October 20, 19 10] 



NATURE 



491 



extending to about 250 pages. The description of the 

 fossil representatives is preceded by a sketch of the 

 recent members of the group, illustrated by some 

 excellent figures both of the habit and the anatomy. 

 This method, which the author extends to all groups 

 which have living' representatives, is eminently suit- 

 able for a book which is intended for geologists as 

 well as for those who have had a botanical training. 

 The best known of the fossil Lycopods were trees ; 

 recently, however, we have learnt a good deal about 

 herbaceous club-mosses which also flourished in 

 Palaeozoic times. These are fully described, and it 

 is interesting to note how closely, in habit and re- 

 productive methods, some of them approached the 

 living genus .Selaginella. On the other hand, we 

 still have no proof of the early occurrence of Lyco- 

 podium, which on theoretical grounds is regarded as 

 tlie more primitive type. 



The great family of the Lepidodendreje is admir- 

 ably treated, with equal regard to external features 

 and internal structure. .Some progress has now been 

 made in correlating the two, and the results are some- 

 times a little surprising. Thus the well-marked ana- 

 tomical species LcpiJodciidroii fiiUginosttiv, William- 

 son, is shown to correspond to at least three species 

 based on external characters — a Lepidophloios and 

 two species of Lepidodendron proper. Thus the two 

 genera last-mentioned are indistinguishable, even 

 specifically, by their anatomical characters, and it is 

 very doubtful whether their claim to generic rank can 

 be upheld. 



Until quite recently we had but little knowledge 

 of the structure of the well-known genus Sigillaria, 

 but now, thanks to the work of Bertrand in France 

 and of Kidston and .Arber and Parkin in our own 

 country, we have become acquainted with the anatomv 

 of a number of species. The new evidence, together 

 with the characters of the fructifications, has finallv 

 confirmed Williamson's view of the close affinity 

 between Sigillaria and Lepidodendron, and of the 

 cryptogamic nature of both alike. .At the same lime, 

 facts have come to light which might once have 

 been regarded as favouring Phanerogamic affinities, 

 for in two genera of Palaeozoic Lvcopods organs 

 closeh' analogous to true seeds have been discovered ; 

 in both cases the plants, one herbaceous, the other 

 probably arborescent, are in all other respects tvpical 

 members of the Ljcopodiales. Their seed-like repro- 

 ductive bodies are regarded by many palsobotanists 

 as a striking instance of parallel development; the 

 author, however, is inclined to see in them evidence 

 of a genetic connection between the Pateozoic Lvco- 

 pods and certain Conifers. 



The second half of the volume is devoted to the 

 Fern-like plants, and here the effect of recent dis- 

 coveries makes itself felt in an even greater degree 

 than elsewhere. .As is well known, a large proportion 

 of the Palaeozoic Ferns, formerly so-called, are now 

 under well-founded suspicion of not having been real 

 Ferns at all, but seed-bearing plants of fern-like 

 habit. In several cases this has been definitely proved, 

 and in a majority of the plants in question all the 

 available evidence points towards their spermophvtic 

 NO. 2138, VOL. 84] 



affinities. .At the same time, the resemblance to 

 Ferns, which for so long misled investigators, is by 

 no means wholly fallacious, but, as shown by ana- 

 tomical and other evidence, indicates a real relation- 

 ship to the Fern-stock, while on the other hand the 

 connection with the Cycad type of seed-plant is mani- 

 fest. The practical difficulty is to distinguish between 

 these "Ferns with seeds" and the true cryptogamic 

 Ferns which no doubt really existed side by side with 

 them ; habit is no criterion, anatomy is only avail- 

 able in exceptional cases ; even where the sporangia 

 are present it is often impossible to say whether thev 

 were cryptogamic sporangia or pollenrsacs. The author 

 takes a moderate and reasonable view of the difficult 

 position ; he recognises a considerable group of 

 generalised Ferns, his Ccenopterideae (the Primofilices 

 of Mr. .Arber), in which our knowledge of structure 

 and of reproductive processes is sufficient to establish 

 the Fern kinship ; he is further inclined to admit a 

 certain number of Palaeozoic Marattiales (highly 

 organised Ferns now only represented by a small 

 tropical family), though here the evidence is a good 

 deal more doubtful. 



The Pteridosperms, as such, do not come into the 

 present volume, but the last chapter is devoted to 

 a number of genera of more or less uncertain position, 

 most of w-hich will no doubt prove to be seed-bearing 

 plants, while a few may retain their traditional posi- 

 tion among true Ferns. The clearly established types 

 of Pteridosperms will be considered in vol. iii., which 

 is to be devoted to seed-plants. This part of the work 

 will be looked forward to with keen interest, for a 

 number of questions of the utmost importance for 

 the theory of evolution will then have to be con- 

 sidered. 



The present volume, in its full and impartial treat- 

 ment of habit and structure, of morphology and dis- 

 tribution, is beyond question the best handbook extant 

 for the important fossil groups of which it treats. 

 It is abundantly illustrated, and provided with an 

 excellent index, and with a bibliography; the latter, 

 taken in connection with that of vol i., is singularly 

 complete, so far as works of any importance are 

 concerned. 



Botanists are certain to appreciate Prof. Seward's 

 work ; we hope that it may receive equal recognition 

 from geologists, who, even more than their botanical 

 colleagues, stand in need of a modern text-book of 

 fossil plants. D. H. S. 



THE COLLECTED WORKS OF HUYGEXS. 

 (Euvres completes dc Christiaan Hiiygciis. Publiees 

 par la Society Hollandaise des Sciences. Tome 

 douzieme, Travau.x de .Mathematiques pures 1652- 

 1656. Pp. vi + 296. (La Have: Martinus Xijhoff, 

 igio.) 



THIS volume is the second one which contains 

 reprJnts of published writings of Huygens, the 

 ten first volumes having been devoted to his corre- 

 spondence. During the years 1652 to 1656 Huygens 

 had still to divide his attention between his scientific 

 work and the study of l;iw, but the extracts from his 



