METHODS OF RESEARCH 153 



A second plan was immediately successful and in quite a 

 wholesale way. In the late afternoon we suspended a light 

 net from the outside eaves, so that it hinig downward over the 

 entrance to the "battery." In an hour vampires began to 

 fly out and become entagled in the meshes. One after the 

 other we freed and examined them, liberating all but the 

 very young ones. The net was later removed, the colony 

 remained intact, and we had achieved our desires. 



These and scores of other tricks of the trade were 

 learned by constant experience. At first all we could do was 

 to walk silently through the underbrush or squat motionless 

 at the foot of some great tree in a likely looking spot. And 

 even after years of jungle observation I still resort to these 

 two methods again and again. They are the ones where piu'e 

 luck enters in, and every carefully taken step is a gamble, 

 every passing minute of waiting is filled with ex])ectancy. 

 Silence and apparently lifeless surroundings may be the re- 

 ward, or suddenly there may be perceived some new strange 

 creature or some unimagined habit. It was while taking 

 shelter from the rain in a great hollow tree on the present 

 expedition that I first saw a tinamou — one of the large spe- 

 cies — moimting a slanting tree-trunk. And this was the final 

 proof which was all I wanted to put the seal of certainty 

 upon the careful investigation Avhich I had undertaken. 



NOTE — Jungle pit No. 5 (Fig. 35, page 149), was the scene of the won- 

 derful ant battle which I have described elsewhere (Atlantic Monthly, April, 

 1917, page 514). 



