470 MORPHOLOGY OF THE ANGIOSPERMS 



have been established very early in angiosperm history, or was already 

 present in ancestral stock.) Sporangial development is eusporangiate in 

 angiosperms and the more primitive ferns. 



The description, in 1956, of apparently bisexual fructifications in the 

 fernlike Mesozoic genus Glossopteris, commonly considered a pterido- 

 sperm, may be important to the theory of angiosperm derivation from 

 fern stock. Though details of the reproductive structures are still un- 

 known, there seems no question but that there are two kinds in each 

 fruiting structure, and resemblance of the Glossopteris leaf to simple 

 fern leaves is close. Glossopteris is perhaps closer to the eusporangiate 

 ferns than to the pteridosperms; if so, it may form a link between the 

 ferns and the angiosperms. 



Ancestry for the angiosperms should probably be sought far back in 

 the eusporangiate line, as far as the Permian, when beetles are first 

 known. (The Permian flora was rich in pteridophytes and pterido- 

 sperms.) Seed bearing is perhaps not an essential character in the 

 ancestral forms. (Seedlike structures have developed independently in 

 several major vascular lines.) It is difficult to interpret the simple ovule 

 of the angiosperms in terms of the complex pteridosperm seed, and the 

 pteridosperms should perhaps be dropped as a possible ancestral stock 

 of the angiosperms. 



Further evidence for great age in the angiosperms comes from 

 ontogeny, from the story of carpel closure and sealing. The living angio- 

 sperms show all stages in the closure of the carpels; many that appear 

 closed are unsealed or sealed only in the late stages of carpel matura- 

 tion. In many taxa, the carpels are congenitally fused; they develop 

 from ring-shaped primordia. Carpel sealing, with the establishment of a 

 primordium of an entirely different shape, is apparently a slow process, 

 slower than reduction in ovule number and the development of the 

 achene, because the achenes of Sparganium and Flataims are still un- 

 sealed at flowering. 



The search for the ancestral stock of the angiosperms must continue. 

 Progress has been made in the twentieth century, though it is chiefly 

 in the elimination of many suggested ancestral lines. 



BIBLIOGRAPHY 



Phylogeny 



Andrews, H. N.: On the stelar anatomy of the pteridosperms, with particular ref- 

 erence to the secondary wood, Ann. Mo. Bot. Card., 27: 51-118, 1940. 



Arber, A.: The interpretation of the flower: a study of some aspects of morpho- 

 logical thought, Biol. Rev., 12: 157-184, 1937. 



Arber, E. A. N., and J. Parkin: The origin of angiosperms, Jour. Linn. Soc. Bot., 

 38: 29-80, 1907. 



