XIX] COMPARISON ^7 



in the roots of the young but absent in the adult plant. In Botrychium it 

 has been seen in twelve species, but in varying abundance, and especially in 

 the young. It is thus inconstant in its occurrence in the family, but it is seen 

 to prevail in those species of Ophioglossiim which are regarded as the most 

 specialised. The mycorhizic condition does not necessarily entail a marked 

 state of reduction, though in extreme cases this may be seen: but it appears 

 to be commonly associated with a leathery type of foliage, such as that of 

 most of the Ophioglossaceae. 



Comparative Treatment of the Ophioglossaceae. 



This very distinctive and circumscribed family has been related syste- 

 matically with the Filicales by most writers, a position which has been 

 definitely confirmed by a large body of facts recently acquired. An alternative 

 relation with the Lycopodiales was suggested by Mettenius {^Bot. Zeit. 1867, 

 p. 98); it was seriously considered by Strasburger {Bot. Zeit. 1873, P- 5)> ^^^cl 

 by Celakovsky {Pringsh. Jahrb. 1884, p. 291), while Graf Solms Laubach 

 {Biiit. Ami. Vol. IV, p. 186) withheld a definite opinion in view of the 

 incompleteness of the facts. The case for a Lycopodinous affinity was 

 further developed on the basis of many new facts in 1896 (Bower, MovpJi. of 

 Sp07-c- Producing Organs. Dulau, London, 1896), and fully argued in 1908 

 {Land Flora, pp. 430—494). The question turns upon the morphological 

 nature of that characteristic organ the "fertile spike." Is this an up-grade 

 part illustrating a progressive evolution from a part similar in nature to the 

 Lycopodinous sporangium, or is it a foliar part showing modification, mostly 

 of the nature of reduction? In the former case the affinity would be with the 

 Lycopodiales : in the latter with the Filicales. Naturally the facts of form, ana- 

 tomy, development, and of physiological probability must all be brought into 

 the final decision. Unfortunately Palaeontology is silent, except in so far as it 

 supplies important comparisons with early fossils not recognisable as belong- 

 ing themselves to the Ophioglossaceae. The first point will then be to decide 

 the morphological character of that most distinctive part, the fertile spike. 



Taking first the Lycopodinous alternative, the adaxial position and the 

 spore-producing function are the same for the sporangium and for the spike. 

 By comparison of the simplest spikes known it is possible to construct an 

 up-grade series leading by successive steps of amplification, septation, and 

 branching from the Lycopods and Psilotales to the most elaborate of 

 Ophioglossaceous spikes. In such a series Ophioglosstun would take its place 

 as relatively primitive, and the extreme would be seen in Botrycliinm 

 virginiannni. As a purely morphological tour-de-force this is possible ; but 

 physiologically is it probable that it should happen in a family marked 

 strongly by mycorhizic nutrition.' Further, how does this view accord with 

 the anatomical facts? Taking first the vascular supply of the spike of the 



