422 Mr. Woods on the Species of Fedia. 



1. Locusts, with one or two empty cells, and a gibbous, corky, or spongy 

 mass at the back of the fertile one. 



2. PsiLoccEL^. The two empty cells each reduced to a hollow nerve. The 

 description of the genus assumes the existence of empty cells ; otherwise, per- 

 haps, it would be better to say that the fruit in this section had only one cell. 

 The nerve is not always sensibly hollow ; and a similar nerve sometimes exists, 

 in F. Auricula, for instance, on the surface of each empty cell. 



3. PLATvc(ELi5E. Two empty cells, as large, or nearly as large, as the fertile 

 ones. Section of the fruit rounded. 



4. Selenoccel^. Section of the fruit crescent-shaped, with two empty cells. 



These divisions once pointed out cannot be neglected by succeeding bo- 

 tanists ; but we may be permitted to introduce some modifications in the 

 divisions themselves, and in their arrangement, and some alterations in the 

 species assigned to each. 



Fedia Cornucopice (Fedia of De CandoUe). This plant seems not to be fre- 

 quent, though widely scattered, on the coasts of the Mediterranean. I have 

 not seen the fruit in a perfect state. A second species, F. scorpioides, with 

 which I am unacquainted, has stalked leaves and unilateral spikes of flowers. 

 It is a native of Tangier. 



Division 1. Locusts. 



De Candolle separates this into two sections, the first having only one, and 

 tlie latter two barren cells. This character is hardly sufficient, since in f^. oli- 

 toria (fig. 1.), which is at the head of the first division, we not unfrequently 

 find the trace of a dissepiment separating, more or less completely, the empty 

 cell into two parts. Reichenbach says that the fruit of this species is some- 

 times hairy. I have never met with it so. But this is a character which seems 

 very variable in the genus. 



Two other species are enumerated. Falerianella radiafa, which the author 

 suspects to be an American variety of P^. olitoria, but which, from specimens 

 shown me by Mr. Bentham, seems rather to be a name for several European 

 species when they have been carried over to America ; and F. exscapa, a plant 

 of Caucasus, described as having two fertile cells. 



The second subdivision of the Locusts, where the separation of the cells is 

 uniformly complete, contains three names : f^. turgida, V. gibhosa, and l^. co~ 



