80 flora indica. [Magnoliacece. 



fusco-sericei. Flores flavi vel aurantiaci, suaveolentes. Sepala et petala 15-20, 

 1^-2-pollicaria, exteriora oblonga cuneata aeutiuscula, interiora multo angustiora 

 lineari -oblonga acuta. Carpella in spicam 3-4-pollicarem congesta, subsessilia. 



The Champaca of Rheede and Rumphius, adopted by Linnseus and all following 

 authors, and universally recognized, notwithstanding the brevity of the original de- 

 scription of Linnseus, is only known as a cultivated tree. Indigenous trees, how- 

 ever, have been described by Walli 7 , _ _ ., „ 



gated and illustrated with so much success by these botanists, which very closely re- 

 semble the cultivated tree, differing only, it appears to us, in such characters as are 

 chiefly affected by cultivation. In all, the flowers have the same structure, and the 

 leaves the same shape and degree of variation. The pubescence is much more con- 

 siderable in the wild plants described by Wallich and Blume than in the cultivated 

 Champaca ; and though Wight describes his M. Eheedei as glabrous, his specimens 

 were in fruit only, whilst flowering ones in our possession from the same localities 

 are quite as pubescent as M. Doltsopa from Nipal. Blume has recognized the affi- 

 nity of his M. pubinervia with 3L Doltsopa, Wall., while at the same time he fully 

 admits its close affinity to the cultivated Champaca of Java, by relying on characters 

 for its separation which are of very subordinate importance. For these reasons, after 

 a very careful examination of all the specimens to which we have access, we have 

 convinced ourselves that all the synonyms adduced above are referable to one species. 

 $L rufinervis of De Candolle (not of Blume) is a cultivated Mauritius plant ; a spe- 

 cimen in Herb. Hook., which agrees exactly with the description, is a luxuriant 

 young shoot, with copious brown silky pubescence, but with leaves like those of M. 

 Champaca. De Candolle's specimens were also without flowers, and probably of the 

 same ao-e. It is more difficult to decide whether the Doltsopa of De Candolle and 

 Don be^the same as that of Wallich, as the descriptions given by the two former 

 authors of M. Doltsopa and 3f. Kisopa are very brief, and so obscure that they can- 

 not be referred with certainty to either species, but partake of the characters of both. 

 In these circumstances, as the original specimens are not available, having been dis- 

 persed with the Larabertian Herbarium, we have thought it advisable to follow Wal- 

 lich in the use of the names Doltsopa and Kisopa, considering him m fact as the 

 authority for the species, which he was the first to characterize in a satisfactory 



manner. 



3. M. excelsa (Blume, Fl. Java3 Magn. 9, in adnot.); foliis ob- 

 loner is vel oblongo-lanceolatis acutis superne glabris subtus fusco-sen- 



ceis setate . 



Wall. Cat. 6494 ! Wight, III. i. 14. Magnolia excelsa, Wall 



Nap. 5. t. 2. 



Hab 



petalis 



ib In Himalaya oriental! temperata, alt. 6-8000 ped. : Nipal, 

 Wall./ Sikkim ! Bhotan, Griffith/ et in Khasia, alt. 5000 ped., Simons/ 

 (Fl. vere.) (v. v.) 



Arbor excelsa, ramosa. Hamuli rugosi, grisei, punctis callosis conspersi. Gemma 

 fusco-pubescentes. Folia coriacea, acuta vel acuminata, superne mtida, subtu^ U£ 



ultra medium petiolum externa. Alabastn subsessues, aense lusco-^xucut^., » 

 JomJes e^lfra, spathis pluribus deciduis involuti. Sep***, obovata, cona £ 

 Petala 9-10, anguste obovata, interiora sensim angustiora et brev,ora ■<**«* 

 secus rhachin 4-8-poUicarem lase disposita, subsessiha, |-polhcana. Setmna 1-4. 



4 M. lanuginosa (Wall.! Tent. M. Nap. 8. t. 5) ; foliis oblongis 

 vel lanceolatis superne nitidis glabris subtus dense ^reo^tomentosis, 

 floribus albis, sepalis petaliscum U.-Wall. Cat. 6493 ! Wtght, III. I 

 14. M. velutina, DC. Prod. i. 79. 



