KUNGL. SV. VET. AKADEMIBNS HANDLINGAE. BAND 55. N:0 5. 



95 



axils next below the primary spike spike-bearing branches are often met with or the 

 young branches occurring there are transformed into turios, which appear partly as 

 wholly transformed short branches, partly as branches with meta- 

 morphosed tops. 



The stem is compressed in the relation of 3 : 1 with rounded 

 margins. Its anatomy corresponds nearly with that of P. joliosus. 

 Central stele terete and consisting of a compound bundle (circular 

 diagram), subepidermal strands as the type, no pseudo-hypoderma 

 and no interlacunar strands. Epidermis-cells of the upper part 

 of an internode 2—5 times as long as broad, in the lower portion 

 more stretched. — The 'peduncle, flattened like the stem and 

 20 — 65 mm in length, lacks seldom, if ever, a few subepidermal 

 strands. The typical pseudo-hypoderma is present and the bun- 

 dles, 4(— 6 a 7) in number, run separately as is shown in the fig. 

 37, A (XX, accidentally occurring bundles). 



The leaf presents a row of large-celled channels on either side 

 of the midrib, lacunulse, and usually four strands of bast-bundles. 

 The regular occurrence of an exterior side nerve and its course 

 are already known from old times and correctly figured by Cha- 

 Misso and Schlechtendal in Linnsea, 1827, t. IV, f. 5. In the 

 same work is also pointed out the »rete cellularum» of the upper 

 leaves »quod totum occupat interstitium nervum inter primarium 

 et laterales». The involucral leaves of this species, namely, are 

 often transformed into a sort of primitive floating leaves by a 

 metamorphosis of the upper half or third part in such a manner 

 as is described under P. ruiilus. »Rete cellularum» is not a cellular 

 tissue but the small air cells or lacunulse of the leaves. But no 

 stomata can be observed in this species either. Front-field of 

 the ligules 11 — 13-fibrous. 



The stem-leaves of this species are usually not much 

 stretched, commonly 50 — 75 mm, but in a variety from Siveden, 

 Gestricia (see below) I have found a leaf-length of as much as 100 

 mm (f. horealis: folia caulina longissima). 



The upper part of the style is protracted backwards (out- 

 ward) into a little beak. The fruit has about the same shape as 

 in P. panormitanus, but is a little larger, 2,2—2,5 mm X 1,5 mm, 

 and with a more obvious keel. However, it varies with conspi- 

 cuous bosses on the lid, especially at the base, by which it chances 

 to coincide more with the pusillus-iruit. 



Besides it varies with narrower or broader leaves (ff. angusti- 

 folius FiscH. et latijolius Ruthe ap. Fischer) and with more obtuse apex (f. ohtusior 

 FisCH.). A small-leaved form is f. minor Tis., Pot suec. exs. f. Ill, no. 101, 1897. 



Fig. 37. P. niucfonatus 

 ScHR. J, Transverse sec- 

 tion of the peduncle show- 

 ing the situation of the 

 bundles, att', bast strands, 

 \^. B, Transverse section 

 of stem-leaf below the 

 middle, midvein portion 

 with 4 strands of snbepid. 

 sclerenchyma (str), X signi- 

 fies accidentally occurring 

 strands nearer to the leaf- 

 base or in broader leaves, 

 «', n", lateral nerves, y^. 



C, Pistil, side-view, ^^. 



D, Top of a leaf, showing 

 the nervation and form of 

 the apex, f-. E, Transverse 

 section of a lignle in young 

 state, y, r, ridge, ic, front- 

 field. 



