OBSERVATIONS ON TERMITES IN JAMAICA 207 



either remained within the nest or ran only in the main roads 

 that were laid down in the night. 



The movements of a lost termite liberated in the middle of 

 a wooded floor, where no termites had been, seemed quite ran- 

 dom and not orientated by light or presence of table or walls. 

 It ran rapidly and constantly in complex curves, often crossed 

 its own trail without returning on it or appearing to be affected 

 by it. sometimes running up the leg of the table and down again 

 and later repeating this, sometimes in the shadow of the table 

 and sometimes outside it; finally coming to a crack in the floor 

 too wide to reach across, it ran rather steadily along this to 

 the wall and under this. 



When in company with other termites spreading out from the 

 nest the termite finds the trail, or follows the track of the others, 

 even when running alone and many inches from any others that 

 have passed before. But an old track is not followed with the 

 speed and directness of a new one constantly traversed. 



The trail is followed not only on the solid bark or stone but 

 on the loose shifting sand where the particles are very large as 

 compared with the feet of the termites, being as small boulders 

 to a man. But on smooth glass in an aquarium the termites 

 do not progress readily; they cover the trail with excretions 

 and in this way we saw them finally escape up the side of a jar 

 as was discussed by Beaumont in Panama. In running along 

 the trail the termites hold the antennae wide apart and down- 

 ward, with oscillating movements. The palps also hang down 

 close to the trail beneath the mouth. 



Some suggestion of the nature of the means by which the 

 termites follow the track of others may be got from the results 

 of breaking the trail and the way the trail is found again. The 

 following rude experiments were made. 



A clean cover glass i8 mm. square was placed on a trail on 

 horizontal boards when no termites were passing. Immediately 

 the centrifugal or nest leaving workers came to the edge of the 

 glass they bounded back and ran aw^ay toward the nest. The 

 centripetal workers on the other side stopped and shoved against 

 the glass. The soldiers stood by the edge of the glass and stretched 

 out their antennae over its edge and also ran out laterally and 

 investigated the neighborhood. No termites got upon the glass. 

 Very soon one wandering soldier chanced to get to the accumu- 



