177 



stigmas. Fig. 6, Capsule. Fig. 7, Section of do. Fig* 

 8, Valve of the capsule, showing the attachment of the 

 seeds, and the central axis partly free. Fig. 9, Seed. 

 Fig. 10, Section of the seed to show the embryo: — all 

 more or less magnified. 



4. Macraea crenata; foliis profunde crenatis marginibus 

 reflexis subtus niveo-tomentosis, floribus subcorymbosis, 

 (petalis albis.) 



Hab. Cum priore. D. Cruickshanks. In Andibus Chilensibus 

 prope " La Guardia" in valle fluminis Aconcagua: alt. 

 5000 ped. Gillies. 



Until after the publication of Mr. Lindley's Memoir, I was 

 only acquainted with the present and foregoing species : the 

 former having the leaves quite entire, the margins plane, the 

 flowers of a beautiful rose-colour; the latter having deeply 

 crenate leaves, with the margins revolute, the petals white. 

 But I am fearful that these characters may not be constant, 

 and I am equally in doubt with regard to the characters 

 of the rest of the genus. The glands, extremely minute in 

 themselves, are very variable on different individuals and 

 even on different leaves of the same specimen. The quantity 

 of pubescence on the under side of the leaf is scarcely to be 

 relied on, and thus the nerves beneath become more or less 

 conspicuous. The length of the peduncles and colour of the 

 flowers I fear cannot much be depended upon. All might, 

 probably, without violation to nature, be referred to varieties 

 of the same species. 



The genus is deservedly named in compliment to Mr. 

 James Macrae, who made valuable collections for the Horti- 

 cultural Society both in the Brazils, at the Sandwich Islands 

 and in Chile ; and who is now entrusted with the charge of 

 the Botanic Garden at Ceylon. 



N 2 



