24 Saxelby . — The Origin of the Roots in Lycopodium Selago. 
sections (see Fig. 2) this row of intermediate cells is seen to be directly 
outside the first xylem elements (v) which are differentiated in the stem 
plerome, and would therefore correspond in position to the pericambium of 
Bruchmann. 
In central transverse sections of a slightly older root origin, the inner 
and radial walls of the cells of the innermost layer of the stem cortex 
stained slightly with fuchsin-ammonia, though not with cotton red, iodine 
green, or aniline-water safranin (see Fig. 3). At this stage the endodermis 
appears continuous with the outermost layer of the root origin, but this 
appearance would be produced if the root formed in an inner layer had 
begun to grow out, and the pressure or secretions from the root from one 
side of the stem plerome had in some way hindered thickening of the 
endodermal cell-walls in that region. The cortical cells immediately 
outside the root origin are preserved in places, and show no thickening of 
cell-walls. Those around the apex of the root are not well preserved in 
these sections, owing to the absorptive nature of the outermost layers of the 
root. Fig. 4 shows a transverse section near the apex of the stem of an 
older plant. The stem is cut above the last point of dichotomy, through 
the part bearing sporangia. The section shows the root base surrounded 
by mucilage when only slightly projecting from the circumference of the 
stem stele, so that even if the stem endodermis originally surrounded the 
root base, as Van Tieghem figures for L. inundatum , in L. Selago it would 
be disorganized by the root-cap before the root grew out as far as that 
figured by Van Tieghem. This agrees with Bruchmann’s statement for 
L. inundatum , that the endodermis of the stem is not seen covering the 
root base ; Bruchmann, however, as stated above, found the endodermal 
cells — with darkly staining walls — passing up to the centre of the sides of 
the root base in a vertical section. He uses as an argument in favour 
of the cortical origin of the root the fact that the cortex and the epidermis 
of the root pass laterally into the cortex of the stem, not reaching the 
vascular cylinder of the latter, and also only the few outer layers of the stem 
cortex are destroyed by the outward passage of the root. 1 These points 
he says are noticeable, not only in L. inundatum , but also in L . clavatum , 
L. annotinum , L . complanatum , and L. alpinum . 2 That this is not the 
case in L. Selago is seen from Fig. 5, where the periblem of the root is 
continuous with the plerome of the stem, and also from Fig. 4, where the 
stem cortex is destroyed round the root up to the point where the latter 
begins to curve out from the stem plerome. 
From the study of such sections it seems clear that the outermost 
layer of the young root is derived from an intermediate layer which, as 
stated above, corresponds probably to the endodermis. The layer of the 
plerome of the stem lying within this gives rise to one meristematic layer 
1 Bruchmann ( b ), loc cit., p. 77. 2 Ibid., p. 80. 
