70 
JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY.. 
the stomata, or transpiration pores, are on the upper instead of on the 
under surface. In Hippuris, which grows partly above water, the stem 
has a cuticle, and, though its vascular bundles are " cauline," ?;.c. are 
not at first united to those of the leaves, they do contain some vessels 
with lignified walls ; but whilst the short thick leaves above water have 
a cuticle, stomata on both surfaces, palisade tissue, and several vascular 
bundles, the thinner and paler submerged ones have neither stomata nor 
cuticle, much less mesophyll, with no distinct palisade tissue, and with 
but one vascular bundle. The completely submerged Elodea, on the- 
oiher hand, has no stomata, no vessels with lignified Avails, and a 
vascular bundle in the leaf consisting of but little more than two or 
three bast-fibres.* The typical hydrophyte, in fact, being without 
transpiration, is destitute of water-conducting tissue, of stomata, trachea^-,, 
or tracheids. 
Air, on ihe other hand — or perhaps rather various gases, including 
those that constitute air — is specially secreted, not only by true hydro- 
phytes or aquatics, bat also by marsh plants. This secretion takes place 
in numerous and extensive intercellular spaces without openings to the- 
exterior, sometimes forming a tissue that may be termed a'crencliyma.. 
More than one function would seem to be served by these air-cavities : 
(i.) they serve to store up gaseous food, which is gradually taken into the- 
protoplasm of the actively living cells in its anabolic processes by dif- 
fusion, as I have already described (p. 68) ; (ii.) they serve as floats ; 
or (iii.) they serve to introduce oxygen for respiration to parts submerged 
in water or mud. The well-known air-bladders in the Bladder- wracks 
[Fucus vesiculosus, &c.) and the inflated hollow leaf-stalks in the rosettes 
of leaves of Trapa natans and Eicliornia {Pontcdcria) speciosa would 
seem to serve only as floats bringing the plants to the surface, and so- 
facilitating their assimilation. It was formerly erroneously supposed 
that the traps of the Bladderworts {Utricularia) were also floats. It 
may well be supposed that the very large symmetrically disposed air- 
cavities in the mesophyll of the floating leaves of Salvinia and those irt 
Alisma and between the stellate cells of Juncus, together with the- 
extensive tabulation of both petioles and peduncles in Water-lilieS; and 
the spaces, partitioned off at intervals by diaphragms one cell thick, in the 
stems of Hippuris and Myriopliyllum, serve other uses beyond merely- 
acting as floats, secreting the gases utilised in metabolism. The most 
interesting and least understood tissue, however, in this connection is 
that secondary tissue, originating in a phellogen near the surface of roots 
or stems, to which the name " aerenchyma " has been specially applied. 
This is a loose corky tissue containing large intercellular spaces, and is 
more characteristic of marsh plants or semi-aquatics than of true hydro- 
phytes. In the deciduous Cypress of the Mississippi {Taxodium disticJmm, 
Rich.), for instance, hollow ascending branches known as " knees " rise 
from the roots above the surface of the swamp, and in various genera 
known as " Mangroves " but belonging to different Natural Ordeis, in- 
Bruguiera in Bhizophoracece, Sormeratia in Blattiacem, Sbiid Avicennia 
in VerhenacecB, we have similar roots ascending to the air and developing 
* F. 0. Bower, Practical Botany, 3rd edition (1891), pp. 110-12, 138-9, 171-2, 
180. 
