I IO 
Notes . 
The seed, as is well known, possesses a very thick two-layered testa, the inner 
zone of which, the endotesta or sclerotesta, consists of sclerotic tissue many cells 
in thickness; its more external elements are commonly elongated in the radial 
direction. The sclerotesta bears three sharp and prominent longitudinal ridges 
on its outer surface with corresponding furrows on the inner side, marking the 
position of the three sutures; in each space between two main ribs are two, or 
more commonly three, less prominent ridges, the whole number of projections 
thus varying from nine to twelve. The degree of prominence of the various ridges 
is very variable ; in some cases only the three main ribs are clearly marked. The 
thickness of the sclerotesta between the ridges is about 1-1*25 mm * 
On the outside, the sclerotesta is abruptly succeeded by the delicate tissues 
of the sarcotesta, the inner layers of which consist of a continuous thin-walled 
parenchyma, passing over towards the exterior into a lax tissue, with large and 
definite lacunae between the cells. This zone, owing to its weak construction, 
is seldom well preserved, and its usually collapsed condition prevents us from giving 
any accurate estimate of the total thickness of the sarcotesta in the natural state, 
which must certainly have exceeded 2 mm. The lacunar zone had not previously 
been distinguished. The whole is bounded externally by the limiting layers to 
which Mr. Wild first called attention. They consist of a hypoderma formed of 
one or sometimes tw r o series of columnar cells, resembling a palisade-tissue. Beyond 
this is the true epidermis, composed of small, often nearly square cells. This layer, 
which had not been noticed before, appears to have been always present, though 
it is only here and there that it is found well preserved. Mr. Wild compared the 
limiting layers of the sarcotesta in Trigonocarpon as then known (i. e., without 
the small-celled epidermal layer) with the similar tissues on the exterior of the 
petiole in Medullosa anglica, and indicated the general agreement in structure between 
the two. It must, however, be pointed out that in the petiole the small-celled 
epidermal layer has not yet been detected, and it appears that the epidermis itself, 
as well as the hypoderma, had a palisade-form. The agreement between the two 
structures thus seems, on our present knowledge, less close than Mr. Wild believed 
it to be. 
The very long micropylar tube, triangular in section in its lower part, is bounded 
immediately by thick-walled tissue forming the prolongation of the sclerotesta ; 
outside this there is a great development of the sarcotesta, appearing, as seen in 
the transverse sections, as a wide flat wing on either side of the micropyle (see 
the series of figures in Mr. Wild’s paper). This is limited externally by the 
usual hypoderma and epidermis. How far the flattened, wing-like form was 
natural is difficult to determine. There has certainly been some loss of tissue, 
so that the form in the natural state was less flattened than it now appears. It 
is doubtful, however, whether in this part of the seed the sectional form could 
ever have been circular. 
One of the principal points to which our attention has been directed, is the 
distribution of the vascular supply through the seed. We find that there was 
a double bundle-system present. The outer system is situated in the sarcotesta, 
the bundles lying opposite the smaller ridges of the sclerotesta ; they have 
