258 J. O. HAGSTRÖM, CRITICAL RESEARCHES ON THE POTAMOGETONS. 



Although this variety p, in the locality stated, probably is only one cross, 

 it nevertheless appears in two rather different forms: one with large, broad leaves, 

 in shape more perfoliatus-\ike, 90 — 130x40 — 55 mm: 



f. Yarraingii Baagöe in herb. J. O. Hagström 1902: — Folia magna lata P. 

 perfoliati similiora. 



The other has narrower but comparatively more elongated, i. e. more prcelongus- 

 like leaves, 110-130 x 30—35 mm: 



f. subpra*longus Baagöe, 1. c. 1897: — Folia quam in pra?cedente angustiora 

 et longiora, iis P. prcelongi similiora. 



This var. occurs small-leaved in Norway, Hindöen, 10, Notö (hb. Christian.). 



The Canadian specimens, again, determined by Ar. Bennett under the name 

 of P. i nter mixtus, should be ranked under P. cognatus as a var. or f. intermixtns, 

 provided that they really are P. perf. X prcel. 



Xeither the Canadian specimens nor those from »Waddainsee bei Serbigal 

 (Kupffer, in M. v. zur MOhlen, Pot. Ostbalt., 1906, 172) are described as 

 regards the leaf-margin. 



It is striking that P. prcelongus and perjoliaius, the stigmas and pollen of 

 which are much the same, so very rarely produce mongrels between themselves, 

 whereas P. perfoliatus and grammens, for inst., dissimilar in the afore-named respects 

 scarcely can grow together without producing crosses. This rare occurrence is also 

 stated respecting the other preelongus-hyhrids (alpinas X prcel., crispus x prcel., gra- 

 mmens X prcel., and lucens X prcel.). The explanation of this peculiar fact lies in the 

 early florification of P. prcelongus, b}' which the pistils of P. prcel. are prevented 

 to receive pollen from its coexistents. Only the last emitted pollen" of P. pral. can 

 be thought to hit the stigmas of its companions. 



P. buplenroldeS Fernald in Rob. and Fern., Gray's New Manual of Bot., 1908, 75. 



According to specimens (hb. Stockholm.) from the lakes Fresh Pond and Mystic 

 Pond, Mass., U. S. A., near the Atlantic coast, and a specimen from Connecticut, 

 leg. A. Cray (hb. Christiania) this species is characterized by a stem with very short 

 and very short-jointed branches, and by short and rounded leaves like those of the 

 f. rotundifolius of P. perfoliatus. Their apex and basis, nervation and serrulation 

 as in perfoliatus. Ligules as to form and nervation as in perfol. Peduncle, spike and 

 pistil scarcely deviating. Fruit smaller, 2 ? 5 X 2 mm (perfol. 3.26X2,8 _..-> mm), 

 rounded at the apex and with a thin rostrum. Anatomy like that of perfoliatus. 



Without fruit this species will be very difficult to dietinguish from P. perfol. v. 

 rotundifol. WALLB. as you havc only the short-jointed, small-leaved branches to con- 

 clude from. FERNALD states, it is true, that the leaves are »flat, not crisped», but 

 also the said form of I\ perfol. has such leaves, at least sometimes, which is even 

 acknowledged by Fernald when saying of P. perfol. »leaves . . . usually obtuse and 

 crisped (1. c. 74). 



