POOR HILL 43 
Fig. 3. Photograph of Poor Hill (P5/227) after clearing, looking south-west. 
amphitheatre eastern set (see “terrace divison” on Fig. 2). The area of each terrace was 
measured and the terraces put in rank order on the basis of their size. The difference 
between the two terrace groups was subjected to a statistical test based on these ranks 
(Mann-Whitney “U” test). The probability that the difference between the two groups 
arose by chance is less than 0.01, which is significant (G. Law pers. comm.). This does 
not take into account the changes in slope angles (cross-section line A-B runs from 
about 10° to 16° down the slope) which may have influenced terrace construction. 
This seems unlikely, however, because terraces on the eastern side of the amphitheatre 
run from near the top of the ridge to near the bottom regardless of the slope angle and 
all these terraces follow the amphitheatre pattern. Gorby, in his notes, also 
commented that the terraces fell into two distinct groups and suggested that they could 
represent different social units. There could be a number of other reasons for the two 
contrasting groups such as changes in function or a time difference. There is, however 
no evidence that the pa and the amphitheatre terraces were not contemporary. 
The site is undated. A search, after the area had been disced, revealed no trace of 
European artefacts which might have been expected on a site say from 1810-1830 
onwards and none have been found during subsequent farming operations. 
