1 8 University of Michigan 



the wood. Notwithstanding- the size of the colony there was 

 no outside evidence of its presence. 



Many other colonies were found. On the sixteenth a small 

 nest was uncovered in the low hardwood forest near the pond. 

 It was in a very cool, shaded place and the major part of the 

 nest was in the damp earth, with only a few passages in a 

 small rotten log that was partially buried in the ground and 

 covered the nest. No eggs, larvae nor pupae were in this nest. 

 but all of the other colonies found in the same habitat con- 

 tained one or all three of these in more or less abundance. 

 Many colonies occurred along the sandy beaches. Here the 

 species sometimes constructed small circular crater mounds one 

 to six inches in diameter and one quarter to two inches high, 

 with a single opening at the bottom of the crater. The burrows 

 led straight down from the entrance for an inch or two before 

 turning, but the extreme looseness of the sand prevented fur- 

 ther accurate observation. This looseness of the sand also- 

 undoubtedly accounts for the fact that the mounds were not of 

 the turret-like type that is so common in more compact soiL 

 Sometimes no mounds were constructed and the nest would 

 be located beneath logs and other beach debris. Frequently 

 if this shelter were punky from decay, it would be utilized for 

 a part of the nest and here the pupae would almost invariably 

 be found. No other colonies were found that approached the 

 size of the one first noted. 



On the nineteenth and twentieth winged males and females 

 became abundant. They were found everywhere on the island, 

 and females which had just shed their wings were collected in 

 greater or lesser numbers in all habitats. Large numbers of 

 them fell on the water all around the island and were then 

 extensively preyed upon by the common terns. The stomachs 

 of several of these birds taken at that time were gorged with 



