diff --git a/skimage/color/colorconv.py b/skimage/color/colorconv.py
index 74a9a520..86353e0b 100644
--- a/skimage/color/colorconv.py
+++ b/skimage/color/colorconv.py
@@ -56,7 +56,7 @@ from __future__ import division
 from warnings import warn
 import numpy as np
 from scipy import linalg
-from ..util import dtype
+from ..util import dtype, dtype_limits
 
 
 def guess_spatial_dimensions(image):
@@ -715,13 +715,16 @@ def rgb2gray(rgb):
 rgb2grey = rgb2gray
 
 
-def gray2rgb(image):
+def gray2rgb(image, alpha=None):
     """Create an RGB representation of a gray-level image.
 
     Parameters
     ----------
     image : array_like
         Input image of shape ``(M, N [, P])``.
+    alpha : bool, optional
+        Ensure that the output image has an alpha layer.  If None,
+        alpha layers are passed through but not created.
 
     Returns
     -------
@@ -734,11 +737,39 @@ def gray2rgb(image):
         If the input is not a 2- or 3-dimensional image.
 
     """
-    if np.squeeze(image).ndim == 3 and image.shape[2] in (3, 4):
+    is_rgb = False
+    is_alpha = False
+    dims = np.squeeze(image).ndim
+
+    if dims == 3:
+        if image.shape[2] == 3:
+            is_rgb = True
+        elif image.shape[2] == 4:
+            is_alpha = True
+            is_rgb = True
+
+    if is_rgb:
+        if alpha == False:
+            image = image[..., :3]
+
+        elif alpha == True and not is_alpha:
+            alpha_layer = (np.ones_like(image[..., 0, np.newaxis]) *
+                           dtype_limits(image)[1])
+            image = np.concatenate((image, alpha_layer), axis=2)
+
         return image
-    elif image.ndim != 1 and np.squeeze(image).ndim in (1, 2, 3):
+
+    elif image.ndim != 1 and dims in (1, 2, 3):
         image = image[..., np.newaxis]
-        return np.concatenate(3 * (image,), axis=-1)
+
+        if alpha:
+            alpha_layer = (np.ones_like(image) * dtype_limits(image)[1])
+            return np.concatenate(3 * (image,) + (alpha_layer,), axis=-1)
+        else:
+            return np.concatenate(3 * (image,), axis=-1)
+
+        return image
+
     else:
         raise ValueError("Input image expected to be RGB, RGBA or gray.")
 
diff --git a/skimage/color/tests/test_colorconv.py b/skimage/color/tests/test_colorconv.py
index c9febb5f..71e2d849 100644
--- a/skimage/color/tests/test_colorconv.py
+++ b/skimage/color/tests/test_colorconv.py
@@ -454,6 +454,23 @@ def test_gray2rgb_rgb():
     assert_equal(x, y)
 
 
+def test_gray2rgb_alpha():
+    x = np.random.random((5, 5, 4))
+    assert_equal(gray2rgb(x, alpha=None).shape, (5, 5, 4))
+    assert_equal(gray2rgb(x, alpha=False).shape, (5, 5, 3))
+    assert_equal(gray2rgb(x, alpha=True).shape, (5, 5, 4))
+
+    x = np.random.random((5, 5, 3))
+    assert_equal(gray2rgb(x, alpha=None).shape, (5, 5, 3))
+    assert_equal(gray2rgb(x, alpha=False).shape, (5, 5, 3))
+    assert_equal(gray2rgb(x, alpha=True).shape, (5, 5, 4))
+
+    assert_equal(gray2rgb(np.array([[1, 2], [3, 4.]]),
+                          alpha=True)[0, 0, 3], 1)
+    assert_equal(gray2rgb(np.array([[1, 2], [3, 4]], dtype=np.uint8),
+                          alpha=True)[0, 0, 3], 255)
+
+
 if __name__ == "__main__":
     from numpy.testing import run_module_suite
     run_module_suite()