diff --git a/README.md b/README.md
index edfec71..36ed4f6 100644
--- a/README.md
+++ b/README.md
@@ -12,7 +12,7 @@ This classifier is suitable for coarsely classifying 18S sequences to order or c
 
 Created from the SILVA 138 SSU Ref Nr99 (Preusse et al., 2007), the most recent curated dataset, with modifications described below.
 
-We checked for the presence of species with more than one unique taxonomic lineage using the check_for_SILVA_inconsistencies4.plx script in the CheckSilva138Taxonomy directory.  We found that 2,495 / 98,776 (2.5%) *unique* species were annotated with more than one taxonomic lineage due to the way SILVA curates taxonomic lineages using an automated phylogeny-based method (combined with manual curation for certain groups).  This affects 335,983 / 510,984 (66%) of all the sequneces in the SILVA fasta file.  
+We checked for the presence of species with more than one unique taxonomic lineage using the check_for_SILVA_inconsistencies.plx script in the CheckSilva138Taxonomy directory.  We found that 2,495 / 98,776 (2.5%) *unique* species were annotated with more than one taxonomic lineage due to the way SILVA curates taxonomic lineages using an automated phylogeny-based method (combined with manual curation for certain groups).  This affects 335,983 / 510,984 (66%) of all the sequneces in the SILVA fasta file.  
 
 This version of the 18S classifier has been trained to make assignments to the **genus** rank only.  This was done for 3 reasons: