384 Watson. — On Mesostrobus , a New Genus of 
Lepidostrobus Bailyanus Sch., which is of the ordinary Lepidostroboid type, 
and is heterosporous, but appears to be verticillate. 
The very close general resemblance between the sporophylls of Both- 
rodendron mundum , Miadesmia , and Selaginella is an interesting feature. 
It is fairly certain that Selaginella and Bothrodendron are not very closely 
related, and I do not think that there is much more connexion between 
Miadesmia and Selaginella. 
The evidence connecting Miadesmia and Selaginella is only that they 
are both herbaceous forms of ligulate Lycopods, with heterosporous cones 
and short sporophylls. Miss Benson (’08) in addition points out that the 
Miadesmia stem resembles that of some Selaginellae in its wood, and in the 
presence of trabeculae, probably representing the endodermis. 
The wood certainly resembles fairly closely that of some vertical Sela- 
ginellas, but it resembles equally closely that of many small Lepidodendroid 
twigs. 
The layer of trabeculae does not show up very well in any of the fairly 
numerous sections of Miadesmia stems that I have now examined, and it 
might apparently represent the middle cortex of the Lepidodendraceae, some 
types of which it considerably resembles. Heterospory combined with 
a ligule and a short sporophyll is met with in Bothrodendron mundum in 
a form which greatly resembles that of Miadesmia. 
There is, I think, no doubt that the Lepidodendraceae were a dominant 
group in the Carboniferous period, the very great specialization reached by 
Lepidocarpon , and shown generally by the Halonial branches and Uloden- 
droid scars is, I think, proof of this. 
It is a general rule that a dominant group contains members of all sizes 
and fitted for many different conditions, for example, the Deinosaurs, the 
dominant group of land animals during the mesozoic, are represented equally 
by the giant Sauropods, Diplodocus and Atlantosaurus , 80-100 ft. long, and 
by the little Compsognathus about as big as a rook and quite as lightly built. 
In just the same manner one would, a priori , expect the Lepidoden- 
draceae to be represented by herbaceous forms as well as by forest trees, 
and it is possible that in Miadesmia we have one of the herbaceous forms 
we should thus theoretically expect. So far as I know there is no evidence 
to prohibit this view, although there is little in its favour ; at any rate the 
evidence is as strong as for its relationship to Selaginella. 
If these three forms are not closely allied then the striking resemblances 
between their cones must be due to one of three causes : — 
1. That their distant common ancestor had a cone of this type, and 
they have descended from it without modification. 
%, Heterogenetic homoeomorphy, i. e. that their cones were once 
distinct, and have reached their present resemblances by convergence during 
their evolution. 
