Join separator field

Hello!

I am pround to use join for doing bioinformatics up to now. But when I 
wanted to separate field by tab character, I failed. May be it's a bug...

Im working under cygwin. Fields are separated by tab characters. When I 
specify tab char as separator, 2 lines are missing, for the 2 next the 
joined field is modified. When I specify # as separator, it works. I 
think I will use such a character as separator, but I hope one day I 
could use tab.

Regards,
--Samuel

bash-3.00$ more aa.txt bb.txt
::::::::::::::
aa.txt
::::::::::::::
a
c
image1000000
image:1000031
::::::::::::::
bb.txt
::::::::::::::
a       aa
b       bb
c       cc
d       dd
e       ee
image1000000    Hs.57732
image:1000031   Data not found
bash-3.00$ join -j 1 -t '' aa.txt bb.txt  > cc.txt
bash-3.00$ more cc.txt
a aa
c cc
image1000000 Hs.57732
image:1000031 Data not found
bash-3.00$ join -j 1 -t \011 aa.txt bb.txt  > cc.txt
bash-3.00$ more cc.txt
image1000000000000      Hs.57732
image:1000031000031     Data not found
bash-3.00$ tr "\t" "#" < bb.txt  > dd.txt
bash-3.00$ join -j 1 -t "#" aa.txt dd.txt  > cc.txt
bash-3.00$ more cc.txt
a#aa
c#cc
image1000000#Hs.57732
image:1000031#Data not found