diff --git a/include/ceres/autodiff_cost_function.h b/include/ceres/autodiff_cost_function.h
index a106e3e..60946fd 100644
--- a/include/ceres/autodiff_cost_function.h
+++ b/include/ceres/autodiff_cost_function.h
@@ -30,7 +30,7 @@
 //
 // Create CostFunctions as needed by the least squares framework, with
 // Jacobians computed via automatic differentiation. For more
-// information on automatic differentation, see the wikipedia article
+// information on automatic differentiation, see the wikipedia article
 // at http://en.wikipedia.org/wiki/Automatic_differentiation
 //
 // To get an auto differentiated cost function, you must define a class with a
@@ -90,7 +90,7 @@
 //                            Dimension of x ---------------+  |
 //                            Dimension of y ------------------+
 //
-// In this example, there is usually an instance for each measumerent of k.
+// In this example, there is usually an instance for each measurement of k.
 //
 // In the instantiation above, the template parameters following
 // "MyScalarCostFunctor", "1, 2, 2", describe the functor as computing a
@@ -138,7 +138,7 @@
 namespace ceres {
 
 // A cost function which computes the derivative of the cost with respect to
-// the parameters (a.k.a. the jacobian) using an autodifferentiation framework.
+// the parameters (a.k.a. the jacobian) using an auto differentiation framework.
 // The first template argument is the functor object, described in the header
 // comment. The second argument is the dimension of the residual (or
 // ceres::DYNAMIC to indicate it will be set at runtime), and subsequent
diff --git a/include/ceres/internal/autodiff.h b/include/ceres/internal/autodiff.h
index 92542a4..a863e3c 100644
--- a/include/ceres/internal/autodiff.h
+++ b/include/ceres/internal/autodiff.h
@@ -30,7 +30,7 @@
 //
 // Computation of the Jacobian matrix for vector-valued functions of multiple
 // variables, using automatic differentiation based on the implementation of
-// dual numbers in jet.h. Before reading the rest of this file, it is adivsable
+// dual numbers in jet.h. Before reading the rest of this file, it is advisable
 // to read jet.h's header comment in detail.
 //
 // The helper wrapper AutoDiff::Differentiate() computes the jacobian of
diff --git a/include/ceres/jet.h b/include/ceres/jet.h
index 3c53088..f6fe4ad 100644
--- a/include/ceres/jet.h
+++ b/include/ceres/jet.h
@@ -31,7 +31,7 @@
 // A simple implementation of N-dimensional dual numbers, for automatically
 // computing exact derivatives of functions.
 //
-// While a complete treatment of the mechanics of automatic differentation is
+// While a complete treatment of the mechanics of automatic differentiation is
 // beyond the scope of this header (see
 // http://en.wikipedia.org/wiki/Automatic_differentiation for details), the
 // basic idea is to extend normal arithmetic with an extra element, "e," often
@@ -49,7 +49,7 @@
 //   f(x) = x^2 ,
 //
 // evaluated at 10. Using normal arithmetic, f(10) = 100, and df/dx(10) = 20.
-// Next, augument 10 with an infinitesimal to get:
+// Next, argument 10 with an infinitesimal to get:
 //
 //   f(10 + e) = (10 + e)^2
 //             = 100 + 2 * 10 * e + e^2
@@ -124,7 +124,7 @@
 //
 //   x = a + \sum_i v[i] t_i
 //
-// A shorthand is to write an element as x = a + u, where u is the pertubation.
+// A shorthand is to write an element as x = a + u, where u is the perturbation.
 // Then, the main point about the arithmetic of jets is that the product of
 // perturbations is zero:
 //
@@ -253,7 +253,7 @@
   // We allocate Jets on the stack and other places they might not be aligned
   // to X(=16 [SSE], 32 [AVX] etc)-byte boundaries, which would prevent the safe
   // use of vectorisation.  If we have C++11, we can specify the alignment.
-  // However, the standard gives wide lattitude as to what alignments are valid,
+  // However, the standard gives wide latitude as to what alignments are valid,
   // and it might be that the maximum supported alignment *guaranteed* to be
   // supported is < 16, in which case we do not specify an alignment, as this
   // implies the host is not a modern x86 machine.  If using < C++11, we cannot