I 1907] CHR YSLER~PO TA MOCETONA CRA E 173 



or even absent. The peduncle shows a structure similar to that of 

 the stem figured; no circle of separate bundles such as in Potamo- 

 geton is found, but the central cylinder consists of two united 

 bundles. 



The well-defined cortical system of this plant is a characteristic 

 feature which at once separates the genus from Potamogeton. In P. 

 NiMallii^ however, we have a suggestion as to how the condition in 

 Zostera may have arisen, as described above (p. 167) ; the lateral bundle 

 e'htering the stem from the petiole and stipule may give off a branch 

 which descends through the cortex. The chief difi'erence between 

 this condition and that in Zostera is that in the latter genus the cortical 

 strand is continuous up and down through a number of intcrnodes. 

 Whether a leaf trace at once penetrates into the central cylinder, or 

 descends at the periphery of the cylinder, or descends altogether out- 

 side the cylinder, appears to be a feature readily modifiable in the 

 monocotyledons, as has been pointed out by the writer (6), and in 

 almost every family of monocotyledons there are members in which 

 at least some of the leaf traces pursue a cortical course for the length 

 of an internode before joining the central cylinder. Apparently this 

 condition is derived from that in which all bundles of the leaf trace 

 Join the stele directly. 



According to Sauvageau (23), Z. MneUeri and Z. tasmanica show 

 two to five cortical bundles on each side in place of the single one of Z. 

 marina. These bundles anastomose with one another at the node, 

 and send a branch to the central cylinder, while receiving a single. 



* 



i 







from 



traces. 



The affinities of this genus arc by no means clear. Like Potamo- 

 geton it has an elongated inflorescence which here reaches a high 

 degree of specialization. The totally immersed habit of the plant, 



pollination 



mentous 



The vascular structures 



may have been derived from the condition found in Potamogeton, as 

 is suggested above, but the resemblance is not striking, especially 

 when we consider the eight radially arranged bundles of the central 

 cylinder. The process of evolution has apparently separated 

 far from its relatives. 





Zostera 



