MICHIGAN ACADEMY OF SCIENCE. 147 



titineae, however, show so mauy points of similarity that Wiohmd'* 

 ill lyuO called attention to this group as being possibly related to 

 the ancestors of the Authophyta. It was Arber and rariviu in 1*JU7'^ 

 who elaborated this suggestion of Wielaud's and worked out an 

 elaborate hypothesis as to the ancestry of the Flowering Plants 

 from the vicinity of this interesting group. The more inii)ortant of 

 these suggestions are given below with the addition of a few minor 

 points by the writer. 



As revealed by the marvellously patient and detailed investiga- 

 tions of Wieland, we find the Bennettitineae to have been, in the 

 main, woody i)lants with short, stout trunks, unbranched or with 

 the crown divided into several very short branches. In general 

 structure, the trunk as well as the leaves resembled closely the 

 modern Cycads. Not all of the group, however, were unbranched 

 or shortly branched, for one species {Willi<i)nsonia august i folia) 

 has been found in which the plant is much branched (PI. XIII, 

 fig. 12). It is noteworthy that in this type the leaves are much 

 smaller and not so deeply pinnately-cleft, wdiich fits in Avith Arber 

 and Parkin's theory that with the acquirement of a branching habit 

 megaphylly becomes transformed to microphylly. It is, however, 

 the reproductive strobilus that is the most interesting, (PI. XIII, 

 fig. 13.) This consists of an axis arising in the axil of a foliage 

 leaf. The up])er parts of the axis is occupied by spirally arranged 

 structures tenninating in ovules and interspersed with sterile bracts. 

 Apparently, we have here much reduced megasporophylls, each with 

 f-ne ovule, the bracts perhaps representing megasporophylls that 

 have become sterile and used for protection purposes. The seed is 

 dicotyledonous. (PI. XI Y, fig. 14.) Below the ''gynoecium" is a 

 whorl of microsporophylls, just where the stamens belong in a 

 rtower, and lielow these several to many large spirally arranged 

 bracts which overtop the "flower," forming a sort of "perianth." 

 The microsporophylls are of great interest as they point forward by 

 their jjosition in the "flower" to the flower of the Magnolia type 

 while by their structure they point backward to the Pteridospermeae 

 or "Seed Fenis." Each microsporophyll is a large, pinnately-com- 

 pound structure along whose pinnae are arranged numerous pairs 

 of microsporangia. The Avhole sporophyll is very fern-like and re- 

 veals an undoubted relationship of the Bennettitineae to the fern- 

 like, woody Pteridospenneae. In this latter group, however, the 

 strobilar arrangement of the sporophylls is lacking, and the mega- 

 .an<l microsporophylls are more nearly of the foliage-leaf t^vpe of 



18. Wieland. G. R. American Fossil Cycads. 190G. 



19. Arber. E. N. and Parkin. .T. On the Origin of tlie Angiosperms. 1907. 



