886 THE STELLATE TRIBE. 



whorl are real leaves with buds in their axils, the others, although 

 precisely similar, are in fact stipules), rarely 2 only, the buds and 

 branches always opposite. Flowers small, in terminal or rarely 

 axillary panicles or heads. Calyx combined with the ovary, 

 either entirely so or rarely with a border of 4 or 5 teeth. Co- 

 rolla monopetalous, with 4 or 5 spreading lobes. Stamens as 

 many, inserted in the tube. Ovary inferior. Style 2-cleft at the 

 top, with a capitate stigma to each branch. Fruit in dehiscent, 

 small, dry or rarely succulent, usually separating into 2 seed-like 

 carpels with one seed in each. Albumen horny, with a small 

 embryo. 



The Stellates are widely diffused over the globe, especially in tem- 

 perate regions : in the tropics they are more rare, except in moun- 

 tainous regions. They form a considerable and very natural tribe 

 in the great Natural Order of Rubiacece, otherwise unrepresented in 

 Britain or even in Europe. It is one of the most extensive ones 

 within the tropics, distinguished by opposite leaves, interpetiolar sti- 

 pules, an adherent calyx, and a monopetalous corolla, and includes 

 trees and shrubs as well as herbs. Many are cultivated in our stoves, 

 greenhouses, or flower-beds, including the genera Cqffea, Gardenia, 

 Luculia, JPentas, Manettia, Bouvardia, etc. 



Corolla rotate, the tube very short or indistinct. 



Fruit fleshy. Corolla usually 5-lobed 1. Madder. 



Fruit dry. Corolla usually 4-lobed 2. Galium. 



Corolla with a distinct tube, as long as or longer than the 

 lobes. 

 Fruit crowned by the 4 teeth of the calyx. Flowers in 



heads, surrounded by an involucre 4. Sherardia. 



Calyx not distinct. Flowers in panicles ...... 3. Asperule. 



I. MADDER. KUBIA. 



A genus only distinguished from Galium by the rather larger suc- 

 culent fruit. The European species have also larger leaves, of a firmer, 

 more shining texture, and the flowers have often four instead of five 

 parts, but these differences scarcely hold good in the South American 

 species. 



The species are not numerous, and might rather be considered as 

 forming one or two sections of Galium, the South American species 

 being intermediate between the two genera as now established. 



