HYBRIDS AMONG BRITISH PHANEROGAMS 273 



is frequently the case with wild roses, and usually so with those 

 which are under cultivation. 



Pyrus Aria Ehrh. has long been supposed by some botanists to 



produce hybrids with P. Ancuparia Ehrh. and P. torminalis Ehrh, 

 Prof. Koehne is an advocate of several forms in this Sorbus section 

 resulting from hybridity. Dr. Focke says (p. 145), " P. AucupaHa 

 X scandica Bab. is Sorbus hybrida L., one of the few hybrids which 

 were rightly determined by Linne"; and he gives other examples. 

 The Rev. A. Ley, who knows the British forms in the field better 

 than any of us, holds, on the contrary, that none of our Sorbus 

 forms are real hybrids, adducing the fertility of our various plants 

 as conclusive evidence. Judging, however, from what is stated 

 above with reference to hybrids of Geum and Rosa, we must not 

 assume the fertility of a plant to be conclusive evidence against a 

 hybrid origin ; and, on the whole, I am inclined to believe that 

 some at least if not all the forms which have been put under 



tifid 



?/< 



P. are P. Aria x torminalis. Perhaps some botanic garden will 

 experiment. Gratagus oxyacantha x monogyna is only a cross 

 between varieties. 



Droserace^. — Drosera anglica x rotundifolia (D. obovata Mert. & 

 Koch), produced only on very soft boggy ground in my experience. 

 It is remarkable that D. rotundifolia L. and D. intermedia Hayne 

 should produce no hybrid, though often growing abundantly together. 



Onagrace^:. — Epilobium has produced numerous combinations 

 which are correctly stated in the London Catalogue, ed. 9. E. hir- 

 sutum L. crosses with E. lanceolatum, montanum, obscurum, and 



ifiorum ; E. parvifl 



varvijl 



ifl 



roseum ; E. adnatum Grisebach with E. Lamyi, montanum, obscurum, 

 palttstre, parviflorum, and also, as a garden hybrid at Milford, Surrey, 

 with E. lanceolatum', E. obscurum Schreb. crosses with E. palustre, 



'fl 



Ifl 



florum and roseum; E. alsinefolium Vill. with E. anagallidifolium, 

 montanum, obscurum, and palustre; E. anagallidi folium Lam. with 

 E. obscurum and jmlustre. Mr. Marshall has no additions to make 

 to this list, and he tells me that he believes Epilobium hybrids to 

 be, as a rule, completely sterile. Mr. G. C. Druce states very 

 positively that they are fertile. I must confess that I am rather 

 puzzled by this statement, for I incline to Mr. Marshall's view. I 

 have seen few in the right condition for forming an opinion, but 

 those few appeared to me to be producing no seed. . 



Kubiace^e. — Galium Mollugo X ventm has occurred in several 

 places, and is, I think, our only hybrid in the genus. M. Kouy 

 (Fl. France, viii. 19, &c.) gives G. ochroleucum Wolf. ap. Schweigg 

 et Kcerte FL Erlang. (1811) as the name for an aggregate hybrid 

 between forms of these two, with many named varieties. He also 



Journal of Botany. — Vol. 45. [July, 1907.] x 



