Gaye:

     You might take a look at my dissertation. I built representative
     assemblages for family households, bachelor male households, saloons,
     brothels, restaurants, hotels and the military, then estimated the
     relative contributions of each type of activity to a city trash dump.
     The model was based on collections from 1880-1920.

     Citation:

     Catherine Holder Blee, 1991 Sorting Functionally-Mixed Artifact
     Assemblages with Multiple Regression: A Comparative Study in Historical
     Archeology. PhD dissertation, University of Colorado, Boulder. Available
     through University Microfilms, Inc.

     The hotel/restaurant assemblage might be similar to your teahouse. While
     the method requires a fairly rigourous sampling regimen, you might get
     some ideas that could help you out.


     ______________________________ Reply Separator
     _________________________________
     Subject: Re: 20th Century Refuse Patterns
     Author:  Gaye Nayton <[log in to unmask]> at np--internet
     Date:    8/12/00 11:35 AM


     I have been following this thread with interest as I have just
     unexpectedly (on the afternoon of the last day of course) recovered a
     lot of artefacts (a dustbin full) from a monitoring job at a footie
     oval. In the rush of recovering them I thought the dump was of domestic
     rubbish used as fill to extend the back of the bank.  The stuff has been
     burned as per this thread. However, on a slightly closer inspection I am
     not sure it is domestic in origin, the range of materials is narrower
     then what I would expect, it might be from the tea rooms and pavilion
     that used to be in the oval grounds.

Has anyone out there done any work on the percentages of different materials
or the range of artefact classes in 20th century town or household rubbish
dumps? Please keep in mind that my client is not interested in the results of
the monitoring and is not going to pay for me to catalogue that amount of
material let along analyse it so any methodology I try to match has to be
fairly simple and based on basic sorting.