DIAGNOSIS áUD CLASSIFICATION.
valuable. Ths component forms of these groups do not shade into each other in Pedimla,
is so gradually as they do in the other genera alluded to, so that there is less tendency
on the part of authors to reduce species prematurely than there is to treat intermediate
forms, that ai-e probably quite stable, as hybrids.^
The arrangement of groups in higher subdivisions—aeftojij of Bentham, Series of
Maximowicz—is a still more difficult matter, but the greatest difficulty of all is
experienced in determining what are the approximately natural primary divisions.
Linnieas and Willdenow arranged their species according to characters derived from
habit, -while Steven, experiencing the inadequacy of such characters as the number of
species increased, relied, in defining five of his primary diidsions, viz. Permmta:,
FamUmtes, Rmtratce, Biempidatce, and Bdentulx, on characters derived from the structure
of the corolla; only one of his divisions, Vertieülcita!, being, as its name indicates, based
on a character derived from habit. It is somewhat curious that while the other
five tribes have been recast and modified so that only the outline of ono (^Rostrata;)
has been retained, and not one of them even in name continues as a primary division, the
Verticillatai have continued in every subsequent revision of tho genus to liold primary
rank and to retain the original or an equivalent name.
Tlie time appears to have now come for the abandonment of this last Steveiiian
section, for though the primary division based on the constant possession or the reverse
of "whorled or opposite leaves is a highly natural one, and gives rise to two subdivisions,
viz.
(1) opposite-leaved species, forming 35-7 per cent, of the whole, and
(2) alternate-leaved specie«, forming 64-3 per cent, of the whole,
this is only one step towards subdividing the genus, and we have to seek in floral struct
u r e the evidence necessary to ensure further subdivision. The most sti-iking floral
character midoabtedly is the presence or absence of a beak to the corolla. Wlien however,
we enquire what relationship this bears to a division based on phyllotaxy, we find
it one of practical indifference, as the appended table shows : —
TABLE IV.— General relationship of phyllotaxis to floral structure.
Specie« mtb Corollas Ireakad. Corollaa boailess.
leaves alternate 45 per cent. 55 per cout.
„ opposite 55 „ 45 „
The cleavage plane indicated by phyllotaxis is thus at right angles to that indicated
b y floral structure; and when on further subdividing the genus according to floral charact
e r s it is found that there is not a single type of corolla amongst alternate-leaved species
that is without a parallel amongst opposite-leaved ones it becomes evident that phyllotaxis
and floral structure cannot both continue to be employed to indicate subdivisions of
' Steininger in U?iheo>-m, Sot. Centralhlatt, xxx, p. 22, who clearly shows tlie liybrid character of seTonil separable
European f.>rms, treats P airoruhens in this fashion, and considers it equal to P. incttrnala x recuti.ia. That ¡1, ¡g
the intermediate form in that group to which the presumed parents belong is certain ; tlmt it is a hybrid is, I believe
dyiilitfiil. In this case, too, Herr Steiuinger docs not adyaucc any direct proof of his couleuLion.
HIGHER SUBDIVISIONS. ^^
equal rank. The question is which of the two is to be abandoned as the primary
character. And the answer seems clear; phyllotaxis in future, except in the formation
of artificial keys to species, must take a subordinate position. The reasons for this view
a r e the following. In the first place the proportion of beaked to beakless corollas -
Corollas beaked
without beaks ...
49-4 per cent, of tlie whole,
60 6 ,,
shows that this character, while quite as natural as the phyllotaxic one, divides the genus
much more evenly. Again, both the beaked and beakless divisions may be further sub.
divided by floral characters, while a primary division by phyllotaxis once effected leads
to nothing beyond. Finally, a subdivision based on floral structure may be hoped to
throw light on the developmental history of the genus: one based on habit cannot do so,
or can only do so indirectly.
At first sight it would appear that the most natural as well as most simple method
of subdividing the genus must be into—
(1) groups wherein the corolla is beaked ;
(2) ,, „ „ is not beaked.
But the occurrence of groups that are in the highest degree and in the most obvious
manner natural, such as the Aeaules, Remtilce. Carnosce, Tristes, and many more, which
nevertheless within their particular limits exhibit a transition from a beakless to a beaked
condition as complete as that shown by the genus as a whole, makes the recognition of
a n intermediate subdivision absolutely necessary. By recognising this we obtain three
divisions, clearly not quite natural in their limits, and therefore not deserving to be called
sub-genera, but apparently in accordance with the genealogical arrangement of the
groups, and so ¡possessing the advantage of being capable of a serial disposition. Arranged
so as to pass progressively from the more to the less specialised groups, these are :—•
Division I.—LONGIBOSTEES;—Corolla tube straight, slender, uniform ; lip large, sessile,
more delicate i n texture than the firm, long-beaked hood.
II.—ADUNCiE;—Corolla tube curved, cylindric, slightly enlarged towards
t h e top; lip sessile or stipitate, more delicate in textui-e than the
beaked or beakless hood.
I I I .— EKOSTRES;—Corolla tube infract, infundibuliform; lip stipitate, base
erect, S-cristate above, lobes patent of same texture as beakless
hood.
But this arrangement, while it accords with facts of structure, implies a simplicity of
natural disposition that a closer study of these facts dispels.
Thus the division LOHGIROSTRES contains two highly natural sections:
§ SiPUONANTH^; corolla tube usually much elongated; filaments inserted between
middle and top of tube and usually just below the apex; beak in bud
circinnate, with dorsal aspect apposed to median line of lip and apex
dii-ected to throat of corolla.
ANN, ROY. Box. GAED. CALCUTTA, VOL. III.