NA TURE 



[NOVEMHEK 5, 1908 



of the life-history; elsewhiMc, liowover (as on p. i5>)), 

 the words are employed in .-i different sense, namely 

 to indicate the author's theory that the asexual 

 generation has been intercalated in the life-cycle, and 

 is therefore newer than tlie sexual phase or gameto- 

 phyte. To avoid confusion it will be best to speak 

 of this view as the " intercalation theory," though 

 the term " antithetic " has conic to be identified with 

 it. The significance of the title, " The Origin of a 

 Land Flora," lies in the fact that the sexual genera- 

 tion retains, at least throughout the archegoniate 

 cryptogams, the primitive method of fertilisation by 

 spermatozoids, requiring the presence of water, while 

 on the other hand the asexual phase, with its wind- 

 scattered spores, is essentially adapted to a terrestrial 

 life. Hence the author speaks of the alternation as 

 ^' amphibious," an appropriate phrase which may be 

 readily adopted, whatever view be taken of the origin 

 •of the two generations. The asexual sporophyte, 

 however it may have arisen, conquered the dry land; 

 the gametophyte, with its conservative adherence to 

 traditional methods, remained dependent on a more 

 -or less watery environment, until the seed-plants came 

 to be evolved. Then the prothallus became a mere 

 .parasite on the sporophyte, enclosed within the mega- 

 sporangium, so that fertilisation could take place 

 on the plant itself. Spermatozoids were retained in 

 the more primitive types (cycads, Ginkgo, and no 

 doubt many fossil seed-plants), but their swimming 

 was now confined to a water-drop secreted within the 

 •ovule; in the rest of the Spermophyta they have 

 ■dropped their now useless motility, and fertilisation, 

 like the other vital processes, has become thoroughly 

 adapted to terrestrial conditions. 



All this is admirably told in Prof. Bower's book, 

 and it is to him that the credit belongs of realising 

 the essential biological significance of alternation of 

 generations as it exists in the higher cryptogams. 



The question at issue relates to the origin of the 

 alternating generations. On the intercalation hypo- 

 thesis, maintained by Prof. Bower in agreement with 

 Celakovsky and some other morphologists, the 

 sexual generation represents the original plant, which 

 alone existed in the presumed ancestor, while the 

 asexual sporophyte is a new development, an inter- 

 calation, arising from the elaboration of the fertilised 

 ovum or zygote, first into a mass of spores, and 

 •ultimately into a complex sporogonium on the one 

 hand or a spore-bearing plant on the other. 



The strength of the intercalation theory lies in the 

 evidence afforded by certain liverworts (Ricciacese), in 

 which the sporogonium actually consists of nothing 

 but a spherical mass of spore-mother-cells, enclosed 

 in an ephemeral epidermis. So simple a body might 

 well have arisen as a new formation, as a fruit-body 

 replacing an oospore, a development for which various 

 apparent analogies have been traced among thallo 

 phytes. From the Ricciaceae there are found suffi- 

 ciently continuous series of forms, leading up to the 

 fully differentiated capsules of the higher liverworts and 

 the mosses. Hence the intercalation theory appears 

 quite credible for the Bryophyta, and some botanists 

 have accepted it for that class while rejecting it for 

 the Pteridophyta. 



NO. 2036, VOL. 79] 



Even as regards the Bryophyta, however, every- 

 thing depends on the primitive nature of the Riccia- 

 ceous sporogonium, and this is open to doubt. As 

 the author himself s.ays (p. 237) : — " It m.iy be a ques- 

 tion whether the absence of a nutritive system is 

 due here to reduction, or is itself the primitive state." 

 Though " the latter is the view usually accepted," 

 there is good evidence for reduction in related liver- 

 worts (Cyathodium, pp. 237 and 263), and in Riccia 

 itself the transitory nature of the sporogonial wall 

 (p. 257) may well indicate a secondary loss or change 

 of function, as we see in the case of the nucellus of 

 so many angiospermous ovules. There are good 

 grounds for holding that far-reaching reduction has 

 gone on even among the higher Bryophyta, and, on 

 the whole of the evidence, the idea of ascending 

 series within this class, starting from the simplest 

 form of sporogonium, cannot be considered as by 

 any means established. In fact, the Bryophyta, which 

 have long been regarded as affording the clue to the 

 interpretation of the life-cycle of the higher plants, 

 themselves stand in need of interpretation, even more 

 than other groups. 



Among the Vasculares, the sporophyte is always 

 (even in Lycopodiiim Selago!) a highly organised 

 plant, and no one would dream of attributing its 

 origin to an intercalation, if it were not for the 

 analogy of the bryophytes. 



During the last fifteen years the cytological distinc- 

 tion between the two generations has played an 

 important part in the controversy as to their nature. 

 In all normal cases the asexual generation is 

 "diploid," its nuclei having twice as many 

 chromosomes as those of the " haploid " sexual 

 phase. Reduction takes place in the spore-mother- 

 cell, at the initiation of the gametophyte. This side 

 of the subject is very ably treated by Prof. Bower, 

 who continues to attach considerable importance to 

 the cytological distinction, in spite of the exceptional 

 cases recently brought to light, where it has been 

 shown with certainty that the gametophyte genera- 

 tion may be diploid, and, with great probability, that 

 the sporophyte may be haploid. Such cases are 

 associated with the occurrence of apospory (suppres- 

 sion of spore-formation) and apogamy (suppres- 

 sion of sexual reproduction) in the same Kfe-cycle, 

 as happens in various anomalous ferns. These 

 observations prove that there is no necessary connec- 

 tion between the number of chromosomes and the 

 morphological characters of the alternating genera- 

 tions, but " cannot be held to invalidate the view that 

 the cycle as above stated existed in all probability 

 throughout the earlier phases of descent of the Arche- 

 goniatae " (p. 62). 



The cytological distinction was at one time regarded 

 as supporting the opinion that the two generations 

 were distinct in origin, and thus as favouring the 

 intercalation theory. This can no longer be main- 

 tained, since it has been shown by Lloyd Williams 

 and Motlier that in the alga Dictyota there is a 

 regular alternation between the haploid sexual and 

 the diploid asexual generation, generations which in 

 all morphological respects are perfectly similar to one 

 another. There can be no question of intercalation 



