6 
Psyche 
[February 
that this mixed nest was in reality a portion of a parasitized 
colony (like number 10), that happened not to include any 
parasites. 
The following experiment also suggests that raids are made 
by H. americanus. On August 2 an island was made by inverting 
a large pan and surrounding it by a moat. Three small piles of 
earth were placed on the island, and on each of these was placed 
a bit of sponge and a few bits of paper. Nest number 9 was 
placed on one pile of earth, and on each of the others was placed 
a pure nest of L. curvispinosus. The three colonies all became 
established, in the soil under their respective sponges. By August 
10 there was very little loss by drowning in the moat, and rel- 
atively seldom were fights between foraging workers seen. Each 
Leptothorax nest had a single dealated queen, but these and the 
Harpagoxenus were practically never seen except when the nests 
were examined by lifting the covering bits of sponge or paper. 
The ants were fed chiefly on molasses and Drosophila larvae, 
and all seemed to be thriving. But by August 21 one pure colony 
had disappeared and the other was very weak. On September 14 
careful search revealed only the one flourishing parasitized 
colony. At this time there were observed to be some callow 
workers of Leptothorax present. While no raid was actually 
observed, it seems probable that both pure colonies were plun- 
dered by the Harpagoxenus (or perhaps by the auxiliaries) of 
the mixed colony, which did not change its position on the island 
during the course of the experiment. 
Another reason for thinking that raids occur is that pupae 
and callows of Leptothorax are so regularly present in the para- 
sitized nests. 
Winged queens are rare in H. suhlcevis. Adlerz failed to find 
any in Sweden, and in Germany Yiehmeyer has found only a 
few. Their place is taken by worker-like forms that usually 
possess one or more ocelli. Adlerz has observed that these 
ergatoids mate, and has shown by dissection that they possess 
spermathecse and well-developed ovaries. Such ergatoids have 
not been described in H. americanus , and the data show that 
true queens are common. Wheeler found one dealated queen, 
and six are recorded in this paper, together with nine winged 
