diff --git a/Makefile b/Makefile
index e53205a..ab77c8a 100644
--- a/Makefile
+++ b/Makefile
@@ -6,7 +6,7 @@ TARGETS += pyesc
 TARGETS += xor 
 TARGETS += unhex
 TARGETS += pcat
-TARGETS += drop
+TARGETS += slice
 TARGETS += hex
 
 all: $(TARGETS)
diff --git a/README.md b/README.md
index 0bb9057..47266f5 100644
--- a/README.md
+++ b/README.md
@@ -75,21 +75,19 @@ The "-x" option treats values as hex.
 	cbcbcb
 
 
-### drop: discard octets
+### slice: slice octet stream
 
-Throws away some octets from stdin,
-and sends the rest to stdout.
+Slices up input octet stream,
+similar to Python's slice operation.
 
-You could use `dd` for the same purpose.
-
-    $ echo 01234567 | drop 0 3
-	34567
-	$ echo 01234567 | drop 4 7
-	01237
-	$ echo 01234567 | drop 4 6
-	012367
-	$ echo 01234567 | drop 3 9999
-	012
+    ~/src/fluffy $ printf '0123456789abcdef' | slice 2; echo
+	23456789abcdef
+	~/src/fluffy $ printf '0123456789abcdef' | slice 2 6; echo
+	2345
+	~/src/fluffy $ printf '0123456789abcdef' | slice 2 6 8; echo
+	234589abcdef
+	~/src/fluffy $ printf '0123456789abcdef' | slice 2 6 8 0xa
+	234589
 
 
 ### pcat: print text representation of pcap file
@@ -133,6 +131,14 @@ writing to output.
 The opposite of `unhex`:
 encoding all input into a single output line.
 
+This differs from `hexdump` in the following ways:
+
+* All input is encoded into a single line of output
+* Does not output offsets
+* Does not output glyph representations of octets
+
+In other words: you can feed `hex` output into `unhex` with no manipulations.
+
 	$ printf "hello\nworld\n" | hex
 	68 65 6c 6c 6f 0a 77 6f  72 6c 64 0a
 
diff --git a/drop.c b/drop.c
deleted file mode 100644
index b0a156a..0000000
--- a/drop.c
+++ /dev/null
@@ -1,43 +0,0 @@
-/*
- * drop octets -- 2017 Neale Pickett <zephyr@dirtbags.net>
- *
- * This file is in the public domain.  I make no promises about the functionality
- * of this program. 
- */
-
-#include <stdio.h>
-#include <stdlib.h>
-
-int
-main(int argc, char *argv[])
-{
-	unsigned long int start;
-	unsigned long int end;
-	unsigned long int pos;
-
-	if (3 == argc) {
-		start = strtoul(argv[1], NULL, 0);
-		end = strtoul(argv[2], NULL, 0);
-	} else {
-		fprintf(stderr, "Usage: %s start end\n", argv[0]);
-		fprintf(stderr, "\n");
-		fprintf(stderr, "Drops octets from input\n");
-		return 1;
-	}
-
-	for (pos = 0; ; pos += 1) {
-		int c = getchar();
-
-		if (EOF == c) {
-			break;
-		}
-
-		if ((start <= pos) && (pos < end)) {
-			/* drop */
-		} else {
-			putchar(c);
-		}
-	}
-
-	return 0;
-}
diff --git a/slice.c b/slice.c
new file mode 100644
index 0000000..e326d06
--- /dev/null
+++ b/slice.c
@@ -0,0 +1,55 @@
+/*
+ * slice up octet streams -- 2017 Neale Pickett <zephyr@dirtbags.net>
+ *
+ * This file is in the public domain.  I make no promises about the functionality
+ * of this program. 
+ */
+
+#include <limits.h>
+#include <stdbool.h>
+#include <stdio.h>
+#include <stdlib.h>
+
+int
+main(int argc, char *argv[])
+{
+	int argn = 1;
+	unsigned long int drop = true;
+	unsigned long int next;
+	unsigned long int pos;
+
+	if (1 == argc) {
+		fprintf(stderr, "Usage: %s start [end start...] [end]\n", argv[0]);
+		fprintf(stderr, "\n");
+		fprintf(stderr, "Slices input, keeping specified ranges\n");
+		return 1;
+	}
+
+	next = strtoul(argv[argn], NULL, 0);
+
+	for (pos = 0; ; pos += 1) {
+		int c = getchar();
+
+		if (EOF == c) {
+			break;
+		}
+
+		if (next == pos) {
+			drop = !drop;
+		    argn += 1;
+			if (argn == argc) {
+				next = ULONG_MAX;
+			} else {
+				next = strtoul(argv[argn], NULL, 0);
+			}
+		}
+
+		if (drop) {
+			/* drop */
+		} else {
+			putchar(c);
+		}
+	}
+
+	return 0;
+}