538 



NATURE 



[September 27, 1900 



relatively complete method propounded by Antoine Laurent de 

 Jussieu in his " Genera Plantarum," just lOO years later. 



The nineteenth century opened with the struggle for pre- 

 dominance between the Jussiean and the Linnean systems. In 

 England the former soon obtained considerable support, notably 

 that of Robert Brown, whose " Prodromus Floroe Novoe Hol- 

 landise." published in 1810, seems to have been the first English 

 botanical work in which the natural system was adopted ; but it 

 did not come into general use until it had been popularised by 

 Lindley in the 'thirties. 



Meantime the Jussiean system had been extended and ini- 

 proved by Augiiste Pyrame de Candolle (1813-24). It is 

 essentially the Candollean classification which is now most 

 generally in use, and it has been immortalised by its adoption in 

 Bentham and Hooker's " Genera Plantarum," one of the great 

 botanical monuments of the century. In Germany, however, 

 it has been widely departed from, the system there in vogue 

 being based upon Brongniart's modification (1828, 1850) of de 

 CandoUe's method as elaborated successively by Alex. Braun 

 (1864), Eichler (1876-83) and Prof Engler (1886, 1898). It 

 must be admitted that for the last fifty years the further evolution 

 of the natural system, at any rate so far as Phanerogams are 

 concerned, has been confined to Germany. 



One of the most important advances in the classification of 

 Phanerogams was based upon Robert Brown's discovery in 1827 

 of the gymnospermous nature of the ovule in Conifers and 

 Cycads, which led Brongniart (1828) to distinguish these plants 

 as " Phanerogames gyinnospermes " ; and although the system- 

 atic position of these plants has since then been the subject of 

 much discussion, the recognition of the Gymnospermae as a 

 distinct group of archaic Phanerogams is now definitely 

 accepted. 



Moreover, the greatly increased knowledge of the Cryptogams 

 has involved a considerable reconstruction in the classification of 

 that great sub-kingdom. One of the most striking discoveries 

 is that first definitely announced by Schwendener (1869) con- 

 cerning Lichens, to the effect that the body of a Lichen consists 

 .,of two distinct organisms, an Alga and a Fungus, living in 

 symbiosis ; a discovery which was so nearly made by other con- 

 temporary botanists, such as de Bary, Berkeley and Sachs, and 

 which can be traced back to Haller and Gleditsch in the 

 eighteenth century. 



But the discoveries which most affected the classification of 

 the Cryptogams are those relating to their reproduction. Whilst 

 it had been recognised, almost from time immemorial, that 

 Phanerogams reproduce sexually, sexuality was denied to 

 Cryptogams until the observations on Liverworts and Mosses by 

 Schmidel and by Hedwig (of whom it was said that he was born 

 to banish Cryptogamy) in the eighteenth century ; and even as 

 late as 1828 we find Brongniart classifying the Fungi and Algae 

 together as "Agames." But in the middle third of the nine- 

 teenth century, by the labours of such men as Thuret, Pringsheim, 

 Cohn, Hofmeister, Naegeli and de Bary, the sexuality of all 

 classes of Cryptogams was clearly established. It is worthy of 

 note that, although the sexuality of the Phanerogams had been 

 accepted for centuries, yet the details of sexual reproduction 

 were first investigated in Cryptogams. For it was not until 1823 

 that Amici discovered the pollen-tube, and it was more than 

 twenty years later (1846) before he completed his discovery by 

 ascertaining the true significance of the pollen-tube in relation 

 to the development of the embryo ; whilst it remained for 

 Strasburger to observe, thirty years later, the actual process of 

 fertilisation. 



The discovery of the reproductive processes in Cryptogams 

 not only facilitated a natural classification of them, but had the 

 further very important effect of throwing light upon their relation 

 to Phanerogams. Perhaps the most striking botanical achieve- 

 ment of the nineteenth century has been the demonstration by 

 Hofmeister's unrivalled researches (1851) that Phanerogams and 

 Cryptogams are not separated, as was formerly held, by an 

 impassable gulf, but that the higher Cryptogams and the lower 

 Phanerogams are connected by many common features. 



The development of the natural classification, of which an 

 account has now been given, proceeded for the most part on the 

 assumption of the immutability of species. As Linnaeus ex- 

 pressed it in his " Fundamenta Botanica," " species tot numer- 

 amus, quot diversse formce in principio sunt creatae." It is 

 difficult to understand how, with this point of view, the idea of 

 affinity between species could have arisen at all ; and yet the 

 establishment of genera and the attempts at a natural system 



NO. 1 6 13, VOL. 62] 



prove that the idea was operative. The nature of the prevalent 

 conception of affinity is well conveyed by Linnaeus's aphorism, 

 " Affines conveniunt habitu, nascendi niodo, proprietatibus, 

 viribus, usu." 



