116 



J. O. HAGSTEOM, CRITICAL EESEAECHES ON THE POTAMOGETONS. 



tomy is chiefly the same as in the next foregoing and following species. The devia- 

 tions lie in the fact that the vasal bundles of the central cylinder distinctly divide 

 into two lateral and two median bundles, that a single cell-layer or hypodermatic 

 supporting stratum sometimes exists in the borders of the stem, where, in such a 

 case, the strands also appear in a greater number, finally that rarely one or a few 



faint interlacunar strands occur on 



str 



■vh 



.0. 



o 



.OoOQO 





Fig. 51. P. obtitsifolius M. & K. 

 tion of a stem-leaf at the middle, Y. 



■i. Transverse sec- 

 B, Top of a stem- 



Transverse section of ttie rhizome, 



] 



T • 



^, 



QG 



(7/ 



leaf, f 



Pistil, «, side-view, h, from the back, ^y*. E, Transverse 

 section of the stem, Y. * accidentally occurring bast-bundles. 

 m l'\ surface section of the stem-epidermis, \^. G, Surface 

 sect, of the epidermis of the peduncle, \^. H, Transverse 

 section through the leaf and ligule at the very base 

 immediately above the insertion point and before the 

 development of the bast-bundles in the upper surface of 

 the leaf and in the intercarinal part of the ligule has 

 taken place. The subepidermal strands of the stem con- 

 tinue in the lower surface of the leaf and in the wing- 

 iield of the ligule. Cf the fig. 27!, \''; m, midvein, n', 

 lateral nerve, str, bast-bundles, /, lacunte, cc, central stele, 

 ic, intercarinal part (front field), r, ridge-part. K, Young 

 ligule, y. 



both sides of the central cylinder 

 in the border-segments of the cross- 

 section. In the peduncle the bundles 

 (5q| run separate from each other, the 

 aQQi-cc two median ones approximated, the 

 lateral ones more remote from the 

 centre. No partition of the lateral 

 bundles has been observed. A sup- 

 porting layer is present, whereas 

 the subepidermal strands are absent. 

 The peduncle has the same shape as 

 the stem but is a little thicker. 

 The rootlets lack sclerenchymatic 

 tissue and air-channels. 



The rhizome is A^ery slightly 

 compressed in this species also. The 

 bundle system of the central cylinder 

 is reduced to a compound bundle, 

 surrounded by an endoderm of rather 

 thin -walled 0-celIs. In the cortex 

 the lacunar system is reduced to 

 nearly a single circle of narrow la- 

 cunae, outside which the cell-tissue 

 is more compact. Subepidermally 

 there are scattered faint sclerenchy- 

 matous strands (see fig. 51, C!).* 



Respecting the anatomj^ of the 



leaf I refer to the figure. In the 

 upper part of the leaf the lacunas along the midrib gradually disappear. The midrib 

 and the lateral nerves finally grow very faint and join in the very apex. In the broader 

 leaves there is not seldom a faint nerve within or without the proper lateral nerve, 

 in which latter case the leaf bears a certain resemblance to a mucronatus-\esii. Those 

 secondary lateral nerves, however, usually anastomose with the chief lateral nerves 



' The rhizome described and figured by C. W. Fontell in his Beitruge etc. (Taf II, f. 55) is the basal 

 portion of a lurio and not a true rhizome, whicli, besides, is a very rare occurrence in tliis si)ecies as well as 

 in all its allies. 



