Mr. Woods oti the Species of Fedia. 423 



stata, A comparison of the specimens from the Chev. de Steven, in the herba- 

 rium of Sir J. E. Smith, witii the description in the Moscow Transactions, has 

 convinced me that the first of these belongs to the Selenoccelce, and is a spe- 

 cies which I have gathered at Rome. The second, found by Gasparini on the 

 mountains of the Madonia, and published by Gussone in the Flora; Siculas 

 Prodromiis, is supposed by De CandoUe to be nearly allied to V. turgida. The 

 figure of the fruit, however, which he has given in the Memoire sur la Famille 

 des FalMan^es, is hardly distinguishable from that of V. oUtoria, from which 

 this plant seems chiefly to differ by its quite entire bracteae. The third is 

 from the South of Tauria. It is described as smaller than the two preceding, 

 and as having a deep furrow on each barren cell. I have seen no specimen of 

 either of the two last. 



Div. 2. PsiLOCCEL^; 



In following the order of the Prodromus, we now come to the Psilocoelce, 

 although some of the Selenoccelce appear to be more closely allied to the Lo- 

 cushe both in habit and in artificial character. De CandoUe has two subdivi- 

 sions of Psilocoelce. The first, with recurved teeth, contains F. uncinata (fig. 2.) 

 and F. echinata. The former is a plant from Caucasus, which has two distinct 

 barren cells at the base of the fruit, but much smaller than the fertile one. A 

 section of the upper part of the seed-vessel exhibits, besides the fertile cell 

 which extends into the crown, three other openings filled with a white pith- 

 like substance. That near the base shows also a pith-filled opening on the 

 side of each barren cell. In F. echinata, the second cell is nearly as large as 

 the fertile one, and it is uniformly this cell, and not that containing the seed, 

 which is prolonged into the largest horn ; — the three horns which terminate 

 the fruit being in this species a prolongation of the cells, and not a distinct 

 calyx. This description seems inconsistent with the admission of this plant 

 among the Psilocoelce, where it is nevertheless placed by Soyer Willemet as 

 well as by De CandoUe. Of the five species forming the subdivision marked 

 by an erect calyx, F. Morisonii var. /3. leiocarpa, is according to De CandoUe 

 the Fedia deiitata of Engl. Bot. t. 1370.; but he also cites Reichenbach, PI. Cr. 

 t. 62. (fig. 3.), and the same work, t. 63., for the F. Morisonii a. with hairy 

 fi-uit. Both these figures appear to me to represent varieties of F. eriocarpa, 

 while that of Engl. Bot. is either F. dentata, or its variety F. mixta. The latter 



