362 NATURAL SCIENCE. j,,^^. 



centripetally-developed strands of xylem, nor any alternating strands 

 of phloem. There is secondary growth in thickness, but it does not 

 originate in the manner which is so characteristic of roots, but rather, 

 so far as can be seen, in the way which prevails in stems. Hence it 

 follows that anatomically there is scarcely a single character in the 

 axes of Stigmayia Jicoides, as found in this country, which, to a 

 morphologist, is suggestive of a root. 



Anatomy of the Appendages. — Some of those who have most 

 confidently maintained the view that Stigmaria ficoides is a root, have 

 naturally attached much importance to the anatomy of the appen- 

 dages. I say naturally, because it cannot be denied that in some 

 respects that anatomy bears some resemblance to what we find in 

 some rootlets. Williamson has especially drawn attention (6) 

 repeatedly to the fact that the single vascular bundle of the appendages 

 has a likeness to the monarch bundles found in Ophioglossnm, Lycopodinm , 

 Selaginella, and Isoetes. Now, if it could be shown that this bundle is 

 really homologous with a true monarch bundle, it would be a strong 

 point in favour of the root hypothesis, so far as it afjects the appendages. 

 But there are difficulties in the way of doing this. In fact it 

 may be a collateral bundle, and this seems to be the opinion of Count 

 Solms (7). Secondly, its mode of origin differs widely from that of 

 existing monarch bundles, which, according to Van Tieghem, arise in 

 one of three ways (8). In most species of Lycopodinm, by repeated 

 dichotomy of the root trunk, the central vascular cylinder becomes 

 reduced to the diarch type. At the next dichotomy each branch 

 carries off half the xylem and half of each of the phloem strands, and 

 these latter uniting, an arc of phloem is formed on one side of the 

 xylem, and the monarch bundle is constituted. In Lycopodimn inun- 

 datnni, L. selago, some species of Ophioglossnm and Isoetes, the monarch 

 bundle appears to arise from a diarch bundle by suppression of the 

 phloem on one side of the xylem, while in Selaginella it sometimes 

 arises by the suppression of one of the two xylem strands and the 

 union of the two strands of phloem. Similar views, it may be added, 

 appear to be held by Russow (9). Thus, however abnormal these 

 roots may be in some respects, they aye perfectly normal in the 

 penultimate stage, presenting the diarch structure so common in Ferns, 

 Equisetum, and other vascular Cryptogams. Now in Stigmaria 

 ficoides the so-called monarch bundle of the appendages does not arise 

 in any of these ways, nor does it originate from a diarch bundle. As 

 described by Williamson (10), each bundle arises at the inner apex of 

 a vascular wedge of the axis, and is presumably a derivative of one of 

 the primary vascular bundles. In some cases, if not in all, the 

 bundles share in the secondary increase of the axial bundles from 

 which they arise, a peculiarity by no means of common occurrence. 

 Apart from this peculiarity, the mode of origin of these bundles 

 approaches much more closely to that met with in the bundles of 

 leaves than to that found in rootlets which arise from a root. From 



