F. IV. Oliver. 
148 
(endotesta) much specialised towards the apex of the seed. 1 Below 
the level at which the nuccllus ceases to he free the canopy dies 
out. The tissue which originally occupied the chambers of the 
canopy is not preserved with the exception of a single tracheal 
strand (v) in each chamber. These strands are the upward pro¬ 
longations of vascular strands which run in the zone common to 
integument and nucellus and which take their origin from the 
supply bundle which enters the seed at the chalaza. It is not, 
however, the purpose of the present paper to elucidate the obscure 
functions of the canopy or to consider the homologies of the 
vascular strands. 
Pachy testa. 
The immense seeds which attained the form and dimensions 
of a hen’s egg, were amongst the most remarkable that have been 
preserved from pernio-carboniferous times. Our knowledge of their 
organization is due to the investigations of Brongniart 2 and 
Renault. 3 Separated from the thick testa by a considerable interval 
there stands the nucellus, raised upon a conspicuous and peculiar 
chalazal peduncle. The nucellus just as it narrows down to this 
peduncle is produced considerably beyond the place of its insertion, 
so that in a longitudinal section the nucellar base might be de¬ 
scribed as cordate or even reniform. At the summit of the nucellus 
is a large pollen-chamber, but within the body of the nucellus no 
prothallium has been found preserved. The vascular supply enters 
the seed at the base, giving off two successive series of bundles as it 
traverses the testa. These are very numerous and form a double 
system near the outside surface of the exotesta. After its passage 
through the nucellar peduncle the bundle expands into a tracheal 
plate, the margins of which are continued around the periphery of 
the nucellus as a number of distinct bundles. The remarkable 
features of the seed reside in that part which lies between the 
nucellus and the exotesta, and are only realized on examining 
transverse sections. 
1 Where the seed-coat is differentiated as a hard outer layer 
with an inner (often soft) lining, the terms exotesta and 
endotesta are employed ; where, as in Cycas , there is a fleshy 
outer layer, an intermediate stony layer, and a soft inner 
layer, they may be distinguished as sarcotesta, sclerotcsta 
and endotesta. Whether an aril, when present, is homo¬ 
logous with sarcotesta is an open question. 
2 Fes graines fossiles silicifiees, p. 25, pi B., figs. 4 and 5, and 
pi. xviii.—xxi. 
3 Flore fossile d’Autuu et d’Fpinac, p. 389, and pi., lxxxiii. 
and lxxxiv., figs. 1 and 2. 
