SPECIAL MORPHOLOGY: AGAM^E. 167 



Gymnosporce. The propagating cell is usually either wholly de- 

 veloped to a new plant, or it remains partially in the external mem- 

 brane and dies off. 2. In the remainder, the propagating cell is 

 developed into tubular utricle, one of whose extremities remains in 

 the outer membrane of the spore, and decays afterwards, whilst at 

 the other end cells are formed, which range themselves into a pecu- 

 liar formation (the germ, proembryo). At one part of the latter, a 

 stem-likes tructure is developed from a denser group of cellular 

 * tissue ; and from this a bud, if we may use the word, arises, from 

 whence the new plant is unfolded. Here, however, this essential 

 difference presents itself, either (a) that this axial structure is only 

 capable of development in an upward direction, as in the rootless 

 Agamcz (the Mosses and Liverworts), or (b) that it is developed both 

 upwards and downwards (Agamce with roots, the remainder, Lin- 

 naeus's Filices^ with the exception of the Rhizocarpece}. All agamic 

 plants present the remarkable peculiarity of having the sporangium 

 absorbed soon after the development of the spores, which always 

 occur in a quaternary combination *, so that the ripe spores remain 

 free in the sporocarps. They differ essentially from the Angiosporce, 

 their distinction from which is further grounded on the equally im- 

 portant resemblance presented between the sporocarps and the 

 anthers of the Phanerogamia. On this account the sporangia are 

 generally termed parent cells. 



The Conferva-like proembryo in the rootless Agamce, and the Ulva-like 

 proembryo of the others, afford a characteristic by which the groups are 

 most intimately connected ; whilst the Liverworts approximate on the one 

 side to the Angiosporce by their germination, and on the other to the 

 Gymnosporce by the formation of their propagating cells. The sporocarps 

 of the Kicciece and Blasice might perhaps present some analogies with the 

 sporocarps of the nucleolar Lichens. At the same time, however, the 

 proembryo, by its different modes of development into a plant, furnishes 

 us with a ground of separation into two distinct groups. In the 

 schematic ambiguity of the word " root," which we find in the place of a 

 clear idea among most botanists, it has escaped them that Mosses and 

 Liverworts have no analogue of a root, and that the bud rising from the 

 Conferva-like tissue of the proembryo, only morphologically limited 

 in an upward direction, developes itself into definite forms, stems and 

 leaves (fronds), while below it is lost in the Conferva-like threads 

 of the proembryo, and is thence incapable of all further development 

 in this direction in a morphologically distinct manner. We might say, 

 with C. F. Wolff, that there is only one punctum vegetationis present 

 here, whilst in the remainder two, an upper and a lower punctum vegeta- 

 tionis, manifest themselves, being divided by the intervening primary cells, 

 which cause the course of development into two opposite directions, 

 upward to the stem and downward to the root. On this account almost 

 all perennial rootless Agamce decay away below in proportion as 

 they develope upwards, whilst the other kinds are developed in both 

 directions and are thus able to increase in mass. A physiological dif- 

 ference corresponds remarkably with the morphological one mentioned. 



Compare H. Mohl, in the Flora of 1833, p. 33. 

 M 4 



