diff --git a/examples/1D_image.py b/examples/1D_image.py
index ae1d6c1..befaa15 100644
--- a/examples/1D_image.py
+++ b/examples/1D_image.py
@@ -5,8 +5,8 @@
 pn = PerlinNoise(num_octaves=7, persistence=0.1)
 data = []
 
-t = [i for i in xrange(length)]
-for i in xrange(length):
+t = [i for i in range(length)]
+for i in range(length):
     data.append(normalize(pn.fractal(x=i, hgrid=length)))
 
 fig = plt.figure()
diff --git a/examples/fbm_image.py b/examples/fbm_image.py
index 96ef6c2..34c9723 100644
--- a/examples/fbm_image.py
+++ b/examples/fbm_image.py
@@ -6,13 +6,13 @@
 sn = SimplexNoise(num_octaves=7, persistence=0.1, dimensions=2)
 data = []
 
-for i in xrange(size):
+for i in range(size):
     data.append([])
-    for j in xrange(size):
+    for j in range(size):
         noise = normalize(sn.fractal(i, j, hgrid=size))
         data[i].append(noise * 255.0)
 
 # Cast to numpy array so we can save 
 data = np.array(data).astype(np.uint8)
 img = Image.fromarray(data, mode='L')
-img.save('./fbm_example.png')
\ No newline at end of file
+img.save('./fbm_example.png')
diff --git a/examples/noise_image.py b/examples/noise_image.py
index 42af2c0..d23f7eb 100644
--- a/examples/noise_image.py
+++ b/examples/noise_image.py
@@ -7,13 +7,13 @@
 sn = SimplexNoise(num_octaves=7, persistence=0.1, dimensions=2, noise_scale=noise_scale)
 data = []
 
-for i in xrange(size):
+for i in range(size):
     data.append([])
-    for j in xrange(size):
+    for j in range(size):
         noise = normalize(sn.noise(i, j))
         data[i].append(noise * 255.0)
 
 # Cast to numpy array so we can save 
 data = np.array(data).astype(np.uint8)
 img = Image.fromarray(data, mode='L')
-img.save('./noise_example.png')
\ No newline at end of file
+img.save('./noise_example.png')
diff --git a/simplexnoise/noise.py b/simplexnoise/noise.py
index 11dc89d..43a4ecd 100644
--- a/simplexnoise/noise.py
+++ b/simplexnoise/noise.py
@@ -1,7 +1,7 @@
 from __future__ import division
 import math
 import random
-from geometry import Point
+from .geometry import Point
 
 # Constants to avoid magic numbers
 DEFAULT_NOISE_SCALE = -1  # Check noise_scale against this
@@ -41,17 +41,17 @@ def __init__(self, num_octaves, persistence, noise_scale=DEFAULT_NOISE_SCALE):
         else:
             self.noise_scale = noise_scale
 
-        self.octaves = [PerlinNoiseOctave() for i in xrange(self.num_octaves)]
-        self.frequencies = [1.0 / pow(2, i) for i in xrange(self.num_octaves)]
+        self.octaves = [PerlinNoiseOctave() for i in range(self.num_octaves)]
+        self.frequencies = [1.0 / pow(2, i) for i in range(self.num_octaves)]
         self.amplitudes = [pow(persistence, len(self.octaves) - i)
-                           for i in xrange(self.num_octaves)]
+                           for i in range(self.num_octaves)]
 
     def noise(self, x):
         noise = [
             self.octaves[i].noise(
                 xin=x * self.frequencies[i],
                 noise_scale=self.noise_scale
-            ) * self.amplitudes[i] for i in xrange(self.num_octaves)]
+            ) * self.amplitudes[i] for i in range(self.num_octaves)]
 
         return sum(noise)
 
@@ -61,7 +61,7 @@ def fractal(self, x, hgrid, lacunarity=DEFAULT_LACUNARITY, gain=DEFAULT_GAIN):
         frequency = 1.0 / hgrid
         amplitude = gain
 
-        for i in xrange(self.num_octaves):
+        for i in range(self.num_octaves):
             noise.append(
                 self.octaves[i].noise(
                     xin=x * frequency,
@@ -78,9 +78,9 @@ def fractal(self, x, hgrid, lacunarity=DEFAULT_LACUNARITY, gain=DEFAULT_GAIN):
 class PerlinNoiseOctave(object):
 
     def __init__(self, num_shuffles=DEFAULT_SHUFFLES):
-        self.p_supply = [i for i in xrange(0, 256)]
+        self.p_supply = [i for i in range(0, 256)]
 
