Mr. Woods on the Species of Fedia. 425 



also a tooth opposite to this, and, consequently, the number of teeth in the 

 imaginary regular type must always be even. When the contrary is the case, 

 it is because an additional intermediate tooth is developed on one side more 

 than on the other, producing some degree of irregularity ; or, perhaps, we 

 should rather say that one of the lateral segments— for it is there principally 

 that the irregularity takes place— is more subdivided than its opposite one. In 

 the Psiloccelce the case is exactly the reverse. The middle tooth of the anterior 

 face is wanting, and the number of teeth in the type must be considered as 

 uneven. But the lateral teeth in this division are usually small, and often 

 unequally developed, so that there is one tooth, or perhaps even two, more on 

 one side than on the other. Thus an even number of teeth may occasionally 

 occur, but it ought hardly to enter into the specific character. Reichenbach's 

 figures often attribute to the Psiloccelce a tooth in the centre of the anterior 

 face. I can only say that I have never met with such a circumstance. 



Div. 3. Platyc(el^. 

 We now pass to the Platyccelce. In several species of this division the 

 barren cells are contiguous at top and bottom, so that a section of the fruit 

 would there be nearly round, but they are separate in the middle. In F. Au- 

 ricula (fig. 15.) they are contiguous for their whole length ; and in F. echinata 

 (fig. 16.) they may be said to touch in their whole length, but without having 

 a common partition. The two cells are not in any part united in any of them. 

 F. uncinata and F. echinata, both of which I incline to place in this division, 

 have been already mentioned. The former was brought by the Chev. de Steven 

 from Caucasus, but has not, I believe, hitherto been found in Europe ; the 

 latter is frequent in the neighbourhood of the Mediterranean. In F. Auricula 

 the inner sides of the barren cells sometimes shrink away in drying, and give 

 to the fruit somewhat of the appearance of that of the Selenocoelce, and well 

 characterized by the expression "fructu untied rimoso," used by De Candolle 

 to some of that tribe. F. tridentata (fig. 17-) of Reichenbach is a variety of 

 F. Auricula. The terminating tooth in this species is often very small, thin, 

 membranous and fragile. Gaudin describes it 3—6-dentata. I have never 

 observed more than 3 teeth. De CandoUe's expression, " calycis limbo acute 

 auriculi/ormi" would suit some of my specimens, but the tooth is not always 



