— 935 — 
with a result, which agrees with Murbeck’s, i. e. that the egg- 
cell becomes an embryo. No doubt other botanists have made 
experiments of the same kind, but I have not found more mentioned 
in the literature !). 
The results of the castration experiments may be 
summed up thus: In the genus Hieracium we have apogamic and 
non-apogamic species, together with transitions between both kinds; 
the three subgenera are in this respect not quite alike, the subgenus 
Stenotheca representing the most primitive stage with typical fertili- 
sation; the subgenus Pilosella being intermediate, as it comprehends 
both apogamic and typically fertilisating species, nevertheless mostly 
apogamic; and the subgenus Archieracium representing the most 
developed stage with nearly all species apogamic, only excepting 
the H. umbellatum-group. The genus Taraxacum has gone a little 
farther, as all its species are apogamic — as far as we at present 
know. Corresponding to these results the cytological investigations, 
which have been made, give a graduation in the abormal develop- 
ment of the eggcell. The power to make hybrids in Pilosella and 
Archieracium agrees well with the apogamic or non-apogamic 
development, but still here much remains to be done in both sub- 
genera and more too in the subgenus Stenotheca; the following 
experiments with hybridisation in the subgenus Pilosella will, taken 
together with the earlier experiments by F. Schultz, G. Mendel, 
and A. Peter, throw some light upon several perplexing facts. 
III. Hybridisation Experiments. 
When I did my first crossmg experiment by bringing the pollen 
of H. aurantiacum to the stigmas of H. pilosella, I thought, it 
would be of no result at all, as I had lately discovered the power 
to set fruits after castration in the two species here in question. 
Therefore in my first note in Ber. D. bot. Ges. (1904), p. 380 I 
mentioned the experiment as one without any probability of success. 
But a few months after (1904, 2) I could annonce (pp. 538—539), 
that this very simple experiment had given rise to 19 individuals 
of which one was an unquestionable hybrid, having characters 
intermediate between those of the parents. 
1) Perhaps I may here note, that I have tried castration with some other 
Composite, but with negative results, viz. Calendula- and Dimorpho- 
theca-species, further Aster- and Eupatorium-species from North-America. 
