ON ERICA DABOECIA. 263 



a capsular fruit by pronouncing it of four locula- 

 ments, many seeded and opening by four valves; 

 but the observer of natural affinities, not satisfied 

 with this character, inquires into the form and 

 structure of the valves, their mode of union, the 

 situation and insertion of the seeds ; and by com- 

 bining these several observations, derives more 

 accurate results, and is thus often enabled to fix 

 immutably the place which a plant ought to occupy 

 in the natural order. 



The plant in question is a very striking instance 

 of what has been advanced. In the present scries 

 of Dicotyledones with monopetalous and perygy- 

 nous corolla (or corolla attached to the calyx), we 

 find two families, which formerly were united in 

 one, and adulterated befides by the introduction of 

 other genera in the Or dines naturales ofLiNNwEUS,* 

 in the family of Vaccinia of Ad an son, and also in 

 the series of orders in the garden of Trianon. 

 These two families, which of late have been 

 determined by characters no ways ambiguous, but 

 sufficiently distinct to keep them for ever separate, 



* Ordo xviii. 



