ILLUSTRATIONS OF INDIAN BOTANY. O7 
‘¢ Tribe 4th.—Savuravusrm.—Calyx deeply 5-parted, furnished with 2-3 bracteas. Petals 
alternating with the sepals, more or less connected together at the base. Stamens numerous, 
adhering to the base of the corolla. Anthers incumbent, inserted by the back, not adnate. 
Styles 3-5, distinct from the ovary. Seeds wingless. Albumen fleshy.” 
wo genera are referred to this tribe Saurauja and Apatelia the former, nearly, altogether 
of Asiatic, the latter of American origin. Of Saurauja Wallich enumerates 4-continental 
species, and one from Penang: Blume has no fewer than 9 from Java alone. None have yet 
been found in the Peninsula. 
“ Tribe 5th.—Lartacem.—Calyx bractless of 3-5 sepals, sometimes 5-parted. Petals 
usually 5, distinct. Stamens numerous, free, or connected at the base. Anthers adnate or 
versatile. Styles equal in number to the cells of the ovary, joined in 1, crowned by man 
stigmas. Truit 3-5 celled. Albumen fleshy or wanting. Seeds compressed or winged, 
rarely cochleate.” 
To this tribe, the largest of the order, only one Indian genus is referred, namely Cochlos- 
permum, which was long supposed, on account of the woolly covering of its seeds to be a Bombaz, 
whence, the only species found in this part of India had received the name of B. Gossypium, 
under which name, it is well described in Roxburghs Flora Indica. 
« Tribe 6th.—Gorponim—Sepals 5, free, oF joined together at the base. Petals usually 
connected at the base. Stamens numerous, monadelphous at the base. Anthers ovate, oscilla- 
“ Tribe 7th.—Camettia.—Sepals 5-9. Petals 5-7-9, usually cohering at the base. Sta- 
mens numerous, monadelphous, or polyadelphous at the base. Anthers versatile, 2-celled. 
Styles 3-5, connected at the base. Fruit 3-celled, 3 valved, few seeded ; valves with a desse- 
piment in the middle (loculicidal dehiscence) or bent in at the margins ‘so much as to form 
dessepiments (septicidal dehiscence.) Albumen wanting.” : : 
& This tribe forms in DeCandolle’s Prodromus, a distinct order, including two genera 
Camellia and Thea. These genera have hitherto been kept distinct on account of a supposed, 
difference in'the dehiscence of their fruit : the one Camellia being said to have a loculicidal 
dehiscence, that is, the valves splitting along the back, midway between the septa or partitions 
which then form a projection in the middle of the valves ; the other Thea, a septicidal dehiscence, 
that is, the partitions themselves splitting. These septa being composed of two inflexed coher- 
ing lamine, often separate, at the period of maturity, and form two thin marginal partitions 
between the cells of the capsule. oe 
The researches of Mr. Griffith with species of both genera in all states of fructifi cation 
before him, have convinced him, that there is not ths least foundation for the distinction whict 
has here been attempted to be established : the dehiscence of both being valvate ani the fruit 
a 3-celled capsule, in short that Camellia and Thea form but one genus. Cambessedes, who, 
recently wrote a monograph of the order, and, as above remarked, came to the conclusion that 
