— 255 
Further Hybridisation involving A. fastigiata and 
A. CORYMBOSA. 
In 1889 farther experiments were carried out. Flowers 
of the above hybrid (hybrid a) were pollinated by pollen 
from the inner stamens of A. corymbosa. It acted with the 
greatest certainty, a full yield of seed being obtained. The 
seeds were sown, and the new hybrid (hybrid (3) flowered 
when 14months old. It was quite intermediate between the 
parents, and did not show any special reversion to the original 
species. It was interesting to place plants of A. fastigiata . 
hybrid a, hybrid (3, and A. corymbosa in a row, and 
observe the gradations in the foliage from the one species 
through the intervening hybrids to the other species. If 
A. fastigiata had been in flower, the same gradations 
would have been seen in the floral structure. The bulb of 
hybrid (3 (figs. 14, 15) is 1| inch in diameter when it 
produces its first flower-stem. The relatively great thick¬ 
ness of its tunics indicates its close relationship with 
A. corymbosa. A further series of hybrids was secured 
during the past autumn, viz. hybrid a X hybrid [3, hybrid 
(3x A - corymbosa , A. corymbosa X hybrid[3, A. corymbosa 
X hybrid a. 
By representing the names of the original species by 
their initial letters, the following will show the parentage 
of the new hybrids : — 
2(fXc)x d* (fXc x c). 
« ^ _ 
$(fXc X c) X a* c. 
x c X a* {fXc X c). 
$ c X a* {fX c ). 
The development of progeny having so complicated a 
