330 MORPHOLOGY. 



The PolygalacecB are also referable here, although I have not yet suc- 

 ceeded in finding out the earliest structure of the flower. The earliest 

 condition of the bud which I have yet been able to arrive at, exhibits five 

 free foliar organs in one circle, and within that, apparently placed in 

 another circle, five other leaves, at that time also entirely free ; of these 

 the undermost becomes the pitcher-shaped, fringed petal ; the two upper- 

 most the bilobed petals ; the two remaining side leaves are four-lobed ; 

 and each of these lobes is a perfect anther. The question remains to be 

 solved whether this entire inner circle in truth originated as a penta- 

 merous circle, whose two side members each represent a four-lobed 

 stamen, since there is no other way of reducing the flower to regu- 

 larity ; or whether abortion had already occurred at a very early period. 



c. The third point to which I wish particularly to draw attention is, 

 that all foliar organs of the flower, though they may subsequently 

 unite in growth, first arise entirely free parts ; and if they belong to one 

 circle, they are, at their earliest rudiments, and for some longer or 

 shorter time after, exactly like each other ; so that the coherence of these 

 several members, and their symmetrical development, is a later process. 

 I have been able readily to trace the most irregular flowers up to the 

 condition of bud in reference to this ; as, for instance, the flowers of the 

 LeguminoscB, of the Labiates, the Scrophulariacece, and the species of 

 Aconitum, and these fully established the laws laid down here. One of 

 the most remarkable instances in this respect occurred in the stem of an 

 Orobanche, which had not yet risen above ground, and which I came 

 upon by happy chance, in digging for another plant. This exhibited 

 such surprising regularity in tetra-merous circles distinctly alternating 

 with each other, that nothing could appear more elegant. I have not 

 yet succeeded in following out the perfect history of the development of 

 the very irregular flowers. 



I refer again to the flower of the Grasses (see Plate III. figs. 21 23., 

 with the explanation, of Agrostis alba), and the Carices (see Plate III. 

 figs. 24 26., with the explanation, of Carex lagopodioides.) 



C. OF THE TRUE FOLIAR ORGANS OF THE FLOWER. 



a. Of the floral Envelopes. 



148. As among the floral envelopes are usually reckoned the 

 perianth, the calyx, and the corolla, I also include the epicalyx 

 here, and I circumscribe the term perianth in the narrowest 

 sense, so that under it only those foliar organs fall, which, at 

 least two in number, are applied closely to the flower, and upon 

 one level ; so that all individual foliar organs on the axis of the 

 flower, which only enclose stamens or germens, are to be termed 

 bracts. All these bracts have this in common, that they are merely 

 foliar organs peculiarly modified ; and, consequently, all the pecu- 

 liarities of form which occur in the latter naturally appear in 

 the former also. The few distinctions depend upon what here 

 follows : 



As for all other foliar organs, so also for these all forms are 



