HYBRIDISING OF MONSTROSITIES. 73 
ally six, and regards this as an example of increase of the female organs 
by hybridisation. I found upon my hybrids a fair number of flowers with 
six, and a few with seven pistils. I thereupon examined my stock plants 
of Lychnis vespertina glabra and found the same deviation among them. 
I had therefore simply overlooked it the previous year, and it is not 
to be doubted that, in this case at least, the malformation has not resulted 
from the crossing, but is simply inherited from one of the two parents. 
It results, however, from this that the same explanation may be true 
in other cases: the hybrids are, as a rule, examined more closely than 
their parents, and hence more malformations are remarked in them.* 
I found furthermore, in my bed of hybrids, tricotyledonous and hemitri- 
cotyledonous plants, divided leaves, triple leaves, fasciate stalks and forked 
ones, four-petalled flowers (K,,C,Mg), and other malformations. Most 
of these divergences, however, I found on careful examination to exist 
on one or both of the parental species. They were, therefore, simply 
inherited. 
The inheritance of malformations by hybrids I have often found con- 
firmed by experiments with species of other genera: it appears to me to 
form a rule which so far has been too much overlooked as an explanation 
of hybrid characters. 
The colour of the flowers in my culture bed varied greatly. The great 
majority were purple, some were quite white, others dark red, but, as it 
appeared to me, not so deeply red as the mother plants in 1892. Between 
these three principal colours there were numerous grades of intermediate 
tints. 
Gaertner also (J.c. p. 241) found the colour of the flowers of his hybrids 
to vary: by far the greater number were purple ; a few, however, were white- 
flowered. 
It appears to me to be very unlikely that the cause of this variability 
- is to be found in the crossing. I would rather assume as an explana- 
tion that the examples fertilised by me in 1892 in this connection 
were not of pure origin. They were selected from a varied mixture of 
colours as the finest examples. Regarding their integrity, I had at the 
time when I fertilised them no cause for doubting it. 1f they, however, 
were hybrids between the warious forms of the mixed lot in question, 
one would naturally expect white-flowered offspring as well as red to be 
yielded by the seed. 
On this view the circumstance, that my hybrids of the first generation 
were variable as regards the colour of the flowers, justifies the supposition 
that they have simply inherited this variability from their mothers. And 
this assumption may possibly embrace the essence of an explanation of 
the variability of so many hybrids in their first generations, since only 
in the rarest cases have the parental forms, in hybridising experiments, 
been carefully tested as to their constancy. It does not suffice that they 
appear to be of a constant type. Many hybrids are exteriorly hardly 
to be distinguished from one or other of the parents, and therefore 
many hybrids may easily be mistaken for true species. In many other 
* It has been so in other cases. When I began to give special attention thereto 
I found twistings and ascidia to be much more frequent than one would expect from 
existing literature. Compare for the first Ber. d. d. bot. Ges. vol. xii. 1894, p. 25; 
and for the latter, Dodonea, 1895, vol. vii. p. 129, Over de erfelijkheid van Synfisen. 
