56 



SCIENCE. 



[N. S. Vol. XX. No. 497. 



suits of the two methods the presence of P. 

 campestris is shown on the seeds of all four 

 of the seed plants examined. 



Full details of these studies will soon appear 

 in Bulletin No. 261 of the New York Agri- 

 cultural Experiment Station. 



H. A. Harding, 

 r. C. Stewart. 

 Geneva, N. Y. 



a notable paleobotanical discovery. 



The term Cycadofilices proposed by Potonie 

 for the group of synthetic genera, which in 

 Paleozoic times were abundant and widely dis- 

 tributed, and which so curiously combine 

 filicinean and cycadean anatomical characters 

 with foliar organs entirely fernlike in form, 

 seems likely to become a penuanent acquisition 

 to taxonomy; although some students, follow- 

 ing so eminent an authority as Zeiller, still 

 regard the Cycadofilicales as merely a special- 

 ized group of ferns. 



The announcement a few months ago* by Pro- 

 fessors Oliver and Scott that the gymnosperm- 

 ous seeds named by Williamson Lagenostoma, 

 were borne by a species of Lyginodendron is 

 most interesting. This identification is based 

 on the identity of the capitate, often staUved, 

 glands, which clothe the enclosing envelope, 

 with those clothing the vegetative organs of 

 Lyginodendron Oldhamium. These glands 

 show the closest possible agreement in size, 

 foiTQ and structure, and no other known Car- 

 boniferous plant possesses them. In addition, 

 the vascular system of the enclosing cupule 

 was well developed, the structure of the large 

 bundle agreeing with the petiolar strand in 

 Lyginodendron; and the minute characters in 

 the former are in close agreement with those 

 of the xylem of the foliar organs of the latter, 

 so that the evidence presented is • fairly con- 

 clusive. 



Now Lyginodendron with a number of 

 species under a variety of names ranges from 

 the lower Carboniferous into the Permian. 

 The calcified remains, often beautifully pre- 

 served and including roots (Kaloxylon) , stems 

 (Lyginodendron) and leaves (Sphenopteris) 

 have been associated with the carbonaceous 



*Read before the Eoyal Society, May 7, 1903, 

 and reprinted from the Proceedings. 



impressions of large finely divided fronds of 

 the Sphenopteris type. 



The minute striicture of these parts is as 

 well known as in living forms, as are the 

 corresponding structures in Heterangium, an- 

 other genus, considered in a general way an- 

 cestral to Lyginodendron and exhibiting a 

 balance of filicinean characters; while in the 

 latter, though many filicinean characters re- 

 main, the eycadean features are more pro- 

 nounced, the primary bundles are isolated and 

 arranged around a large pith, the vascular 

 bundles both primary and secondary are sim- 

 ilar to those of the peduncles in living cycads, 

 and the roots, while agreeing with those of the 

 Marattiacese when young, after their second- 

 ary thickening are entirely gymnospermous. 



The fructifications of this curious plant 

 have hitherto remained unknown, although 

 certain poorly preserved iilicinean remains of 

 the Galymmatotheca type have been so re- 

 garded.* 



The structure of the seed as now set forth 

 furnishes a preponderance of gymnospermous 

 features, at the same time emphasizing the 

 combination of transitional characters which 

 distinguish the Cycadofilicales, and furnishes 

 good ground for supposing that many of the 

 plants which furnished the elegant fronds of 

 AlethopteriSj Neuropteris,-\ Sphenopteris, etc., 

 had already advanced in seed, bearing as far 

 as, if not farther than, the modern Cycads. 



The seed under consideration, as does the 

 similar one of Jjagenostoma ovoides, ap- 

 proaches the cycadean type in that the integu- 

 ment and nucellus are distinct in the apical 

 region only, where the former becomes massive 

 and complicated, consisting of nine chambers 

 radiating from the micropyle. The free por- 

 tion of the nucellus tapers upward reaching the 

 exterior and plugging the micropylar opening; 

 the separation of the nucellar epidermis from 

 the underlying parenchymatous body of the 

 free part of the nucellus forms a bell-shaped 



'■'Described by Benson (Ann. Bot. 18: 161-177, 

 pi. 11, 1904) and regarded as the microsporangial 

 sori of Lyginodendron. 



t In a recent note Professor Oliver ( New Phy tol- 

 ogist, 4: 32, 1904) records seeds on the fronds of 

 Neuropteris heterophylla. 



