CONCLUSIONS 



indicate that in the distant past the cytological state of the Pteridophyta as a whole must 

 have been more hke that of the Flowering Plants of to-day than is now the case. 



Of the various types of mechanism enumerated in Chapter 2, all have been seen to be 

 operative, though the degree of completeness in our knowledge about them varies very 

 greatly, as was to be expected. It may be of interest to pass each briefly in review. 



(i) Hybridization has been met with unexpectedly often and in a great variety of 

 groups. Outstanding examples were Equisetum litorale, trachyodon and Moorei, perhaps 

 Lycopodium Selago, and the various ferns, the latter ranging from frequently reformed 

 hybrids such as Asplenium germanicum, Polystichum illyricum, Dryopteris uliginosa and so on, 

 to very ancient hybrids now ranking as species owing to their loss of sexual reproduction, 

 such as Pteris cretica and the other apogamous ferns. 



(2) Polyploidy is present in almost bewildering profusion and has reached levels not 

 yet touched by any other group of plants. It is only necessary to recall the 205 chromo- 

 somes of pentaploid Dryopteris Borreri, the approximately 204 of tetraploid Psilotum, the 

 216 of Equisetum (the sporophytic and not the reduced numbers are, of course, now being 

 quoted), the 222 of hexaploid Poly podium, the 252 of hexaploid Cystopteris, the 400 odd of 

 Tmesipteris, the 500 odd of Ophioglossum vulgatum, to realize how much further these 

 nuclear processes have gone here than in the mere 81 or 120 chromosomes which con- 

 stituted the maximum numbers quotable in the Cruciferae. The most probable con- 

 clusion to draw from this fact is, as already suggested, that high chromosome numbers 

 are not primitive but a sign of antiquity. The undoubted tendency for outstandingly 

 high numbers to accumulate most conspicuously in the most ancient groups (Psilotales, 

 some Lycopods, Equisetum, Ophioglossum) will then become intelhgible and may further 

 be found to denote a measure of senility in these groups. 



It is not always possible to assess with certainty the grade of polyploidy involved. In 

 favourable cases where the monoploid state is a prime number, as in Dryopteris (« = 41) or 

 Polypodium {n = 37), we may diagnose, say, hexaploidy with certainty, but in many other 

 instances we are clearly deahng with only the upper members of series whose bases have 

 been lost. In these we can infer the existence of polyploidy and some facts about it only 

 by the indirect evidence of the arithmetical attributes of the numbers {Cystopteris, 

 Ophioglossum), or the uncertain evidence of comparisons external to the groups 

 (Equisetum) . By all these means, however, a strong suspicion is raised that the rather 

 rigid hmitations on nuclear increase encountered in the experimentally induced auto- 

 polyploid series of Osmunda, which ended in sterility at ^n, and in similar series in other 

 parts of the plant kingdom (see Chapter 3), has been far exceeded by the degree of 

 polyploidy actually achieved in nature. The reason for this is uncertain, but it may at 

 least suggest to us that there are some effects of longevity which are not exactly 

 reproduced when these processes are imitated in the laboratory. 



With regard to the type of polyploidy it is noticeable how little sign of autopolyploidy 

 has been discovered except in the artificially induced series of Osmunda (Chapter 3). 

 Psilotum is the only clear case in which simple autopolyploidy is suspected on positive 

 grounds. It should, however, be remembered that the only positive criterion for the 

 diagnosis of autopolyploidy is multivalent pairing, and we do not know for certain 

 whether in the course of thousands or millions of years this power might not become lost. 



285 



