1839.] Linnean Society. 25 



than the sporule, whilst the root, although more backward in its de- 

 velopment, pierces its sheath before it is as long as the sporule. The 

 sheaths are not distinct, but communicate with each other ; and the 

 only point of connexion between the sheath (there being in fact but 

 one) and the germ is around its base close to the sporule, so that 

 nearly the whole of the germ is inclosed in this sheath. Besides 

 this sheath wliich embraces the upper part of the root, there is an 

 exceedingly delicate expansion, which so closely embraces the ex- 

 tremity of the root like a cap, that it is only by a careful examina- 

 tion that it can be discovered. I am not aware that this has ever had 

 any connexion with the sheath through which the root bursts, but, 

 on the contrary, I believe it to be a distinct formation. After the 

 leaf has grown many times the length of the sporule, or about 2 

 lines long, another leaf grows from the germ close to the first, to 

 which it is in all respects similar, and then a bud begins to be de- 

 veloped from some indefinite part of the germ, but like the leaves 

 and root from within the sheath, which is now frequently much lace- 

 rated. This bud is covered by a peculiar kind of jointed hairs, whose 

 attachments are lateral, at a short distance from their bases, and 

 they contain a few colourless granules. This bud sometimes appears 

 after the first leaf, in which case there is no second primordial leaf 

 formed, and is the rudimentary stem, the first growth from it being 

 a leaf which exhibits, although in a small degree, the first evidence 

 of gyration, and shortly after a root which is furnished with its own 

 sheath. As I have not seen more advanced specimens, I am unable 

 to describe the succeeding steps; but as, up to this j^oint, my obser- 

 vations were made upon several hundred examples, I may safely 

 affirm that the instances were sufficiently numerous for my purpose. 

 All the leaves after the primordial ones, or those which grow direct 

 from the germ, are developed in a similar manner to ferns, and even 

 the running stem partakes in a slight degree of the same gyrate evo- 

 lution. The roots are all formed in sheaths, through the apices of 

 which they ultimately burst ; the sheath continuing to embrace the 

 base of the root, whilst a distinct and far more delicate sheath 

 closely embraces its point. Transverse sections of the stem, root, 

 and leaves show them all to be hollow with the cavity divided longi- 

 tudinally into separate channels. In the stem these longitudinal par- 

 titions are about fifteen or sixteen in number, and in the leaf and 

 root they are about ten or twelve, which in the latter are arranged 

 in pairs. These partitions radiate from a central column of enlarged 

 cells which surround a bundle of minutely dotted ducts that may be 

 unrolled spirally, and the channels between these partitions are fre- 



