380 ACCorxT OF a nev/ genus of plants,
may be stated to depend either on a reduced or increased development of the parts enumerated, on differences in the manner of bursting, or on the confluence of two or more anthercC.
Reduced development may consist merely in the approxi- mation of the tliecae, consequent on the narrowing or entire absence of the connecting portion of the filament, which is one of the most common states of anthera ; in their partial confluence, generally at the npper extremity ; their paral- 213] lelism either continuing, which is also not unfrequent ; or accompanied by various degrees of divergence, as in many genera of Labi at a ; in their complete confluence while they remain parallel, as in Epacridea, FoJjjgalecB, and in some genera of AcanfliacecB ; and lastly, in the imperfect production or entire suppression of one of the thecse, as in Westrlngla, Anisouides, and Marantece.
Increased development may in like manner be confined to the dilatation, elongation, or division of the connecting portion of the filament, of which exan] pies occur in many Scitamine^B, OrcJddet^ and Acantliacece ; it may consist in 211] the elongation of the thecae either above or below the connecting filament ; in an increased number of divisions of each theca by longitudinal, transverse, or oblique processes of the receptacle of the pollen, as in several genera of OrchidecB and Laurince ; or in the persistence of part of the cells in which the pollen is formed, as in jEgkevas.
Reduced and increased development of different parts may co-exist in the same organ,' as in the bifid or incum- bent anthera with contiguous thecse ; in the extraordinary dilatation of the connecting portion of the filament, Avhile one of the thecse is abortive or imperfect, as in the greater number of Salvice ; or in the thecse being confluent, while
whether in the simple or compound state, they become always inclosed^ and, before fecundation at least, are completely protected from the direct action of light and of the atmosphere.
In Conifera and Ci/cadea, however, according to the view I am disposed fo take of them {Tuckefs Congo, append. p. 454, cnitc,pA2>^), this is not entirely the case. But tliese two families will perhaps be found to differ from all other pliaino- gamous plants in the more simple structure both of their ovaria and antherae.
�� �
