AFFINITIES OF BALANOPHORE^. 7 



point ; that of inducing such a diseased action on tlie root as gives the appearance of 

 the latter forming growths witlun the rhizome is a secondary one, and varies in amount ; 

 from Lojihoplujtum, in whicli it seems to be none, to Rhopalocnemis, in which tlic bulk of 

 the vessels in the rhizome are confluent with those of the root. As a general rule, the 

 older the root attacked by the parasite, the fewer are the branches which it appears to 

 send into the parasite ; and as all my specimens of LopJiophytum are on very much older 

 and larger roots than are those of any other species, and indeed on wood of many years' 

 growth, it is quite possible that in the case of its attacking younger and feebler roots it 

 may develope the same power. 



Under this view, the propriety of considering the rhizome of Balanophora and its con- 

 geners to be an intermediate body, as suggested by linger, seems, as Gocppert has pointed 

 out, to be erroneous ; indeed, there are stronger objections to it than have hitherto been 

 urged, derived from the development of that body. 



I am imable to confLrm Gceppert's observations on one extremely difficult point, namely 

 the presence of the two wholly iudependent and unconnected systems of vascular tissue. 

 This author maintains, 1. that no free vascular bundles origruate in the rhizome previous 

 to the formation of flower-buds, but that the root gives off bundles to the rhizome, within 

 which they ascend, prolonging, increasing in diameter, and branching, M'ith the corre- 

 sponding development of the celltdar system of the rhizome. 2. That on the formation 

 of the peduncles (floral organs), free and independent vascular bundles are developed in 

 them, which ascend as the peduncles elongate, and also descend into the rhizome, occu- 

 pying a position between the vascidar bundles of the latter, -with which they do not unite. 

 3. That these independent vascular systems present anatomical characters by which they 

 may constantly be recognized, at any rate in the individual species. These positions I 

 shall examine consecutively, premising that it is with considerable diffidence that I venture 

 to dissent from the conclusions of tliis eminent author, since though I possess the advantage 

 of having repeated my observations, both on living and dead plants of several species, I 

 cannot regard these as entitled to more consideration than M. Gceppert's known skfll 

 and accm'acy*. 



1. "With regard to M. Gceppert's first observation, it must be remembered that he never 

 had the opportunity of examining very young specimens, the importance of which 

 desideratum he fully admits. In the section in Plate VI. figs. 7 & 8, which represents the 

 independent formation of vascular tissue in a germinated Balanopliora iniolucrata (and 

 in other similar cases), I find in the axis of the rhizome pale transparent lines consisting 

 of elongated cells, which contain no wax or cytoblasts, surrounding rudimentary vascular 

 bundles. I have never examined a very young specimen in which these bundles were 

 found to have descended to the vascular system of the root, but I infer that they do so, 

 and, becoming incorporated with the vascular bundles of the root, present the appearance 



* The diflBculty of investigating these points is further far greater in living than in dead specimens : this is owing 

 to the rapid sphacelation of the parts when cut, and the quantity of viscid Balanophorine (the term applied by Goeppert 

 to the peculiar waxy secretion of Balanophora) contained in their cellular tissue, which prevents dissection with any 

 approach to nicety : impediments so great, that I have no hesitation in saying, that in many cases better results may 

 be obtained from specimens preserved in acid or spirits, than from living ones. • - 



