Arnica peduncalala 
Si 
ABNORMAL RAYS IN COLLARETTE FLOWERS 
1. Ray of the first collarette dahlia (President Viger) showing the additional 
lobes described in the text. 
2. Ray of the collarette coneflower with long, narrow, additional lobes. 
3. Arnica pedunculata var. tubularis: variation in rays, showing appendages 
at throat of tube. (Fig. 20.) 
flower, was described by DeCandolle 
under the name pulcherrima as early 
as 1836. This year Miss Hazel Andrews 
has found at Boulder a variety with 
pale yellow rays, like the ‘‘primrose’”’ 
sunflower, and if this is crossed with 
pulcherrima, the F, will certainly in- 
clude a vinous Rafibida. Varieties of 
Ratibida columnifera with twin or 
double heads, with short rays, with 
tubular rays, and with rays incised at 
the end, all exactly parallel variations of 
the Helianthus annuus group. 
There is no reason for ascribing all 
these parallel variations to “‘reversion”’ 
to some common ancestor possessing 
such characters; we must rather suppose 
that the genetic composition of the 
whole group of genera is such that these 
particular changes arise from time to 
time, without reference to the environ- 
ment. Whether, in any given case, 
they are due to original variations of the 
germinal substance, or are due to the 
cropping out of characters for which 
determiners have existed in the chro- 
mosomes for ages, may be extremely 
difficult to decide. 
Annual Meeting of the A. G. A. 
The next annual meeting of the 
American Genetic Association will be 
held in connection with the American 
Association for the Advancement of 
Science, in New York City in the week 
following Christmas. It is anticipated 
that that week will bring together one 
of the largest gatherings of men of 
science that has ever met in the United 
States. Members who desire to present 
papers at the meeting of the American 
Genetic Association should notify the 
secretary as far in advance as possible. 
Papers which are suitable will be pub- 
lished in the JouRNAL OF HEREDITY; 
others may be published by the authors 
elsewhere. Further details will be an- 
nounced as soon as possible. 
Breeding Citrous Fruits 
All possible combinations of citrous 
fruits are being made at the Citrus 
Experiment Station, Riverside, Cal., 
according to Director H. J. Webber. 
Among the most important are: (1) 
Combinations of the mandarin and the 
tangerine with the good varieties of the 
common orange to obtain large fruited 
varieties with the easily removable 
skin of tangerine and the quality of 
common orange. (2) Cross of tangerine 
and blood oranges to secure blood tan- 
gerines. (3) Crosses of lemon with 
blood oranges to secure blood lemons. 
(4) Crosses of blood oranges with pomelo 
to secure blood pomelos and the like. 
431 
