I've been thinking about it and the models had to be trimmed for center of bouyancy (CB) instead of center of gravity. This usually meant adding weight to the tail because the volume behind the desired location of the CB was greater than the volume ahead of the CB. When the model was trimmed to glide at a high angle of attack (slow glide) the vortices were bigger than when trimmed at a lower angle of attack (fast glide). By putting a third pellet at the center of the wing, the downwash angle could also be directly observed.

If I remember correctly, the density of water is about 700 times greater than the density of air in a standard atmosphere. That would mean that the speed in water times the chord in water would be about 700 times greater than for another chord and speed in air for the same reynolds number to apply. If, for example, the chord in air were ten times the chord in water, then the speed in water would have to be 1/70th of the speed in air for the same reynolds number in the two cases.