THE GUESTS OF ANTS AND TERMITES. 73 
former protects its owners from the attacks of Insectivora on account 
of their resemblance to warrior ants, and is also to be found in other 
species besides the true guests, e.y., Clerus formicarius and the large 
tropical Cicindelidae of the genera Trycondyla, Myrmecoptera, &c. As 
a rule this kind of mimicry is only found in animals that live in the 
neighbourhood of ants’ ne sts——species of bugs of the genera Systello-. 
notus, Alydus, Nabis, Myrmecoris, &¢., are examples. These latter are, 
in many cases, ant-robbers, and draw a twofold advantage from their 
dwelling-place, ¢.e., of protecting themselves from their enemies, and 
of having easy access to their prey. A third kind of mimicry exists, 
the purpose of which is to deceive the ants themselves, ant mimicry, in 
fact, in the true sense of the word. This kind of mimicry is only 
found in the trae ant guests, and is consequently a trustworthy 
character by which to recognise them. This third kind of mimicry varies 
in connection with the host’s powers of sight, whether they have well- 
developed eyes or are nearly blind. In the case of well-developed sight 
in the hosts, the mimicry, being intended principally to deceive the 
host’s eyes, begins with a similarity of colour and reaches perfection 
in the imitation of the structure of the host’s body. This rests more 
on deceptive reflections of light than on any real similarity of form, 
which is explained by the fact that the eyes of the ants are not strong 
enough to perceive shapes and forms. Lomechusa strumosa (pl., fig. 2), 
ecurring with the blood-red robber-ant (J’ormica sanguinea), gives us 
one of the most beautiful examples of this kind of mimicry. A less 
perfect example is seen in the glossy-black Myrmedonia funesta and 
laticollis that live with the glossy-black ant, Lastus fuliginosus. It is 
certainly not by chance that the most like this ant in colour of the six 
species living with it is the one most often found. These examples 
show that mimicry calculated to deceive the ant is found both in the 
true guests (Lomechusa) and in the hostile lodgers (Myrmedonia). In 
the former case it furthers the friendly intercourse between guest and 
host, and in the latter it hides from the hostile watchfulness of the 
host the rascal, whose size would otherwise make him noticeable, and 
who now appears as a wolf in sheep’s clothing. This difference in the 
mimicry of Lomechusa strumosa and Myrmedonia funesta is explained by 
the totally different intercourse between the two species and their 
hosts. The similarity is more perfect in the Ayrmedonia, because it 
hides itself like a murderer in the nest, and only creeps up to isolated 
ants, whereas the Lomechusa is accustomed to live amongst the ants as 
their spoilt pet. An isolated Lomechusa has hardly any resemblance to 
an ant, being much broader and plumper, but when it sits in the midst 
of ants its resemblance is so illusive that it can hardly be found. The 
light which is reflected from the concave sides of the thorax appears to 
the eye like the narrow back of the ant, and the rolled-up abdomen of 
the beetle reflects the light in the same way as the rounded abdomen 
of a fatant. Such reflections of light would be useless in Wyrmedonia 
funesta, as it is not in the habit of sitting in the midst of the ants. Its 
mimicry limits itself to such items as are of advantage to an 
isolated beetle, z.c., the glossy-black colour, narrow shape, and rolled- 
up abdomen. ‘This mimicry is intended to deceive the ants alone, as 
when the Myrmedonia finds itself confronted by a stronger enemy it 
protects itself in quite a different manner by rolling itself up and _ pre- 
tending to be dead. It then looks like a small lump of earth. 
