If that is a wrong assumption, then I really don't understand how this afocal projection works I'm afraid.
Generally best to think in terms of two different
methods... one is projection
and the other afocal
Projection into a lens-less camera body:
1) This can be "direct projection" from the objective onto the sensor. (Not very common with most compound microscopes. With many it is not even possible).
2) Projection via a dedicated, designed projection photo-eyepiece. These are different from "viewing" eyepieces. They were made specifically to mount in a trinocular tube to project the objectives intermediate ("real") image onto film or a sensor. They typically were made with lower magnifications such as 1.67X, 2X, 2.5X, 3.3X, 4X. The power (magnification) selected for use depended largely on the camera format (film or sensor size).
3) Projection using a "normal" viewing eyepiece. A viewing eyepiece produces "virtual image" that requires the lens of the eye (or a camera lens if using a camera) to form a real image on the retina (or sensor). But.... by raising the "viewing" eyepieces some amount from the position it normally occupies you effectively "change" it into a projective eyepiece. The more it is raised from it's "base" position the smaller the projected "real" image (and the closer to the eyepiece that projected image occurs). This is what we are trying to accomplish with your adapter.
method is different from the "projection "method. With the afocal method, the camera tube is set up with an eyepiece that produces an image viewable by eye, (This type of image is referred to as a "virtual image"... it requires another lens to be seen or recorded). Ideally it will be in focus (to the eye) at the same microscope focus setting where the viewing eyepieces are in focus. The camera now has a lens attached (usually focused at "infinity"). The camera lens takes the place of lens in your eye. The camera with lens are positioned to "look" through the eyepiece. It then produces a "real" image onto the sensor.
It would really be helpful know if there is any way to have the eyepiece you are using placed in the adapter you have so that it is in focus (by eye) simultaneously with the viewing eyepieces. I am trying to determine if the adapter you have elevates the eyepiece too much even at the "lowest" location of the eyepiece.
I have a 5D MK3 and the photo eyepiece is a 10X and that pretty much has the same magnification and crop as looking through the regular viewing eyepieces.
Wouldn't a 2.5X yield a much smaller scale and visual representation?
No. As soon as you "elevate" the eyepiece above it's "base" location (the location where it is in focus -by eye- simultaneously with the viewing eyepieces), the 10X magnification no longer applies. The actual magnification into the camera now depends on how much it has been elevated. With a full frame camera (24x36mm sensor) a 2.5X projection of the intermediate image will provide nearly the exact field you see when using typical 10X viewing eyepieces.
The "intermediate" real image formed by the microscope objective is typically a circle about 20mm in diameter. Your 10X eyepieces see an 18mm diameter circular section of that intermediate image. (That's what the FN 18 marked on the eyepiece refers to). The diagonal of your camera sensor is 43mm. If you magnify that 18mm intermediate image 2.5X into the camera you have enlarged it to a 45mm diameter circle at the camera sensor. An excellent "fit" for a sensor with a 43mm diagonal. If you were to actually magnify that 18mm intermediate image 10X at the camera sensor it would be a circle of 180mm diameter. The sensor is only 43mm diagonally so it would "see" and record less than 1/4 of the image you are seeing through the viewing eyepieces... your pictures would look like a severe crop of the view you are looking at. In the camera you would be "wasting" more than 75% of the image produced by the objective.
If you have not seen this PDF you might find it helpful: