KUNGL. SV. VET. AKADEMIENS HANDLINGAE. BAND 55. N:0 5. 237 



petioles (5 mm) and shorter ligules; peduncle incrassate. It wholly coincides with 

 specimens from France, St. Denis (hb. Lund.). 



This variety, of which I never saw fruit, has often appeared to me as a 

 planta duhia. In some respects it reminds of the hybrid P. lucens X nutans. On 

 closer examination, however, I have not been confirmed in such a supposition. The 

 topmost leaves never show the least tendency of getting a structure like the floating 

 natans-\eSi{. Nor does the stem-anatomy properly approach to P. natans. I must 

 regard it as a singular, characteristic river-form of P. lucens. The narrow shape of 

 stem- and branch-leaves and the sterility must be owing to local conditions. Prof. 

 K. R. KuPFFEE, who is also somewhat hesitating as to this form, has expressed a 

 supposition that it might be a proper species (Riga Korr. Bl. 1906, 161 — 162). The 

 rare and sporadic occurrence and the sterility seem, however, to contradict this 

 thought. I know it only from three localities outside the Westrussian district, viz. 

 from the above named station in France, one locality in Denmark (Skalsa), and from 

 the Lena River in Sibiria, Kirensa, ca. 58° N. lat. (hb. Petersb.).^ The original habitat 

 was »la riviere de Bapaume* in the Northeast France to the south of Lille. Greniek, 

 in his Flore de la Chaine Jurassique, II, 1869, says on P. lucens: »Hab. toutes 

 les mares et rivieres, depuis la plaine jusque sous les sommets, ou il prend ord. la 

 form p» (P = P fluitans Coss. et Germ. = [3 longifolius DC). According to this 

 statement the form would belong to the rather high-situated mountain brooks. A 

 mistake may, however, possibly have been made between this variety and the river- 

 form of P. nodosus which often extremely resembles var. longifolius of P. lucens. Prof. 

 G. Fischer hesitates between the possibilities of P. lucens x alfinus, hicens x natans, 

 and lucens x nodosus. But its occurrence in Sibiria, where P. nodosus does not grow 

 is against the last combination and against the two first placed partly the leaf- 

 serrulation, partly, among others, what is said of it above. Already in 1827 Chamisso 

 wrote: »Speciminibus inspectis P. longifolium Gay nil aliud quam heec forma longi- 

 folia* (i. e. P. lucentis), and this view is no doubt the only correct. The weakened 

 serrulation does not derive from a hybrid origin (in which case rudimental teeth 

 were rather to be expected) but from the abating width of the leaves. In specimens 

 with reduced blade in some leaves we always find the strongest serrulation in the 

 leaves (stem- and branch-leaves) which have their full width, while the narrower 

 have a comparatively weaker serrulation. The local conditions may perhaps also 

 play any part. Besides we must not forget that in older leaves, and then especially 

 from running water, the serrulation to a great part or even completely can be abraded, 

 so that the margin seems to be quite smooth. 



P. lucens occurs also in N. America and in Asia, Kashmir, 92, Duthie (hb. 

 Kristiania), Khiva, 99, Paulsen (hb. Haun.), but the distribution is not yet fully 

 known. A revision of lucens-ViX^Q forms is highly wanted. Especially is that the case 

 as to the African plants which have been included with P. lucens {azoricus, vaginans 

 etc.). True P. lucens is not seen by me from the regions to the south of the equator. 



^ I have seen a similar form, but with longer-stalked stem-leaves, from Pcud Oreille River, Orcgou, 

 U. S. A., leg. Dr. Lyall 1861 (hb. Stockholm., Borolin.). 



