— 277 — 



belonging to the P. pol i/(/OHif(>liiis-gvou\:) and 3) one or several 

 barren forms, especially one in Europe rather common form which 

 according to its liabit and inner structure is to be looked upon as 

 R Jiicens x nofaus as far as the bastard question can at all be 

 settled without experiments of cross-fertilization. 



With the greatest facility we are now able to separate the 

 two forms by only a single transverse section of the stem and the 

 petioles. How impossible it has been, even to Fotamo(jeton-s\.i[denis, 

 to class a given plant is seen from the following uttering of 

 Bennet^): „A specimen from „Varde leg. Hempel" which (except 

 that it has no fruit) might well have been gathered in the United 

 States as F. LoncJiltes'' (1. c. pag. 296) ; but P. fuitans from Vai'de 

 rivulet has just nothing to do with /-'. Lonchites but is /'. hicens 

 X natans and has numerous bundles in the bark of the stem 

 while P. Lonchites, just as is the case with the Neckar-plant, is 

 without bundles in the bark. 



Finally the question is only about the names of these two 

 forms. Which of them are we to give the name of P. 

 fluitans, or ought this name not at all to be kept? Roth's first 

 description is the following-): „P. lluitans, P. foliis inferioribus 

 longissimis, lanceolatis, acuminatis, membranaceis ; superioribus 

 ovali-lanceolatis, coriaceis: omnibus petiolatis." From this, however, 

 is not to be seen which form Roth has had to deal with. The 

 fact that the fruits are not described indicates that this plant has 

 been without fruits but even if his plant was without fruits it is 

 not a matter of course that it could not fruit at all. 



If Roth has seen both forms he has most likely considered 

 them to belong to one and the same species as most of the later 

 authors have done. Therefore we might think it the best to throw 

 out the name of P. fluifans and to choose two new names or at 

 least one and give this one to the Neckar-plant seeing that for the 

 bastard a connection of the names of the parent species would be 

 sufficient. The good precept of not increasing the number of 

 names more than absolutely necessary, however, makes 

 me propose to keep the name of P. fluitans for the one form, and 

 as it is the right thing to designate a bastard with only a connec- 



') Bennet, Aitliur, Notes on i'ondweetls. Jouin. of Botany, 1893, page :J94 



— 1>!)7. 

 ') Rotli, A. G., Tent. Fl. Genu. I. Pa;,'. 1± 



