208
ILLUSTRATIONS OF INDIAN BOTANY.
this membrane bursts interiorly or towards the centre of the flower and sheds the pollen, after which the membranous valve withers and almost disappears, leaving the fleshy body of the anther in its place apparently as large or even larger than when the flower first expanded. I first had an opportunity of observing the anther of R. muc.ronata about 10 years ago, and endeavoured to represent it before and after dehiscence. The attempt was not very successful but such as it was I have recently published it tab. 238 of my Icones. Since then (in 1835 ?) Mr. Griffith published a paper on the Rhyzophoreae in the transactions of the Medical and Physical Society of Calcutta, accompanied with delineations of the anther of this genus, accompanied by some ingenious remarks on its formation. This structure, which is confined to the species of this genus, he very properly considers of itself a sufficient reason for separating them as a genus from the rest of the order even supposing there was no other, which however is not the case, hence this cannot but be considered a good genus. Passing over these peculiarities, which do not extend to all the species of the order, its nearest allies seem to be, Memecyleae, Lethrarieae and Mijrtaceae, and though the relationship does not appear to be very close, we yet find the same plant in one or two instances referred to different orders, thus we find, according to Arnott, that, Petalosoma of D.C. doubtfully referred by him to Mijrtaceae, is not distinct from Carallia: and Olisbia D.C. placed without a doubt in Rhyzophoreae, referred by Lindley and Arnott under the name Guildmgia to Memecyleae.
Geographical Distribution. I mentioned above that plants of this order occupy a wide range, inhabiting salt, marshes, especially near the mouths of great rivers, in tropical countries. The following rather long extract from Dr. Arnott's paper gives a more precise account of their distribution which is my reason for introducing it here.
"Martius in his observations on India plants in the f Algemeine Zeilung' for January 1834, and ' Ann. des Sc. Nat.' n. s. i. p. 250, mentions that there are eleven species of Rhizophoreae in East India according to Wallich's list : there are however only eight noticed by Dr. Wallich, but this number may be made up of the three species of Carallia described in DeCandolle's Prodromus, of Rhizophora conjvgata, gymnorrhiza and cylindrica of Linnaeus ; R. caryo- phylloides, Jack ; R. candel and candelaria of DeCandolle, with R. parviflora and decandra of Roxburgh, mentioned in Wallich's list. While one of these has been unnecessarily spilt down, others, very distinct, appear to be confused under the same names, and Blume's are alto- gether omitted. 1 have now enumerated fifteen species of Rhizophoreae verae from which we must deduct one from America and another from New Holland not found in East India. Of the remaining thirteen, one from Cochin China is scarcely known ; a second extends throughout the Indian ocean, from the Mauritius to the Moluccas or perhaps Timor, and as far north as Arabia Felix and Bengal; a third is found in Malabar, Timor, and New Holland; three appear to be confined to the Eastern Islands; two to Malabar ; one is common to Malabar and the mouth of the Ganges ; two found at the mouth of the Ganges occur probably also further east and in Penang, and the remaining two seem to have been observed (unless different species are alluded to by different authors) in Ceylon, Malabar, Bengal, Java, and Amboyna. To these thirteen have to be added, Roxburgh's two species of Carallia with serrated, and the four spe- cies or varieties I have mentioned with entire leaves : so that although Loureiro's Rhiz. hexa- gona were reduced, the number is about double of that given by Martius. This order is thus concentrated in India or its Islands ; the only known exceptions consist in the two or perhaps three species of Rhizophora and one Carallia that occur further west, and one Ceriops and one Bruguieria that are found in New Holland ; but of these, two are also natives of India."
Properties and Uses. On this head no very precise information exists, the bark of several species is astringent and has been used as a febrifuge. In the Antilles the fruit of Rizoph: mangle is said to be sweet and edible and the juice is fermented to form a light wine. When growing within low water mark the lower branches and stems are sometimes covered with oysters which are esteemed, by those who have an opportunity of procuring them, a great deli- cacy. The wood of R. mucronata is said by Roxburgh to be of a dark redish colour, hard and durable while that of Bruguieria Rheedei is described as being of a yellowish colour, hard and durable, but is chiefly employed for burning and for posts with which the natives construct their houses.
