AFFINITIES OF BALANOPHORE^. 19 



made by authors, probably in all cases arises from their having only examined immature 

 specimens. 



The individual cells of the homogeneous embryo are angular, with very thick transparent 

 waUs, and small cavities filled with a few chlorophyll-granules. I have never found starch 

 in the embryo of any species ; the contents of their cells being browned by iodine. Oil 

 abounds in the exalbmninous species, and in the embryos oiCijnomoriuin Sinmii/stropetalon. 



Hitherto the true nature of the exalbuminous, so-called exembryonate seed of ^«Za«o- 

 phorce has eluded aU research; nor till its germination has been traced*, is it probable 

 that this point will be satisfactorily cleared up. In the mean time it may involve less of 

 hypothesis to assume that the embryo is a homogeneous mass, in so far as any evident 

 distinction of cotyledons and radicle is concerned, than to regard it as an albumen in which 

 the embryo has not yet been discovered. Much may be said on both sides of this ques- 

 tion ; for instance, analogy with Cynomorium, in which the embryo is oily and the albu- 

 men not so, is in favour of the seed of Balanophora being considered to be embryo ; on 

 the other hand, if the cellular ovule of Balanophora so perfectly resembles the embryo of 

 Cynomorium, it appears reasonable to conclude that the albumen of Cynomorium is endo- 

 spermic, and developed in the same delicate sac with the embryo itself; against which view 

 there appears no theoretical objection!. 



* Impressed with the great importance of this point, I endeavoured, when in India, but uniformly in vain, to induce 

 the seeds of Balanophora and RhopalocHetnis to germinate. 



f Amongst the many Natural Orders whose homogeneous seeds or embryos present more or less analogy with those 

 of Balatiophoiece, none have so close a similarity as those of THuridece. I have examined a species of this Order iu 

 a living state in the Khasia Mountains (East Bengal) : its ovule (which has not hitherto been described) is manifestly 

 anatropous, and consists of one integument and nucleus ; offering one out of many proofs that the structure and posi- 

 tion of the oride in no degree influence the after-development of the embryo : in other words, that the development 

 of the embryo, so far as its form and structure are concerned, is in a great measure irrespective of the presence or 

 absence of envelopes to the embryonary sac. 



Mr. Miers, in his valuable and elaborate paper on THuridece (Linn. Soc. Trans, xxi. p. 51) considers that it is con- 

 sistent with the simplicity of the structure of other parts of the plants belonging to that family, " to expect a nucleus 

 equally simple in its nature, formed merely of an aggregation of cytoblasts, which, imder favourably-exciting circum- 

 stances, are endowed with the faculty of self-development." The true nucleus of the ovule in Triuridece is however in 

 no way different in structure or position from that of ordinary Phsenogamic plants, from which it follows that although 

 the embryo appears amorphous, its radicular extremity must be a determinate point with relation to the seed, and that in 

 germination that end will elongate, and perform the function of the roots. The term " Protoblastus," therefore, as indi- 

 cating an embryo that germinates from no determinate point, cannot under this view be adopted for that oi Triuridece, 

 though, if it were proved that the germination oi Balanophorm (the structure of whose ovules does not reveal the posi- 

 tion of the radicle) were from an indeterminate point, it might be more appHcable to them. In both Cynomorium and 

 Mystropetalon however, the radicular end of the embryo is very evident, and as there can be no doubt that the embryos oi 

 most or all Balatioplwreee germinate whilst still within the pericarp, it may be inferred that the radicle will protrude 

 from a given ruptured point of the latter, and not mdifferently through any part of its walls. Under these circum- 

 stances, I hesitate to adopt a term, which, in the present state of the inquiry, and as far as regards this order, implies, 

 not that the germmating point is indeterminate, but that this, and the whole process of germmation, are absolutely 

 unknown. 



With regard to other embryos which would come under the definition of a Protoblastus, that of Orchidea evidently 

 germinates from a given point ; and Caspari's beautiful observations on Orohanche show that the same is the case 

 in that genus. Blume describes the embryo of AmorphophaUus as throwing out plumulary leaves from several 

 points at once, which probably indicates a development of several much-reduced intemodes crowded together, 



i>2 



