413 



I have only had material of ascending stems for inve- 

 stigation. With the exception of some small modifications 

 their structure resembles that of the examined species of Ra- 

 nunculus. The cells of the epidermis are much smaller than 

 those of the cortex; as many as seven layers of the latter 

 occur and they are somewhat more closely placed than are the 

 cortical cells in those species of Ranunculus which have been 

 investigated (Fig. 50, B). The 3 — 5 bundles are each sur- 

 rounded by an endodermis which is slightly lignified in the 

 older stems. As is also the case in other species of Batra- 

 chium the woody mass in the bundle frequently encloses a 

 large air-space, produced by the breaking down of young 

 vessels ; remnants of the walls are often seen projecting into 

 the space. In the material from Iceland (January) these canals 

 had just begun to develop in a young internode which was 

 nearly one cm. long. The continuous pith-cavity which occurs 

 in full-grown stems had not yet begun to develop in the inter- 

 node in question. 



The leaves resemble in structure the divided leaves of other 

 species of Batrachium\ the segments of the aquatic leaves are 

 cylindrical with an axial bundle surrounded on all sides by a few 

 layers of homogeneous, elongated non-prosenchymatous cells 

 (Fig. 50, D). The epidermis contains a larger quantity of 

 chlorophyll than the inner layers; the radial walls of its cells 

 are straight. Only one or two stomata occur at the apex of 

 the segments (Fig. 50, C); they resemble in form the water- 

 stomata figured for R. nivalis. The apex of the leaf usually 

 bears 1 — 4 hairs; but sometimes these are absent. 



When the plant becomes dry during summer Icf. Rosenvinge) 

 it develops, in common with other species oï Batruchinm., leaves 

 which in form and certainly in internal structure differ from those 

 of the aquatic form (cf. Askenasy, Bot. Zeitung, 1870). Such 

 leaves were not found upon the Arctic specimens which I 

 have had for investigation; but a land-form of the nearly- 

 XXXVI. 27 



