34 2 ON THE GENUS BRAIXEA. 
and observed : — " It should surely hold the same rank among the 
Graminitacece, section Hemlonitidemy of PresI, which Woodwardia and 
Doodia do among Blechiacere of the same author." Professor Mette- 
nius (Fil. Lechlerianae, part ii. p. 9, a, 1859) considers its natural 
place in the system to be in the neighbourhood q{ AcrosticJiece. Finally, 
in 18G4, Sir ^^illiam Hooker (Sp. Fil. v. 161) definitively locates the 
genus in his suborder GrarnmUidea ^ which includes Jamesonia^ Notho- 
lana, Monogramme^ Gymnogrammey Brainea^ Memscium, AntropTiyumy 
Vitta)na^ Tcenitis, Drymoglossum, and Ilemionitis. 
Although this suborder seems to me to contain some heterogeneous 
elements, yet I certainly think Gymnogrammea (as it would be more 
correctly called) the true and natural station for Brainea ; wdiilst the 
resemblance to Lomaria and its allies, insisted on bv Mr. Smith, and 
recognised by Mr. Moore, appears to me to be rather analogical tlian 
indicative of direct affinity. It would be difficult to produce a more 
perfect instance of paralleHsm between two tribes than that show^n in 
the following diagram, in which the opposite genera exactly corre- 
spond. 
LomariecB, Gymnogrammece. 
Blechnum .... f Gy^^nogramme, 
L § Coniogranime. 
Sadleria • Brainea. 
Woodwardia 1 / Gytnnogramme, 
§ Lorinsoria J ' I § Dictyogramiae, 
And, still more singularly, both Sadleria and Brainea are remarkable 
for their stout arborescent caudex ; in which respect they stand alone 
amongst their immediate allies. 
No sharply defined limits exist between the tribes Acrostichea^ 
Gymnog7'amme(£y and PolypodiecB, 
if oil 
Splitg., has evidently as much claim to rank as an aberrant Aero- 
atichim as it has to be united with Sejnionilis or Antrophymn^ which 
two latter appear to me in no wise distinct. And, whilst the costular 
veinlcts of Gymyiogramme Japonica are, contrary to M. Fee's assertion, 
frequently soriferous, I have specimens of Brainea in which the entire 
under surface of the frond is just as densely covered wuth sporangia as 
4;]iat of Acrodichnm (StenocJdcEna) scandens, L. In Polypodium tridac- 
tylon. Wall., the sori very frequently extend along the veins, and even 
anastomosfe occasionally quite freely ; in which case the plant is, of 
course, Gywnogrammoid. In the Niphoholus group of Polypodium^ 
