12. R. DISCOLOR. 103 



this difference is not so apparent: then the stems of R. 

 discolor, which naturally only rise into a low arch at their 

 base, finding su})port, continue to raise themselves higher 

 and higher until rather late in the season of growth, when 

 they make a vigorous attempt to reach tlie ground by 

 sending off one or more slender quick-growing shoots from 

 their extremity. When in this supported state they may 

 easily, but erroneously, be supposed to possess the same 

 tendency to rise as exists in those of E. thyrsoideus; but 

 even in these cases the difference in habit is apparent to a 

 careful observer. The stems of R. discolor seem to lie along 

 the top of the hedge or bush; those of R. thyrsoideus to 

 stand of themselves. The stem of R. thyrsoideus is usually 

 much the most sulcate ; that of R. discolor being often only 

 angular. The panicles present to the eye considerable 

 difference, although it is nearly impossible to describe in 

 what it consists. The edges of the leaflets of this plant have 

 a tendency to turn downwards so as to render the leaflet 

 convex, and often do so in a very marked manner; those of 

 R. thyrsoideus, if not flat, turn their edges upwards. The 

 colour and consistence of the felt on the under side of the 

 leaves is very different in the two plants. The petals are 

 different in colour and shape. The styles purple in one are 

 green in the other. 



The R. abruptus (Liudl, ed. 1) is combined without 

 remark with his R. fruticosus in edition 2. In this he is 

 doubtless correct. The specimen named it. ahruptus by 

 Lindley, from the Hort. Soc. garden and one from the same 

 garden (in the Herb. Borr.), which was called R. cunei- 

 folius (Lindl.) in 1829, is clearly the plant named R. ab- 

 ruptus by him in the Synopsis. It is remarkable that a few 

 years afterwards he should have given the name of R. 

 rhamnifolius to specimens of the same plant sent to him by 

 Leisfhton. 



