There were a couple of egregious bugs in kanova() that caused testing
for interaction to give spurious results.
These bugs have now been corrected.  The corrected version, 0.3-20, is
now on CRAN.
The bugs *probably* have caused no harm, since testing for interaction
(against an additive model) in the K-function context is unlikely to be
of interest.
cheers,
Rolf Turner
-- 
Honorary Research Fellow
Department of Statistics
University of Auckland
Stats. Dep't. (secretaries) phone:
         +64-9-373-7599 ext. 89622
Home phone: +64-9-480-4619