PHANEKOGAMIA : FLOWERS. 347 



(as in Canna and Salvia), or the excessive development of the 

 connective, either altogether, so that the thecae are more or less 

 removed from each other (as in Lacistema and Sfalvia), or at its base 

 (as in Stachys sylvatica), at its upper part (as in Berberis and 

 Humirium), or on the under surface, so that the thecae have the 

 appearance of lying upon the upper surface (anther & anticce, introrsce), 

 or on the superior surface, so that they have the appearance of 

 lying upon the under surface (anther ce posticce, extrorsce) ; or several 

 of these modes of excessive and disproportionate development may 

 occur together. Further, we find very irregular development of 

 the connective, and of the thecae dependent upon it ; for instance, 

 in the serpentine form (in many of the CucurbitacecR\ the thecae 

 rolled inwards like the Corinthian volute, in Phitydrum, &c., all 

 originally starting from the same structure, and only gradually 

 assuming these forms. Besides the forms already mentioned, other 

 irregular growths of the connective are presented, especially upon 

 the under surface, where they assume strange forms of spurs, hoods, 

 &c., as in Asclepias, &c. : all these varieties are generally thrown 

 into a heap with things of the most different nature, under the 

 name of corona. On the thecae also occur, sometimes in the upper 

 and sometimes in the under part, processes and appendages of very 

 various kinds (as in the Ericaceae). The connective expands in a 

 very peculiar manner on the back of the anther, beyond it, but pro- 

 jecting especially above and below, and encasing it as a coat, as 

 in many of the Apocynacece. 



Many varieties occur in the mode of union of the anther with 

 the filament ; sometimes no filament is formed. When it exists, it 

 sometimes again merges into the connective, which appears some- 

 what broader than it, and the base of which is not surpassed by 

 the base of the loculi; or the latter grow further out beyond it, 

 so that the filament seems to be inserted between the thecae corre- 

 sponding to the folium cordatum or sagittatum ; or the thecae are 

 developed out in a similar manner beyond the base of the connec- 

 tive, and become blended in the course of their formation, corre- 

 sponding to the folium peltatum : this is termed anthera dor so affixa, 

 or, as it is usually unsteady upon the slender filament, anthera versa- 

 tilis. Again, the filament corresponding to the petiole offers a mul- 

 titude of varieties ; sometimes it is linear or flat (band-like), or may 

 be developed thick and fleshy, exhibiting all kinds of appendages 

 both upon its upper and under side, and especially such as corre- 

 spond to appendages on the leaves : thus, for example, like the 

 ligule (in Cuscuta and some species of Zygophyllum) ; and in par- 

 ticular appendages corresponding to the stipules (as in many 

 Lauracece, Amarantacece, and species of Allium, Alyssum, and 

 Campanula)., which is the more remarkable, since no other foliar 

 organ of the flower exhibits anything similar. 



A true articulation in the continuity of the same stamen I have 

 nowhere found ; in the Composites, it is certain that no such thing 



