I plotted such references in a 2008 article titled: Memetics death report exaggerated.
I revisited Google Scholar and made another plot today. This time I used "memetic OR memetics" as my search term, which seems to be the most reasonable thing to search for to me.
Here are the results:
As before, it is best to ignore the last couple of years: Google hasn't indexed all the papers from those years yet.
This time I also did a "control" plot: I searched for "gold":
Here's my raw data:
Others on the internet seem to be noticing downturns in recent years too. For example, this guy appears to have concluded that we are at "peak haplotype" - on similarly flimsy evidence.