diff --git a/content/variables.md b/content/variables.md
index f3ea091..6db20d2 100644
--- a/content/variables.md
+++ b/content/variables.md
@@ -7,9 +7,10 @@ Nimrod supports three different types of variables, `let`, `var`, and `const`. A
 
 ``` nimrod
 proc getAlphabet(): string =
-  result = ""
+  var accm = ""
   for letter in 'a'..'z':
-    result.add(letter)
+    accm.add(letter)
+  return accm
 
 # Computed at compilation time
 const alphabet = getAlphabet()
diff --git a/content/variables/result.md b/content/variables/result.md
index dfc249c..dbccdaa 100644
--- a/content/variables/result.md
+++ b/content/variables/result.md
@@ -3,13 +3,13 @@ title: Result
 ---
 # Result
 
-The `result` variable is a special variable that serves as an implicit return variable, which exists because the control flow semantics of the `return` statement are rarely needed. The result variable is initialized in the standard way, as if it was declared with `var result: T`.
+The `result` variable is a special variable that serves as an implicit return variable, which exists because the control flow semantics of the `return` statement are rarely needed. The result variable is initialized in the standard way, as if it was declared with `var result: ReturnType`. For example, the `getAlphabet()` function could be rewritten more concisely as
 
 ``` nimrod
-proc `**`(number, power: int): int =
-  result = 1
-  for i in 1..power:
-    result *= number
+proc getAlphabet(): string =
+  result = ""
+  for letter in 'a'..'z':
+    result.add(letter)
 ```
 
 A possible gotcha is declaring a new variable called `result` and expecting it to have the same semantics.