Counted twice in 1870

So I'm going through records on the Family Search Pilot website looking for documents to confirm data that I had already known from elsewhere (but didn't make note of the references when I added the information last time).  I'm finding some new information, like a few new siblings to family members that I already knew.  But in the 1870 U.S. census, I found records that show what looks like a double counting for at least one family.

I'm working back through the Holmes and Pullen lines this week, with the ultimate goal of finding how my Holmes line connects to the famous Oliver Wendell Holmes.  Family legend states that Clark B. Holmes (born 6 October 1843, New Jersey) was Oliver Wendell Holmes Sr.'s nephew.  Clark married Phoebe Pullen (born January 1847, New Jersey) in 1868.  Phoebe had a sister named Anna M. Pullen (born December 1842, New Jersey) who married Stephen B. Bergen (born 14 December 1840, New Jersey) in December 1863.  From what I'm seeing in census records today, it looks like Stephen Bergen's family was counted twice in the 1870 census.

Here's the entry from the first time the enumerator came by the Bergen household on 11 July 1870:

A couple weeks later, the enumerator came back to the Bergen home and recorded the data again on 29 July 1870:

So the obvious question is whether the family was counted twice or not.  I can think of a few reasons why the enumerator might need to come a second time, but without any further evidence, it looks like this family was counted twice.  At least the data is consistent on both records!