76 REVUE BRYOLOGIQUE 



drawing of Porella by Dillenius. This drawing is one 

 more link in the chain of wliat is to me wholly conclu- 

 sive evidence that we know absolutely what the Porella 

 of Dillenius was and that the suspicion once expressed 

 by M. Lejolis (1), that there had been some derangement 

 of labels in the herbarium at Oxford is without sufficient 

 foundation. This original drawing shows, in addition 

 to the detached « capsules b and leaves simply an outline 

 sketch of the ramification with about twenty of the leaves 

 indicated. None of the so-called « capsules « are repre- 

 sented in attachment with the stem. The arrangement of 

 the leaves confirms Dickson's observation that « in his 

 [Dillenius's] original drawings, in the possession of Sir 

 Joseph Banks, the leaves, so far as they are represented, 

 are placed in the manner 'as in the [Dickson's] annexed 



figure » (2). 



The completion and correction of an imperfect or par- 

 tially erroneous diagnosis by a study of the original, or 

 at least an authentic specimen is not a new thing in the 

 history of taxonomy, and instances are not wanting where 

 botanists even of the more conservative kind have chosen 

 to use a name with a meaning thus amended in preference 

 to one of a later date whose meaning had been more 

 accurately defined. We find even Nees von Esenbeck, who 

 was on the whole not an adherent of the priority prin- 

 ciple, preferring (3) the older specific name fragraiis of 

 Balbis for the plant which his friend Bischoff had, with a 

 very accurate and detailed description and beautiful 

 drawinss launched into the scientific world as GrimahUa 

 harbifrons. We also find Nees here acknowledging his 

 indebtedness to authentic specimens, « von Balbis selbst 

 bezeichnet » in determining the true application of the 

 name Marchantia fragrans Balbis, It may be objected 

 that the comparison of this case with that of Porella is 

 not exact that the original description of the Marchantia: 

 fragrans of Balbis contained no such errors as that of Po- 

 rella, I have never seen the original diagnosis in question 

 and do not know how that may be. Nees says that Balbis 

 < beschreibt sie gut, bildet sie aber mittelmassig I'.But 

 the comparison is accurate in so far as assistance from an 

 authentic specimen in getting at the true meaning of a 

 name is confessed and the comparison is accurate also in 

 the maintenance of this name by Nees in preference to a 

 later one more carefully defined. To what extent we 



(1) M6in. Soc. Sc. Nat. et Matli. de Clierhouru:, XXIX, p, 142, 1894. 



(2) Trans. Linn. Soc, m, 238, n!)7. 



(3) Naturgesch. Eur. Leberm., IV, p. 239, Aiimerkung. 4, 1835. 



