CH.xvii] ORGANOGRAPHIC COMPARISONS 337 



which is held to be a derivative state in Descent, as it is actually seen to be 

 in the individual life of Helminthostachys , Datiaea, or Christensettia. 



Terminal branching of the shoot is not uncommon (Frontispiece). As 

 a rule it is dicJiotonioiis in primitive types: but the shanks are frequently 

 developed unequally, so that a sympodial shoot-system results. If the 

 inequality is great it may lead to the weaker shank appearing as an appendage, 

 either in the axil of the nearest leaf (Ophioglossaceae, Hymenophyllaceae, 

 Ankyropteris), or in various other positions, most frequently as an abaxial 

 bud on the leaf-base {Lopkosoria, Metaxya, etc.). Various grades may be 

 seen between equal dichotomy and monopodial branching, with axillary 

 branching as a special case. But all may be referred to dichotomy, which 

 appears to have been the primitive mode of branching ; that is, if branching 

 occurred at all in the most primitive forms. In many no such branching is 

 seen. 



The shoot is fixed in the soil by numerous roots, of which all the later 

 are clearly adventitious. The nature and origin of the first root may be 

 open to question. That origin is not constant in time or place for all Ferns. 

 The root may actually be absent in Salvinia. Its emergence where there is 

 a suspensor present is lateral. These facts indicate that it is accessory to 

 the shoot, as are all those which follow accessory to the shoots which bear 

 them. Accordingly it is possible to contemplate a primitive Fern-sporophyte 

 as a simple, upright, radial, rootless shoot, either unbranched or showing 

 dichotomy. 



The leaves of living Ferns are bifacial, and most of them have two rows 

 of pinnae, one on either side of the rachis : but in certain Zygopterids the 

 higher appendages were in alternating pairs, forming four longitudinal rows, 

 and giving a radial construction. Thus the bifacial leaf was not universal. 

 In Chapter IV it is shown how the whole leaf of Ferns is traceable in origin 

 to an elongated rachis with dichotomoiis distal region, and frequently with 

 stipular growths at the base. All manner of steps of sympodial development 

 can be traced in the distal branch-system, so as to establish a dichopodiiim, 

 which being continuous with the rachis constitutes \.h& phyllopodinm. Upon 

 this in advanced cases the earlier pinnae may arise monopodially, the later 

 being dichotomous. In many primitive types the segments are all separate, 

 each containing a single vein. This is held to be the original state. Webbed 

 leaf-expansions are held as derivative, but they may still retain the dicho- 

 tomous venation, with free vein-endings. A further advance was the looping 

 of the veins to form a reticulum. Thus a primitive type of Fern-leaf may be 

 figured as long-stalked, with a distal dichotomy of narrow, separate, single- 

 veined segments, arranged either radially or bifacially. 



Such Ferns as existed in the Primary Rocks were mostly, and perhaps 

 exclusively Eusporangiate. It is shown in Chapter v that corresponding 



