withering’s fern. 
213 
rior ones : this discrepancy between superior and inferior pin- 
nules gradually diminishes, and altogether ceases with the sixth 
pair, which are of equal length : a somewhat similar discrepancy 
is observable in the pinnules of the second pair of pinnae, but 
beyond these it becomes scarcely observable : some of the lower 
pinnules are deeply pinnatifid, almost pinnate : the lobes of the 
pinnules are toothed, and the teeth terminate in short but dis- 
tinct spines; each serrature with its accompanying spine has a 
decided curvature towards the apex of the pinnule. The entire 
frond is characterized by a flat surface, having neither the con- 
vexity which distinguishes the species immediately following, 
nor the concavity which is characteristic of the next but one. 
The veins in the pinnules are alternately branched, each sys- 
tem of branches entering a division of the pinnule, and the an- 
terior branch bearing a circular cluster of capsules just within 
the sinus which occurs between each two divisions : this cluster 
is covered by a flat reniform involucre, the margins of which are 
sinuate, generally entire, and always without stalked glands, a 
character which, as far as my observation has extended, is con- 
stant, and is of great importance in distinguishing this species 
from those which follow, (see the left-hand figure at page 236). 
This character sufficiently distinguishes the present plant from 
the spinulosa of Swartz, Willdenow, Schkuhr and Francis, all 
of which 1 imagine to be mere varieties of the next species. 
Owing to the constant position of these clusters on each pinnule 
they form a regular double line, the midvein of the pinnule pass- 
ing up the centre : but when the pinnule is completely divided 
into lobes, each branch of the vein usually bears a cluster of cap- 
sules : this is more frequently the case in those pinnules which 
are nearest the main stem of the frond ; and it may be observed 
that the clusters on all except the usual cap side -bearing branch 
I are of smaller size. The seed, as pointed out by Linneus, Berk- 
enhout and Withering, is confined to the upper portion of the 
frond: exceptions to this are of rare occurrence. The clusters 
are usually distinct and perfectly separate, indeed I may say al- 
ways so when grown in woods, but on heaths they are more 
crowded, and I have seen specimens from Bawsey and Holt, in 
I which they are quite confluent. Each branch of the vein enters 
