RHIZOGENS. 87 



banclie and Orcliisworts. But with regard to OroLancbe, that plant has a 

 sHghtly two-lobed embryo lying in a mass of albumen, so that I do not see 

 how it can be brought into comparison with that of Rhizogens ; and as to 

 Orchisworts, we have no right to say that their embryo is essentially different 

 from that of common Endogens, except in its smallness. 



More recently, Ml"- Griffith has adopted the views of Brown, and endeavoured, 

 by new arguments, to show that Rhizogens cannot be regarded as a pecidiar 

 class in the Vegetable Kingdom. He is of opinion that " in the construction 

 of the group, a remarkable diversity of characters has been sacrificed to an 

 appearance resulting from parasitism on roots, and to an assumed absence 

 of an ordinary form of vegetable embryo." He asserts, that these plants 

 are not similar in their parasitism, and that in those which he has 

 examined there would appear to be two remarkably different types of 

 development of the embryo. He thinks, moreover, that such a class is 

 opposed to the system of Nature, a chief point of the plan of which consists 

 in an extensive interchange of characters, either positively by structure, or 

 negatively by imitation of structure. The want of uniformity in opinion of 

 the founders of the group regarding its rank or value, is incompatible with 

 any group of the system of Nature. And he is persuaded that Rhizogens 

 are an entirely artificial class, not even sanctioned by practical facihty, 

 which is the only merit of an artificial association, and " a retrograde step 

 in the course of philosophical botany." This being the case, it was of 

 com'se necessary to show where the genera of Rhizogens can be stationed, 

 if they are not collected into one common class, as is here proposed. 

 Accordingly, Mr. Griffith suggests, that the genus Mystropetalon may be 

 " the homogeneous-embryo-form of that order, which he takes to include 

 Proteacese, Santalacese, &c., and which nearly agrees with Professor Lind- 

 ley's alliance Tubiferse." The tendency of Sarcoj^hyte is, he thinks, 

 towards Urticacese, and he also considers Balanophora as the homogeneous- 

 embryo-form of Urticaceae, forming a direct passage in one, and usually the 

 more perfect, structure to Musci and Hepatic^e." Finally, he stations his 

 genus Thismia between Taccacess and Bm-manniaceae. — [Proceedings of the 

 Linncean Society, No. XXII., p. 220.) 



I think that this kind of argument affords strong evidence in favour of 

 the propriety of constructing the class here again proposed. The 

 forced resemblances which are sought after by both Dr. Brown and Mr. 

 Griffith, in themselves indicate the weakness of the arguments by which it 

 is proposed to do away with the class of Rhizogens. The fact is, that the 

 species which constitute it have no real relation to any other parts of the 

 system. It is true that the genera differ very much from each other in 

 the details of their fructification ; though not at all more than the genera 

 of other classes ; but the character of the order does not depend upon the 

 fructification. It depends wholly upon the great peculiarity in the manner 

 of growth, already pointed out ; and the fructification is connected with 

 questions of quite a subordinate degree. All the classes of plants depend 

 equally upon such considerations ; and, therefore, Rhizogens are a class. 

 I am indeed surprised, that so acute a botanist as Mr. Griffith — one of 

 the very few men who combine with minute accuracy of observation great 

 general \aews — slioidd not himself perceive how much his position is weak- 

 ened by comparisons like the following, the justice of which, however, I am 

 far from disputing. He particularly directs attention to the resemblance 

 between the pistil of Cynomoriums and that of Mosses, or more especially 

 to that of some evaginulate Liverworts, and to the effects produced by the 



