256 MR. MIERS ON THE STRUCTURE OF THE SEED 
that of the arillode, as indicated by the interesting researches of Dr. Planchon, requires 
to be confirmed by a series of more extensive and careful observations, before it be admitted 
as a settled fact. The most instructive and conclusive evidence of the origin and subse- 
quent extraneous growth of the arillus over the ovule, has been adduced by Cambessèdes* ; 
he found among the ripe seeds of Casearia grandiflora many that were incomplete; and - 
here, although the anatropal ovule showed itself in a state of complete abortion, the 
arillus had grown over it to its full state of development, proving that where the ovule 
had ceased to grow, the increment of the funiculus was not stopped in its progress of 
extraneous production. - | 
Among the instances cited and figured by Dr. Planchon of the development of the 
arillode, is that of Olusia flava, where he describes its ovules as presenting two short 
cupulary membranaceous appendages, one covering the fourth part of the ovule, the other - 
much shorter and superimposed ; and while he inquires, whether one of these cups be due 
to an expansion of the funiculus, and the other of the exostome, he seems inclined to 
infer that both proceed from an enlargement of the foramen of the ovule. Such an 
inference is opposed to the facts described in the foregoing pages, but his observation is 
worthy of attention, although it is more probable that the appearances he describes are 
those of a true arillus in progress of its development, the exterior swelling being perhaps 
that enlargement which I have described as the stipitate foot of the arillus. According 
to the views of Dr. Planchon, the exterior coating existing in the seeds of the Clusiee 
must be a true arillus, because the micropyle, which I have mentioned as existing near 
the hilum of the testa, is wholly covered by that envelope. It is proper to notice that 
Cambessédes distinctly asserts that the seed of Clusia Criuva is enveloped in its apex by a 
scarlet cupuliform arillus, in the work last referred to (p. 317. pl. 65. fig. 8), where it is 
figured upon the extremity of the seed contrary to that of the hilum, and connected with 
it by a raphe: this assuredly must be an error, made perhaps in transcribing the notes of 
St. Hilaire, who by such swelling probably intended to figure the cupuliform caruncular 
process surrounding the diapyle, which I have shown to exist in the same position in the | 
summit of the testa: that such a mistake evidently exists, is proved by the circumstance 
of Cambessédes having described and figured the arillus in the apex of the seed, instead 
of being at the dase, as it would have been had it been in existence. Von Martius (Nov. 
Gen. et Sp.. iii. 166) describes the ovarium of Quapoya (Schweiggera) as containing 
“ovula basi arillo vaginata,” and Endlicher, upon no other authority than the above, 
states that the seeds of Havetia are “basi arillo subcarnoso lax? cupulæformi cincta,” 
and he assumes upon no better evidence the same in regard to those of Quapoya, which 
in that genus Aublet affirms to be “ pulpa rubra involuta.” y 
The facts which I have here brought together, relative to the structure of the seeds in 
this family, must in a considerable degree change our views of the affinities of the Clu- 
siapec. They serve to bring the Order into close proximity with the Rhizobolacee, a 
relationship founded upon analogies in the floral structure, long ago pointed out by Cam- 
bessèdes +, but now rendered still more evident by the great similarity observed in their 
* St. Hilaire, Flor. Bras. Mérid. ii. 232. pl. 126. fig. 11. 
+ St. Hilaire, Flor. Bras, Mérid. i, 323. 
