330 - . BOTANICAL GAZETTE (april 



of the xylem. In the other modern pteridophytes it is apparently the 

 former which has taken place, secondary xylem appearing rarely and 

 in small amounts; and, as Hill (15) infers from a comparison of the 

 examples of pteridophytes showing the phenomenon, it is more prob- 

 ably an example of reduction than a new development. In Isoetes, 

 however, the reduction in the bundle has not been limited to the 

 xylem portion, but has extended to the whole of the phloem, bo>h 

 primary and secondary. 



* 



The position of Isoetes 



The phylogenetic connections of Isoetes have been discussed in all 

 recent papers, with the great weight of evidence in favor of a lycopod 

 ancestry. The evidence as to its relationship afforded by its anatomy 

 has been taken up recently by Bower (2), on the basis of the inter- 

 pretation given by Scott and Hill, with the conclusion that Isoetes 

 is in its anatomy a lycopod, with a stem structure which can be 

 explained by regarding it as a stunted lycopod. In his eagerness to 

 show a unity of structure in the Lycopodiales, Bower makes t e 

 following statement (p. 339): "Throughout the ^^^TT^ 

 foliar traces are inserted peripherally, and with only a sag 

 disturbance upon the periphery of the cauline xylem core." n 

 of the questionable existence of a cauline portion in the xy em w ^ 

 there seems to be little justification for so sweeping a statemen . 

 is in this very thing that Isoetes differs markedly from ot er a^ ^ 

 lycopods, although the difference is not of such a character^ ^ 

 make the relationship doubtful. The difference is correla ^ ^ 

 the stunted habit, and such differences of body habit h * ven J^ ' .^ 

 admitted to have great weight in determining the larger S 10 ^' 

 The stunted habit of stem is not limited in Lycopodiales 

 but it occurs also, though of very different type, in rn)u 6^.^ of 



Even among those who recognize the strong lycopo ^ t0 

 Isoetes, it is suggested occasionally that it might e a ^^ 

 separate Isoetes from the lycopods and establish a new or e ,^ ^ 

 The present tendency seems to be toward a multipUca 1 ^ 

 so that it may be well to consider the desirability of it in ^ 



The closest connection of Isoetes is, as has been freq ^ 



out, with the Lepidodendreac, although it has many pom 



; in 



