of the Orobanchacefe. 31 



of the roots arise from the bundles of the tuber and are of the 

 reticulated kind; the cortical layer of the roots is continuous 

 with the cortical parenchyma of the tuber. 



These tubers of Orobanche propagate by subterraneous buds. 

 It is well known that the plants often occur three or four ad- 

 hering together^ but I believe the reason has not before been 

 shown. I found growing plants with the decayed tubers of the 

 preceding year still adherent, and others which had completed 

 their flowering, that had buds growing out from the base of 

 the tuber. These buds were not axillary in appearance, for they 

 arose quite below the lowest scales of the tuber, but it is reason- 

 able to sup))Ose that they had originated from the axils of scales 

 which had decayed. 



The most important point remains, viz. the mode of attachment 

 of the pai'asite to the foster-plant. I have only observed this in 

 O. minor ; here the root of the Trifolimn was traced into the sub- 

 stance of the tuber ; its fibro-vascular structures become sepa- 

 rated, and lose themselves by ramifying in the substance of the 

 parasite. The union is completely organic, and in one speci- 

 men examined the tuber had grown so much that the root of the 

 Trifolium, which was curved round the tuber, lay imbedded in 

 a groove formed by the growth of the latter, but actual union 

 only existed at the apex of the root which penetrated into the 

 substance of the tuber. 



The point which has always struck me most in observing the 

 parasitism of Orobanche is the small size of the root to which 

 they are always attached, and it appears to me that there is much 

 yet to be explained both in this tribe and in all the other root 

 parasites. The presence of proper roots would seem to indicate 

 that the parasites are not wholly nourished from the foster-plant, 

 a conclusion which irresistibly presses upon us when we see a tall 

 Orobanche some two feet high and three-quarters of an inch thick 

 attached to a slender root not measuring a quarter of an inch 

 in diameter. Their own proper roots in Orobanche are small and 

 few in number, and I believe that in O. rapum at least, the whole 

 tuber with its scales is an absorbing organ. My reason for this 

 supposition is the condition of its tissues. The tuber and scales 

 are composed almost wholly of succulent cellular tissue; the epi- 

 dermis resembles the epithelium of roots, and like it dries up and 

 becomes discoloured very rapidly on exposure to the air. These 

 structures are manifestly as well adapted to the absorption of 

 fluid nourishment as the aerial roots of the epiphytic Orchi- 

 dacese, and I see no objection to the assumption that they are so 

 employed. 



The question of the parasitism is not interfered with by the 

 above proposition ; but we have to account for the assimilation of 



