PEELIMINAEY NOTE ON THE PEOTECTIVE VALUE 
OF COLOUE IN MANTIS EELIGIOSA. 
By a. p. di CESNOLA, Queen's College, Oxford. 
During last snramer vacation, at Professor Woldon's suggestion, I undertook a 
small series of experiments upon protective coloration from a statistical standpoint. 
The form chosen was Mantis religiosa, which occurs in Italy (where the experiments 
were made) in tAvo forms, a green and a brown. It is interesting to notice that the 
green form is always found upon green grass, the brown form upon grass burnt by 
the sun. The green form is characterised by its more sluggish behaviour ; the brown 
form is more active. 
For the purpose of this experiment I collected 110 specimens of Mantis religiosa, 
45 green and 65 brown. Each individual was tied by a silk thread about six inches 
long to a plant, the thread being tied round the animal's thorax, and each Mantis 
being attached to a separate plant. 
The individuals were divided into four groups, two green and two brown ; of 
the green individuals, 20 were tied to green plants in a place covered with green 
herbage, the remaining 25 being tied to brown plants in a place where the herbage 
generally was burnt and brown. In the same way, 20 of the brown individuals 
were tied to brown plants in a burnt-up spot, and 45 were tied to green plants 
on a green piece of ground. The whole series was tied in this way and exposed 
on August 15th, and observations were made daily during seventeen days. 
The death-rate during the period of observation may be gathered from the 
diagrams Figs. 1 and 2, and it will be seen that during the whole period the 20 
green individuals exposed among green grass, and the 20 brown individuals exposed 
among brown grass were untouched by enemies, so that all of them survived. 
Of the 25 green individuals exposed upon brown grass, the last was killed by 
August 25th, eleven days after the commencement of the experiment. Of the 
45 brown individuals, exposed upon green grass, ten were left on September 1st. 
On the evening of September 1st a gale occurred, which ultimately swept away the 
remaining individuals, and brought the experiment to a conclusion. 
Of the individuals which died, neai-ly all were killed by birds ; but of the 25 
green individuals exposed on brown grass, five were killed by ants. 
It will of course be necessary to repeat the experiments upon a larger scale ; 
but they seem to show in a fairly convincing manner the value of protective 
coloration. 
