OUBLni NATI7AAL HISTOBY SOCIETY. 191 



ferns. It is by a character of the annulus that OsmundaceaB and Schiz83- 

 acc'tt) are distinguished from PolypodiacefiB, and, except by the annulus, 

 Todca does not materially differ from Gyranogramma. But, leaving the 

 annulus aside, instances in plenty may be adduced to show the variabi- 

 lity of all the other characters of the fructiiicatipn, viz., variations in the 

 position, insertion, and shape of the sorus, and the presence or absence, 

 form and insertion, of the involucre. In our common Harts- tongue the 

 sorus is often found on the upper surface of the frond, showing that it 

 varies in position. Such is the case occasionally in several other species, 

 and in a Ceylon fern {Polypodium anomalum, Hook.) the sori are almost 

 always foimd on the upper surface. This last-named species varies in 

 a still more remarkable manner, its son ha\'ing sometimes a peltate invo- 

 lucre, and being sometimes quite naked, so that it has been described 

 both as a Polystichum and a Polypodium. Similar instances of the in- 

 constancy of involucres, especially in the genera Polypodium, Polysti- 

 chum, and Lastnea, are not unfrequent among tropical ferns. "Lots of 

 instances," Sir "William Hooker writes to me," might be given where an 

 involucre is sometimes absent and sometimes present, in the same species 

 of fern, and this all recent pteridologists bear witness to. On the same 

 specimen I have seen laterally inserted and peltate involucres. " That 

 is, Sir William has seen a single frond which; technically, by its in- 

 volucre, was partly a Polystichum and partly a Lastrsea. And if invo- 

 lucres are found to vary in this wholesale way, can we be sure that they 

 do not vary in the tootliing and non-toothing of the margin ? Their 

 major characters do vary ; are their minor characters immutable ? But 

 there][is also a South African fern which is even yet more uncertain in 

 its fructification, being sometimes a Blechnum, with a solitary medial 

 sorus, and sometimes a Scolopendrium, with numerous divergent and 

 opposite sori along the lateral nerves; and sometimes it is neither 

 clearly one thing nor the other. Of this plant Sir William Hooker writes : 

 " Look at your compatriot, the Cape Blechnum pundulatum, and if you 

 have specimens from Dr. Atherstone, from Grahamstown, you may see 

 tlie long sori breaking up and shooting off at right angles on each side 

 the costa, working themselves along with the needful receptacular vein, 

 arranging themselves in pairs, facing each other, and so forming Sco- 

 lopendrium Krehsii. I can swear to this, and I only wonder others have 

 not suspected it long ago." 



After a case of irregularity like this, I do not think I need multiply 

 instances to prove further that the characters of fructification, among 

 ferns, as well as those of frond, are subject to extraordinary variation ; 

 and if Dr. Kinahan sets aside the annulus because of its uncertainty, 

 I am entitled to set aside the sorus and involucre. But I am very far 

 from advocating any such sweeping conclusion. The only point I de- 

 sire to establish is this : seeing that in some ferns the annulus, sorus, 

 and involucre do vary by their major or generic characters in the same 

 species, it is not beyond the bounds of probability that they may also 

 vary in their minor or specific characters. Therefore, if truthfril and 

 shai-p- eyed, witnesses, such as Mr. Bentham and Dr. Joseph Hooker, tell 

 me that they have seen involucres on our Hymenophyllums, which are 



