Bob, the routines from Microchip's AN-526 work fine. I use the Parallax
assembler but the routines are in Microchip's format. They are:

ACCaLO  DS      1               ; Accumulator A LSB
ACCaHI  DS      1               ; Accumulator A MSB
ACCbLO  DS      1               ; Accumulator B LSB
ACCbHI  DS      1               ; Accumulator B MSB

; Double Precision Subtraction ( ACCb - ACCa -> ACCb )

Dsub
        CALL    neg_A

; Double Precision Addition ( ACCb + ACCa -> ACCb )

Dadd    movf    ACCaLO,w        ; Addition
        addwf   ACCbLO          ; add lsb
        btfsc   STATUS,CARRY    ; add in carry
        incf    ACCbHI
        movf    ACCaHI,w
        addwf   ACCbHI          ; add msb
        retlw   0

neg_A
        COMF    ACCaLO
        INCF    ACCaLO
        BTFSC   STATUS,Z_bit
        DECF    ACCaHI
        COMF    ACCaHI
        RETLW   0
--------

   - Tom

At 09:06 PM 3/28/97 -0500, you wrote:
>Greetings,
>
>I am afraid that my binary math knowledge has waned over the years...  Can
>anyone provide me with a sixteen bit subtraction routine example ???
>
>The quibble of course is how to manage the carry (borrow bit).....
>
>Bob Segrest
>
>