1.  To begin with, Mark and Moses are not brothers of James and Martin.  Mark HOPKINS and Moses HOPKINS are known to be sons of Mark HOPKINS (Sr.) & Anastasia Lukens KELLOGG.  Both Mark and Moses were born in Henderson, Jefferson Co., NY, not North Carolina.  These HOPKINS'es descend from a well known immigrant to Cambridge, MA, whose descendants migrated across the northern tier of states, from Massachusetts to New York, to Michigan and Canada, then points westward.  These are not the HOPKINS'es of the southern United States.

The genealogy of these HOPKINS'es is well known via the publication of Timothy HOPKINS's 1932 book, John Hopkins of Cambridge, Massachusetts, 1634, and Some of His Descendants (Stanford Univ. Press, Stanford, CA).  Given the availability of this book two decades before LATTA wrote hers, this blatant error in her claims caused me to look further into her other claims by "doing" her genealogy.  The results of that effort are shown in the pedigree chart below and the family group sheets linked to it. 

2.  Estelle and her near ancestors are to be found in four adjacent NC counties:  Granville, Person, Orange, and Durham — though mainly Person Co. — not in Randolph [or Montgomery] Co., NC (see map).  See also North Carolina HOPKINS census records.

3.  With regard to Estelle's pedigree, below, we are already past the generation where her pedigree should have intersected with those of Ned HOPKINS's sons, James HOPKINS (b. 1802/3) and Martin HOPKINS (b. 1808), and it doesn't.  Therefore, Estelle cannot be descended from either one of them. 

There were HOPKINS'es in the Person County area who may very well be kin to Estelle (e.g., the Alexander HOPKINS who married Sarah COTHRAN), but I've yet to find a connection between the Person County HOPKINS'es and either the Randolph County ones or the northern ones.

4.  A 1965 case in the U.S. District Court of Northern California (viz., SEC v. Latta — scroll down) clearly shows Estelle LATTA was engaged in an elaborate fraud regarding the estate of Mark HOPKINS.

So, please do your homework on this one, and definitely avoid using her book as documentation for anything — the history in it is as flawed as the genealogy.  Please see this discussion of her claims and this message at GenForum, though I would disagree with this message's opening suggestion that there is anything "accurate" about LATTA's contention that her family was defrauded.  She wasn't related to Mark HOPKINS, nor were the Randolph County HOPKINS'es related to him, so how can she or they possibly have been defrauded of something that was never remotely theirs? 

Please note that even if the "blue" connections in this pedigree were to fail, it still shows that Estelle (COTHRAN) LATTA cannot have descended from James HOPKINS (b. 1802/3) or Martin HOPKINS (b. 1808).  As she surely would have known the identity of her own grandparents, there can be no argument that she was simply an incompetent genealogist.  She had to have known her claims were false.