282 
Journal of Agricultural Research 
Vol. XXXI, No. & 
none of them showed infection in any stage. Although the writer 
has had these plants growing for two years, none of them has pro¬ 
duced seed. 
Allium heldreichii Boiss. Only three plants came up, and none of 
them was infected. 
Allium macranthum Baker. About 50 plants were obtained from 
the single planting. No lesion was found at any time on any of 
them. 
Allium suhhirsutum L. About 100 plants came up. No smut 
was observed in them at any stage. 
Allium roseum L. A very few plants came up and soon died, but 
no smut was observed. The writer is of the opinion that further 
tests should be made with this species. 
EXPERIMENTS WITH SPECIES WHICH REPRODUCE BY BULBLETS 
It has long been known that when onions are started from sets 
(small bulbs) they are entirely free from smut. In a previous pub¬ 
lication, the writer has shown that all infection of onions takes place 
through the cotyledons (i). As there are no cotyledons when a plant 
starts from a bulblet, it is easy to understand why onions from sets 
are immune. There are many species of Allium which never produce 
seeds but reproduce by small bulblets which form in dense heads in 
a position much like the seed clusters of seed-bearing species. There 
are other species which produce both bulblets and seeds, and all 
gradations may be found between these two methods of reproduc¬ 
tion. The most common native wild onion in New England (, Allium 
canadense) reproduces by bulblets only. Although the writer has 
found plants which were producing flowers and seemed to be setting 
seed, he has never found, that the seed matured in this section of the 
country. Bulblets of this species have been planted and many 
plants grown to maturity, but no smut has been found at any time. 
The results were the same using bulblets of A . roseum. The 
writer has not been able to obtain viable bulblets of other species, 
but there is every reason to believe that the results would not be 
different. Probably none of the bulblet-producing species are ever 
infected. In view of the widespread tendency toward the bulbiferous 
habit throughout the genus AUium, it is an interesting speculation 
as to the influence which the smut fungus may have had in eliminating 
the seminiferous strains and forcing T>y natural selection the devel¬ 
opment of bulbiferous species. 
EXPERIMENTS WITH THE WINTERRECK ONION 
Some seed labeled u Winterbeek Zwiebd” were received from 
Haag & Schmidt, Erfurt, Germany. The writer is not sure whether 
this should be considered a distinct species or only a variety of 
Allium cepa. It comes up more quickly than the common onion, 
and grows more rapidly, but does not develop a large bulb, and the 
bulb divides more readily, somewhat with the same general habit as 
the Welsh onion and the multiplier (or Egyptian) onion. Unlike 
the latter, however, it does not produce a head of bulblets but repro¬ 
duces entirely by seeds which form in heads, as in the case of the 
common onion. It is a stronger grower and more hardy than the 
common onion, less susceptible to damping-off and thrips. It m 
