PIIANEROGAMIA : FLOWERS. 327 



them (the circles and their members opposite). But it by no 

 means is to be assumed that the members of each circle are always 

 of equal number in a flower. The number of members often in- 

 creases up to the stamens, and from thence diminishes : it is rare 

 for the circle of the carpel to contain the greatest number, as in 

 the Malopea and Maluacece. The generality of Monocotyledons with 

 perfect individual flowers* have regular hornomerous circles through 

 the entire flower : in the Dicotyledons this is relatively rarer ; the 

 outermost and innermost circles have usually fewer members. 

 Again, respecting the number of circles which follow one another, 

 no general statement of importance can be given. Seven different 

 forms of foliar organs may possibly exist in the same flower, 

 namely, the epicalyx, calyx, corolla, accessory corolla, the stamens, 

 accessory stamens, and the carpels ; yet I know no flower in which 

 all occur in conjunction. All these foliar organs may be pre- 

 sent in one or more circles, with the exception of the epicalyx, 

 in which I know no example of a double circle. Perianth, calyx, 

 corolla, accessory corolla, and carpels occur in one, or more rarely 

 in two circles. Stamens may be present in one, two, three, or 

 possibly even four circles ; more circles than this are not exhi- 

 bited in the normal condition of the flower. If the number is 

 increased, which seldom happens, except in stamens and carpels, 

 as in the Ranunculacecs and Dryadece, the Magnoliacecs, &c., they 

 stand no longer in circles, but in a spiral. In Monocotyledons 

 with perfect individualized flowers, with the single exception of 

 some ScitaminecB, five trimerous circles of foliar organs of the 

 flower appear to be formed in those where a second circle of petals 

 exists. The greatest multiplicity of forms occurs in the Dicoty- 

 ledons. Lavatera, for example, has an epicalyx, calyx, corolla, 

 stamens, and carpels in five circles, with increasing numbers of 

 members ; those of the calyx and corolla alone are equal. Gnidia 

 virescens has perianth, stamens, accessory stamens, and carpels, but 

 in eight circles, which are throughout composed of two members 

 each. It is by no means necessary that all the parts of a circle of 

 floral foliar organs should be ultimately developed in the same 

 manner ; and many floral structures which have hitherto been ap- 

 parently inexplicable may probably, by keeping this truth in mind, 

 and following out the history of the development, be readily traced 

 back to the original type. 



The duration of the individual parts of the flower is very various ; 

 the axial organs, so far as they support the rudiment of the fruit, 

 or aid in its formation, persist naturally, at least until the ripening 

 of the seed, then fall away with it ; or if it becomes disengaged 

 from them, die away with the remainder of the plant. When 

 axes bear only male organs or flowers, their duration is different ; 

 sometimes they are cast off at a true articulation, sometimes they 

 remain upon the parent plant, and gradually die away. The 



* Perhaps the Grasses and Cyperacece alone excepted, in which only one carpel 



k- 1 c f c 



Y 4 



