G. R. Wieland — Williamsonian Tril>c. 463 



(fig. 17C) may quite possibly be unisexual, though the absence 

 of a peduncle leaves the point in doubt. Considerable variance 

 in the explanation of such fossils can arise, whence to avoid 

 confusion the use of the terms cast, mould, surface, and 

 imprint requires to be exact. 



Thus the attempt to interpret a very beautiful and undoubtedly bisjuoran- 

 giate specimen, No. 2406 of the Paris Museum Yates Collection, has led to 

 some difference of opinion between Professor Lignier and myself. In the 

 first place, following my paper of 1901 (12), Lignier (15) described this speci- 

 men as the rugose surface of an ovulate cone with indication of axial pro- 

 longation as a more or less pyriform sterile leafy appendage. To this I took 

 exception in my American Fossil Cycads, p. 152, interpreting Lignier's figure 

 as that of the upper portion of a bisporangiate bud showing the short sterile 

 apex of an ovulate coiie enveloped by the microsporophylls, the curved ra- 

 chides of which it was supposed had broken away so as to expose the proximal 

 synangia en masse just outside the decurved rachidal tips. But later, when 

 visiting the Paris Museum. I found that the figure on which I had based my 

 opinion is not, as inferred from the legend, that of the specimen itself, but 

 of a mould obtained from it. 



Since then Lignier has, in a separate note (19), insisted on the virtual cor- 

 rectness of his original view that the rugose marking is only ovulate, but of 

 course agreeing with me that in any event the series of converging segments 

 iudicates a staminate disk of 20 or 21 microsporophylls (not bracts or parts 

 of a leafy apical prolongation). With a cast of the specimen before me now, 

 as kindly given me by Professor Lignier on the occasion of a visit at his 

 house, I still fail to agree fully with his explanation of this fossil. Nor yet 

 can I be sure of its exact interpretation, whether indicating a portion of the 

 base or the summit of a bisporangiate bad, — and in either alternative, whether 

 or not we see the ovulate surface, or either the adaxial, or finally abaxial sur- 

 face of the microsporophylls, which latter might be rugose as in Cycade- 

 oidea dacotensis. 



To fully discuss the various possibilities of casts and moulds of surfaces 

 involved would require too much space ; but in the appended fig. 18 of the 

 cast (exactly equivalent to the specimen itself) the disk of 20 or 21 members 

 plainly appears, as does also the central axis. The fruit is in my judgment 

 still a " litigious " one, since it is most difficult to read seed characters into 

 the sculpturing on the segments, or to see just how the surfaces of the micro- 

 sporophylls were appressed to the ovulate center. Equally the character of 

 the latter is in doubt, as but a single large seed might have been present ; 

 while Lignier's idea of pyriform extension simply lacks direct evidence. If 

 the rugosity is staminate. a different genus or even family from Williamsonia 

 might be represented. This I believe the more likely. 



In closing this brief resume" of floral characters in William- 

 sonian types, the great extent of sex diversity found fixes our 

 attention less by far as a phenomenon per se than because of 

 the immense variation it not only theoretically permits, but 

 must actually have involved through very long periods of time. 

 The uniform dioecism of existing cycads might be conceived 

 of as long persistent or not ; but not so the bisporangiate stro- 

 bilus, which must have been a potentially bisexual axis far back 

 toward the time when the ancestral forms attained heterospory. 

 And this floral form, as was insisted upon when the first descrip- 

 tions were given, must in itself be exactly the one capable of 

 giving rise to the most varied phases of monoecism and dioecism. 

 As showing the value of reasonable attempts at interpretation 



