202 ME. G. BENTHAM ON EUPHORBIACE^I. 



correct, if grouped into about six principal ones, very fairly 

 characterized as to the great majority of species, but all more or 

 less confluent through intermediate ones. These are: — 1. Aniso- 

 phyllum, usually prostrate, much-branched herbs from most parts 

 of the globe, with all the leaves opposite and stipulate, and 

 the glands of the involucre almost always furnished with petal- 

 like appendages; 2. Adenopetalum, American species with the 

 petal-like appendages of Anisopliyllum, but with a varying habit, 

 the stem-leaves below the branches alternate except in two or 

 three species, the stipules usually replaced by small glands or 

 deficient ; and four sections without the petal-like appendages : 

 3. Poinsettia, American species distinguished by the coloured 

 bracts below the involucres, and by a greater or lesser obliquity 

 in the involucres, showing an approach to Pedilanthus ; 4. Ere- 

 mophyton, distinguished from Tithymalas by the usually dicho- 

 tomous not umbellate inflorescence ; 5. JZuphorbium, by the 

 thick and succulent or rarely slender stem and branches, almost 

 or quite without leaves ; and 6. Tithymahis, with all or most of the 

 primary branches forming a terminal umbel, the leaves of the 

 main stem always alternate, without stipules. Of these six sec- 

 tions, Anisophyllum and Tithymalus, and sometimes even Poin- 

 settia, are by some botanists considered distinct genera, but with 

 characters derived solely from habit, and not definite enough 

 for generic separation. All attempts to take into account minor 

 modifications of inflorescence, the form of the glands of the 

 involucre, the degree of union of the styles or their branches, 

 &c, have only resulted in isolating single species or collecting 

 several together which bear evidently but little general affinity 



to each other. Many of Boissier's sections therefore, however 

 useful in practice, can scarcely take rank but as artificial divi- 

 sions of the above six. 



Of the other genera composing the tribe, Pedilanthus is an 



w llarity of the involucre, 

 giving it so peculiar an aspect that the genus has been univer- 

 sally adopted, although a few of its fifteen supposed species come 

 very near to such of the Poinsettia section as have a decidedly 

 oblique involucre. On the other hand, Synadenium, Anthostema, 

 and Calycopeplus are three genera of two or three species each, two 

 of them African, the third Australian, which afford some assist- 



ance 



measure connecting the tribe with some genera of Crotone© 



