Up until the late 1950s, each airplane company building carrier-based jets for the U.S. Navy was responsible for providing the ejection seat in accordance with a Navy-furnished specification. (See http://thanlont.blogspot.com/2011/02/transition-to-martin-baker-ejection.html)
The specification reflected some of the features developed by Martin-Baker, the most notable being the use of a face curtain for the final step of ejecting. (There originally was a preparatory step, which in part served as a safety to assure the pilot seating on the seat that it would not fire accidentally.)
The Navy tested the contractor seats by ejecting them from the rear compartment of a Douglas JD.
This is an early Douglas seat with a separate seat bucket inside the seat frame.
This is a Grumman seat photographed by the Navy test unit in July 1949. Note the large headrest and "figure-8" face curtain handle.
While taking a closer look at pictures of the F9F-3, I was surprised to see that the ejection seats in ones taken in 1949 appeared to have a different headrest than later Panthers: the bottom of the cushion was notably lower. The shoulder harness attachment to the seat also appeared to be lower as a result. While the headrest was not the same as in the picture above, it was similar in terms of the size of the cushion behind the pilot's head.
The F9F-3s used to carrier qualify pilots on Boxer in 1949 definitely appear to have the larger headrest. I can't be sure about the ones that deployed with the air group on Valley Forge. For more on the F9F-3, see http://tailspintopics.blogspot.com/2012/01/f9f-3-panther.html