KUNGL. SV. VET. AKADBMIENS HANDLINGAE. BAND 55. N:0 5. 235 



fails and a more primitive state appears with two median xylem canals. Anything 

 such like more often takes place with the lateral bundles which not rarely are two 

 (or even three) free from one another, although the typical is a compound bundle 

 on either side, see the figure 1. In F. gramineus the conditions are much more 

 fixed and the lucens-tyT^e perfectly carried out, while the specialization of the leaves 

 here is more advanced. But in this respect, again, a regress to the prototype can 

 often be seen in a reduction of number and size of the floating leaves and a simplifying 

 of their structure to a greater resemblance to the submersed leaves. The anatomical 

 conditions of P. lucens are quite analogous with the morphological facts of P. gra- 

 mineus. — In the bark there are 2—3 interlacunar circles of bundles present and 

 commonly scattered 2 a 3 (0 — 10) subepidermal strands in the strengthening layer. 



The stem presents little intention of prolonging itself by branches above the 

 primary spike, while quite the reverse is the case with P. gramineus. Usually there 

 is only one single spike-bearing branch, type A, fig. 2. For the purpose of keeping 

 the spike over the water-surface, it stretches the middle internodes of the stem. 

 The upper ones, again, and the peduncles are rarely elongated (f. ea;ce/!s2fs Hagstr.). 



For the rest the species does not vary much except as to the size of the stem- 

 leaves. The largest stem-leaves I have seen have measured 200 mm in length by 

 60 mm in width. — The most remarkabe variety is v. longijolius DC. (P. macro- 

 phyllus WoLFG.), which has long, narrow stem-leaves and also narrow branch-leaves. 

 — A form with long stem-leaves (as much as 300 x 56 mm) and branch-leaves of the 

 usual sort is f. insignis Tis. The ligules of the stem-leaves as far as 110 mm long. — 

 The form with rather short and broad leaves has been called f. ovalifolius M. K. 

 which, however, at first seems to have comprised similar forms of the hybrid P. Zizii, 

 too. — The main-form is f. vulgaris Cham. (= f. lancifolius M. K.). — P. acuminatus 

 was identified as a lucens-iovra in the year 1816 by E. Fries in Nov. Fl. Suec. and 

 by J. H. Wallman in Liljeblad's Flora. Which of these works was first published 

 it not known to me. — P. lucens p nitens Cham, is a Zizii-tovm, and the same is 

 true about the P. lucens p lucescens Tis. Pot. suec. exs. n:o 59 and the following 

 numbers 60 — 67 and 144, see also under P. Zizii, above! — What is meant by v. 

 jluviatilis Thore and v. rivularis Thore is till now not quite disentangled. Var. 

 lacustris Thore is very likely a form of P. nodosus. — P. lucens v. jloridanus Ar. 

 Bennett, Notes etc. in The Journ. Bot. 1907, 374 is not at all P. lucens to 

 conclude from the description (see also Graebner, Potamog. 1907, 161 : Leaves 50 

 X 16 mm, stipules ca. 2,6 cm etc.). 



Figures. For figures I refer to Flora Danica, t. 195, t. 1384. — Linnaea 

 1827, t. V, f. 16. — Reichenbach, Icones etc. vol. VII, 1845, t. XXXVI, f. 64, 

 t. XL, f. 69, 70. — MoRONG, Najadacea; etc., 1893, t. XXXVIII. — Schenck, 

 Vergleich. Anat., 1886, t. VII, f. 38: cross-section of central cylinder. — Sauvageau, 

 in Journ. de Bot., 1894, p. 117 — 122, ff. 23—27: turios, cross-sections of stem and 

 turios. — Raunki^r, Danske Blomsterpl., 1896, 47, f. 17, C: cross-section of endo- 

 dermis, 55, f. 20, A: vernation, 58, f. 23, A: leaf-part, nervation, 60, f. 25, D: top 

 of leaf, 70, f. 33: cross-section of leaf, 81, f. 44: turios. — Graebner in Kirchner, 



