15 



130 fossil species have been found in the coal shale in the 

 same area. Dr. Carruthers also tells us that a group of ferns 

 has entirely passed away with a stem-structure fundamentally 

 different from any now in existence. All these distinctions 

 are equally prominent in the still older Devonian remains. 

 So far back as we can trace the three great'groups of Vascular 

 Cryptogams they move in parallel and not in converging lines. 

 The importance of this fact is so enormous that it seems, to 

 dispose of the question for ever ; for there is really not t 1 me 

 enough left before the Devonian beds to allow a primitive 

 cryptogamic form to vary into three such strongly-marked 

 and highly specialised groups of descendants. Then, again, 

 as Lyell remarks, it is astonishing how little ferns have 

 altered since their first appearance, so that possibly even the 

 genus Pteris is a survival from the carboniferous age. If they 

 have varied so little during such an enormous period of time, 

 why should they be supposed to have varied immensely just 

 before the commencement of that time ? And is it not a 

 singular fact that all the remains which would support the 

 theory of the derivation of the three groups from an older form 

 have been lost ? 



The same story is told by the other vegetable remains of 

 the coal measures : thus the Conifers are represented by 

 the Taxinece, or Yew alliance, a highly specialised form. 

 FQV the present the opponent of tbe Theory of Descent 

 may take up an impregnable position behind his fortress of 

 coal. 



9. Do Synthetic Types prove Evolution ? Synthetic types, 

 i. e., those which are supposed to combine the characteristics 

 of separate orders or classes, are considered by many as 

 a proof of Evolution. Let us bring this assumption to the 

 test of fact. I suppose the Cycads are a synthetic type. 

 They resemble ferns in the circinate vernation of the leaves 

 and in the sorus-like aggregation of pollen-sacs ; in their 

 dioecious, entirely naked flowers, crowded into cones, they 

 partly resemble Conifers and partly Equiseta. In the processes 

 of germination they resemble the higher Vascular Cryptogams. 

 In their general habit they are like Palms. Here, I imagine, 

 we have what is usually called a synthetic type. Now, 

 according to the Theory of Descent it ought to have been 

 prior in time to the Ferns, Conifers, and Palms, the charac- 

 teristics of which it combines. As a matter of fact it is later 

 than Ferns and Conifers. We ought, according to theory, to 

 trace a series of diverging forms starting from it. As a 

 matter of fact, we find it an isolated group throughout all its 

 existence. We see the first scattered indications of its coming 



