THOMAS I FLORIDA. 331 



early settler built a small liouse, wbieh at one time served to accoimno- 

 date the occasional travelei-. All traces of this are now gone and, in 

 fact, the part of the mound on which it stood is believed to have been 

 entirely dug away. The nearly vertical face from which excavations 

 have been made otters an excellent means of inspecti^ig the structure 

 of the mound. The sides and base are buried in a talus almost exclu- 

 sively composed of Viripara t/eorgiana, Lea, which have weathered out 

 of the general mass, and owing to their form and strength have re- 

 sisted decay. To the casual visitor this talus would give the idea that 

 the mound was composed of clear Viripara shells, which would be a 

 very erroneous notion. After clearing away the talus it was evident 

 that the body of the mound is formed of mud and marl resembling 

 that previously described as underlying the orange grove and which I 

 am convinced was brought to the spot from the swamp to build the 

 mound. Land fi-om the beach would be liable to be washed or blown 

 away at any time and the marl was but a few yards away. The main 

 mass, especially toward the base of the mound, is composed of this 

 material unstratitied, and by the percolation of lime water rendered 

 almost as hard as stone. At about half the height of the mound slight 

 indications of stratification are apparent; here and there small layers 

 of clean shells, Viripara or AmpiiUaria, are visible, an inch or two 

 thick and a yard or two long in section, as if the shells from a repast 

 had been thrown out. Bits of charcoal, occasional fish, and other 

 bones are more abundant as we ascend. I did not succeed in fintlinga 

 single artificial article of aboriginal origin in all the exposed area and 

 talus after a careful search. About 2i feet below the surface, in the 

 compact material, I found one or two pieces of glass which had been 

 subjected to the action of fire, and which by age had become beauti- 

 fully iridescent. It had been originally quite thin and of i)ale green- 

 ish color, like that used for cheap looking glasses, such as ai-e used in 

 Indian trade. It may, however, have been a relic of the early white 

 settlers before referred to, though the depth to which it was buried is 

 adverse to this idea. 



" 1 collected of the rough material composing the mound, about 4 feet 

 below the surface, enough to fill a box such as holds 100 cigars. This 

 weighed about 5.^ pounds, and ^ pounds of it were broken up, the con- 

 tained shells were sorted and identified, with the following result, the 

 identifiable shells of each species being counted: 



V'nnpara georgiana, Lea .313 



Melania etouaheiisis, Lea 109 



Amnicola, wp. iudet 1 



UnUt buckleiji, Lea (valves) 30 



Unio ( valves) : _ . . 5 



Ameria acaJaris, Jay 4 



Ghindiua triincata, Say 1 



Helix ( Poljigyra) auriformis, Bid 1 



Zoiiites mhiiiscula^ Binuey 13 