-        for i in xrange(num_shuffles):
+        for i in range(num_shuffles):
             random.shuffle(self.p_supply)
 
         self.perm = self.p_supply * 2
@@ -121,12 +121,12 @@ def __init__(self, num_octaves, persistence, dimensions, noise_scale=DEFAULT_NOI
 
         if DIMENSIONS_2D == dimensions:
             self.octaves = [SimplexNoiseOctave2D()
-                            for i in xrange(self.num_octaves)]
+                            for i in range(self.num_octaves)]
             self.noise_scale = DEFAULT_2D_NOISE_SCALE
 
         elif DIMENSIONS_3D == dimensions:
             self.octaves = [SimplexNoiseOctave3D()
-                            for i in xrange(self.num_octaves)]
+                            for i in range(self.num_octaves)]
             self.noise_scale = DEFAULT_2D_NOISE_SCALE
 
         else:
@@ -137,9 +137,9 @@ def __init__(self, num_octaves, persistence, dimensions, noise_scale=DEFAULT_NOI
         if DEFAULT_NOISE_SCALE != noise_scale:
             self.noise_scale = noise_scale
 
-        self.frequencies = [pow(2, i) for i in xrange(self.num_octaves)]
+        self.frequencies = [pow(2, i) for i in range(self.num_octaves)]
         self.amplitudes = [pow(persistence, len(self.octaves) - i)
-                           for i in xrange(self.num_octaves)]
+                           for i in range(self.num_octaves)]
 
     def noise(self, x=0, y=0, z=0):
         noise = [
@@ -148,7 +148,7 @@ def noise(self, x=0, y=0, z=0):
                 yin=y / self.frequencies[i],
                 zin=z / self.frequencies[i],
                 noise_scale=self.noise_scale
-            ) * self.amplitudes[i] for i in xrange(self.num_octaves)]
+            ) * self.amplitudes[i] for i in range(self.num_octaves)]
 
         return sum(noise)
 
@@ -158,7 +158,7 @@ def fractal(self, x=0, y=0, z=0, hgrid=0, lacunarity=DEFAULT_LACUNARITY, gain=DE
         frequency = 1.0 / hgrid
         amplitude = gain
 
-        for i in xrange(self.num_octaves):
+        for i in range(self.num_octaves):
             noise.append(
                 self.octaves[i].noise(
                     xin=x * frequency,
@@ -182,7 +182,7 @@ class SimplexNoiseOctave2D(object):
     unskew_factor = (3.0 - math.sqrt(3.0)) / 6.0
 
     def __init__(self, num_shuffles=DEFAULT_SHUFFLES):
-        self.p_supply = [i for i in xrange(0, 256)]
+        self.p_supply = [i for i in range(0, 256)]
 
         self.grads = [
             Point(1, 1, 0),
@@ -191,7 +191,7 @@ def __init__(self, num_shuffles=DEFAULT_SHUFFLES):
             Point(-1, -1, 0)
         ]
 
-        for i in xrange(num_shuffles):
+        for i in range(num_shuffles):
             random.shuffle(self.p_supply)
 
         self.perm = self.p_supply * 2
@@ -277,7 +277,7 @@ def hashed_gradient_indices(self, points_ij):
     def calc_noise_contributions(self, grad_index_hash, points_xy):
         """ Calculates the contribution from each corner (in 2D there are three!) """
         contribs = []
-        for i in xrange(len(grad_index_hash)):
+        for i in range(len(grad_index_hash)):
             x = points_xy[i].x
             y = points_xy[i].y
             grad = self.grads[grad_index_hash[i]]
@@ -299,7 +299,7 @@ class SimplexNoiseOctave3D(object):
     unskew_factor = 1.0 / 6.0
 
     def __init__(self, num_shuffles=DEFAULT_SHUFFLES):
-        self.p_supply = [i for i in xrange(0, 256)]
+        self.p_supply = [i for i in range(0, 256)]
 
         self.grads = [
             Point(1, 1, 0), Point(-1, 1, 0), Point(1, -1, 0), Point(-1, -1, 0),
@@ -307,7 +307,7 @@ def __init__(self, num_shuffles=DEFAULT_SHUFFLES):
             Point(0, 1, 1), Point(0, -1, 1), Point(0, 1, -1), Point(0, -1, -1),
         ]
 
-        for i in xrange(num_shuffles):
+        for i in range(num_shuffles):
             random.shuffle(self.p_supply)
 
         self.perm = self.p_supply * 2
@@ -412,7 +412,7 @@ def hashed_gradient_indices(self, points_ijk):
     def calc_noise_contributions(self, grad_index_hash, points_xyz):
         """ Calculates the contribution from each corner (in 2D there are three!) """
         contribs = []
-        for i in xrange(len(grad_index_hash)):
+        for i in range(len(grad_index_hash)):
             x = points_xyz[i].x
             y = points_xyz[i].y
             z = points_xyz[i].z