OKTHOBBHYXCH.E. 



139 



Plate 20. D,E; figs. 1 1-17.-Pedicularis gracilis Wall. var. typica. 



D-Specimen of form "vera" from Nepal (Wall., Cat. n. 413 in Herb. Calcutta. 



,, „ approaching to P. porrecta from Jumnotri, near the snows 



(Duthie n. 810 in Herb. Saharanpur). 



Plate 21. A-E; figs. l_i 3 .-P e dicularis gracilis Wall. v AE . macrocarpa Prai» 



and vak. knasiana Hook. f. ■ 



A, C— Specimen of var. macrocarpa from Jongri; Chooabainia. 



B 

 D 



E 



" » n » Singalelah (Clarke n. 12732). 



n ii 



khasiana from Toungyi, Slum 1 1 .11$ (Colh tt n. 8). 



» ii ii ,i Kala Pani, Khasia 1 1 ill- (Giiflith n. 742), 



The present account of this widely-distributed and very variable Himalayan specie 

 is based on the examination of 164 specimens with dissections of over 2 corolla*. 



Of these specimens 104 have been of var. typka (form "vera" 25. " ricta" 08 and 



"prostrata" 11), 36 have been of var. macrocarpa, and 24 of var. khasiana. 



The species is most nearly allied to P. porrecta, which is usually very different 

 habit. The leaves and calyx of P. porrecta are subcoriaceous (in this respect, howef 



var. khasiana approaches it), its calyx teeth are distinct and serrate at t lie apices (in 

 which respects var. macrocarpa approaches it), while it is much shorter and smaller and 

 is hardly branched (though even in these characters it is approached by an extreme 



form of var. typica; Duthie n. 810 in Herb. Saharanpur, and specimens in Herb. 



G. Watt.). Perhaps P. porrecta itself will ultimately have to be regarded as only an 



extreme variety of P. gracilis. 



The forms that are included here under P. gracilis are as a rule very eai 



distinguished, but connecting specimens of all kinds occur, and they pass into each 



other on every hand. 



Most easily distinguished of all is var. khasiana, with small thick cauline leaves, 



hardly larger than those on the branches, sub- coriaceous calyx with cartilaginous sub- 

 dentate margin, and with a distinct geographical area ; (it is confined to the Indo- 

 Chinese Division of the Himalayan province.) But Mr. Duthie has collected in Jaun- 



Bahar (n. 1200!) specimens with the membranous calyx and sinuate hardly 



leaves of vak. 



On"the other hand, Mr. Clarke's specimens from Kegwima and my own fron 



sar ±>ahar (n. ±zw ij apcuiiuciia v»^ — ~ j 



dentate margin of var. typica, form « stricta," and with the small 



khasiana 



Konoma, two localities in the Naga hills, have the calyx and puberulous stems c 



but have the cauline leaves shortly petioled, and therefore connect vak. 



khasiana 



khasiana with' the other two varieties. The Shan hill plant collected by General Collett 



is exactly the Khasian variety. 



The other varieties are confined to the Himalaya proper. Of these, vae. macrocarpa 

 has to be recognised as separate from the typical plant by characters of almost as great 

 weight as thos^ that give P. porrecta specific rank. These are its ovate serrate calyx- 

 teeth, its higher staminal insertion, and its much .arger capsule. In ^ a is gather- 



three forms of 



gs 



typ 



very well agree ; and though it grows side by side with all 



there are really no intermediate states. Since however, it is probable that 

 luac J - * • n 7 - I have not ventured 



ultimately even P. porrecta will require to be reduced to P. S raciU 

 to accord more than varietal rank to this plant. 



Akn. Kov. Boi. Gard. Calcutta, Vol. III. 



