316 CIV. PROTEACEJE. 
thin, rarely coriaceous or hard ; — pe with fleshy cotyledons 
d ort inferior radicle.—Shrubs rarely unde rshrubs or 
even perennialherbs. Leaves ocn or inm ina very few genera 
strictly opposite or verticillate, but often crowded under the inflorescence 
80 as RÀ appear verticillate, usually coriaceous, often vertical with stomata 
on both sides, or in the same genera horiz zontal or narrow and terete, 
entire toothed or variously divided, without paa Flowers axillary 
or terminal, solitary or in racemes or spikes, 0 condensed in 
heads or cones, each flower or pair of flowers aeiae bya [se very 
deciduous in some ra and perhaps sometimes really deficient, the 
ovg — without — 
to Ja out rder are a 
in Australia. Of the first four, constituting the Nucamentace æ, the two principal -— 
Proteez and Personiez, are also in South Africa, but represe in d by different que 
the nine Australian ones being, as well as the four prd Sr the small tribes nd 
an i xception of a single Zeala 
merica, where it is accom by two nearly allied genera, and Sten carpus to ^ 
Caledonia ; B two remaining ey Xu ra the deo which constitute the tribe 
Banksieæ, are en Adonio f in Australia lly de- 
The clavate fusiform or i shaped end of the style in Proteacez is usua J 
scribed e stigma, and where it is more or less constricted it is 
am, but I I have never found an -— eal articulation h 
y be an essential aid in the ctos or dissemination of the pollen, 
b stigm So the real stigma being usnally very small, either ou the poin 
the st cnp or in the centre of the disk, or quite lateral. e 
t : : 
wn 3 
tion of dried specimens, wou d be an interesting study for local botanists i 
eans of examining and bett: the plants hie in their native stations 
In the satiation f the numerous species of this most natural he admi 
genera, and sections, I fond only had to follow, with slight modifications, ina Meissner; 
arrangement proposed by Brown and further developed by Zndlicher and ust n0b 
on wi 
and the infloresceno ce, neither of them again strictly constant, we dors 
two lar oe groups which are both natural and to a ventaced? alone 
geographical. For although both are a aura in Au vitm rie Nucamen 
in Africa, and the Fol ularss alone in Asia and Am 
thin each br 
Tut 
SusonbER 1. Nucamentaceæ.— Fruit an indehiscent nut or drupe. 
itary i aet 
Trine 1. Proteese.— Anthers all perfect or very rarely the upper rant 12 
2 parallel cells "ntes to the connectivum, inserted at the base i “a wee 
lamine of the perianth. Ovule 1 or rarely 2. Dipole terminal. 
1 
3 
