38 THE GENUS FUMARIA IN BRITAIN 



with teeth pointing towards the apex ; by a laterally compressed 

 upper petal, with wings produced backwards a considerable distance 

 towards the spur ; and by fruits which, if mature, become dis- 

 tinctly rugose when dried. In the living plant the most marked 

 of these features is probably the laterally compressed and long 

 winged corolla, which is quite different from that of every other 

 British species. 



The form differing most widely from typical F. Bastardi, as I 

 understand it, seems to be F. Gussonei, which normally bears 

 smaller flowers, with a dark tipped upper petal and smaller fruits, 

 more rounded above and narrower below. Boissier says of it : 

 " Habitus et florum magnitudo F. officinalis, quae longe differt . . . 

 fructu majore . . . ," and its relationship to typical F. Bastardi 

 may be said to be somewhat analogous with that of F. muralis and 

 F. Borcsi. But, on the whole, I think the two rugose-fruited 

 forms present less essential differences, although the colouring of 

 their flowers looks very distinct, and so I place F. Gussonei as a 

 variety rather than a subspecies. F. Gussonei appears to be the 

 prevalent form of the species in the southern and eastern parts of 

 its range, and the most extreme state that I have seen is from 

 Greece, near its most easterly Hmit (Heldreich & Halacsy, Fl. 

 ^gea. Insula Melos, 1889). 



F. Jordani Guss., so far as I can judge, hardly differs from 

 F. Gussonei except in the absence of the dark tip of the upper 

 petal, and I suspect F. serotina Guss. to be a shade-form of the 

 same plant. 



Jordan's types of F. vacjans, on the other hand, are nearer to 

 the restricted F. Bastardi or F. confusa, from which I can only 

 distinguish them by their smaller, more acute, and less broadly 

 necked fruits. I hesitate to separate these even as a variety. 

 Spanish specimens in my herbarium, labelled F. vagans, differ 

 again in having a dark-tipped upper petal like F. Gussonei. 



In Haussknecht's Monograph one variety only of F. Gussonei 

 is recognized, yo^v. patens, which seems to be a modification of the 

 true plant of Boissier. The species is otherwise divided by the 

 monographer into four forms : a. typica, which includes F. affinis 

 of Hammar ; b. Jordani, which is said to be the F. Jordani of 

 Gussone ; c. diffusa, with which original specimens of F. Bastardi 

 Bor. and F. confusa Jord. are identified ; and d. umhrosa, to which 

 other examples of F. confusa and authentic specimens of F. vagans 

 Jord. are referred. In the case of the last two forms the pale 

 coloured flowers are mentioned as a character, but as the size of 

 the corolla in both cases is shown as less than in a. typica, it is 

 evident that the specimens examined were abnormal, and either 

 cultivated or shade-grown, in which fight Haussknecht apparently 

 regarded them, erroneously supposing the lack of the dark corolla- 

 tip a contingency of environment. That this feature was essenti- 

 ally characteristic of these forms the material at his disposal 

 seems to have afforded no evidence, and he thus mistook a second 

 important point in the corolla of these plants from want of 

 good specimens. In this connection the monographer's remark. 



