Iris Breeding 
“On ripening, the seeds are removed 
from the pod, described, recorded and 
dried. I have the seed-bed prepared 
quite a time in advance so as to have it 
moderately firm; a plan is made, and 
the seed is planted about half an inch 
deep; the result of each cross in its 
respective square; and a square of wire 
mosquito netting is pinned down over 
it to protect it from accidental dis- 
turbances by birds or animals, as this 
is in August and the seed does not 
germinate, as a rule, before the following 
April. When about an inch high they 
are transplanted to a bed for seedlings, 
being placed one foot apart in the lines. 
Each is recorded on the plan of this bed 
and here they remain until they bloom. 
They require little attention (weeding 
and a few inches of leaves for covering 
in the fall) until the following year 
when many will bloom and the descrip- 
tions have to be written and records 
completed. 
“Fortunately any gaps in these de- 
scriptions may be filled, or additional 
information as to interesting points 
may be obtained, in subsequent years. 
Comparison can also be made with the 
parents, or offspring, when plants are 
mature, as records are kept of their 
permanent positions. 
“Although I have not grown enough 
generations to be able to foretell 
results, I have obtained some unique 
individuals; some as good as or better 
than their parents; and many similar 
to the well-known garden forms.”’ 
Miss Sturtevant’s irises show abun- 
dant evidence of segregation of charac- 
ters, and it is therefore fair to suppose 
that Mendelian heredity would be found 
in the German irises, if they were prop- 
erly studied. Such does not seem to be 
the case in hybrids between wholly dis- 
tinct species, however. 
For many years it has been an open 
question whether the Mendelian type of 
heredity applied to crosses between 
distinct species, or only to crosses within 
a species, or between hybrid forms. To 
throw light on this point, W. R. Dykes 
made a number of hybridizations, and 
describes the result in the Gardeners’ 
Chronicle of London (Vol. LVIII, 
pp. 196-197). With numerous pairs of 
503 
characters, the results showed no dom- 
inance but only blending. 
Tris boissiert, bulbous, with the beard of the 
sepals in the form of long straggling golden 
hairs 0.117 to 0.234 inch in length, crossed with 
I. tingitana having no trace of hair gave a 
hybrid with hair distinctly visible to the naked 
eye but less than 0.0585 inch in length. 
I. tectorum (which has a tuft in place of hair) 
crossed with J. cengialtu (a hairy type) gave a 
hybrid with a light violet coloured tuft bearing 
a short hair. 
I. xiphium (without perianth tube) crossed 
with J. tingitana and J. filifohia (having 
perianth tubes 0.975 and 0.507 inch, respec- 
tively) gave hybrids with perianth tubes 
respectively 0.507 and 0.234 inch long. 
I. clarkei with solid stems crossed with 
I. chrysographes with the internal cavity of the 
stem occupying about half the diameter, gave 
a hybrid intermediate with central hollow 
almost but not entirely closed with pith. 
I. pallida with papery spathes which become 
entirely white and dry before protruding from 
the floral opening, crossed with J. variegata 
with green and herbaceous spathes gave a 
hybrid with spathes green in the lower portion 
and parchment-like in the upper portion. 
The hybrid between J. reticulata and I. 
bakeriana is intermediate between the parents 
as regards leaf shape. 
Also with regard to the coloring of the 
petals many hybrids are intermediate between 
the parents of various species, e. g., I. pallida * 
I. variegata; I. trojanXTI. variegata; I. boissiert 
xT. juncea; I. fulua * I. foliosa; I. forresti X 
I. sibirica. 
With the exception of I. chrysographes X I. 
forrestit and also possibly of J. pallidaXlI. 
variegata and of I. fulva XT. foliosa all the above 
hybrids were sterile both with respect to their 
own pollen and that of both parents. The two 
possible exceptions are cases in which the 
parents are somewhat related whilst the fertile 
hybrid has more definitely related parents. 
Another interesting species-cross, 
which to some extent confirms the above 
conclusions, is reported by S. Mottet. 
in the Revue Horticole of Paris (87, pp. 
582-583). Iris pumila, a species which 
flowers early (beginning to end of April) 
with J. germanica, of which the earliest. 
flowers appear about the middle of May, 
yielded numerous varieties flowering in 
the first half of May, thus enabling 
growers to have a continuous supply of 
iris for about three months. Mottet 
has given the new forms the horticul- 
tural name of Iris ‘interregna and 
describes them as intermediate between 
the two parent species not only in date 
of flowering but for height, leaves and 
dimensions of the flowers. Miss Sturte- 
vant adds that interregna flowers are 
often larger than those of either parent. 
