444 Scott and Hill.—Structure of Isoetes Hystrix. 
reduction (in relation to aquatic habit) from some more 
complex type, and probably from some very highly organized 
form of Lycopod, as indicated by the secondary growth, 
the marked heterospory, and the somewhat complex organiza¬ 
tion of the leaves and of the root-bearing portions of the 
axis. 
Our investigations have convinced us very strongly of 
the close Lycopodinean affinities of the genus, and of the 
absence of any special relationship to the Ferns. It will be 
necessary to consider briefly the points bearing upon this 
question. The Filicinean affinities of Isoetes have been 
maintained principally by Vines (1888), Farmer (1890), and 
Campbell (1891 and 1895). 
So far as the asexual plant is concerned, there can be no 
question that the sum of characters points clearly to Lyco¬ 
podinean affinities, while, with the single, very doubtful 
exception of the velum, there is no single point in which 
the Ferns are approached. The habit of the plant, with its 
crowded simple and narrow leaves, is more suggestive of 
Lycopods than of any other group. The fact that the stem 
is stunted appears to us to have no bearing on the question of 
affinities. Elongated and shortened forms of stem occur both 
in Lycopodiales and Filicales, and throughout the Vegetable 
Kingdom we find the greatest variety in this respect among 
closely related plants, and often within the limits of a single 
genus. Neither is the stunted form of stem by any means 
specially characteristic of the Ferns. The usually un¬ 
branched habit (which is not without exception) is no doubt 
correlated with the shortened form of stem 1 . 
The anatomy of the stem, with its solid stele, from which 
densely crowded, small and simple leaf-traces pass off, is just 
what we should expect to find in a stunted Lycopod, and bears 
no real resemblance to anything in Ferns, though a general 
similarity between the simplest forms of any two groups is 
easy enough to find. Secondary tissue-formation is a character 
1 It should be remembered, however, that there is some evidence that many of 
the Palaeozoic Sigillariae branched little or not at all. 
