It's a good question Fiona, and one that needs answering, but I'll
warn you you won't get the answer here.  But I thought I would put in
some of my thoughts.

First, I don't think there is any doubt that women can get Candidal
infections of the nipples.  The incidence is probably even increasing,
because we are setting up conditions which are just dandy for Candida
to live.  First of all, women do have an increased incidence of
Candidal infections during pregnancy.  Second, it is virtually
impossible for a mother or baby to get out of hospital in 1997 in
North America without getting antibiotics.  (This is a research
project I've proposed for maternity for years, and it does not look as
if anyone has taken me up--just count the percentage of mothers and
babies who get antibiotics in the perinatal period.  I would be
surprised if it doesn't approach 50% of all mother baby pairs).
Furthermore, candida grows better in damaged skin, and converts to its
pathogenic form in the presence of serum (oozing in cracks, say).
It's a perfect setup and I have no problems believing in nipple
candidiasis.

But is there such a thing as Candida in the ducts?  Your points about
milk and anticandidal antibodies and all that are just.  The pain in
the breast, the pain in the back could obviously be referred pain (and
not even just from candidal infection of the nipple, but any source of
nipple pain).  Treatment of the nipples often does get rid of the
breast pain.  But I have also seen women with breast pain only, who
respond so well to fluconazole, so quickly, that I cannot be help
believing that there is something to this.  Interestingly, gentian
violet has helped on occasion even when the mother has no nipple pain,
but breast pain alone.

The research needs to be done.  Several years ago, I proposed such
research to our Candida "expert" at the Hospital for Sick Children,
Toronto, and she was not the least bit interested. Maybe Lisa Amir has
some ideas.

Jack