diff --git a/docs/disc-mod.md b/docs/disc-mod.md
index aabfe9b6..64d6caf8 100644
--- a/docs/disc-mod.md
+++ b/docs/disc-mod.md
@@ -249,7 +249,7 @@ sits above a bumper wrapper's near-1.0 size share and below a branch body's.
 
 **Known residual:** a wrapper that pads runtime only modestly AND withholds
 most of its bytes from the body clip is ambiguous by shape alone. The corpus
-title-selection gate is the backstop across the disc hoard, and
+title-selection gate is the backstop across the test corpus, and
 higher-precedence nav/authoring signals override the composite key.
 
 ## aligned_unit_keys_validate
diff --git a/docs/mux-timeline.md b/docs/mux-timeline.md
index 4e7f0e32..a72c994d 100644
--- a/docs/mux-timeline.md
+++ b/docs/mux-timeline.md
@@ -149,7 +149,7 @@ Which clip owns feed byte `b`, by BINARY SEARCH.
 
 `spans_trusted` guarantees the spans tile the feed contiguously in order, so
 this is a partition point rather than a scan. That matters: discs in the test
-hoard reach 900 clips, and this runs once per frame per track — a linear scan
+corpus reach 900 clips, and this runs once per frame per track — a linear scan
 would be ~900 comparisons on every one of millions of frames, which is real
 time spent for no reason.