Okay, here's my take on it.
For good measure, I gave the cantus firmus a try as the treble voice too (transposed up an octave), which let to a problem in bar 2-3: In order to prevent consecutive perfect consonanses in similiar motion, intervals over a 10th and unwanted repetitional notes at the octave jump, I found the best way around was to let the bass voice jump a fifth, from a third under to a third over the CF, thus making a smooth transition to an 8th. I'm not sure what Jeppesen says about this, but it was accepted by Fux and certainly commonplace in the Palestrina style. Furthermore, the 8-10-10-8 progression in bars 3-4 may not be the best theoretically, but I found that it sounded the best.
(edit: found a small error, fixing now)
YC2.pdf
YC2.sib