298 REPORT OF THE CONFERENCE ON GENETICS, 
ordinary biennial stock, we find that the black seeds have transparent 
seed-coats and the colour is entirely in the endosperm layer, the result 
seen, therefore, was undoubtedly due to double fertilisation, and not 
xenia. 
The other effect which has been taken for xenia is that of Mendelian 
dominance. In Piswm this error was easy to make, and all early records 
of xenia in this genus are open to this objection. 
In the experiments detailed below, which I have made myself, every 
precaution has been taken to ensure that sources of error shall have been 
eliminated. The flowers were emasculated before dehiscence of the 
anthers and carefully bagged, and pollen taken also from bagged flowers. 
MAIZE. 
Many experiments have been made with this subject, and cases of 
supposed xenia have been frequently recorded. 
In 1902 I obtained a race with a red seed-coat, which was mixed in 
with the commercial yellow variety. This was cultivated, and without 
exception came true from seed. 
This was crossed with the yellow variety, the pigment being held in 
this case also in the seed-coat, but no xenia was visible. Seeds of the 
cross were sown the following year, but owing to climatic conditions no 
seeds were produced. 
Webber has made many interesting experiments with maize, and 
much care was taken to ensure that the varieties chosen reproduced their 
seed characters truly. In some few cases the colour of the endosperm 
layer was changed as a result of double fertilisation, but the changes of 
form figured seem likely to be due to Mendelian dominance. 
De Vries, in an interesting paper in the “ Revue Générale de Botanique,”’ 
April 1900, quotes Vilmorin as haying observed changes of colour only 
in his experiments and no changes of form of seed, and on his own part 
suggests, as did Koernicke, that those changes occur only when an older 
character meets a younger, an idea similar to that once entertained with 
regard to Mendelian dominance. I can find no trace of any influence 
of the pericarp having been noted in maize. 
BEANS—RUNNER AND DWARF. 
I have tried very many experiments with these varieties of beans, but 
have had no success at all, the operation of emasculation being extremely 
difficult to perform, and the light colour of the pollen renders it almost 
impossible to ascertain if the stigma is pollinated or not. It is, however, 
notable that in these widely-grown plants, where in all cases the pigment 
is contained in the pericarp, I can find no recorded cases of xenia. 
Pras (Pisum). 
In the case of peas we come to the stronghold of xenia, as in this | 
genus the evidence is most plentiful. 
My own experiments were unfortunately mostly carried out with 
‘William the First’ as a pollen-parent, and as I found out too late the 
