475 
Influence of Carbon Dioxide. IV. 
normal untreated seeds, the seed-coats were removed after ten hours’ 
soaking; germination and growth of the bare embryos proceeded immediately. 
Eight experiments were conducted with embryos from inhibited seeds, 'the 
results of which may be summed up as follows (see Table XVI). In all cases 
a certain number of seeds, usually about 50 per cent., started to germinate 
after removal of the testa with a short lag amounting to about twenty hours. 
Where germination did not ensue in this way almost immediately, a definite 
continuation of the dormant condition became apparent. The period of this 
dormancy varied in these experiments from three to thirteen days. The 
cotyledons and radicle enlarged to a considerable extent by inhibition of 
water, and, where exposed to light, the former assumed their green colour. 
The whole embryo, however, had the appearance of a fully mature organ 
without the capacity for growth. Neither the hypocotyl nor the radicle 
showed the least sign of elongation by growth in the normal fashion. The 
photographs (Plate XXIII, Figs. 1 and 2) show the marked contrast between 
the embryos in this dormant condition several days after the removal of the 
testa and those seedlings in which growth had started almost immediately. 
With regard to the appearance and rate of growth after germination 
had once commenced, no marked difference was observed between the 
embryos of control untreated seeds and those of seeds showing secondary 
dormancy. 
These experiments afford the second piece of positive evidence 
in this research with regard to the causes underlying the phenomenon of 
secondary dormancy in White Mustard. In the first section of this paper 
it was shown that changes in the embryo resulting in injury prohibited the 
occurrence of secondary dormancy. It now appears that in the case of 
seeds in which secondary dormancy has been successfully induced, changes 
occur which render them less sensitive to normal growth conditions than 
the tissues of a newly-swollen untreated seed. 
The interpretation of secondary dormancy must now be, not that 
any change has occurred in the seed-coat, but that the power of the embryo 
to rupture the testa and germinate has been reduced. As has already been 
pointed out, the rupture of the seed-coat in Brassica alba results from 
a process of growth. In other words, the power of the embryo to respond 
to growth conditions and to germinate under the limitation of the seed- 
coats decreases during the primary period of inhibition in the presence of 
carbon dioxide. 
On the other hand, the broad fact that by the removal of the seed- 
coats the germination of seeds showing secondary dormancy can be induced 
must not be lost sight of. In the first place, it is impossible to remove the 
seed-coats with a fine sharp needle, the method employed in our experi- 
ments, without almost certainly causing some injury to the superficial cells 
of the embryo, or, at any rate, giving the whole tissue considerable 
K k 
