— 561 — 
on nearly all the rock fragments with the lattor (14). The two species 
are by far the most abundant in the collection. A large number 
of the seeds, like that seen in Pl. x, fig. 8, are deprived of sar- 
cotest. In fact this condition is so common as to afford ground for 
the conclusion that the outer skin was separated from the sclerot- 
est by a sarcotestal tissue so soft and easily removed as readily to 
permit the escape of the hard contents at an early stage of mace- 
ration. The example, fig. 8, shows the sclerotest in the upper part 
of the seed, though the lower part has suffered lateral deforma- 
tion. Additional seeds are in figs. 3 and 4 of this Plate. 
Specimens of the seeds showing the wing are illustrated in 
figs. 5, 6 and 7. In fig. 7 the apex of a seed with the wing lobes is 
seen on the left, while on the right is shown the lower part of 
another seed exhibiting the broad chalaza and the slight basal pro- 
longation of the sarcotestal wing on either side. The wing is here 
slightly uneven or torn in appearance. The torn or lacerated cha- 
racter of the basal separation of the seed is frequently observed and is 
better illustrated in fig. 6. In theoriginalof fig. 5, it descends along 
the cuneate chalaza which seems to become diffused in passing down 
to the point of fracture. The ordinary aspect of the apical wing lobes 
is shown also in figs. 6 and 7,o0n Pl. viii. In the example drawn 
in fig. 7, Pl. x, the lobes are unusually obtuse and divergent. The 
thin substance and,the wide dilation, especially at the base, of these 
seeds conform to the definition of Samaropsis. 
Seeds of the Samaropsis group are reported from various regions 
of Gangamopteris flora. Among these the forms nearest Samaropsis 
Seivast are those illustrated, without specific name, by Feistmantel (2) 
from the Karharbari of India, and by Johnston, (3) as Carpolithes 
Tasmanicus from the Permo Carboniferous of Tasmania. 
Both of the latter are much larger, theIndian seed being cordate 
and not divaricately emarginate, the Tasmanian also larger, with 
narrower wing of equal width the whole length of the seed. The 
Cardiocarpon(?) sp. figure by Feistmantel (4) from the Karharbari 
presents a very marked difference in the proportions of the wing and 
in its adaptation to the nucellus. 
(1) See pl VIII, figs. 2, 5, 6, 7. 
(2) Fl. Gondwana Syst., vol. iii, 1881, p. 59 pl. XXVII, fig. 8. 
(3) Geol. Tasmania, 1888, pl. VII, fig. 7 (not descrile!). 
(4) Op. cit. pl. XXIV, fig. 5. 
5560 60 
