One day an army of carnivorous ants invaded our quarters while we were busily occupied preparing the specimens collected during the morning. The first intimation we had of the arrival of the ravaging host was when scores of cockroaches suddenly appeared and frantically ran up the walls of the room. Not long after, several centipedes eight inches long joined the fleeing cockroaches, and before long a number of scorpions followed in their wake, hotly pursued by the multitude of ants. There was nothing for us to do but follow the lead of the panic-stricken insects, so we hurriedly transferred our collections to a zone of safety outdoors, and waited a few hours until the ant army had completed its work and gone on its way. The natives welcome these visits as the ants act as scavengers and rid the house of vermin. While at Puerto Valdivia we were presented with a young night monkey not larger than a good-sized mouse. It was a most interesting pet, and readily took to a diet of condensed milk, which it drank from a spoon. My companion, to whom the little animal belonged, kept it on the window-sill, from which point of vantage it took a lively interest in all that occurred within its range of vision. It so happened that there was a very small crack in the sill, and this proved to be a matter of the utmost concern to the tiny monkey. Hundreds of times each day it crept timidly to the crack and peered down into it anxiously, although there was only darkness below. When we held the pelt of an animal near it paid no attention whatever to it, with the single exception of the skin of one of its species, which it recognized immediately, and to which it clung tenaciously. When we left the hot climate of the Lower Cauca and started on the return journey to Medellin the little creature was unable to withstand the cold of the higher altitude and died. The purpose of our zoological exploration of this section of Antioquia was to secure material that would throw light on the geography of the country farther north; for, beyond the general knowledge that the junction of the Cauca and the Magdalena mark the breaking down of the Cordillera Central, we knew comparatively little of a definite character about this part of Colombia. It was not until several months later that our work farther west--on the Paramillo and the Rio Sucio--provided the material which, viewed from a distributional standpoint, furnished the clews that aided very materially in solving our problem. CHAPTER IX ASCENT OF THE PARAMILLO--COLLECTING ON THE RIO SUCIO The return to Medellin from Puerto Valdivia occupied five days. We again went to our former headquarters, the “Gran Hotel,” and spent a few busy days packing the large collections brought from the Lower Cauca. Then we began to gather provisions and cargo mules for a second expedition.