268 MR. J. B. VAIZEY ON THE ANATOMY AND 
due to the absence of an apophysis, makes the relation of the 
tissues in the seta to those in the theca easily intelligible. The 
epidermis is, of course, continuous from one to the other. The 
cortex of the theca consists immediately within the epidermis of 
a layer of close-celled parenchyma with a very few small inter- 
cellular spaces, which together with the epidermis forms the wall 
of the theca; internally to this there is a zone of trabecular 
tissue consisting of long parenchymatous cells, separated by 
very large intercellular spaces. This trabecular tissue connects 
the wall of the theca with the layer of cortical cells forming the 
outer wall of the spore-sac. The outer wall of the spore-sac 
consists of a layer of cortical cells two or three deep, without 
intercellular spaces derived from the exomeristem. Internally 
to the archesporium is the tissue of the columella, which consists 
of an outer layer of cubical parenchymatous cells, concentric with 
an inner axial strand of elongated parenchymatous cells. These 
two forms of tissue abut on one another without any cells of 
intermediate form. 
Tracing the connexion of the tissues of the theca with those 
of the seta, the central strand of the seta is seen to correspond 
with those tissues of the columella which, according to Kienitz- 
Gerloff (12), are formed from the endothecium (my endo- 
meristem *), namely the archesporium and all the tissues internal 
to it. 
The most detailed account which has hitherto appeared of the 
anatomy of the seta of the Musci is the account given by Lorentz 
(17) of Atrichum undulatum; but he only studied transverse 
sections, so that his account is imperfect. He also seems to have 
been mistaken as to the connexion of the central strand with 
the cortex, for he states that it is only held in place by a single 
strand of cells, so that in transverse section it appears to be 
attached to the cortex by a single cell: with fig. 28 compare 
Lorentz (17), Taf. xxviii. 88. 6. This he accounts for by saying 
(although he gives no evidence for the statement) that some of 
the cells of the internal layers of the cortex are destroyed and 
absorbed, and that this result is partly arrived at by the growth 
in thickness of the seta. I have carefully examined sete in all 
stages of development, but detected no sign of any such process, 
although the gradual separation of the cortical cells to form 
the intercellular air-passages was easily seen. It was indeed 
found, when the sections were made with due care, that in the 
* Cf. footnote on p. 277. 
