Field Studies of the Non-Social Wasps 403 
struction, so a record of the proportions and materials 
will suffice for comparison. In the bottom of one 8%- 
inch tunnel was a half inch of broken pith, a cell with 
a cocoon of mud with papery or glazed lining and an 
empty space equal to the length of the cocoon; above 
this was a papery partition, and on top of this a layer 
of dirt which had been dropped and lightly packed in 
—not plastered like masonry—a plug of pith % inch 
thick, an empty cell 1 inch long, another papery roof 
over this, and a mud plug of 8g inch depth on top of 
this. This upper cell had been left empty and termi- 
nated this small nest. Another, in an elder stem this 
time, began at the bottom with the usual pack of pith 
chips 4% inch deep, then a cell containing its cocoon and 
Some additional space, a roof over this and a pack of 
pith 14 inch thick, topped with a layer of mud %4 inch 
thick; another cell with its roof this time of mud % 
inch thick, and a layer of pith 5 inch, another layer of 
mud 5/16 inch and again pith 14 inch; a third cell 
% inch long covered by a mud roof %4 inch thick, an- 
other cell with a roof of pith % inch, and beyond this 
the open hollow of the stem for several inches, showing 
all along its length the horizontal jaw-marks where the 
pith had been torn out. 
Any attempt to formulate rules or system from such 
chaotie data is obviously absurd. It may develop later 
that the mother wasp oviposits and provisions the cell 
and then fills in the upper part with dirt which the 
larva at pupating time pulls down around itself to make 
its cocoon, and that sometimes it tears down the whole 
supply, clear to the pith or papery partition, and some- 
times it leaves a part of the dirt adhering to the roof, 
forming a mud layer. On the other hand, it may be- 
come more apparent that this wasp is a versatile little 
