1896.] D. Praia — Some additional Fumariaceee. 29 



Clarice nn. 12585 ! 12710! Khasia ; Shillong, etc., Hooker and Thomson! 

 Clarke n. 7300 ! n. 44266 ! Mann ! Gallatly ! Murdoch ! 



Stems weak, much bran-lied, leafy. Leaves membranous, flowers yellow £ in. 

 long ; posticous petal &•* long as its spur. Capsule \ in., style as in C. Casimirinna. 



On Dr. Royle's original sheets of C. filiformis this is the plant distributed. 

 And it would appear as if Dr. Royle had distinguished this from the preceding species 

 for, though he does not mention C. Casimiriana under any name in III. Him. he has 

 named it C. longipes in Herb. Saharaupur. Dr Wallich's n. 1433 is a mixture of this, 

 of C. Casimiriana and of his own C. diphylla ; the latter fact probably explains 

 the use of the name G. longipes by Don to designate G. diphylla. This is also the 

 plant from Kamaon and Khasia referred by Mr. Maximowicz, doubtfully as to itself. 

 and to the exclusion of C. longipes Maxim, not of DC. (0- Casimiriana Duthie & 

 Praiu), to G. sibirica. When preparing the present paper the writer came to the 

 same conclusion as Mr. Maximowicz, viz : — that the species with capsules in which 

 the seeds are 1-seriate must be G. longipes DC, since M. de Oandolle has described 

 C, longipes as having linear capsules. This view possessed the great advantage of 

 enabling the use of Royle's name 0. filiformis — regarding which, owing to the existence 

 of authentic specimens, there was no dubiety possible — for the plant with ovate cap- 

 sules. And when duplicates were distributed from Calcutta to the great European 

 Herbaria, the species with linear capsules was issued as O. longipes and that with 

 ovate capsules as C. filiformis. But before publishing this paper, the writer took 

 the Liberty of referring the matter for final decision to Mr. C. de Candolle. He 

 and Mr. Baser have most kindly compared specimens of both plants with the type- 

 specimen of C. longipes in the Prodromus Herbarium. The result of their examina- 

 tion is that the original description of the capsules of G. longipes does not accord 

 with their actual condition ; the true G. longipes is in reality the same as C. filiformis 

 Royle. In consequence of this the plant with linear capsules is still unnamed and 

 Mr. Duthie and the writer have named it C. Casimiriama as a slight recognition of 

 the obligation under which Mr. C. de Candolle's kindness has placed them. Students 

 of the genus should therefore note that sheets issued from Calcutta as G. longipes 

 are in reality C. Casimiriana ; those issued as C. filiformis should be known as- 

 C. longipes, that being the oldest name. 



These two species, along with C, tongolensis and G. gracilis, form a very natural 

 group of species that perhaps only differ from each other as species of secondary rank.* 



* This is not tbe only instance in Corydalis where two species repeat practi- 

 cally every vegetative character and only differ slightly in flower and more consider- 

 ably in fruit. A good example of the same parallelism among Eastern Asiatic 

 species is exhibited by the well-known Eastern Chinese and Japanese species CL 

 incisa Pers. and a species from Central China which has been named in Herb. Paris 

 and Herb Calcutta by M. Franchet and the writer,- the following is a brief diagnosis. 



Corydalis Hemsletana Franchet ■ Sf Prain ; rootstock rather slender dividing 

 at apex, crowned with radical leaves and emitting numerous slender flexuous stems; 

 leaves alternate long-petioled, twice ternate, segments ovate-oblong acutely incised; 

 bracts oblong-cuneate incised shorter than the pedicels; sepals laciniate; outer petals- 

 both crested, and with explanate margins ; fruit wide-ovate acute at both ends. 



Centra l China : Hupeh, Henry n. 3729 ! 



Very near C. incisa Pers. from which it differs in its smaller size (stems 8' 

 .instead of 20 in.), larger flowers with spur rather longer than lamina, and shorter 

 wider fruits. 



