diff --git a/GPy/kern/rational_quadratic.py b/GPy/kern/rational_quadratic.py
index 15200fd3..561ea065 100644
--- a/GPy/kern/rational_quadratic.py
+++ b/GPy/kern/rational_quadratic.py
@@ -73,8 +73,8 @@ class rational_quadratic(kernpart):
         if X2 is None: X2 = X
         dist2 = np.square((X-X2.T)/self.lengthscale)
 
-        dX = -self.variance*self.power * (X-X2.T)/self.lengthscale**2 *  (1 + dist2/2./self.power)**(-self.power-1)
-        target += np.sum(dL_dK*dX)
+        dX = -self.variance*self.power * (X-X2.T)/self.lengthscale**2 *  (1 + dist2/2./self.lengthscale)**(-self.power-1)
+        target += np.sum(dL_dK*dX,1)[:,np.newaxis]
 
     def dKdiag_dX(self,dL_dKdiag,X,target):
         pass
diff --git a/doc/tuto_creating_new_kernels.rst b/doc/tuto_creating_new_kernels.rst
index 8ebf8b8f..24003ba2 100644
--- a/doc/tuto_creating_new_kernels.rst
+++ b/doc/tuto_creating_new_kernels.rst
@@ -136,8 +136,8 @@ Computes the derivative of the likelihood with respect to the inputs ``X`` (a :m
         if X2 is None: X2 = X
         dist2 = np.square((X-X2.T)/self.lengthscale)
 
-        dX = -self.variance*self.power * (X-X2.T)/self.lengthscale**2 *  (1 + dist2/2./self.power)**(-self.power-1)
-        target += np.sum(dL_dK*dX)
+        dX = -self.variance*self.power * (X-X2.T)/self.lengthscale**2 *  (1 + dist2/2./self.lengthscale)**(-self.power-1)
+        target += np.sum(dL_dK*dX,1)[:,np.newaxis]
 
 **dKdiag_dX(self,dL_dKdiag,X,target)**