1917] Wheeler—Social Parasitism of Lasius’subumbratus Viereck 173 
observations of July 9. Nevertheless these observations throw no 
light on three interesting problems, which may be briefly consid- 
ered: 
1. The first problem relates to the elimination of the host queen, 
evidently a conditio sine qua non of successful colony formation by 
such temporary social parasites as swhumbratus. The elimination 
is supposed to be effected in one of three ways: Either adoption 
succeeds only in host colonies that have already lost their rightful 
queen through death by accident or old age, or the queen is killed 
by her own workers after and because the young parasitic queen 
has secured their allegiance, or the latter assassinates the host 
queen. The first hypothesis seems to be supported by the above 
described adoption of a subumbratus queen in a depauperate colony 
of neoniger, although it is by no means certain that this colony 
was queenless. The large sitkaénsis colonies containing recently 
adopted subumbratus queens, however, would seem to support the 
second or third hypothesis. In regard to this problem in the EFuro- 
pean umbratus, Donisthorpe (1915) says: ““No cases are known 
of the host queen and the parasite living together in a nest, so 
unless a female can only be adopted by a queenless colony, it must 
sometimes happen that a female is accepted by a colony already 
possessing a queen of its own species. In such a case the intruder 
must either kill the rightful queen herself, as the female of (non- 
British) Bothriomyrmex kills the queen of her host Tapinoma or the 
workers of the host species must themselves assassinate their own 
queen, as do the workers of Tetramorium cespitum when they have 
accepted a female Anergates atratulus.’”’ He adds, however, that 
“Crawley confined several queens of wnbrata with queens of nigra 
and the latter were always killed by the former, which although 
a little the smaller, is stronger and possesses more powerful mandi- 
bles.” It seems to me that the morphological peculiarities sug- 
gested by Donisthorpe, namely the much broader and larger head 
and stronger mandibles so characteristic of the female of all 
umbratus forms as compared with the same sex of niger, indicate 
very clearly that the parasite actually decapitates the host 
queen in much the same manner as Bothriomyrmex decapitans was 
seen by Santschi (1906) to decapitate the Tapinoma erraticum 
queen and the queen Wheeleriella santschii, the queen of Mono- 
morium solomonis (Forel 1906). 
