568 TRANSACTIONS OP SECTION K. 



The examination of the nucelhis and integument of older ovules of Bowenia 

 suggests that the inner vascular system is to be regarded as nucellar and not 

 integumental — a further point in favour of comparison with seeds of the Medul- 

 bseae. 



Stages in the development of spermatozooids have been studied in Bowenia 

 which show that this genus agrees with more fully studied Cycads. 



2. The Structure of a New Type of Synangium from the Calciferous Sand- 

 stone Beds of Pettycur, Fife, and its bearing on the Origin of the 

 Seed. By Miss M. J. Benson, D.Sc. 



The Synangium is attributed on structural grounds to Heterangium Grievii, 

 and probably represents the pollen synangium of that plant. It differs from all 

 hitherto described synangia in the variety and large proportion of its sterile 

 tissue. This sterile tissue shows the sclerotic plates characteristic of the inner 

 cortex of Heterangium Grievii. Numerous vascular bundles with hydathodal 

 ends occur, and irregular longitudinal dehiscence was brought about by the 

 swelling of hygrosoopic fibres. Another wholly new feature is the occurrence 

 of both central and peripheral loculi (four central and twelve peripheral). The 

 loculi agree in size and form with those of the incrustation fossil, Diplotheca 

 stellata (Kidston), which is identified as the same synangium in a dehisced 

 phase. The discovery of the structure of this early synangium adds fresh con- 

 firmation to the synangial theory of the seed, which may be restated as follows : 

 The Palaeozoic ovule of the Lagenostoma type may be regarded as the product 

 of the elaboration of a synangium comparable with the above — the megaspore or 

 embryo sac being derived from the central group of loculi, and the canopy and 

 peripheral part of the ovule from the peripheral part of the synangium with its 

 envelope, twelve loculi, cortical tissue, and vascular bundles. 



3. A Palaeozoic Fern and its Relationships (Zygopteris Grayi, Williamson). 

 By Dr. D. H. Scott, JP.i2.iS., Pres. L.S. 



The simpler Palteozoic Ferns (Primofilices of Mr. Arber, Ccenopterideas of 

 Professor Seward) have received much attention of late, especially in the fine 

 memoirs of M. Paul Bertrand. 



Zygopteris Grayi, a species founded by Williamson in 1888, on somewhat 

 imperfect material, occurs both in roof and seam nodules of Lancashire coal- 

 beds, but is very rare. Besides the specimens described by Williamson, there 

 is a much better one, the sections of which are partly in his collection ; this has 

 been figured by the author in 1900, by M. Paul Bertrand in 1909, and by Mr. 

 Kidston in 1910, but never adequately. 



Last year a fine series of sections of an entirely new specimen from Shore, 

 Littleborough, was received from Mr. Lomax, and forms the basis of the present 

 communication. 



The new specimen shows the general characters of the Z. Grayi type : a 

 five-rayed stellate stele, the corresponding § phyllotaxis, leaf-trace bundles with 

 axillary shoots, scale-leaves or aphlebiae, and adventitious roots. The charac- 

 teristic internal xylem, consisting of narrow tracheides embedded in paren- 

 chyma, is particularly well shown, both in the main stem and in the axillary stele. 



This specimen affords clear evidence that it belongs to the genus Ankyropteris, 

 as defined by P. Bertrand. The leaf-trace and foliar bundle show perfectly 

 the peripheral loops of small-celled xylem characteristic of Ankyropteris. The 

 loops begin to be differentiated long before the leaf-trace separates from the 

 stele. This confirms P. Bertrand's own view ; he found peripheral loops in 

 Williamson's specimens, where, however, they are usually very obscure com- 

 pared with those in the Shore plant. 



Stenzel's species, Z. scandens, which Williamson was at first inclined to 

 identify with his Z. Grayi, is probably also an Ankyropteris, but does not 

 appear to be identical with the Shore fossil. 



The latter is somewhat peculiar in the form of the leaf-trace, which is 



