Diagiiiii for Indian pottenj and u-eapons. This rock-slielter of Fairfield County, 

 Connecticut, illustrates one of the most inferior of Indian shelters discovered, formed 

 merely by an overlianging rocky ledge but in favorable position at the junction of two 

 streams. The Indian used rock dwellings as temporary habitations, choosing the ones 

 which combine southern exposure with a near supply of water 



broken pottery, either on the ground or a Httle below the surface. Still 

 it is quite possible that some of these shelters, especially those M'hich were 

 favorably situated, may have been occupied by whole families for months 

 at a time, and this particularly during the winter season. 



Wherever man dwells he lea\'es traces as evidence of his whilom presence. 

 If scrutinized, these mute witnesses of days gone by furnish much informa- 

 tion of archaeological import. They tell us something al)out the mode of 

 living and the degree of skill attained in fashioning tools. Each rock- 

 shelter that I have in\estigated has presented features of its own. The 

 arrow point has been met with in nearly all of them, a notable exception 

 being an Indian cave near Paterson, N. J., where nothing was found but 

 flint blades and bones of many species of animals. Scrapers and chips are 

 of common occurrence, a fact showing plainly that the ancient occujiants 

 of these rocks devoted much of the time spent there to replenishing their 

 stock of weapons. As already stated, pieces of pottery are generally found 

 only under tho.se rocks which are easy to reach. Again, some shelters are 

 remarkable on account of the profusion of bones scattered throughout the 

 soil, with also many indications of ancient fires — fireplaces, heat-cracked 

 pebbles and smoke-stained rocks. The bones belong predominantly to 

 deer l)ut also to many kinds of ujiiiiials Inuited by the redman, bears, 

 opossums, rabbits, woodchucks, nuiskrats and wild turkeys, some now Ix'ing 



64 



