diff --git a/docs/index.md b/docs/index.md
index ccf3c08..d1bd815 100644
--- a/docs/index.md
+++ b/docs/index.md
@@ -8,4 +8,5 @@ The reference implementation can be found on [GitHub](https://github.com/Ratstai
 
 [Reserved Words](reserved-words.md)  
 [Operators](operators.md)  
+[Types](types.md)  
 
diff --git a/docs/types.md b/docs/types.md
new file mode 100644
index 0000000..acb702f
--- /dev/null
+++ b/docs/types.md
@@ -0,0 +1,71 @@
+# Types
+
+When a variable name is declared, you may specify what kind of value can be stored within - this is known as its "type". If you attempt to store a value of a different type, an error will be raised.
+
+```toy
+//syntax
+var name: type = value;
+
+//example
+var answer: int = 42;
+```
+
+Specifying a type is optional, in which case, the type is set to `any` by default ([see below](#any)).
+
+```toy
+//omit the type and set an integer value
+var answer = 42;
+
+print typeof answer; // will print "any"
+```
+
+You may access the type of a variable using the `typeof` keyword. Types can also act as values - the type of any type is `type`.
+
+# null
+
+The type `null` is a special case in the language, as it represents the absence of any meaningful value. It can't be specified as a variable's type, only as it's value.
+
+Unlike the other types, the type of `null` is `null`.
+
+# bool
+
+The `bool` type can hold two meaningful values, either `true` or `false`.
+
+# int
+
+The `int` type can hold any whole number between `-2,147,483,648` and`2,147,483,647`, due to being stored as a signed 32-bit integer.
+
+# float
+
+The `float` type can hold a single-precision floating-point value as defined by IEEE 754, which is the most commonly used method for storing real numbers in 32 bits.
+
+What this means in practice is that floating point errors are possible, but this is still the best option for managing decimal numbers.
+
+# string
+
+TODO
+
+# array
+
+TODO
+
+# table
+
+TODO
+
+# function
+
+TODO
+
+# opaque
+
+TODO
+
+# type
+
+TODO
+
+# any
+
+TODO
+