G. i?. Wieland — American Fossil Cycads. 435 



forms of heterospory and the acquiring of the seed habit by 

 types immediately ancestral to the cycads or other gytnno- 

 sperms, if not also to the angiosperms. 



It is, however, if we clearly apprehend the significance of 

 the above described structures, as hitherto held, difficult to 

 believe that the existing cycads could have been derived 

 directly from the Bennettitacese. The evidence at present 

 wholly favors the belief that the former represent a line of 

 descent from an ancestral stock common to both, which under- 

 went alteration of both macro- and microsporophylls on elon- 

 gate axes resulting in true cones, with the exception of the 

 macrosporophylls of Gycas, which have retained a quite primi- 

 tive type and position. On the contrary, in the Bennetti- 

 taceae, it is the microsporophylls which remain primitive, while 

 the macrosporophylls have undergone an extreme of altera- 

 tion and consolidation, albeit on ancient lines of development 

 finding their nearest known analogy in Cordaites, and perhaps 

 Ginglco. It appears from the facts now at hand that both 

 cycadean branches may not haYe passed through precisely the 

 same evolutionary sequence of hetorospory, bi- or mono- 

 sporangiate, monoecious, and finally dioecious fructification. 



It now becomes possible for the first time to consider the tax- 

 onomic position of the living and fossil cycads in the light of 

 a clear knowledge of those structures and characters of funda- 

 mental importance in classification. It is quite evident that 

 there is more resemblance between these two important groups 

 than would be indicated by classification as the Cycadales and 

 the Bennettitales. The lateral fructification of the latter 

 group, while a distinctive feature, is not a profound difference. 

 For if Williamson's restoration of Zamia gigas be correct, as 

 now appears probable, this genus did not bear lateral fructi- 

 fications, and is hence in this respect intermediate. Nor are 

 the frequent adventitious buds and branches of certain living 

 cycads to be overlooked. 



On weighing the new evidence as carefully as possible, and 

 bearing in mind that the staminate fronds of Gycadeoidea 

 form a connecting link qualitatively of the same value as the 

 carpophylls of Gycas, it becomes wholly clear that these fossils 

 are true cycads and form an order, or perhaps preferably a 

 sub-order, of equal rank with the living cycads. The classifi- 

 cation already suggested by Professor Scott, I, . therefore, 

 believe to be a natural one. He says in his Studies in Fossil 

 Botany, p. 474, " It appears then that there are at present 

 known to us three distinct families of Cycadales. On the one 

 hand the Zamiese and the Cycadese, still existing, and together 

 constituting the order Cycadacege ; and on the other the Ben- 



A.M. Jour. Sci. — Fourth Series, Vol. XI, No. 66.— June, 1901. 

 30 



