If you look at the larger top surface in the photo , the frosting from the scraper is only visible on the centre section,so there is also wear on the flat top part so idealy both the dovetail and the flat both need attention,I visited Tom Seniors many years ago,at the time there was a mill table set up for machining on a vertical mill, they followed machine tool manufacturers usual practice of having a deep cast iron "box" on the mills table,the top of the box was machined flat and the box stayed in place ,and possibly never removed.so the mill table plus the fixture was solid and remained dead flat and could not be easily distorted when clamping the work the be machined,I do not remember seeing a large shaper in the works. so I should imagine they had special size dovetail cutters made to suit the job,there were lots of specialists making cutters in those days. A big vertical mill in good condition is required to machine this job as it is so easy to get the xy axis out of square, and keep the top of the table to the z axis square.

If there isn't space for a dovetail cutter then it may have been done on a horizontal mill, I'll let you imagine the setup but it's not rocket surgery

It was probably originally planed and then ground or scraped.

The wear is quite apparent on the face in view which is concentrated around the ends as you would expect. This will have the effect of 'loosening' the table at the travel extents in the way you describe and coupled to the inevitable wear on the angled vee ways will make the table rock the way it is.

The thing is you cannot fix that by just machining the one vee. You need to machine/grind and/or scrape the top flat AND parallel with the y-axis flat way first and then map the wear in the front and rear vee then figure out where it needs machining/scraping. If you just go at it you'll have no idea if the x travel is going to be perfectly perpendicular to the Y travel both horizontally and vertically.

What you have, is normal wear.

The saddle dovetails wear wider at the ends and the table dovetails wear so they're narrower in the middle.

The top surface of the saddle will also have worn to be convex upwards

And that will be matched by the underside of the table wearing more in its middle.

You will also have some degree of bending of the table, as a result of the peening it has received from repeated clamping of work and fixtures.

_______________________________________

please don't rush in to "machine" metal away. You may well create more geometric problems than you cure, and you will create surfaces which will wear into inaccuracy, far faster, and travel less smoothly than a correctly scraped surface.

The two references for understanding what is going on and how to return the machine to schlesinger limits, are; Connelly, machine tool reconditioning

And

Moore, foundations of mechanical accuracy.

There was a yahoo group on rescraping machine tools, and there is a currently very active forum on practical machinist forum, with participants organising scraping courses on both sides of the Atlantic.