diff --git a/README.md b/README.md
index 409c943..110c342 100644
--- a/README.md
+++ b/README.md
@@ -77,6 +77,17 @@
   than box ASA speeds.  (See the [Updates](#updates) section for
   other ideas.)
 
+* But there is a big downside with this.  Highly dilute development
+  raises acutance at the expense of making grain more visible. This
+  tends to get worse the longer the film sits in developer.
+
+  For larger formats, this typically isn't a big deal, but it's a
+  killer for 35mm if you want very high quality images, particularly
+  with faster films like Tri-X.  For this reason, when processing 35mm
+  film, I use Pyrocat-HD 5:3:500, agitate initially for 90 seconds,
+  for 5 seconds at 7, 14, and 21 minutes, and pull the film at 28
+  minutes.
+
 * When faced with a low contrast or short Subject Brightness range, a
   lower dilution semistand (1.5:1:250) for an hour seems to work best.
   Begin with 2 minutes agitation and one 15 second midpoint agitation