diff --git a/dc.py b/dc.py
index ba37a69..00ac6d6 100644
--- a/dc.py
+++ b/dc.py
@@ -5,55 +5,20 @@
 """
 from collections import defaultdict
 from math import sqrt
+
 import numpy as np
 from scipy.spatial.distance import pdist, squareform
 
+from dcor import dcor
+
 
 def dc(X, Y):
     prob_X, marg_X, prob_Y, marg_Y = distributions(X, Y)
-    dXtoY = dCor(prob_X, marg_Y)
-    dYtoX = dCor(prob_Y, marg_X)
+    dXtoY = dcor(prob_X, marg_Y)
+    dYtoX = dcor(prob_Y, marg_X)
     return (dXtoY, dYtoX)
 
 
-def dcov(X, Y):
-    n = X.shape[0]
-    XY = np.multiply(X, Y)
-    cov = sqrt(XY.sum()) / n
-    return cov
-
-
-def dvar(X):
-    return sqrt(np.sum(X ** 2 / X.shape[0] ** 2))
-
-
-def cent_dist(X):
-    M = squareform(pdist(X))    # distance matrix
-    rmean = M.mean(axis=1)
-    cmean = M.mean(axis=0)
-    gmean = rmean.mean()
-    Rmean = np.tile(rmean, (M.shape[0], 1)).transpose()
-    Cmean = np.tile(cmean, (M.shape[1], 1))
-    Gmean = np.tile(gmean, M.shape)
-    CM = M - Rmean - Cmean + Gmean
-    return CM
-
-
-def dCor(X, Y):
-    A = cent_dist(X)
-    B = cent_dist(Y)
-
-    dcov_AB = dcov(A, B)
-    dvar_A = dvar(A)
-    dvar_B = dvar(B)
-
-    dcor = 0.0
-    if dvar_A > 0.0 and dvar_B > 0.0:
-        dcor = dcov_AB / sqrt(dvar_A * dvar_B)
-
-    return dcor
-
-
 def distributions(X, Y):
     N = len(X)
     unq_X = set(map(tuple, X))