234 MR. JOHN MIERS ON THE LOASACEÆ. 
subject; but his conclusions are ambiguous, and may be questioned with much reason, 
He agrees with others in regard to their anatropous development ; at least, he says * that 
the ovules of Loasa, Caiophora, and Blumenbachia are pendulous and “ epitropous” (a 
term employed by him to denote an ordinary anatropous ovule with a ventral raphe, in 
contradistinction to his term “apotropous’” where, as in Zlex, the pendulous ovule has 
a dorsal raphe +); he states, in addition, that in Microsperma (Eucnide) and in Bar. 
tonia the ovules are heterotropous, a term now seldom used, but which implies that the 
hilum, or point of its attachment, is placed midway between the micropyle and cha- 
laza, or, in more ordinary language, amphitropous. This supposition is at variance with 
the drawing of the seed of Microsperma, given by Sir Wm. Hooker in his ‘Icones; 
plate 234. fig. 5; Zuccarini also says that the seed in this genus has an orthotropous 
embryo, with flat linear cotyledons in the axis of albumen. Prof. Agardh is equally 
mistaken in regard to the seed of Bartonia, for the structure explained in the foregoing 
details is quite opposed to its heterotropous development. In the work just quoted, he 
shows the figure of an ovule of Caiophora (plate 18. fig. 8), where it is anatropous, with a 
ventral raphe, which latter feature is not mentioned in the text; he adds another of 
Loasa (fig. 7), which is anatropous, without any indication of a raphe: in the explana- 
tion of these figures, he repeats that the ovule is nearly straight and heterotropous in 
Microsperma, curved and epitropous in Caiophora, adding that it is amphitropous in 
Blumenbachia, and nearly campylotropous in many species of Loasa. These declara- 
tions must be held doubtful, until they are supported by other authority ; they are 
certainly not consistent with the diagnoses of the several genera of Loasacee in Endli- 
cher's * Genera Plantarum,’ where in every case the embryo is said to be orthotropous, 
or only slightly curved, and anatropous, which I have found to be true in every instance 
that has fallen under my notice, and I have shown in the previous details that the seed 
is not amphitropous in Blumenbachia. 
It may also be noticed that, in 1823, when few species of Loasa were known, Schrader 
divided the genus into two sections :—1°, where the seeds have an arillus, the radicle 
pointing to the hilum; 2°, where they have no aril, the hilum being lateral. I have 
never met with a seed in this family with a lateral hilum; one instance only occurred " 
me where the radicle was slightly curved, but there it still pointed to the hilum, which, 
correspondingly, was removed from the geometrical apex of the seed—a result which ] 
attributed to the effect of pressure during growth. 
We may hence infer, as a general rule, the exceptions to which are doubtful, that 
the ovules and seeds in the Loasaceæ are simply anatropous, with integumental jen 
ings which have no visible raphe, that they are either horizontally attached to the ze 
or suspended from it, and that the embryo is constantly orthotropous, or Very mam 
curv: ed in the axis of albumen, with the radicle turned towards the hilum. i 
Poeni that Gripidea differs in many essential respects from all other QE 
y, t now proceed to enumerate its characters. 
* Than. i 
Theor. Syst, p. 262, + Ibid. Introd. p. Ixxiv. 
