I don't know that there is an easy way to do this within the NCBI website. However, you can get some of the way with a well-defined search and some scripting.

I would start with a taxonomy search for Annelida. Clicking through (just keep clicking on "Annelida") gets to the summary page, showing (at present) 31 606 nucleotide records.

We can then qualify that search further:

txid6340[Organism:exp] AND COI[Gene]

to retrieve 11 208 records.

At this point, you should click on "Send to -> File" and choose an output format. "Genbank" or "XML" are good choices, since they are structured formats and will contain a field for the species. However, this may result in a large download.

Genbank format includes the field ORGANISM, e.g.

ORGANISM  Drilonereis longa

So we can count up the terms that come after the word ORGANISM using grep, cut and awk:

grep ORGANISM sequence.gb | 
cut -d " " -f 4- | 
awk 'BEGIN{OFS="\t"} {n[$0]++} END {for (i in n) {print i, n[i]}}'
> wormcount.txt

Result (first 5 lines only):

Nephtys sp. CMC05    1
Nephtys sp. CMC06    1
Amynthas lini    9
Mesenchytraeus solifugus    64
Satchellius mammalis    11

Unique ORGANISM with COI:

wc -l wormcount.txt
# 1995 wormcount.txt

Thanks a lot, I thought that it might come down to a bit of simple scripting. This seems straight forward, I'll post if I run into trouble. Thanks again!

Should I ever publish this information, I would like to reference you in the acknowledgements. What name would you want me to put in the acknowledgements? Thanks a lot again, this saved me a lot of time!! Sebastian