172 THE EVOLUTION OF PLANTS 



retained permanently within the sporangium, are 

 essential characters of a seed. The fertile leaf, 

 however, formed part of the seed-like organ, 

 and the whole was shed together, like the fruit of a 

 Composite. 



It is curious to find so great an advance in 

 reproductive methods, so nearly approaching 

 the highest types, in a little inconspicuous Lyco- 

 pod of Carboniferous age. If, as seems fairly 

 clear, Miadesmia was one of the Selaginellacese, 

 it rose to a higher level, as regards reproduction, 

 than any known member of its family has done 

 since certainly higher than any of those now 

 living. Here, as in so many groups, the less 

 advanced forms have persisted while the more 

 advanced have disappeared. It must be remem- 

 bered that at that time the seed-method of re- 

 production, though already common, was not so 

 firmly established as it became later. The 

 constant absence of an embryo in Palaeozoic seeds 

 suggests that not all the possibilities of the seed- 

 habit had then been realised. Under such cir- 

 cumstances, it may be that there was more scope 

 for experiment in the seed-direction, and more 

 chance of successful competition with the main 

 line of Seed-plants, than at a later time. 



As regards the particular conditions which 

 favoured the seed-habit in the case of Miadesmia, 

 we are left to conjecture. The specimens of the 



