﻿These have been divided by Dr. Bos well Syme into four sec- 



■; - ! . 



orispus ; (2 ing cU nsiia only ; (8) 6 



■ the species from zosterifolin to <; - 



does P. lanceolcttus belong? Looking at the mature plant, un- 

 doubtedly to Section 1, Plantayiniftd'ia. But if we examine the 

 plant more closely we find one remarkable character, ,,h,uds at the 

 {'"■•"' "/each leaf or leaf -stalk, a character found in no other species 

 in this section, but absent in no species of the Section 8, Gramini- 

 fnlii! Taking this into account, and looking at the young sub- 

 mtr-Hl growths of lane t ,dahis which have the habit of those of 

 Fricsii, I long ago began to regard this form as a possible hybrid. 



'h : , ; .. . , : . ■ . -:;:ua.-:.-r 



belonging to Section 3, all the fruits of which are more or less 

 toothed towards the base. The case in favour of the hybrid origin 

 of /'. lanceolatus is strongly supported by the local distribution of 

 the species. In what locality does it form a general feature in the 

 flora of the district ? Is it not always confined to a single river, 

 stream, or ditch? In Burwell F< pen, for the 



one spot where Mr. Bennett found it might well have been filled by 

 a single, roots tock. 



I have repeatedly searched Burwell Fen for P. lanceolate with- 

 out success ; the ditch where Mr. Bennett found it has now closed 

 up, and its aquatic vegetation is extinct, but a long series of water- 

 courses forming a continuation of that ditch still exist, and they 

 contain no trace of lanceolatus. It is not difficult to form a theory 

 that might explain the occurrence of a rare species in the Isle 

 of Anglesea and in Burwell Fen, but such a theory is destroyed by 

 the Saone and Loire discovery. Again, although all the specimens 

 from the various localities are good lanceolatus, each locality has its 

 own peculiar form ; presenting, in fact, just such slight indefinable 

 differences as we see in such hybrids as decipietu, though in a jar 

 less umrked degree. 



If, then, we regard P. lanceolatus as a hybrid between P. hetero- 

 phyllus and some one or other of the " Graminifolii," both the 

 peculiarities of the species and the difficulties presented by its 

 distribution will be explained. I think the Burwell Fen form is 

 heterophyllus x Friesii, ; possibly the Anglesea form may be 

 hetero^hjllu* X pusUlm. 



My friend Mr. Bennett has kindly supplied me with information 

 which may be thought to weigh against the view I take — such as 

 the apparent absence of the other necessary forms of Potamogeton in 

 some of the British localities of lanceolatus. At one time this 

 absence would have had great weight with me ; but some years 

 ago, in the case of a hybrid violet, Mr. Beeby pointed out to me 

 that a hybrid might long survive the extinction of one or both of its 

 parent species. I almost doubted this, but kept the suggestion in 

 mind, and watched the distribution and growth of that hybrid violet 

 throughout the fens ; I soon found abundant evidence that in 

 localities where through drainage one of the parent plants was 



