MR. JOHN MIERS ON THE LOASACEZÆ. 231 
In Seyphanthus, the structure of the seed is similar to that of Loasa, that is to say, it 
js anatropous, of an oval shape, with a short terminal strophiolar expansion; it is opake 
and deeply pitted or foveolated; and when this outer tunic is removed, the ridges which 
separate the external pits just mentioned are found to consist of cancellated bars, with 
intervening pellicular spaces, analogous to the outer integument of Gripidea; but, as in 
all the previous cases, there is no visible raphe, nor any appearance of a chalaza, either 
upon it or on the delicate integument which invests the albumen. 
In Raphisanthe the seed is of an oval form, rendered trigonous by three narrow longi- 
tudinal ridges; the outer integument, as in some species of Loasa, is very thick, with a 
minutely bullated surface, being rigid and brittle, formed of numerous corneous cells, and 
easily softened in boiling water; it has a strophiolar projection in its apex: the inner 
capacity of this thick tunic has twice the length and nearly double the breadth of the 
inner body seen within it; this inner body is of an oblong shape, and consists of the 
albumen, covered by a pellicular and finely retieulated integument, with a black apical 
mieropylar point, by which it is attached to the strophiolar process, so that it is thus 
suspended in the summit of the vacant space; and it has another minute dark speck at 
is lower extremity, which is far removed from the bottom of the cell of the outer integu- 
ment—a fact of considerable importance in this inquiry; it is covered by an extremely 
pellicular white intermediate envelope, as in the preceding genera, but which in this 
instance, rather adheres to the outer tunic; this envelope, from its great tenuity, might 
easily escape observation. It is worthy of remark, that there is no appearance of a raphe 
R any of these integuments. This genus has a spiral cylindrical capsule, much resem- 
bling that of Gripidea; but its placentation is nearly that of Caiophora, that is to say, 
tt has two lamellar seminiferous plates upon each of the three spiral lines of placentation, 
which nearly reach the centre of the cell. This genus differs from Gripidea in the from 
cee in the number of its stamens, its placentation, and in the structure of its 
In Ancyrostemma and Mentzelia the seeds are oval; in the former, suspended by a 
— rnp hiolar process, which looks like a transparent membranaceous funicle ; their 
Coating, as in Gripidea, is foveolated and divided by prominent bars into elongated 
ia oidal areoles, the interspaces being formed of very reticulated membrane; this 
— as in Scyphanthus, closely invests an extremely delicate inner integument, with 
skin smaller square reticulations, which give it the appearance of being transversely 
| ited, but there is no vestige of any tracheal vessels in either of these translucent 
ther, > Upon the inner one, which is perfectly white, there is a small black spot at the 
diy one at the base so very minute as to be scarcely visible. In both genera, the 
© shaped as in Scyphanthus, is imbedded in the axis of the albumen. 
a = 
M hop Penh was proposed by Lilja, in 1841 (Linnea, xv. 263), upon the Loasa lateritia, Hooker ; 
it er » acknowledged by botanists. Klotzsch was evidently wrong in making it identical with Caiophora; 
i » “Valid genus. I have examined another plant, which corresponds in structure with the type, and 
2 Rank nd species, viz, — 
h » s ks . > . 
lera de Maule e ss :— Loasa Ochagaviæ, Philippi (Linnzea, xxviii. 641). Chile, v. s. in herb. meo, Cordil- 
