%l- IS STIGMARIA A ROOT OR A RHIZOME? 361 



their rootlets radiated in these soils in all directions." This may be 

 accepted as a correct description of the state of affairs, but, as such, 

 shows how much the so-called rootlets differ from those commonly 

 met with in recent plants. In these, their positive geotropism 

 induces them to grow vertically or obliquely downwards, no matter 

 in what position they arise on the parent axis. In the case of 

 horizontal rhizomes which give off rootlets from all sides — that of the 

 common Iris for example — even the rootlets which spring from the 

 upper surface soon curve over and grow downwards, and do not 

 radiate " in all directions " as do the appendages of Stigmaria. 



Finally, Stigmaria ficoides has two modes of branching, viz., 

 by dichotomy and by the formation of lateral appendages at short 

 intervals and on all sides. This is so unusual a phenomenon in the 

 the case of roots that it is doubtful whether another instance can be 

 cited where it occurs. 



On the whole, then, it appears that the external characters of 

 Stigmaria ficoides differ in several very important particulars from any- 

 thing we are acquainted with in existing roots, and these differences 

 do not seem to have been satisfactorily explained by those who 

 advocate the view that it is a root. On the other hand, they are all 

 in harmony with the hypothesis that it is a rhizome, and, indeed, so 

 far as they go, give unqualified support to it. 



Internal Anatomy. — Though not as completely known as could 

 be wished, the internal anatomy of Stigmaria ficoides has been fairly 

 well worked out by several investigators, especially by Williamson in 

 his Monograph, published by the Palaeontographical Society (4). 

 As there described it possesses a parenchymatous pith, which at an 

 early period becomes fistular, and this is surrounded by a zone of 

 wood, made up of wedge-shaped masses of scalariform tissue, which 

 are separated by medullary rays and were somewhat truncate at the 

 apex. In the youngest axes, according to Williamson, the pith is 

 surrounded by " a thin ring of very small vascular bundles," and the 

 secondary increase in thickness is said to have been effected by means 

 of a " meristem ring equivalent to a cambium zone " (5). It is 

 important to observe, however, that the secondary increase proceeds 

 outwards from the primary bundles, so that although the latter form 

 the apices of the wood wedges, there is no line of demarcation 

 between the primary and the secondary elements. Outside the zone 

 of wood there are traces of a bast layer, which, however, is seldom 

 preserved, and outside this again a cortex consisting of an inner and 

 an outer portion, the former of which usually disappears with the bast 

 to which it is contiguous. 



Now, in this and other accounts given of the anatomy of the axes 

 oi Stigmaria ficoides, nothing is more striking to a botanist than the 

 total absence of all those structural details which are everywhere 

 recognised as morphologically characteristic of roots. There is no 

 central cylinder with its pericycle and endodermis. There are no 



