382 MORPHOLOGY. 



ApocynacecB. In those five places lying above the five points, where the 

 epidermis is peculiarly modified, now very soon begins, in the Asclepia- 

 dacece and in Apocynum, the secretion of a peculiar, viscine-like, adhesive 

 substance, and very different forms appear ; in Apocynum forming five 

 little depressed, roundish cushions ; in the Asclepiadacece exhibiting a 

 rather longish body forked in the middle, and, when quite perfect, fur- 

 nished with two arms going from the upper or lower end, which in the 

 different species and genera manifest many small, inessential differences. 

 This body is merely adherent to the subjacent, remarkably sharply and 

 distinctly-developed epidermis ; originally green, it gradually becomes 

 yellow, and at last of a dull brown. Its structure is but very indistinctly 

 cellular, perhaps not at all. Its origin is as yet by no means made out, 

 for the investigation is the most difficult I know. From some original 

 observations on Gomphocarpus and Hoy a, I am inclined to conclude that 

 the outermost borders of the wing-like appendages of the anthers are 

 formed rudimentarily very early here, become firmly adherent and sub- 

 sequently torn away from the anthers, so that each body originates from 

 the cohesion of two fragments of two different anthers. So much is cer- 

 tain, that they are never organically united with the angles of the stig- 

 matic body, for the epidermis,, which was distinctly formed before their first 

 appearance, runs on quite uniformly and uninjured, beneath them. 

 They could, at most, in opposition to the above notion, be regarded 

 as semi-organised secreting points. At the time of the dehiscence of 

 the anthers they always lie in such a manner that one, mostly the upper 

 (in the Stapelice^ the lower) end of the pollen mass must at once come 

 into contact with one arm of this body, and there adhere. Another point 

 which I have been equally unable to decide is, whether the five cords 

 of conducting tissue enter the two styles unequally divided (in two and 

 three), or whether they unite just before into a circle, which is then dis- 

 tributed in two equal portions in the styles. 



b. Of the Spermophore (Placenta). 



160. Since the seed-bud corresponds to a bud which arises 

 immediately from a stem, it follows that the spermophore does 

 not always exist as a special organ : when, for instance, the axial 

 organ from which the seed-bud springs is already defined and 

 named on other grounds. In this case, the spermophore is 

 nothing more than the region in which the seed- buds are attached, 

 and, in the simplest case, this may be limited to the basilar 

 surface of one single seed-bud ; as, for example, in Taxus. But 

 those places on an axial structure on which seed-buds are borne 

 may be formed into such projecting processes that they may be 

 well distinguished as peculiar parts of this axial organ* ; or a 

 particular part of the axis, which is in no other way defined as an 

 organ, may be exclusively devoted to the production of seed-buds. 

 In this way we obtain the following varieties : 



a. Spermophore as a simple region of another organ ; 



* Somewhat as in the projecting ridges on the stem of Echinocactus and Melo- 

 eactus. 



