diff --git a/README.md b/README.md
index a5bf972..929d0fe 100644
--- a/README.md
+++ b/README.md
@@ -1,7 +1,7 @@
 `cidrtrie` is a dumb implementation of a prefix tree and a library for classifying IP addresses based on prefix matching. I know that there other implementations (like [py-radix](http://www.mindrot.org/projects/py-radix/)) that are faster or whatever, but this one is simple and pure-Python.
 
 ## Why? ##
-`bench.py` contains a simple _O(n)_ implementation ("`NaiveCidrClassifier`)of this which I've seen in a few dozen projects (compared with the O(32) prefix tree). Some sample numbers for inserting 100,000 cidrs into the tree and then looking up 10,000 IPs:
+`bench.py` contains a simple _O(n)_ implementation (`NaiveCidrClassifier`)of this which I've seen in a few dozen projects (compared with the O(32) prefix tree). Some sample numbers for inserting 100,000 cidrs into the tree and then looking up 10,000 IPs:
 
     insert CidrClassifier 5.16s
     insert NaiveCidrClassifier 0.11s