426 REPORT OF THE CONFERENCE ON GENETICS. 
CARNATION-BREEDING IN AMERICA. 
By C. Wruuis Warp, Queens, N.Y., U.S.A. 
TxouacH I have devoted a considerable time to the breeding of carnations 
and to the study of them during the past fourteen years, in view of the 
complexity of the subject and my lack of scientific knowledge, I feel 
diffident in placing my views before a conference composed of the most 
advanced scientific talent of the world. I shall not attempt to enter into 
a scientific discussion of the subject, but will endeavour to state in plain 
language what seems to me to have been thus far accomplished. 
As most people know, the original carnation, which has been known 
in history for several centuries before the Christian era, was a five-petaled 
single bloom about one inch in diameter and of a pinkish-mauve colour. 
It was distributed in its wild state over the whole southern half of the 
temperate zone in Europe, but was known more particularly to historians 
as inhabiting France and Northern Italy. It was found in abundance in 
Normandy, from whence it is generally believed to have been introduced 
into Great Britain about the time of the Norman Conquest. Even so 
recently as 1874 it was found in a wild state covering the Castle of 
Fallaise, in which William the Conqueror was born. It was described by 
Theophrastus as early as 800 B.c., and has been frequently mentioned in 
history since that date. 
The carnation of to-day, the subject upon which I am working, is the 
product of several centuries of hybridisation and culture. It is an open 
pollinated species, and mother plants can be chosen from amongst hybrids, 
as well as by inbreeding upon the same plant or upon plants of the same 
variety. A variety once produced from seed is easily perpetuated for a 
certain period by propagation from cuttings which are easily rooted and 
usually-secured in abundance. Improvements in varieties may be made by 
bud-selection, and new varieties are sometimes secured by bud-variation 
(sports), as well as from seed-variation and by hybridisation. 
In raising varieties from hybridised seed very few improvements are 
produced, the proportion being one good variety for every thousand 
hybrids grown, and probably as little as one decided advance in each five 
or ten thousand hybrids grown; and unless some law is discovered 
whereby we can forecast more surely than we can now what certain 
specified hybrid seed will produce, it would seem as if even this low 
proportion of valuable new varieties would decrease rather than increase 
in view of the higher standard which is being demanded from year to 
year. 
Up to the present time I have been pursuing my studies in breeding | 
upon the single subject of colour alone, basing my work upon the theory 
that any laws developed in colour would hold good when applied to the 
development of other desired qualities ; and this hypothesis I still believe 
to be correct. 
