THE "SELAGO" CONDITION IN FERNS 169 



Fern (Onodea Struthioptcris] (Fig. 89). There may be some degree of 

 regularity in the succession of sterile and fertile leaves, which may be 

 correlated with season ; thus in Blechnum boreal e the leaves first expanded 

 in the spring are sterile, and they are followed by a series of fertile leaves. 

 The condition of trie shoot as a whole is, in fact, comparable with that of 

 Lycopodium Selago or of Isoetes, with their successive sterile and fertile zones. 



But the commoner case for Ferns is that where leaves are not sterile or 

 fertile as a whole, but many or even all the leaves of the mature plant are 

 fertile at least in part, and frequently show a correlative reduction of area 

 as compared with the rest of the leaf, which is sterile. In the distribution 

 of the fertile and sterile parts of the individual leaf there is great diversity, 

 and differences may be seen in species of the same genus, or even in 

 individuals ; thus in Osmitnda regalis the lower parts of the fertile leaf 

 are broadly expanded and sterile, the apical region is fertile and correlatively 

 exiguous ; but in O. javanica the fertile region extends irregularly over the 

 lower pinnae, and the apical region is expanded and sterile (Fig. 90). 



It has been shown by Goebel that the mode of development of such Fern- 

 leaves may be experimentally altered : by removing from a plant of Onodea 

 Struthiopteris the foliage leaves which are first expanded in the spring, 

 the later expanded leaves, which are normally sporophylls, were induced 

 to assume the character of foliage leaves. Similar results were also obtained 

 by Atkinson. 



The facts thus briefly summarised for Ferns are evidently comparable 

 with those noted for the Lycopods, and the differences in detail which exist 

 have their relation to the megaphyllaus character. But in Ferns the facts 

 are less cogent ; for though abortive sporangia and imperfect sori are at 

 times found on Fern-leaves, still the evidence that they are vestigial is less 

 clear than in Lycopodium, Isoetes, or Ophiog/ossiem, owing to the less 

 definite position and number of those parts in Ferns. The conclusion that 

 the foliage leaves or parts of leaves in Ferns are phylogenetically sterilised 

 sporophylls, or parts of sporophylls, is therefore based rather on broad 

 comparisons and on analogies with other Pteridophytes than on the direct 

 observation of parts which may be held to be vestigial. That such a 

 transmutation may take place in the individual life is fully demonstrated 

 by the experiments of Goebel above quoted. It seems therefore reason- 

 able to hold for Ferns, as for other Pteridophytes, that sterilisation of 

 sporophyils has been effective in the course of their evolution. 



A converse view to that thus stated has been habitual in the past, 

 and is maintained by some to the present time. By them the evolutionary 

 history is read in direct terms of the ontogeny, and the sterile leaf is thus 

 assumed to be the primitive leaf, which has become a sporophyll by the 

 superposition upon it of sori and sporangia. Those who take this point 

 of view have brought forward in its support the facts that the develop- 

 ment and structure of the sterile and fertile leaves is closely alike, and 

 that intermediate forms exist frequently between the two, so that the 



