= 
1901] : Ames, — Lobelia inflata X cardinalis 297 
flowers produced seeds. These seeds were sown in boxes of earth 
on February 5, 1900, and placed in a greenhouse. In a few days 
they germinated and compact rosettes of ovate leaves were formed. 
On May first the plants were set out in a rock garden and there 
maintained their rosette form throughout the summer. They all 
finally died, however, before the end of winter, only one having sent 
up flowering shoots. Plants which in the fall of the year were trans- 
ferred to a greenhouse lingered along for a short time and then also 
died. 
The plant which bloomed had for the most part the characteristics 
of Z. inflata, L., but the flower shoots were produced laterally from 
the rosettes, with the exception of a terminal one which presumably 
on account of weakness, did not bear flowers. "The structural details 
of the flowers and fruits were the same as in Zo/e/ia inflata, L., but 
the color of the corollas was unusual, being of a sulphur hue at first, 
then changing to a greenish white. As this peculiarity in color and 
the persistency of the compact leaf rosettes were the only convinc- 
ing evidence that some change had been brought about in the con- 
stitution of the plant, it was thought that the unusual conditions in 
which the seeds germinated had exerted an influence. To obtain 
information on this point, seeds of Z: inflata, L. were sown in the 
same greenhouse where the seeds of the supposed hybrid had been 
planted. "This sowing took place on February 5, 1901. The seeds 
germinated in due time and small transient rosettes of leaves were 
formed. When placed in the rock garden the plants sent up their 
flower shoots in the ordinary way and were in every respect character- 
istic of Lobelia inflata, L. A comparison of the leaf rosettes of the 
supposed hybrid and of the species proved of value in forming con- 
clusions. In the hybrid the rosettes were large and vigorous, and 
formed from thirty to forty persistent leaves. In the species the 
rosettes were small and in the course of the growing season disap- 
peared. 
Lobelia cardinalis, L. on the other hand in the first year forms 
rosettes of leaves, though much larger ones than those of the sup- 
posed hybrid, and maintains them through, at least, one growing 
season. 
From these observations, we may conclude, that the attempt to 
impregnate Loédelia inflata, L. with pollen of Z. cardinalis, L. was 
successful. But the behavior of the seedlings was not at all what 
