IX] 



THE LEAF-TRACE 



163 



In the most primitive types such as the Botryopterideae and early Os- 

 mundaceae the leaf-trace, however widened it may be above, is contracted 

 downwards as in Thamnopteris to something like the cylindrical form, and 

 inserted without causing any depression or gap upon the protostele of the 

 axis. But in more advanced types the full width of the leaf-trace may be 

 continued downwards to the actual point of insertion, while the stele of the 

 axis, dilated sometimes with medullation, or more commonly having soleno- 

 stelic or dictyostelic structure, is by such means raised to a size suitable for 

 receiving it. The dilatation of the stele is closely related to the width of the 

 leaf-trace, and both have followed from the demands of an enlarging lamina ; 

 the effect of this has thus spread downwards in the course of Descent. Where 

 solenostely has been arrived at, as described in Chapter Vlll, the leaf-trace 

 comes off bodily as a sector of the vascular ring, leaving a wide leaf-gap as 



Fig. 156. i, ii, iii, acropetal succession of sections of a "runner" of I opho^o) la quadriphiuata 

 (Gmel.), showing the origin of a leaf-trace: i, the solenostele is still complete', but the leaf- 

 trace is indicated by a projecting curve; in ii, the leaf-trace has separated at one margm, thus 

 opening the leaf-gap ; in iii, the leaf-trace has passed off, and the gap is again closed. Slightly 

 enlarged, sclerenchyma dotted, vascular tissue black. 



evidence of the disturbance which it causes, and incidentally allowing com- 

 munication between cortex and pith (Fig. 156). The trace itself may widen 

 out upwards into a broad, more or less gutter-shaped sheet, which in many 

 relatively primitive Ferns is quite continuous. It is completely enclosed in 

 all Leptosporangiate Ferns by endodermis, and in proportion as its margins 

 are near together it approaches structurally to the state seen in a solenostelic 

 stem. Occasionally even the open channel of the gutter-shaped trace is 

 closed, as may be seen in the large leaves of Pteris {Litobrockia) podophylla 

 (Fig. 147, Chapter VIli), so that a vascular ring is completed also in the petiole. 

 On the other hand, in Fferns of relatively advanced type the leaf-trace 

 may be divided into two or more separate vascular tracts ; but these are still 

 so arranged as to represent more or less clearly the underlying horse-shoe 

 curve. In simple cases the trace is divided in a median plane into two equal 

 portions, as in Athyrium Filix-foeuiina, Trismeria, or Aspleniiun adiantJim- 



