PHANEROGAMIA : FLOWERS. 



333 



It is indeed striking that (if my ignorance is not to blame) we have 

 no example of a compound leaf in the floral envelopes, not even in such 

 a way that a single articulation exists, as in Citrus, in the continuity of 

 the same leaf. On the other hand, forms which are relatively rare in 

 the stem-leaves occur here very frequently ; for instance, the hollow ones. 



A different and here especially attractive interest attaches to the dis- 

 tinct separation from the really independent foliar organs, of parts, 

 which, though they appear in such striking and odd forms, are but ap- 

 pendages, and therefore portions of other organs. The terms corona, 

 accessory corolla, nectary, &c., have hitherto been applied by botanists 

 to the most different parts, with a truly inexcusable superficiality ; and 

 how little is dreamt in general of a scientific treatment of the matter, is 

 seen in an expression of Link (I. c. ii. 145.), where he says of the ac- 

 cessory corolla of the Passifloreacce, " We have no double forms by 

 means of which to determine the true nature of this part." A simple 

 examination of the young buds suffices to show that the various filaments 

 and other appendages become developed out from already perfectly- 

 formed foliar organs, consequently the former cannot be independent 

 organs (see PL IV., with the explanation). Double forms, through which 

 Link, for instance, decides upon the nature of the corona in Narcissus 

 (196, h.\ give no results whatever, since the double form is always 



196 



196 Narcissus Icetus. Flower, a, Peduncle ; b, spathe ; c, bud ; d, pedicel ; e, inferior 

 germen ; f, tube of the perianth ; g, limb of the perianth, appearing like six free 

 leaves; h, corona, formed of six coherent ligules of the perianthial leaves. 



