Jeanette-

I think mistaken identification of middens must be pretty common in initial testing because depositional strata can look much like the strata in a single episode of fill when you see them in a shovel test pit. We all have heard horror stories of expensive excavations being begun only to find that the deposits have no integrity. 

Many years ago on one of the Iroquois Pipeline sites I began excavating into what appeared to be a large, clearly defined pit feature filled with some nice midden deposits, including pearlware and a bone toothbrush near the top. A foot or so deeper, however, were a plastic panty-hose wrapper and a curtain rod. Four feet down the feature bottomed out in clearly visible marks from the teeth of a backhoe.

On the Spain's Boarding House Site in Thendara, N.Y. (early 20th c. lumber town boarding house), I identified six remote midden heaps in the woods behind the house foundation and two in the side yard area. One of the side yard middens was identified as such because it looked pretty much like the kitchen midden next to it and had similar artifacts on the surface. No subsurface testing was done on it because it seemed obvious that it was just one more like the others.  During data recovery excavations, however, it turned out to be nothing but a dirt pile. 

On the same site, deposits I had originally identified as nothing more than artifacts in landscape fill turned out to be the edge of a kitchen midden whose top had been shoveled off to fill in a depression right beside it. While it had no real stratigraphy any more, it was sealed under a datable architectural feature and represented deposition from an earlier occupation. It was quite useful for comparison with the later middens.

The Data Recovery report  on Spain's Boarding House is presently being edited for publication in the New York State Museum's Cultural Resources Survey Program Series. It contains an analysis of the different functional types of middens on the site and reasons for their locations. I can send you a copy of the analysis off list if you think you would find it useful.

The gray-lit. references follow:

Pickands, Martin
1997	Cultural Resources Site Examination of the Dowling Site (NYSM# 10240) and the Spain’s Boarding House Site (NYSM# 10241) for PIN 2018.71.121; 94PR 0442 NY 28 at the Adirondack Scenic Railroad, Hamlet of Thendara, Town of Webb, Herkimer Co., N.Y., New York State Museum Anthropological Survey, Albany
2005	Data Recovery Excavations at the Spain’s Boarding House Site and Relevant portions of the Dowling Site, New York State Museum Anthropological Survey, Albany


Martin Pickands
New York State Museum

>>> Jeanette Mckenna <[log in to unmask]> 3/21/2007 2:32 PM >>>
Folks:

I am working on a paper for the SAAs and would appreciate any information
you may have regarding midden.  To be more specific, I am interested in the
identification of midden deposits during surface surveying that ... after
excavations were determined to NOT BE midden.  I do not need a lot of
specifics on the actual excavations, but would like to cite some examples. 
On another approach - examples where midden was not visible on the surface,
but identified during excavations or in natural cuts.

Thanks,

Jeanette McKenna