50 
Proceedings of the Royal Society of Edinburgh. [Sess. 
V . — Observations on the Interruption of the Endodermis in a 
Secondarily Thickened Root of Dracaena fruticosa, Koch. 
By Annette G. Mann, B.Sc. Communicated by Professor F. O. 
Bower, F.B.S. 
(MS. received October 25, 1920. Read December 6, 1920.) 
It is generally accepted that the position and development of any one 
tissue in a plant is to be studied from the point of view of function : 
nevertheless the physiological need for it may alter as the plant develops. 
The endodermis is an illustration of this. Its primary function (1) is to 
restrict the passage of water and soluble substances to certain definite 
channels : it follows that its best development is in those parts nearest to 
the source of supply, e.g. in roots and young stems. The cell-walls are at 
first relatively thin, with the characteristic suberised strip on the radial 
and transverse walls. Its cells are in uninterrupted contact one with 
another, there being no intercellular spaces. This leads incidentally to its 
being also a barrier to gaseous interchange. The ventilating system of the 
cortex is limited by it internally, and cut off from such intercellular spaces 
as may lie within. 
A secondary function which is mechanical is often assumed by it, as 
shown by Schwendener (2). In older roots the cell- walls, particularly the 
radial and inner tangential walls, become indurated and resistant. When 
in this latter condition the endodermis forms a rigid structure enclosing 
the stele, and forming a very effective barrier between stele and cortex. 
It has been shown by Professor Bower (3) that size is a factor which 
has to be considered in this connection. Since the surface of a cylinder 
varies as the square, and the bulk as the cube of the linear dimensions, the 
larger the cylinder is, the greater will be the difficulty of adequate transit 
through the barrier, and a limit may be expected when the barrier must 
be extended or be interrupted in some way or another, otherwise the check 
on interchange through the endodermis would become a serious risk. In 
roots of Dicotyledons the cortex and endodermis peel off as secondary 
growth occurs. In thickening stems it disappears, though the steps by 
which this is carried out have never been accurately followed. But the 
roots of some Monocotyledons, in particular Draccena, offer a favourable 
opportunity for tracing the disruption. Draccena, which has a well- 
marked and indurated endodermis, has also, in some of its largest roots, a 
