diff --git a/docs/content/Interpolation.md b/docs/content/Interpolation.md
index 384e1d71..5f298153 100644
--- a/docs/content/Interpolation.md
+++ b/docs/content/Interpolation.md
@@ -37,12 +37,13 @@ Interpolation on arbitrary sample points
 ----------------------------------------
 
 * *Rational pole-free*: Barycentric Floater-Hormann Algorithm
-* **Rational with poles**: Bulirsch & Stoer Algorithm
+* *Rational with poles*: Bulirsch & Stoer Algorithm
 * *Neville Polynomial*: Neville Algorithm. Note that the Neville algorithm performs very badly on equidistant points. If you need to interpolate a polynomial on equidistant points, we recommend to use the barycentric algorithm instead.
 * *Linear Spline*
 * *Cubic Spline* with boundary conditions
 * *Natural Cubic Spline*
 * *Akima Cubic Spline*
+* *Monotone Cubic Spline*: Monotone-preserving piecewise cubic Hermite interpolating polynomial (PCHIP), based on Fritsch & Carlson (1980).
 
 
 Interpolation with additional data
diff --git a/src/Numerics.Tests/InterpolationTests/PchipSplineTest.cs b/src/Numerics.Tests/InterpolationTests/PchipSplineTest.cs
new file mode 100644
index 00000000..1f74751e
--- /dev/null
+++ b/src/Numerics.Tests/InterpolationTests/PchipSplineTest.cs
@@ -0,0 +1,130 @@
+// <copyright file="PchipSplineTest.cs" company="Math.NET">
+// Math.NET Numerics, part of the Math.NET Project
+// http://numerics.mathdotnet.com
+// http://github.com/mathnet/mathnet-numerics
+//
+// Copyright (c) 2009-2021 Math.NET
+//
+// Permission is hereby granted, free of charge, to any person
+// obtaining a copy of this software and associated documentation
+// files (the "Software"), to deal in the Software without
+// restriction, including without limitation the rights to use,
+// copy, modify, merge, publish, distribute, sublicense, and/or sell
+// copies of the Software, and to permit persons to whom the
+// Software is furnished to do so, subject to the following
+// conditions:
+//
+// The above copyright notice and this permission notice shall be
+// included in all copies or substantial portions of the Software.
+//
+// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+// EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
+// OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+// NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
+// HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
+// WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+// FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+// OTHER DEALINGS IN THE SOFTWARE.
+// </copyright>
+
+using MathNet.Numerics.Interpolation;
+using NUnit.Framework;
+
+namespace MathNet.Numerics.UnitTests.InterpolationTests
+{
+    [TestFixture, Category("Interpolation")]
+    public class PchipSplineTest
+    {
+        readonly double[] _t = { -2.0, -1.0, 0.0, 1.0, 2.0 };
+        readonly double[] _y = { 1.0, 2.0, -1.0, 0.0, 1.0 };
+
+        readonly double[] _tNag = { 7.99, 8.09, 8.19, 8.70, 9.20, 10.00, 12.00, 15.00, 20.00 };
+        readonly double[] _yNag = { 0.00000E+0, 0.27643E-4, 0.43750E-1, 0.16918E+0, 0.46943E+0, 0.94374E+0, 0.99864E+0, 0.99992E+0, 0.99999E+0 };
+
+        /// <summary>
+        /// Verifies that the interpolation matches the given value at all the provided sample points.
+        /// </summary>
+        [Test]
+        public void FitsAtSamplePoints()
+        {
+            IInterpolation it = CubicSpline.InterpolatePchip(_t, _y);
+            for (int i = 0; i < _y.Length; i++)
+            {
+                Assert.AreEqual(_y[i], it.Interpolate(_t[i]), "A Exact Point " + i);
+            }
+        }
+
+        /// <summary>
+        /// Verifies that at points other than the provided sample points, the interpolation matches the one computed by Octave as a reference.
+        /// </summary>
+        /// <param name="t">Sample point.</param>
+        /// <param name="x">Sample value.</param>
+        /// <param name="maxAbsoluteError">Maximum absolute error.</param>
+        [TestCase(-2.4, -0.7440, 1e-15)]
+        [TestCase(-0.9, 1.9160, 1e-15)]
+        [TestCase(-0.5, 0.5000, 1e-15)]
+        [TestCase(-0.1, -0.9160, 1e-15)]
+        [TestCase(0.1, -0.9810, 1e-15)]
+        [TestCase(0.4, -0.7440, 1e-15)]
+        [TestCase(1.2, 0.2000, 1e-15)]
+        [TestCase(10.0, 9.0000, 1e-15)]
+        [TestCase(-10.0, -727.0000, 1e-15)]
+        public void FitsAtArbitraryPoints(double t, double x, double maxAbsoluteError)
+        {
+            IInterpolation it = CubicSpline.InterpolatePchip(_t, _y);
+
+            Assert.AreEqual(x, it.Interpolate(t), maxAbsoluteError, "Interpolation at {0}", t);
+        }
+
+        /// <summary>
+        /// Verifies that at points other than the provided sample points, the interpolation matches the one computed by NAG as a reference.
+        /// Reference: https://www.nag.com/numeric/cl/nagdoc_cl25/html/e01/e01bec.html
+        /// </summary>
+        /// <param name="t">Sample point.</param>
+        /// <param name="x">Sample value.</param>
+        /// <param name="maxAbsoluteError">Maximum absolute error.</param>
+        [TestCase(7.9900, 0.0000, 5e-5)]
+        [TestCase(9.1910, 0.4640, 5e-5)]
+        [TestCase(10.3920, 0.9645, 5e-5)]
+        [TestCase(11.5930, 0.9965, 5e-5)]
+        [TestCase(12.7940, 0.9992, 5e-5)]
+        [TestCase(13.9950, 0.9998, 5e-5)]
+        [TestCase(15.1960, 0.9999, 5e-5)]
+        [TestCase(16.3970, 1.0000, 5e-5)]
+        [TestCase(17.5980, 1.0000, 5e-5)]
+        [TestCase(18.7990, 1.0000, 5e-5)]
+        [TestCase(20.0000, 1.0000, 5e-5)]
+        public void FitsAtNagExamplePoints(double t, double x, double maxAbsoluteError)
+        {
+            IInterpolation it = CubicSpline.InterpolatePchip(_tNag, _yNag);
+
+            Assert.AreEqual(x, it.Interpolate(t), maxAbsoluteError, "Interpolation at {0}", t);
+        }
+
+        /// <summary>
+        /// Verifies that the interpolation supports the linear case appropriately
+        /// </summary>
+        /// <param name="samples">Samples array.</param>
+        [TestCase(5)]
+        [TestCase(7)]
+        [TestCase(15)]
+        public void SupportsLinearCase(int samples)
+        {
+            double[] x, y, xtest, ytest;
+            LinearInterpolationCase.Build(out x, out y, out xtest, out ytest, samples);
+            IInterpolation it = CubicSpline.InterpolatePchip(x, y);
+            for (int i = 0; i < xtest.Length; i++)
+            {
+                Assert.AreEqual(ytest[i], it.Interpolate(xtest[i]), 1e-15, "Linear with {0} samples, sample {1}", samples, i);
+            }
+        }
+
+        [Test]
+        public void FewSamples()
+        {
+            Assert.That(() => CubicSpline.InterpolatePchip(new double[0], new double[0]), Throws.ArgumentException);
+            Assert.That(() => CubicSpline.InterpolatePchip(new double[2], new double[2]), Throws.ArgumentException);
+            Assert.That(CubicSpline.InterpolatePchip(new[] { 1.0, 2.0, 3.0 }, new[] { 2.0, 2.0, 2.0 }).Interpolate(1.0), Is.EqualTo(2.0));
+        }
+    }
+}
diff --git a/src/Numerics/Interpolate.cs b/src/Numerics/Interpolate.cs
index af551b96..7c81c638 100644
--- a/src/Numerics/Interpolate.cs
+++ b/src/Numerics/Interpolate.cs
@@ -201,7 +201,7 @@ namespace MathNet.Numerics
         }
 
