436 Trans. Acad. Sci. of St. Lours 
merely one of instinct. One nest of two layers was placed 
flat upon a table so that the emerging insects could not 
gain their freedom by way of the underside, which was 
partly open; the table merely replaced the board wall 
upon which it had been built. Fourteen wasps escaped 
in the normal way from the top layer and the periphery 
of the lower layer. Seven adults in the central cells could 
not escape, but each one bored through the wall and en- 
tered the adjacent cell, where they were found dead. 
They followed no special direction in breaking out of 
these lower cells; one went through the front wall, and 
the other exits were equally divided on either side. In 
one such dungeon three such dead prisoners were found, 
the original inmate of the cell, and the neighbor from 
either side. Had each one of these seven mature wasps 
had the instinctive courage or energy to push on through 
one more wall, all would have escaped. In another nest 
we found where the same inability had brought death 
to one insect, the only one in the nest that was so situated 
that its exit led to another cell. In a third nest the same 
was true for two wasps. We cannot call this defective 
instinct, but only very simple instinct, for in the insects’ 
normal experience they should have but one wall to pene- 
trate in order to gain their freedom. One would like to 
say at least that it is wonderful that the emerging wasp 
knows how to direct its exit toward the light, but even 
this is not always the case. I have another record of an 
eight-celled, one-story nest in which three individuals 
had bored through the side-wall into their neighbor’s cell 
and died there, instead of breaking through their own 
roof to freedom. So even this instinct of working out 
toward the light is sometimes defective. 
For a long time there was a scarcity of T. politum 
nests about the much-studied region of Wickes, but at 
