1. The present paper describes the observations of chromosome numbers and
karyomorphology of 2 species of 2 endemic genera and I endemic species of Chinese
Ranunculaceae: Asteropyrum peltatum (Franch.) Drumm et Hutch. 2n=16, x=8;
Kingdonia unifolia Balf. f. et W. W. Sm. 2n=18, x=9 and Calathodes oxycarpa Spra-
gue 2n=16, x=8. The chromosome counts of three ranunculaceous genera are repor-
ted for the first time.
2. The morphylogical, palynological and cytological date in relation to the syste-
matic postition of Asteropyrum, Kingdonia and Calathodes within the family Ranun-
culaceae are diseussed and resulted in following conclusions:
(1). On the basis of the basic number x=8 in Asteropyrum, it is further con-
firmed that this genus is distinct from the r elated genera such as Isopyrum, Dichocarp-
um and other allied taxa.
The comparison of Asteropyrum with Coptis shows that they are identical in short
chromosomes, with magnoflorina and benzylisaquinodine type of alkaloides, but dif-
ferent from coptis in the chromosome numbers (T-type), pantocolpate pollens, united
carpels and the dorsi-ventral type of petioles. In view of these fundamental morpho-
logical and cytological differences, Asterop yrum is better raised to the level of Tribe.
However Asteropyrum and Coptis may represent two divaricate evolutional lines of
Thalictroideae.
(2). The systematic position of the genus Kingdonia has been much disputed in
the past. We support the view of Sinnote (1914), namely, the trilacunar in leaf
traces “the ancient type”, appeared in the angiosperm line very early, while the uni-
lacunar of Kingdonia may be derived from the trilacunar. On the basis of the chromo-
some numbers and morphylogical observation, the present writer accept Tamura’s and
Wang’s treatment by keeping Kingdonia in Ranunculaceae instead of raising it to a
family rank as has been been done by Forster (1961). Kingdonia and Coptis are
similar in having short chromosome with x=9, but with one-seeded fruits; therefore it
is suggested that placed into Thalictroideae as an independent tribe, indicating its close
relationship with Coptideae.
(3). Comparing with its allies, Calathodes being with out petals, seems to be more
primitive than Trollius. But Calathodes differs from Trollius with R-type chromosomes in having T-type chromosome with x=8 and subterminal centromere. Those charac-
teristics show that it is very similar to the related genera of Thalictroideae. But as
Kurita already pointed out that most speci es of Ranunculus have usually large long
chromosomes but some species have compar ativelly short chromosomes, therefore we
regard T-type and R-type chromosomes appear independently in different subfamilies
of Ranunculaceae. According to Tamura, G alathodes seems to be closely related to
Megaleranthis, because of the resemblance in follicles. But due to lack of cytological
data of the latter genus, the relationship between the two genera still is not clear pen-
ding further studies. From the fact that the morphology and chromosomes of the
Calathodes differs from that of all other genera of the Helleboroideae, we consider
Calathodes may form an independent tribe of its own with a closer relationship with
Trollieae.