         /// <summary>
-        /// Create an piecewise cubic Akima spline interpolation based on arbitrary points.
+        /// Create a piecewise cubic Akima spline interpolation based on arbitrary points.
         /// Akima splines are robust to outliers.
         /// </summary>
         /// <param name="points">The sample points t.</param>
@@ -221,6 +221,27 @@ namespace MathNet.Numerics
             return Interpolation.CubicSpline.InterpolateAkima(points, values);
         }
 
+        /// <summary>
+        /// Create a piecewise cubic monotone spline interpolation based on arbitrary points.
+        /// This is a shape-preserving spline with continuous first derivative.
+        /// </summary>
+        /// <param name="points">The sample points t.</param>
+        /// <param name="values">The sample point values x(t).</param>
+        /// <returns>
+        /// An interpolation scheme optimized for the given sample points and values,
+        /// which can then be used to compute interpolations and extrapolations
+        /// on arbitrary points.
+        /// </returns>
+        /// <remarks>
+        /// if your data is already sorted in arrays, consider to use
+        /// MathNet.Numerics.Interpolation.CubicSpline.InterpolatePchipSorted
+        /// instead, which is more efficient.
+        /// </remarks>
+        public static IInterpolation CubicSplineMonotone(IEnumerable<double> points, IEnumerable<double> values)
+        {
+            return Interpolation.CubicSpline.InterpolatePchip(points, values);
+        }
+
         /// <summary>
         /// Create a piecewise cubic Hermite spline interpolation based on arbitrary points
         /// and their slopes/first derivative.
diff --git a/src/Numerics/Interpolation/CubicSpline.cs b/src/Numerics/Interpolation/CubicSpline.cs
index f2284506..d2344901 100644
--- a/src/Numerics/Interpolation/CubicSpline.cs
+++ b/src/Numerics/Interpolation/CubicSpline.cs
@@ -210,6 +210,110 @@ namespace MathNet.Numerics.Interpolation
             return InterpolateAkimaInplace(x.ToArray(), y.ToArray());
         }
 
