There's an image on Ancestry. The writing is generally clear but this name is hard to call. It definitely ends "rott" but the first letter is unlike any others by the same hand. It most resembles his "S" but with the loop taken much further round than usual and a strange dot in the middle. It's nothing like the "T" used for the John Trott married the next year.
If it has to be one of the two, then Trott. If Srott is an existing name, then I'd go for that.
ETA. Seeing that Ciderdrinker has come to the opposite conclusion I'll reconsider
That second letter is exactly as "r" was very commonly written in the previous century, but this scribe also uses a more modern form. In that one entry he has the old form in "April" and as the second "r" in "married", the newer form in "Richard" and "Huckleburge" and the first "r" in "married"! If what I thought was an "r" in the surname is actually two letters, then the second could certainly be the old form of "c" (which he does use, e.g. in "Richard"). But the first is not like any "s" I can see. He usually uses the long form, except at the end of a word, but that doesn't fit here. The "E"s elsewhere on the page are very different from the first letter we're puzzling over, and, I would say, not like the one in Carhampton either. The "r" in the Trott entry in 1715 is exactly like the putative "r" here.
Overall I'm less certain than I was, but I still think that's more likely "rott" than "scott".
David