I've just discovered a major flaw in the Broderbund Marriage Index CD for Ohio, 1851-1900 (CD-236).  I've only tested Pickaway County, so there may be other counties on the CD with a similar flaw.

The CD contains only marriages for men whose surnames begin with the letters A through K; letters L through Z are missing.  There are wives with surnames A through Z, but only because they married men with surnames A through K.

So, if you're tearing your hair out because you think someone *should* have a marriage record in Pickaway Co., OH, between 1851 and 1900 — and the husband's surname begins with a letter L through Z — now you know why you can't find them on the CD.  Thankfully, there are post-1850 Pickaway County marriages online at the LDS Family Search web site.

On the 1820 Census Index CD, Coxsackie Twp., Greene Co., NY, has the state as PA.  This is potentially confusing for researchers in both states.  I have not checked to see what other townships or other counties might be afflicted.

.  Census Index:  U.S. Selected Counties, 1820 (Broderbund CD-314):

1820

Ely, John

PA

Greene Co.

Coxsackie Twp.

p. 72

Coxsackie Twp. is in Greene Co., NY, not Greene Co., PA.

Nancy Pfannenstiel has brought to my attention that on Broderbund Marriage Index CD-229, the marriages attributed to Monroe Co., KY, appear to belong to Monroe Co., TN.  The Monroe County Courthouse burned, so there are no extant county marriage records, and there shouldn't be any on this CD.  My preliminary examination confirms the error.

Some townships in Knox Co., OH, wrong on the 1820 census CD. 
Same problem with the 1810 Census of Northampton Co., PA.  Part of the problem is that the original bound census pages were unbound, flattened, and rebound before microfilming and numbering, one result of which is that townships in the index CDs are wrong.  I have made a correct index for Knox Co., OH (link above), but otherwise, you have to see the original film to tell what township someone is really in.  The problem was created by the National Archives — a pox on them for rebinding the schedules out of order! — but should have been spotted and corrected by the indexers, so two wrongs have conspired to create an unreliable index.  Probably the entire state's schedules are affected in both cases, that is, all of Ohio in 1820 and all of Pennsylvania in 1810, and who knows how many other states and years...

I've just discovered that the 1830 census index (Broderbund CD-315) has Hardin Co., TN, as "Yardin" Co. TN.  According to both my Webster's Geographical Dictionary and to Goldbug's AniMap software, there is not now and never has been a "Yardin" County in Tennessee.  It is clear, for that and other reasons, that "Yardin" is Hardin Co.

I've just discovered a record on the 1810 census index (Broderbund CD-313) for Genesee Co., NY, where the numerical data have a "frame shift."  I have no idea, yet, whether this is a fluke or if there is a pattern.  See this page for the record involved.