304 A CENTURY OF PROGRESS IN THE NATURAL SCIENCES 



bryology. The desirability of separating the Psilotales from the Lycopo- 

 diales is stressed, and this point of view, though not in itself original to Hollo- 

 way, has received its strongest support, apart from the fossil evidence, from his 

 work, and is now generally accepted. Knowledge of Psilotum prothalli also 

 dates from 1917 with preliminary descriptions of certain stages independently 

 by Darnell Smith and Lawson. A fuller description comes from Holloway in 

 1939 (preliminary note in 1938). This paper, while confirming the close re- 

 semblance of Psilotum to Tmesipteris in matters of embryology and in particu- 

 lar in showing the all-important character of the young plant as consisting of 

 a dichotomously branching cylindrical axis with apical growth and central vas- 

 cular tissue but without roots or appendages in the early stages other than 

 superficial hairs, adds one other fact of unique interest. The apical growth, di- 

 chotomous branching, and cylindrical form of the subterranean prothalli, while 

 agreeing with Tmesipteris, also recall small pieces of rhizome, and this resem- 

 blance is increased by the discovery that, in all really large prothalli, traces of 

 central vascular tissue were also present. 



This last observation is the reason why it has seemed important to trace in 

 such detail the growth of knowledge of life histories in the more difficult pteri- 

 dophyte groups. The interpretation of the observation was left by Holloway 

 himself, with becoming caution, as suh judice. One suggestion was that the 

 large vascular prothalli were abnormal, and this possibility is a real one since 

 it was later shown (Manton, 1942) that all of Holloway 's prothalli, both vas- 

 cular and nonvascular, were cytologically diploid and derived from tetraploid 

 sporophytes. Haploid prothalli have not yet been found and, until they are, 

 the risk of abnormality cannot be dismissed. There is, however, no precedent 

 for the supposition that teratological structures are necessarily caused by poly- 

 ploidy as such and an alternative possibility must also be kept in mind. This is 

 that we may have here not an abnormal, but a vestigial, structure of a very 

 primitive kind. We are indeed being confronted with an alternative view of 

 alternation of generations in the land plants which is diametrically opposed to 

 that expressed by Bower in The Origin of a Land Flora (1908) and which re- 

 calls very strongly that expressed by Lignier in 1903. The latter postulated an 

 ancestor for all the archegoniates, his "prohepatica," in which both generations 

 were similar except for their reproductive structures and composed of simple 

 dichotomous thalli with apical growth, nonvascular in the ancestor of the bryo- 

 phytes and perhaps vascular at a later stage in the ancestor of the pterido- 

 phytes. This suggestion is obviously of the closest relevance to the facts of life 

 history in the Psilotales as they are at present known. More evidence is needed 

 both from living plants and more especially from fossils before further progress 

 can be made. But in contrasting the views of Lignier and Bower, both of whom 

 have contributed in an essential way to botanical thought, we can epitomize 

 much of the constructive thinking which has been given to the subject of alter- 

 nation of generations in the Pteridophyta in the century which has succeeded 

 Hofmeister. 



One of the more obvious effects of Hofmeister 's work was to remove perma- 

 nently the obscurity about diagnostic criteria for delimiting the main groups 

 of archegoniates, and we may quote in illustration of this another early English 

 textbook, that of Berkeley (1857). In this we find under the heading Filicales, 



