CHAP. ri. 
SEED, 
•207 
viz. 1. the external integument, tunica externa of Willde- 
now, testa of De Candolle; 2. the internal integument, 
tunica interna of Willdenow, endopleura of De Candolle, 
hilofere and tegmen of Mirbel; and, 3, of an intervening 
substance answering to the sarcocarp, and called sarcodermis 
by De Candolle : this last is chiefly present in seeds with a 
succulent testa, and by many is considered a portion of the 
outer integument, which is the most accurate mode of under- 
standing it. 
The Older integument is either membranous, coriaceous, 
crustaceous, bony, spongy, fleshy, or woody; its surface is 
either smooth, polished, rough, or winged, and sometimes is 
furnished with hairs, as in the cotton and other plants, which, 
when long, and collected about either extremity, form what is 
called the coma (sometimes also, but improperly, the pappus). 
It consists of cellular tissue disposed in rows, with or without 
bundles of vessels intermixed: in colour it is usually of a 
brown or similar hue: it is readily separated from the inner 
integument. 
In Maurandya Barclay ana it is formed of reticulated cel- 
lular tissue; in CoUomia linearis and others it is caused by 
elastic spirally twisted fibres enveloped in mucus, and spring- 
ing outwards when the mucus is dissolved ; in Casuarina it 
(or the inner integument) contains a great quantity of spirally 
fibrous cellules. In the genus Crinum it is of a very fleshy, 
succulent character, and has been mistaken for albumen, 
from which it is readily known by its vascularity. According 
to Brown, a peculiarly anomalous kind of partition, which is 
found lying loose within the fruit of Banksia and Dryandra, 
without any adhesion either to the pericarp or the seed, is a 
state of the outer integument. It is said that in those genera 
the inner membrane (secondine) of the ovule is, before fertili- 
sation, entirely exposed, the primine being reduced to half, 
and open its whole length; and that the outer membranes 
(primines) of the two collateral ovules, although originally 
distinct, finally contract an adhesion by their corresponding 
surfaces, and together constitute the anomalous dissepiment. 
But it may be reasonably doubted whether the integument 
