MOEPHOLOGY AND ANATOMY OF THE NTMPIT.EACE.i:. 291 



particular to be observed in the sharpness of the delimitation between the central 

 vascular region and that surrounding it, although in no case is any appearance presented 

 that might be regarded as a central cylinder. This distinction is least apparent in 

 the Nymph seas, indeed it is scarcely to be observed at all in N. alba and 3^. blanda; 

 whereas in Victo7'ia regia it is very clearly defined, owing to the fact that the vascular 

 elements at the periphery of the central mass constantly run in certain definite directions, 

 forming thus a zone continuous but for interruptions due to the outward passage of 

 traces to the leaves &c. The outer components of this peripheral zone are found to be 

 groups of phloem elements running around the central region in obliquely horizontal 

 directions, like so many hoops or girdles. Next within come a number of more or less 

 separate vascular bundles (or sometimes grouj)s of tracheides only), the elements of 

 which run almost vertically. Pinally, there is often another inmost system of tracheides 

 running almost horizontally, although this last is frequently absent. 



The rhizomes of the Nuphars differ from those of the Nymplia?as, and from that of 

 Victoria regia, principally in the absence of that "cavernous zone" described above. 

 On the contrary, the ground-tissue is homogeneous throughout, and consists of a spongy 

 lacunar parenchyma with large intercellular spaces which decrease in size as the periphery 

 is approached. Again, the vascular bundles are not so confused ; a greater jiroportion of 

 them run vertically, especially in the peripheral regions, where certain of them are 

 arranged in a very irregular ring, on the outside of which the ground-tissue is traversed 

 by the small leaf-traces only. Most of the vascular bundles have their xylems turned 

 towards the centre, although some are turned in various other directions. Some again 

 run singly and separately in the ground-tissue, but the majority are fused together by 

 their confluent xylems into groups of from two to four. These groups are due to the 

 fact that the bundles in the course of their anastomosis very often run together and 

 very closely applied to each other for some time before separating again. 



In Cahoniba aquatica, whether in the comparatively short internodes of the sympodial 

 rhizome or in the elongated ones of the floating shoots, the structure, on account of its 

 great simplicity, is very difi'erent in appearance from that described in the above genera. 

 There are only two pairs of bundles present. In the rhizome tlie bundles of each pair 

 are situated almost exactly facing each other; in the floating shoot each is placed along 

 one arm of a V, the apex of which is occupied by a canal representing the disintegrated 

 xylems of the pair. At every node a leaf-trace arises from each pair, being formed by 

 the fusing together of two branches, one from each of the components of the pair. At 

 the same time the two bundles of each pair separate and travel horizontally round the 

 stem in opposite directions to meet the corresponding bundles from the other pair, with 

 which they fuse to form the two pairs for the next internode. 



With regard to Nelumbium speciosimi, the rhizome possesses, in its elongated internodes, 

 a large number of vascular bundles arranged in several concentric circles and running 

 quite vertically. They are all directly oriented, except the members of the third 

 and fiftli inmost circles, which are all inverse. In the outer circles the 1 bundles are 

 undoubtedly free and sejmrate from each other, but the twelve which form the inmost 

 ring's of all are held by Van Tieghem to constitute a central cylinder surrounded by a 