+        /// <summary>
+        /// Create a piecewise cubic Hermite interpolating polynomial from an unsorted set of (x,y) value pairs.
+        /// Monotone-preserving interpolation with continuous first derivative.
+        /// </summary>
+        public static CubicSpline InterpolatePchipSorted(double[] x, double[] y)
+        {
+            // Implementation based on "Numerical Computing with Matlab" (Moler, 2004).
+
+            if (x.Length != y.Length)
+            {
+                throw new ArgumentException("All vectors must have the same dimensionality.");
+            }
+
+            if (x.Length < 3)
+            {
+                throw new ArgumentException("The given array is too small. It must be at least 3 long.", nameof(x));
+            }
+
+            var m = new double[x.Length - 1];
+            
+            for (int i = 0; i < m.Length; i++)
+            {
+                m[i] = (y[i + 1] - y[i])/(x[i + 1] - x[i]);
+            }
+
+            var dd = new double[x.Length];
+            var hPrev = x[1] - x[0];
+            // This check is quite costly as it usually involves a Math.Pow().
+            var mPrevIs0 = m[0].AlmostEqual(0.0);
+
+            for (var i = 1; i < x.Length - 1; ++i)
+            {
+                var h = x[i + 1] - x[i];
+                var mIs0 = m[i].AlmostEqual(0.0);
+
+                if (mIs0 || mPrevIs0 || Math.Sign(m[i]) != Math.Sign(m[i - 1]))
+                {
+                    dd[i] = 0;
+                }
+                else
+                {
+                    // Weighted harmonic mean of each slope.
+                    var w1 = 2 * h + hPrev;
+                    var w2 = h + 2 * hPrev;
+                    dd[i] = (w1 + w2) / (w1 / m[i - 1] + w2 / m[i]);
+                }
+
+                hPrev = h;
+                mPrevIs0 = mIs0;
+            }
+            
+            // Special case end-points.
+            dd[0] = PchipEndPoints(x[1] - x[0], x[2] - x[1], m[0], m[1]);
+            dd[dd.Length - 1] = PchipEndPoints(
+                x[x.Length - 1] - x[x.Length - 2], x[x.Length - 2] - x[x.Length - 3],
+                m[m.Length - 1], m[m.Length - 2]);
+
+            return InterpolateHermiteSorted(x, y, dd);
+        }
+
+        static double PchipEndPoints(double h0, double h1, double m0, double m1)
+        {
+            // One-sided, shape-preserving, three-point estimate for the derivative.
+            var d = ((2 * h0 + h1) * m0 - h0 * m1) / (h0 + h1);
+
+            if (Math.Sign(d) != Math.Sign(m0))
+            {
+                return 0.0;
+            }
+
+            if (Math.Sign(m0) != Math.Sign(m1) && (Math.Abs(d) > 3 * Math.Abs(m0)))
+            {
+                return 3 * m0;
+            }
+
+            return d;
+        }
+
+        /// <summary>
+        /// Create a piecewise cubic Hermite interpolating polynomial from an unsorted set of (x,y) value pairs.
+        /// Monotone-preserving interpolation with continuous first derivative.
+        /// WARNING: Works in-place and can thus causes the data array to be reordered.
+        /// </summary>
+        public static CubicSpline InterpolatePchipInplace(double[] x, double[] y)
+        {
+            if (x.Length != y.Length)
+            {
+                throw new ArgumentException("All vectors must have the same dimensionality.");
+            }
+
+            Sorting.Sort(x, y);
+            return InterpolatePchipSorted(x, y);
+        }
+
+        /// <summary>
+        /// Create a piecewise cubic Hermite interpolating polynomial from an unsorted set of (x,y) value pairs.
+        /// Monotone-preserving interpolation with continuous first derivative.
+        /// </summary>
+        public static CubicSpline InterpolatePchip(IEnumerable<double> x, IEnumerable<double> y)
+        {
+            // note: we must make a copy, even if the input was arrays already
+            return InterpolatePchipInplace(x.ToArray(), y.ToArray());
+        }
+
         /// <summary>
         /// Create a cubic spline interpolation from a set of (x,y) value pairs, sorted ascendingly by x,
         /// and custom boundary/termination conditions.
@@ -398,7 +502,7 @@ namespace MathNet.Numerics.Interpolation
             double t1 = xx[index1] - xx[index0];
             double t2 = xx[index2] - xx[index0];
 
-            double a = (x2 - x0 - (t2/t1*(x1 - x0)))/(t2*t2 - t1*t2);
+            double a = (x2 - x0 - (t2/t1*(x1 - x0)))/(t2*(t2 - t1));
             double b = (x1 - x0 - a*t1*t1)/t1;
             return (2*a*t) + b;
         }