THE GENUS FUMARIA IN BRITAIN 



chalk districts respectively, it might be expected that hybridiza- 

 tion would frequently happen. In my experience, however, this 

 is not so. I have on rare occasions met with isolated plants, 

 entirely or nearly barren, that are evidently hybrids between 

 species growing in more or less abundance around them. But I 

 have never seen intermediate forms varying between different 

 associated types with the appearance of hybrid origin except in 

 one or two uncertain instances that will be dealt with in detail 

 under the respective species. It is noteworthy that in all the 

 cases of hybridity that I have recognized one of the factors has 

 been F. officinalis. 



The measurements appearing in this paper are taken from 

 dried specimens unless otherwise stated, and in the case of the 

 floral organs refer to average fully developed flowers. It may be 

 remembered that the fruits, when dried, are slightly shorter than 

 when fresh, owing to the shrinkage of the fleshy base. 



The foreign distribution is taken maialy from Haussknecht's 

 monograph. 



The specimens quoted as Exsiccata are to be found in the 

 Herbarium of the British Museum, except where the contrary 

 is stated. 



The sign ! implies that I have seen specimens from the locali- 

 ties so marked. Where the sign ! ! appears, I have gathered the 

 species in situ. 



FUMAEIA L. 



(As restricted by Haussknecht, including only SplicErocapios DC. 



Syst. Veg. ii. p. 131.) 



Section I. Grandiflor^ (Latisect^ of Haussknecht). 

 Leaf-segments flat, from oval to lanceolate in outline, never so 

 narrow as linear, at least in the case of any British species. 

 Flowers normally larger than those of F. officinalis L., exceeding 

 9 mm. in length. Upper petal (in good flowers) with margins or 

 wings reflexed upwards. Lower petal with margins or wings 

 little, if at all, dilated towards the apex. Inner petals more or 

 less curved upwards. 



Subsection 1. Cajjreolatce. 



Peduncles relatively long. Pedicels usually much thickened 

 towards the tip, and (in British species) rigidly recurved in fruit. 

 Bracts (except in F. flahellata Gasp, and F. gaditana Haussk.) 

 about as long as the fruiting pedicels. Sepals large, generally 

 entire towards the apex, at least as broad as the corolla, and not 

 less than one-third of its whole length. Fruit very obtuse and 

 obscurely keeled (except in F. viacrosepala Boiss. and F. montana 

 Schmidt), and when fresh (at least in the British species) with a 

 distinct fleshy neck, which is narrower than the dilated lip of the 

 pedicel. 



The plants placed here by Haussknecht form a comparatively 

 well-marked group, notable for their large sepals and usually long- 

 peduncled racemes. i^'.y2a6e//ftia Gasp., however, in the form of 



