Pupal coloration in Papilio polytes. 417 
of brown pupae, though the dark surroundings were not 
regarded as necessary. 
As a result one green pupa was formed, and the remainder 
were intermediate between the two forms. 
Experiment VII. 
Another set of larvae, ten in number, were allowed to 
suspend themselves on the bamboo supports of the cage 
on which brown pupae were almost always formed; they 
were left in this position until 5 p.m. on the day after the 
night on which they had suspended themselves, and were 
then placed in green hght for the short period of daylight 
which remained. 
All formed brown pupae, but in each there were traces 
of green, not usual in normal brown pupae. 
At this point it was found that the interference with 
the pupating larvae had caused an increase in the percentage 
of cripples, and also had resulted in a certain amount of 
confusion between two of the pedigree broods. As the 
breeding experiments were of greater importance than those 
on pupal coloration, the later were suspended until such 
time as a number of wild larvae could be obtained—a time 
which unfortunately never came. The experiments, there- 
fore, were left incomplete, and were not carried out on a 
scale large enough to give results which can be accepted 
in detail without further confirmation. In spite of this, 
however, certain conclusions can be drawn with some degree 
of confidence. In the first place, it appears that the 
pupal colour is not controlled by the quality of the, light 
derived from the immediate surroundings; secondly, 
it is highly probable that the brown pupa is the result of 
an excess of light, while the green pupa is caused by its 
relative absence; thirdly, the critical period during which 
pigment is developed is the day after the suspension of 
the larvae, and, as an artificial curtailment of this day 
produces intermediates graduated in accordance with the 
amount of light lost, it appears that the formation of pigment 
is directly caused by light and is not a process which once 
started can be continued in its absence. 
Finally, if these deductions are true, it is interesting 
from a protection point of view to note that brown pupae 
must be formed in exposed situations, which in a state 
