3/2 The Last Lt'tter of Dr Gray. [zoe 



remainder of the grasses, the ferus, and the small orders allied to the latter, 

 brings this important work to a conclusion -within the limits prescribed. 

 * * * This fasciculus contains twenty plates (making the whole number 

 238). * * * — Sil inidii's JoKniai, xl, 173, Oct. -Dec, 1S40. 



THE LAST LETTER OF DR. GRAY. 



■Sunday Evening, November 27, 1887. 



Dear Dr. Britton — I wish to call your attention either in a personal 

 way or in the "Bulletin," if preferred, to a name coined by you on the 

 223d page of this year's " Bulletin." 



" Conioselinum bipinnatum (Walter, F'l. Car. under Apium), Britton. 

 Selinum Canadense, Miclix , 1830." 



I want to liberate my mind b}- insisting that the process adopted 

 violates the rules of nomenclature by giving a superfluous name to a plant, 

 and also that in all reasonable probability your name is an incorrect one. 



Take the second point first: On glancing at the "Flora of North. 

 America," of Torrey and Gray r, 619, where the name Conioselinum 

 Canadense legitimately came in, you will notice that the name Apium 

 bipinnatum, Walt, is not cited as a synonym; also that the synonymous 

 name of Cnidium Canadense, Spreng. , is cited with " excl. Syn." This 

 Apium bipinnatum, Walt., you might gather was one referred to. Sufficient 

 reason for the exclusion by Dr. Torrey might have been that Michaux's 

 plant was a cold northern one, which nobody would expect in or near 

 Walter's ground — the low and low-middle part of Carolinia. Besides, the 

 preface of that Flora states that Walter's herbarium had meanwhile been 

 inspected by Dr. Torrey' s colleague, who may now add that the Apium 

 bipinnatum is not there. So that the name you adopt rests wholly upon a 

 mere guess of Sprengels, copied by De CandoUe, dropped on good grounds 

 by Torrey, but inadvertently reproduced in Watson's "Index," copying^ 

 De Candolle. I suppose you would not contend that a wholly unauthenticated 

 and dubious (I might sa}-, doubtless mistaken) name, under a wrong genus, 

 should supersede by its specific half a well-authenticated and legitimate name. 

 And I am sure that you will not take it amiss when I say that very long 

 experience has made it clear to me that this business of determining rightful 

 names is not so simple and mechanical as to j-ounger botanists it seems to- 

 be, but is very full of pitfalls. I trust it is no personal feeling which 

 suggests the advice that it is better to leave such rectifications for mono- 

 graphs and comprehensive works, or at least to make quite sure of the 

 ground. 



W^e look to you and to such as yourself, placed at well-furnished 

 botanical centres, to do your share of conscientious work, and to support 

 right doctrines. So I may proceed to say that, upon the recognized princi- 

 ples since the adoption of the Candollian code, your name of Coniof^elinum 

 bipinnatum, even if founded in fact, would be inadmissible and superfluous. 



February 21, 1894. 



