XVIII] ZYGOPTERIDEAE 25 



All the vascular structures hitherto described are primary, but occasion- 

 ally secondary wood of cambial origin has been seen in Ankyropteris 

 corrngata (Scott, Studies, i, p. 303, Fig. 136), and indications of secondary 

 growth have also been observed in Metadepsydropsis by Gordon {I.e. p. 319). 

 But in these plants cambial activity never took such firm hold as it appears 

 to have done in Botrychioxylon, where secondary developments are normally 

 present (Scott, Trans. Linn. Soe. vii, 191 2, p. 373). The plant showed equal 

 dichotomy of its rhizomatous axis, which bore scattered leaves, forked spines 

 or aphlebiae, and adventitious roots. The stele was cylindrical ; it had a 

 central "mixed pith," and protoxylem-groups lay at its outer border. Outside 

 this was a broad zone of radially disposed wood, with cambium at its peri- 

 phery, and a narrow phloem. The leaf-trace departs in the usual way, and 

 the foliar bundle is that of an ordinary Zygopterid, but at its extreme base 

 the trace shows the unusual feature of secondary structure. Such facts 

 indicate that in the Zygopterideae secondary thickening existed, though it 

 was never turned to full account. Their condition in this respect appears 

 to have been similar to that of the Ophioglossaceae, where cambial activity 

 appears in BotrycJiium so similar as to have suggested to Scott the name 

 of Botrychioxylon for this Zygopterid. Though sporadic in each of these 

 families the existence of cambial activity is at least a feature which they 

 hold in common, though it is unusual in Ferns. 



The most prominent characteristic of the Zygopterideae is found in the 

 vascular system serving their peculiarly constructed upright leaves. The 

 origin of the leaf-trace from the stele has been described above in several 

 examples. At first it is monarch, but the single protoxylem soon divides 

 in simple cases into two {Clepsydropsis^, and the two groups then diverge 

 towards the poles of the trace, which has an elliptical outline in transverse 

 section (Fig. 325). 



A central type round which the whole series may be grouped has been recognised by 

 Kidston and Gwynne-Vaughan in their "assumed primitive type" (Fig. 325, e), which was 

 oval in outline, with two protoxylems at the foci of the elliptical solid mass of .xylem. It 

 has been seen in Asterochlaena that these originated from a single protoxylem by fission. 

 This structure is seen with very slight modification in the "Clepsydroid" trace character- 

 istic of Clepsydropsis antiqiia from the Lower Carboniferous (Fig. 328), and it appears in 

 Asteropteris, the earliest recorded of the family (Fig. 324). Kidston and Gwynne- 

 Vaughan show that from their "assumed primitive type" all the Zygopterid traces may 

 be held to have been derived by modifications in two directions, (i) The two ends of the 

 xylem-strand become more or less extended at right angles to the longer axis of transverse 

 section, taking in simple examples a dumb-bell shape, but in more complex forms growing 

 into long arms, more or less curved ; (2) Islands of parenchyma appear in relation to the 

 protoxylems, which also become extended in the same plane as the ends, each protoxylem 

 dividing into two which pass to the anterior and posterior ends of their particular islands. 



In applying their theory it must be borne in mind that there are two types of Zygopterid 

 leaves : first those with two rows of pinnae, one on each side of the rachis (Clepsydroideae, 



