267 



same in every famil}', and hence a too rigorous form of words will be 

 in all cases inapplicable.* Certainly, I think this requires to be look- 

 ed into, for if characters are employed to determine species which are 

 variable in themselves, the fault rests in the employment of this ex- 

 ceptionable character. Thus the involucra were formerly employed 

 to determine the species of the Umbelliferse, and Q^^nanthe pimpinel- 

 loides was described to have a general involucre, while GEnanthe 

 peucedanifolia had not. From this unimportant point being regarded 

 iRimerous errors have arisen, and the two plants became confounded, 

 and the former even erased from the British Flora by Mr. Babington ; 

 and yet their roots show them to be perfectly distinct, and this cha- 

 racter is constantly available, and probably may be most discrimina- 

 tive in all the Umbelliferae, the roots of which are most important to 

 mankind, though in some other orders this character may be of no ac- 

 count. So that whatever may be asserted about the oat changing into 

 rye, I think all the ingenuity of the greatest advocate of transmutation 

 would not be able to effect the change of a parsnip into a carrot, or 

 induce the QE nan the phellandrium to become a celery. Only then 

 find out the character that is really the most important in an order or 

 tribe, and much doubt and confusion is removed, and we find indica- 

 tions of permanent boundaries in Nature there, at any rate. 



In the rose tribe Nature appears most capricious ; root, leaves, 

 armature and fruit all fail us at need as unerring absolute characters ; 

 yet surely the attempt to discriminate between the variable forms that 

 occur is not to be despised, because in the effort truth may be arrived 



* Whatever theory may suggest, practically, botanists are right in separating as 

 species plants of the satne family that have permanent palpable differences in a wild 

 state in some particular charactei'. It is obviously impossible for a travelling collector 

 to make experiments, and any assumption on his part could only be productive of error. 

 Experimental botany should be considered a separate department, and let the experi- 

 mentalist make his claims to regulate or modify specific nomenclature, as the lawyers 

 say " without prejudice." With respect to varieties, there is perhaps more anomaly and 

 ambiguity than even in species, since uo weight appears to be bestowed upon the rela- 

 tive amount of variation, and thus almost every botanical author's "Alpha — Beta — 

 Gamma '' — is different to that of others, giving rise to whole columns of synonyms. 

 Now this really requires emendation. Transient varieties, therefore, should be distin- 

 guished from pennanent ones, and rules laid down for this purpose. A plant with an 

 additional petal or two, a white-blotched or fissile leaf, or a white flower instead of a 

 coloured one, though curious, is rather a sport or luxuriation, than a variety, and does 

 not deserve to be estimated in the same manner as more important and continuing 

 characters would, affecting the appearance of the whole plant. Hence varieties ought 

 to be classed as casual or permanent. 



