1886. ] Notes on the Leaf-cutting Ants of Trinidad. 125 
they were all stopped up by soil washed into them by the delug- 
ing rains that had been falling for several previous days. Cut- 
ting my way through the bushes by means of that useful and 
indispensable part of a forester’s outfit for tropical woods, the 
“machete” or cutlass, I found, some twenty yards up the hillside, 
an entrance from which led, as far as the eye could see, a wide 
smooth path, worn by repeated travel some five inches deep, and 
carefully cleaned of all vegetation, dead leaves and rubbish. A 
few yards from the entrance a huge tree had fallen but recently 
across the pathway, but the industrious insects had dug a tunnel 
six inches in diameter under it in preference to climbing over it 
or making a new path around it. A little farther on I met 
another instance of formic ingenuity. The path led to the edge 
Fic. 1.—An Cicodoma formicarium. The cleared space is forty-five feet in 
diameter, ‘ 
of a ravine where it branched; one branch led directly across the 
ravine, down the precipitous sides of which an oblique path had 
been excavated at an angle of about 45°; the other branch led 
up the edge of the ravine some twenty yards to a fallen tree 
which spanned it. Over this the pathway led to the opposite 
bank, down which it ran to join the direct path below. I subse- 
quently noted that during the rainy season when the ravine held 
a stream of water, the ants toiled up the hillside to their bridge, 
but, as soon as the water dried up they used the nearer path 
directly across the ravine. On looking around the mound I 
found five other entrances to the formicarium, all at some distance 
: from it, and from each of these diverged a pathway through the 
Woods. Along one of these traveled a dense column of ants, those 
