nd was smooth and
bare, but there were no highways or roads leading to them.

Among the workers I saw some ants whose heads and mandibles were very
large. These ants never engaged in any of the agricultural pursuits of
their sisters; they were the soldiers and the sentinels of the community.
One nest migrated while I had them under observation, and I had the
pleasure of witnessing the behavior of these fearless little warriors
when on the march. The ants were moving nearer to their grain-fields,
and were carrying with them their young, etc. The route, from the old
home to the new, was patrolled on either side by soldiers. Every now
and then I saw one of these individuals rush aside, elevate herself on
her hind legs, shake her head, and clash her mandibles. She acted as if
she saw some danger menacing the marching column and would ward it off.
Others climbed little twigs or tufts of grass and scanned the surrounding
country from these elevated and commanding positions. Others hurried up
the laggards and stragglers, and even carried the weak and infirm.

These ants winnow or husk the grain after it has been carried into the
nest. All during the harvesting I observed workers bringing chaff from the
nest and carrying it some distance away. It is said by Texan observers
that the harvesters of that state bring the grain to the surface and dry
it, if, perchance, it becomes wet. I have never observed this myself, but
accept it as an established fact.[84]

[84] I believe that these observations on the presence of the
harvester ant in Arkansas are unique; at least I have been unable to
find any data corroborative of this fact. How did a fecundated queen
arrive at a spot so far from her usual habitat?--W.

The faculty of computing is among the very last of the psychical habitudes
acquired by man, and is an evidence of high ratiocinative ability. Many of
the savage races are unable to count above three,--some not above
five,--thus demonstrating the truthfulness of the above assertion. Yet I
believe that it can be clearly shown that some of the lower animals and
many of the higher animals are able to count.

The mason wasps, or mud-daubers, build their compartment houses generally
in places easily accessible to the investigator; therefore the experiments
and observations which I am about to detail can be duplicated and verified
without difficulty. These interesting members of the Hymenoptera, the
_avant-couriers_ of the