50 
or erect, nor its symmetry, radial or bilateral, seems to be a very .deep- 
seated or firmly fixed character, for both change in many instances witliin 
universally recognized generic or subgeneric limits, as in " Goniopteris" 
and Eu'Polypodhim. The correlation between length of stem and length 
of stipe is too obvious to need any discussion; ferns with scandent or 
wide-creeping rhizomes have short stipes, while those with erect rhi- 
zomes have the tufted stipes long enough to separate the fronds. 
Fleshy rhizomes serving as water-reservoirs are found in Drynaria 
and its relatives, most notably in Folypodium heradeum,^^ and less 
developed in Photinopteris and rolypodium affiiie. The rhizome and 
stipe of Scolopendriiim piiinaium are fleshy, as is the stipe of Antrophijum 
lati folium. 
All rhizomes are protected against loss of water at the apex, and many 
throughout their length, by scales which vary in form, size, and texture. 
Exceedingly harsh paleje are found on Diptcris and Dennstaedtia Wil- 
m r 
Uamsh two ferns with notably stent rhizomes. It is very probable that 
these are protective against animals, such as deer and hogs which are 
very numerous, but Avhich never, so far as I have observed, touch these 
species. Similar scales protect the fleshy crowns of A-arious Cyatheacerv. 
The muricate stems of StenocMaena, and muricate stipes of Dennstaedtia 
ery til r Gracilis, Diplazium pohjpodioides and other species, as well as of 
Aihyrium silvaticum probably have the same function. Dead bases of 
stipes must provide other rhizomes with an unpalatable mantle, but most 
fern stems are too hard to need protection of this kind. 
Many stems contain chlorophyll when exposed to the light. It is 
regularly present in those of Fohjpodium accedens, P. dolichopterum, 
F. commutatnm and P. Schneideri. 
Light correlations. — The correlation between length of rhizome and 
length of stipe has just been mentioned. A similar correlation exists 
between length of one or the other of these and the development of 
the lowest pinnae. Deltoid fronds — that is, fronds with elongate lowest 
pinnae — would seriously interfere with each other's light if they were 
not borne on wide-creeping rhizomes, as is the case in Davallia and 
Humata; or on very long ascending stipes, as is true of most species 
of Aspidium, Nephrodinm sparsum, N. intermedium, Dennstaedtia ery- 
ihrorachis, Adianttim mindanaoense^ Schizostege pachysora, Pteris excelsa 
and P. plaricaudata; or on comparatively short, but more horizontal 
stipes, as in LeptochUus latifolius and Psomiocarpa apiifolia. Fronds 
with short stipes, unless these are very remote, usually have the pinna? 
reduced toward the base; illustrations with creeping rhizomes are Nepli- 
rodium aridiitm, Artliropteris, Lindsaya hjmenophylloides, L. Ilavicei, 
Pohjpodum ceJehicum, F. ohliquatum, Prosaptia, Drynaria (normal 
'•Ooobol: Pflanzcnhiologischc Bchilderungen (1889), 1, 202. However, Profes- 
sor Gotbel is in error when he cites tlie HymenophyUacete, because they have no 
store of water, as being quickly killed by dry air. 
