January 25, 1918] 



SCIENCE 



95 



The epidermis is said to be of " relatively 

 sliglit phylogenetic interest." Yet tlic stoma 

 is probably the most conservative organ of 

 plants. 



The common term medullary rays is repeat- 

 edly condemned, on the g^round that their 

 relation to the pith is only a " semblance," 

 due to obsolescence of the primary wood. 

 This may be true, but the relation is of very 

 old date, for it was already well established 

 in the Calamites and some of the Cycado- 

 filices. From the author's point of view the 

 wide ray is a compovmd one, derived from the 

 aggregate tj-pe of ray; the vascular bundles 

 were not originally separate, and the state- 

 ments of Sanio and Sachs as to the bridging 

 over of the primary gaps by interfascicular 

 cambium are rejected. They are, however, 

 true, as a description of the facts, and hold 

 good for the yovmg Calamite as well as for 

 more modern plants. 



On the general question of the relation of 

 herbaceous to arboreal types, it may be pointed 

 out that there is no proof that our existing 

 herbaceous Lycopods came from arboreal an- 

 cestors ; the herbaceous Selaginellites was con- 

 temporary with the arboreal Lepidodendrese. 

 The siphonostele is held to have primitively 

 possessed phloem on the inner as well as the 

 outer surface. This type of structure, how- 

 ever, is rare among Palaeozoic plants. 



In the chapter on the Microsporangium the 

 author adopts the view that the higher plants 

 arose from forms like the thallose Liver- 

 worts, and quotes Bower's " Origin of a Land 

 Flora " in support of this theory. No mention 

 is made of Professor Bower's subsequent 

 change of view. 



The " Canons of Comparative Anatomy " 

 which the author insists on are three in 

 number — Recapitulation, Conservative Organs 

 and Reversion. The doctrine of recapitula- 

 tion in the development of the individual of 

 the history of the race is well known, though 

 no longer accepted without question. The 

 author points out that negative evidence is 

 of little or no value, but doubts may arise as 

 to what testimony is negative; in a pine- 

 seedling, for example, short-shoots are absent. 



but foliage-leaves on the main stem are 

 present. 



Among conservative organs the leaf is first 

 cited, and then the reproductive axis. The 

 present writer is given the credit for the 

 latter idea; it belongs rather to Solms-Lau- 

 bach, but neither generalized the conclusion, 

 which was confined to the peduncles of Cycads. 

 Floral axes are subject to modifications of 

 their own, and are not necessarily conserva- 

 tive. As regards the root, the primary struc- 

 ture is no doubt highly conservative, but it 

 does not follow that the same is true of its 

 secondary modifications. 



The word " reversion " is used in a peciiliar 

 sense, for certain effects of wounding, be- 

 lieved by the author and some others to be 

 reminiscent of ancestral characters. This 

 doctrine has hitherto been employed only in 

 support of certain controversial opinions, and 

 has not yet been adequately subjected to im- 

 partial criticism. 



The worst of all such " canons " is that 

 every writer applies them as suits his in- 

 dividual views, and treats inconvenient casea 

 as exceptions. 



In the systematic part of the book we first 

 come to the author's well-known division of 

 the higher plants into Lycopsida, without, and 

 Pteropsida, wth leaf-gaps in the vascular 

 ring, a classification widely accepted, tliough 

 it is now realized by many botanists that 

 Sphenophylls and Equisetales have little in 

 common with the Lycopod group. 



The author's doctrine of the cortical origin 

 of the pith is applied even to the Lycopods, 

 where the evidence seems peculiarly unfavor- 

 able to this interpretation. It is a pity that 

 the exact developmental processes involved are 

 not more clearly explained. 



The author's views on the evolution of the 

 Osmundacese are well expounded, and a strong 

 ease made out, which would have been more 

 convincing if the facts on the other side, 

 brought forward by Ividston and Gwynne- 

 Vaughan, had been dealt with. 



The lower seed-plants are divided into Archi- 

 gN-mnospermse, including Cycadofilicales, Cy- 

 cadales, Cordaitales and Ginkgoales, and 



