The column is the most characteristic part of an orchid flower. It is considered
to be formed by the union of stamens with a central style and stigma. In the Apo-
stasieae, for example, the column is rather primitive in the stamens and style only
partially united, whereas in the majority of higher orchids it becomes more advanced
through a eomplete union of them into a single organ. Within the family, indeed,
the column structure is greatly diversified and of great taxonomic significance.
It is interesting to note that a great range of diversity of column structure is
bund in Neottia (sensu lato), a small but widespread genus consisting of 14 species,
about two thirds of which, however, are of local occurance and seem to be little known
to many botanists. In some speeies of this genus we find a very primitive column
structure which is quite unique in the family, while in the others it is much more
complicated. In all, five types of their column structure can be distinguished as fol-
lows: (1) column rather longer; anther erect with a short filament attached to the
back of the column near the apex; stigma terminal; neither clinandrium nor rostel-
lure; (f. 2, 4) (2) as the preceding, 'except for the stigma more or less curved fore-
ward and filament longer; (f. 6, 8) (3) column rather longer with a clinandrium at
its summit, upon which a sessile and incumbent anther sits; rostellum large, hori-
zontally projecting out over the concave stigma situated in the front of the column;
(f. 10, 13, 15, 17) (4) as the preceding, except, for the anther and rostellum almost
erect, and the stigma more or less bilabiate; (f. 19,21) (5) column very short; an-
ther and rostellum erect; stigma lamellate, erect; reflexed and almost clasping the
rostellum. (f.,2g) In these .five types, with the exception of the first one in which the
labellum (the median petal) is very similar to the lateral: petals, they all possess
zygomorphic perianth with labellum bilobed or entire which is quite different from the
two lateral petals.
Here, we see a great change in the column structure from one form with stamen
and style not fully united to another form in which they have been well fused.
Speaking strictly, these are two sorts of entirely different column structure. The
former one, represented by (1) and (2) as stated above, is, in fact, an incomplete or
s very primitive column in having a terminal stigma and an erect stamen with its
free filament attached to the back of the column; and the absence of clinandrium
and rostellum. Furthermore, there exists on the back of the column a thick ridge
with its upper end joined to the filament, with which it is of the same texture and
appearance. In Neottia pantlingii (=Arohineottia pantlingii) the free filament is
even rather longer than the ridge, (f. 6) while in the other three species (f. 2, 4, 8)
they are shorter. It is in my opinion the lower part of the filament adnate to the
compound style or column. This is another fact of interest perhaps not occuring in
any other living orchids. On the other hand, the latter one, represented by (3), (4)
and (5), is a more advanced column structure, in which a higher level of specialisa-
tion with well-developed clinandrium and rostellum is reached. The stigma becomes
shallow depressed on the anterior side of the column, or sometimes in the form of so-
mewhat a bilabiate lip projecting out before or under the long rostellum. This is
apparently a complete column both in structure and function quite different from
the former and, contrarily, much like that of Listera.
Basing upon the facts just mentioned, we may subdivided Neottia (sensu lato)
into two distinct genera, with two and three sections respectively. They are as fol-
lows:
1. Archineottia S. C. Chen, gen. nov.
(1) Sect. Archineottia
1) A. gaudissartii (Hand.-Mzt.) S. C. Chen, comb. nov. (China)
2) A. microglottis (Duthie) S. C. Chen, comb. nov. (India)
(2) Sect. Furciila S. C. Chen, sect. nov.
3) A. pantlingii (W. W. Smith) S. C. Chen, comb. nov. (Sikkim)
4) A. smithiana (Schltr.) S. C. Chen, comb. nov. (China)
2. Neottia Guett.
(1) Sect. Listeroides S.C. Chen, sect. nov.
1) N. listeroides (L.) Rchb. f. (China, Sikkim, Kashmir)
2) N. camtschatea (L.) Rchb. f, (China, Soviet Union)
3) N. megalochila S. C. Chen, nom. nov. (China)
4) N. inayatii (I)uthie) Schltr. (Pakistan, Kashmir)
5) N. tenii Schltr; (China)
(2) Sect. Neottia
6) N. papilligera Schltr. (Chinas: Japan, Korea, Soviet Union, Sikkim)
7) N. nidus-avis (L.) L. C. Rich. (Europe, Iran, Western Siberia)
8) N. brevilabris Tang et Wang: (China)
(3) Sect. Hologlossa S. C. Chen, sect. nov.
9) N. acuminata Schltr. (China, Japan, Korea, Soviet Union, Sikkim)
Inperfeetly known species:
10) N. ussuriensis (Kom. et Nevski) S6o (Soviet Union)
Thus, the subtribe Neottiinae are composed of four genera, namely, Diplandror-
chis, Archineottia, Neottia and Listera. The new genus Archineottia, as one of the
most primitive genera in the family, is of great interest from a phylogenetic point of
view. It shows dose similarity to Diplandrorchis and Neottia in habit, but sharply
distinct from them in column structure. These genera, as indicated By some authors,
also show affinity in some respects with the subtribe Limodorinae, especially to
Tangtsinia and Sinorchis, the other two quite primitive genera in the family. There
is, indeed, a great need of further study of these interesting or relic genera and this,
I think, would go a long way towards solving the problems concerning the origin of
the Orchidaceae.
