219 
iii S. et Z. and Hylomecon vernale, Maxim.), which is distri- 
buted throughout the Far East. The only points which contradict 
my supposition are that this Chinese species has (1) no radical leaf, 
(2) only four sepals but no petals, and (3) biseriate ovules. 
first and second points are most probably due to the imperfections 
of the specimen which came into the hands of these authors, The 
description and figures of the ovary given by the French authors 
are puzzling, since they point to an intermediate character between 
Ranunculaceae and Papaveraceae, One unmistakable character of 
Papaveraceae, however, is well represented by the stamen of the 
plant in question. The filament tapers towards the apex, and does 
not continue to the connective, as the French authors describe and 
figure. Another eine against referring the plant to Glaucidium 
is the absence of a bract below the flower. 
Anyone who compares the figures of Glaucidium pinnatum with 
those of Hylomecon joao published either in the Botanical 
Magazine, tab. 583 Maximowicz’s “ Primitae Florae 
Amurensis,” tab. 3; will Rats agree with my view, even without 
dissecting the flow 
Another vofnarkatils species, G. paradoxum, was described by 
Makino.* This species is distinguished from G. palmatum mainly 
by having four carpels (two of these are said to be abortive), and by 
some of the stamens being malformed and assuming a carpe -like 
appearance. ‘This is certainly a peculiar, but not at all a surprising 
phenomenon amongst Ranunculaceae. It is well known that the 
stamens of various members of this family are often metamorphosed 
into petaloid forms. It would not therefore be at all surprising 
should four normal carpels occur in Glaucidium palmatum, because 
this plant, which normally has two carpels slightly connate at the 
base, has often only a single carpel, and sometimes may have as 
many as three. As the author suggests, the specimen he examined - 
is evidently a monstrous form of G. palmatum, and not a distinct 
valid species. 
Although this pluricarpellar malformation suggests an apparent 
similarity between this genus and the North American Hydrastis, 
these two genera cannot be et since the structure of the 
pam is eae different in the 
The of G. palmatum is easy solitary, but in a vigorous 
plant inate sete tables occur two flowers borne at the summit of the 
stem, and then the bract is common to the two flowers. 
A species of Hydrastis was recorded from Japan, under the name 
of H. jesoensis, Sieb., a very brief description of which is given by 
Miguel.f Prantl,t usually a careful botanist, accepted this. species. 
Huth§ entertained doubts about it, and suggested that it might be 
a species of Glaucidium, though not G. palmatum. Miquel’s 
description of this doubtful plant is very vague, still there one can 
—— a characteristic of prtsasr ium in it, as he says, “. . . car- 
pellis . . basi inter se connatis, . ” This does not occur 
in Fires but does in Blouin " F have no doubt that this 
* Makino, in Tokyé Bot oe Saf p. 71 (1910). 
¢ Prantl, in Engl. u. Pr. Nat irl. Pflanzen-Fam., iii, 2, p. 55. 
§ Huth, in Eng). Bot. Jahrb, xvi, pp. 292, 293. 
