I think one of the areas of this latest hypothesis that deserves attention is the dating of the life of the original Gerard. Some of these dates seem a bit of a stretch.
For example, Ormerod dates the earliest charters mentioning Gerard's name--only by reference to his son, William, though--occur in the reign of King John, ended 1216. He also strongly implies, on what grounds I do not know, that the "avunculus" charter was issued by the Roger, son of Robert II de Montalt, who died in 1232. This seems to make sense to me, since another charter to William, son of Gerard was issued by (a) Robert de Montalt.
But in the pedigree on page 131, Ormerod implies that Emma, daughter of Richard de Kingsley, the widow of the grantee of the 1232 charter, William fitz Gerard, was still alive in 1260 (44 Henry III) or even 1317 (10 Edward II). A bit of a stretch, to say the least, since the birth of that particular Roger de Montalt (~1165) brackets range of possible birth years for William fitz Gerard to between ~1130 and ~1165.
https://books.google.com/books?id=dI...ley%22&f=false
Elsewhere in this same book, Ormerod implies that the nuptials of Emma's sisters, married into the Done and Thornton families, took place around 1216.
Honestly, I think Ormerod may be confusing 2 or even 3 distinct individuals here, and possibly making some unwarranted inferences from the mere mention of these individuals' names. There are no details provided on these documents beyond a date and in some cases the name of the archive in which they were deposited (e.g., "plea rolls"), so it's impossible to feel very confident.
One thing that may contribute to the confusion is the apparent fact that the William, son of Gerard Ormerod lists as marrying Emma de Kingsley was succeeded in turn by two consecutive Williams. If the family had already begun to use "fitz Gerard" as an hereditary surname, instead of a mere ephemeral patronymic, some confusion would be expected.
At the moment, I suspect that it was the 2nd, not the 1st William "fitz Gerard" who married Emma de Kingsley, but I haven't altered my copy of this hypothetical pedigree accordingly in part because I don't want to stray too far from my sources without more concrete evidence, but also because I'm not sure it would have an obvious effect on my conclusion regarding the agnatic origins of the Gerard family.