UNDERWOOD: AMERICAN FERNS 245 
will illustrate the normal progress of one of the species, and it must 
be noted that in the main the diagram represents the plant at its 
simplest condition except at one point (above c’) where a bud of the 
second order of forks is represented as producing a lateral exten- 
sion of the plant, a condition possible at every one of the second- 
ary and tertiary forks. 
On account of these features some of the larger species of 
Dicranopteris do not lend themselves to what has facetiously been 
called ‘lie flat botany” since they occupy too much space in 
every direction to be easily reduced to the limits of the ordinary 
herbarium sheet, and still reveal the distinctive characters of the 
Species. Neither “rough dry botany” nor “bottle botany” will 
succeed much better in transferring the distinctive characters of 
the larger species to the laboratory, and photography, while re- 
vealing habit in the mass, can add little to the details that go to 
separate technically one species from another. They must be lived 
with in their native haunts to impress firmly their distinctive char- 
acters. The ordinary field botanist, without knowing the necessi- 
ties of the case, meets a proposition in the form of a Dicranopteris 
tangle, and simply breaks off one of the small terminal portions 
of possibly a fork of the second or third order with no hint of the 
main stem or habit of growth and brings it to some herbarium for 
preservation as a stumbling-block for the future. Such material 
unfortunately formed the basis for some of the “type specimens” 
of the genus and they can only be elucidated and correlated after 
extensive study of the plants in the field. Field-work of an in- 
telligent character alone will ever disentangle the muddles in this 
genus. The hortus siccus will furnish some of the types but they 
must be interpreted by the field study in their respective type 
localities. 
The terminology of the parts of these peculiar plants requires 
Special notice and can best be explained by reference to the dia- 
gram (FIGURE 1). We commence with a young simple upright 
branch* W7. This normally forks twice, producing a bud at +’ 
' SS eercvercpmereeree ns « 
* This upright branch from the creeping rootstock is circinate when young and is 
‘parently homologous with the “ frond’? or leaf of the ordinary fern. Like the ana- 
logous case in some of the Lygodieae it is utterly impractical touse such a terminology 
here in descriptive work. It would be particularly absurd to sneik of such a tangle 
aS we have mentioned above as ‘‘ leaves” or ‘ fronds’’ and yet such they appear to 
sy Morphologically. 
