diff --git a/README.md b/README.md
index 6a97507..409d07c 100644
--- a/README.md
+++ b/README.md
@@ -438,6 +438,35 @@
 can process normally if needed.
 
 
+## When Should You NOT Use These Techniques?
+
+With over a year's experience under my belt at this point, and having
+tested many film/developer cominations across a variety of subjects,
+I've concluded that Semistand/EMA is a superior development
+strategy for many things.  It yields full box ASA, protects the
+highlights from blowing out, and gives you snappy midtone local
+constrast.
+
+However, there are some subjects that don't respond really well to
+this.  The increased midtone contrast can look gritty and get almost a
+cartoon like quality to it.  Specifically, when an image has a
+textured surface that already has noteworthy local contrast, the image
+can become unrealistically harsh.
+
+This is especially true with very high acutance films like Efke PL100M
+or Adox CMS 100 II. When the actuance of the film is combined with the
+edge effects of extended development, you can get a kind of comic book
+or graphic arts effect.  It's subtle but very real, though it might be
+useful as an aesthetic tool.
+
+The worst offender here is EMA.  The repeated agitations drives up
+edge effects far more so than does the 1 midpoint agitation of
+Seminstand.  Given a subject with a lot of midtone texture and local
+contrast, you might prefer Semistand with a film like Shanghai GP3 or
+Tri-X.  Of course, conventional development is always a thought if the
+dynamic range of the scene can be properly managed.
+
+
 ## Updates
 
 New findings since the document first went live can be found below.