But a conviction had been gradually growing that the assumed 

 fixity of species was not well founded, and that, on the 

 contrary, species are descended from pre-existent species. This 

 view found clear expression in Lamarck's " Philosophic Zoo- 

 logique," published early in the century (1809), but it did not 

 strongly affect public opinion until after the publication of 

 Darwin's " Origin of Species " in 1859. Regarded from this 

 point of view, the problems of classification have assumed an 

 altogether different aspect. Affinity no longer means mere 

 similarity, but blood-relationship depending upon common 

 descent. We no longer seek a " system" of classification ; we 

 endeavour to determine the mutual relations of plants. The 

 effect of this change has been to stimulate the investigation of 

 plants in all their parts and in all stages of their life, so as to 

 attain that complete knowledge of them without which their 

 affinities cannot be accurately estimated. If the classification of 

 Cryptogams is, at the present moment, in a more satisfactory 

 position than that of Phanerogams, it is just because the study 

 of the former group has been, for various reasons, more thorough 

 and more minute than that of the latter. 



Palaeophytology. 

 The stimulating influence of the new doctrine was not, how- 

 ever, confined to the investigation of existing plants ; it also 

 gave a remarkable impulse to the study of fossil plants, inasmuch 

 as the theory of descent involves the quest of the ancestors ot 

 the forms that we now have around us. Marvellous progress 

 has been made in this direction during the nineteenth century, 

 by the labours more especially of Brongniart, Goeppert, Unger, 

 Schimper, Schenck, Saporta, Solms-Laubach, Renault, on the 

 Continent, and in our own country of Lindley and Ilutton, 

 Hooker, Carruthers, and more especially of Williamson. So 

 far-reaching are the results obtained that I can only attempt 

 the barest summary of them. I may perhaps best begin by 

 saying that only a small proportion of existing species have 

 been found in the fossil state. In illustration I may adduce 

 the statement made by Mr. Clement Reid in his recent work, 

 "The Origin of the British Flora," that only 270 species, that 

 is, about one-sixth of the total number of British vascular plants, 

 are known as fossils. Making all due allowances for the im- 

 perfection of the geological record, for the limited area investi- 

 gated, and for the difficulty of determination of fragmentary 

 specimens, it may be stated generally that the number of exist- 

 ing species has been found to rapidly diminish in the floras of 

 successively older strata ; none, in fact, have been certainly 

 found to persist beyond the Tertiary period. Certain existing 

 genera, belonging to the Gymnosperms and to the Pteridophyta, 

 have, however, been traced far down into the Mesozoic period. 

 Similarly, the distribution in time of existing natural orders 

 does not coincide with that of existing genera ; thus the Ferns 

 of the Carboniferous epoch apparently belong, for the most part, 

 if not altogether, to the order Marattiacese, but they are not 

 referable to any of the existing genera. 



Moreover, altogether new families of fossil plants have been 

 discovered : such are, among Gymnosperms, the Cordaitaceae 

 and the Bennettitacece ; among Pteridophyta, the Calamariacese, 

 the Lepidodendracete, the Sphenophyllacese and the Cycado- 

 filices. It is of interest to note that all these newly discovered 

 families can be included within the main subdivisions of the 

 existing flora ; in fact, no fossil plants have been found which 

 suggest the existence in the past of groups outside the limits of 

 our Phanerogamia, Pteridophyta, Bryophyta and Thallophyta. 



It cannot be said that the study of Palaeobotany has as yet 

 made clear the ancestry and the descent of our existing flora. 

 To begin with the angiospermous flowering plants, it has been 

 ascertained that they make their first appearance in the 

 Cretaceous epoch, but we have no clue as to their origin. The 

 relatively late appearance of Angiosperms in geological time 

 suggests that they must have sprung from an older group, such 

 as the Gymnosperms or the Pteridophyta; but there is no 

 evidence to definitely establish either of these possible origins. 

 Then as to the origin of the Gymnosperms, whilst it cannot be 

 doubted that they were derived from the Pteridophyta, the 

 existing data are insufficient to enable us to trace their pedigree. 

 The most ancient family of Gymnosperms, the Cordaitacew, can 

 be traced as^far back as any known Pteridophyta, and cannot, 



