21S [Sept. 



Hills disxecteJ, 

 First hill opened 22il March, 1S42. General outlines very much like those 

 of a hay stack ; situated in a valley. 



Measurement. 

 Circumference at base, - - - - - . 34 ft. 



" at § height from base, - - - 25 ft. 



Height from apex to base on the surface, - - - 13 ft. 

 '*' " " perpendicular, ... 9 jt. 



The work was begun with three men at 20 minutes past 4, P. M., and re- 

 quired 2a hours to accomplish it. 



The material was reil clay obtained about two feet below the surface soil, the 

 latter being a mixture of sand and decayed vegetable matter brought down from 

 the surrounding hills. The surface was highly indurated, receiving a slight im- 

 pression from a single blow of the mattock. 



The order first seen was the workers, who instantly retreated on exposure 

 to the external air. They were succeeded by one, and then another, and then 

 many of the larger class of soldiers, who, rushing out in great rage with jaws ex- 

 tended, threatened vengeance on the intruders. 



The experiment of permitting them to bite was tried several times, when it 

 was perceived that a drop of brownish fluid was exuded upon the part. The 

 sensation was like that of a minute sharp cutting instrument, the jaws moving in 

 cross direction like scissors. 



On breaking several of the upward projections or " turrets," they were per- 

 ceived to be hollow, leading into the " dome," and the main passages in the walls 

 down to the basement. These several passages were smooth, as if by being well 

 worn by constant tread, and it undoubtedly is throucjh them that their food is 

 brought from below to the "magazines." The first fragment of the hill ex- 

 posed numerous apparent perforations, from the size of a shot to that of a dollar, 

 which were increased by every stroke; these were the different passages, run- 

 ning in every direction and anastomosing with each other, keeping up a commu- 

 nication throughout the domicil. 



The walls seemed to be about twelve inches thick, and contained numerous 

 cavities or cells of various sizes and shapes, with young in different stages of 

 growth, extremely white and delicate. They communicated with each other and 

 with the main passages. The number of young contained in them varied from 

 twelve to twenty. When several were found in one cell, they were regularly 

 and closely packed, with their heads converging towards the bottom. The first 

 idea which this arrangement presented to my mind, was that of pigs in an 

 autumnal night, stowed in the angle of a " Virginia fence." 



Having beaten away the wall of the hill, a layer of light-brown spongy sub- 

 Btance was seen, its structure irregularly cellular, and enclosed in red moist 

 clay of corresponding form; the "nurseries" of Smeathman. The cells con- 

 tained young of different sizes; on the surface were visible numerous scattered 

 minute white globular bodies, probably fungi. Messrs. Kirby and Spence sup- 

 pose them to belong to the genus Mucor. But the mucoridei are generated from 

 decayed animal and stercoraceous matter. Without a microscopic examination, 

 they seem to me to be assigned more naturally to the Trichocisti, perhaps Tri- 

 ehia,tkepi?i head fungi, which are known to spring from decayed vegetable 

 substance. It is highly probable that the material of which these nurseries are 

 made, is at base ve.-etable matter. Their extent, as thus observed, is from the 



