch. 



May 17, 1911. (X78 circa.) [A. A.] 



apparently been caught in an algal tangle at the bottom of a 

 ditch, so that they were unable to reach the surface, but un- 

 folded beneath the water. The green colour of these leaves was 

 unusually pale, and a section of one of them revealed the fact 

 that the palisade region was poorly diifferentiated, the cells 

 being scarcely elongated (Fig. 27); it was, in fact, a typical 

 'shade leaf.' The spongy mesophyll was developed normally. 

 In Hydrocharis this tissue is not distributed in the irregular 

 fashion with which we are familiar in land plants, but it takes 

 the form of plates of cells disposed in a polygonal mesh-work 

 over the lower epfidermis, which itself contains a small amount 

 of chlorophyll (Fig. 25 B). Attention has been drawn by 



1 Haberlandt, G. (19 14). 



