The locus of the wing motion needed for flying under water is quite similar
to the motion for flying in air. However, there are a couple of differences.
The wing doesn't have to support the weight of the body underwater, so
a higher thrust coefficient can be utilised, allowing a different stroke
plane and spanwise twist distribution during the wingstroke. Water
is about 814 times denser than air, so the water wing needs less than 1/800th
of the area required to fly in air. For a given body mass, this translates
to a span and chord about 1/30th that of a bird that flies in air.
As a visual example, look at the size of a penguin's wing and consider
the size that it would have to be for the animal to fly in air.
David Marjanovic wrote: (or somebody did)
Another problem:"flying "underwater could need a different movement than that needed for an active flight in the air.