diff --git a/diffrax/__init__.py b/diffrax/__init__.py
index d8aca8d4..7e33c4bf 100644
--- a/diffrax/__init__.py
+++ b/diffrax/__init__.py
@@ -71,6 +71,7 @@
     Midpoint,
     MultiButcherTableau,
     Ralston,
+    Rattle,
     ReversibleHeun,
     SemiImplicitEuler,
     Sil3,
diff --git a/diffrax/solver/__init__.py b/diffrax/solver/__init__.py
index ace213c4..77652d35 100644
--- a/diffrax/solver/__init__.py
+++ b/diffrax/solver/__init__.py
@@ -24,6 +24,7 @@
 from .midpoint import Midpoint
 from .milstein import ItoMilstein, StratonovichMilstein
 from .ralston import Ralston
+from .rattle import Rattle
 from .reversible_heun import ReversibleHeun
 from .runge_kutta import (
     AbstractDIRK,
diff --git a/diffrax/solver/rattle.py b/diffrax/solver/rattle.py
new file mode 100644
index 00000000..5a1e484a
--- /dev/null
+++ b/diffrax/solver/rattle.py
@@ -0,0 +1,146 @@
+from typing import Callable, NamedTuple, Tuple
+
+import jax
+import jax.numpy as jnp
+import jax.tree_util as jtu
+from equinox.internal import ω
+
+from ..custom_types import Array, Bool, DenseInfo, PyTree, Scalar
+from ..local_interpolation import LocalLinearInterpolation
+from ..solution import RESULTS
+from ..term import AbstractTerm
+from .base import AbstractImplicitSolver
+
+
+_ErrorEstimate = None
+_SolverState = None
+
+ConstrainFn = Callable[[PyTree], Array]
+
+
+class RattleVars(NamedTuple):
+    p_1_2: PyTree  # Midpoint momentum
+    q_1: PyTree  # Midpoint position
+    p_1: PyTree  # final momentum
+    lam: PyTree  # Midpoint Lagrange multiplier (state)
+    mu: PyTree  # final Lagrange multiplier (momentum)
+
+
+class Rattle(AbstractImplicitSolver):
+    """Rattle method.
+
+    2nd order symplectic method with constrains `constrain(x)=0`.
+
+    ??? cite "Reference"
+
+        ```bibtex
+        @article{ANDERSEN198324,
+            title = {Rattle: A “velocity” version of the shake
+            algorithm for molecular dynamics calculations},
+            journal = {Journal of Computational Physics},
+            volume = {52},
+            number = {1},
+            pages = {24-34},
+            year = {1983},
+            author = {Hans C Andersen},
+        }
+        ```
+    """
+
+    term_structure = (AbstractTerm, AbstractTerm)
+    interpolation_cls = LocalLinearInterpolation
+    # Fix TypeError: non-default argument 'constrain' follows default argument
+    constrain: ConstrainFn = None
+
+    def order(self, terms):
+        return 2
+
+    def init(
+        self,
+        terms: Tuple[AbstractTerm, AbstractTerm],
+        t0: Scalar,
+        t1: Scalar,
+        y0: PyTree,
+        args: PyTree,
+    ) -> _SolverState:
+        return None
+
+    def step(
+        self,
+        terms: Tuple[AbstractTerm, AbstractTerm],
+        t0: Scalar,
+        t1: Scalar,
+        y0: Tuple[PyTree, PyTree],
+        args: PyTree,
+        solver_state: _SolverState,
+        made_jump: Bool,
+    ) -> Tuple[Tuple[PyTree, PyTree], _ErrorEstimate, DenseInfo, _SolverState, RESULTS]:
+        del solver_state, made_jump
+
+        term_1, term_2 = terms
+        midpoint = (t1 + t0) / 2
+
+        control1_half_1 = term_1.contr(t0, midpoint)
+        control1_half_2 = term_1.contr(midpoint, t1)
+
+        control2_half_1 = term_2.contr(t0, midpoint)
+        control2_half_2 = term_2.contr(midpoint, t1)
+
+        p0, q0 = y0
+
+        def eq(x: RattleVars, args=None):
+            _, vjp_fun = jax.vjp(self.constrain, q0)
+            _, vjp_fun_mu = jax.vjp(self.constrain, x.q_1)
+
+            zero = (
+                (
+                    p0**ω
+                    - control1_half_1 * (vjp_fun(x.lam)[0]) ** ω
+                    + term_1.vf_prod(t0, q0, args, control1_half_1) ** ω
+                    - x.p_1_2**ω
+                ).ω,
+                (
+                    q0**ω
+                    + term_2.vf_prod(t0, x.p_1_2, args, control2_half_1) ** ω
+                    + term_2.vf_prod(midpoint, x.p_1_2, args, control2_half_2) ** ω
+                    - x.q_1**ω
+                ).ω,
+                self.constrain(x.q_1),
+                (
+                    x.p_1_2**ω
+                    + term_1.vf_prod(midpoint, x.q_1, args, control1_half_2) ** ω
+                    - (control1_half_2 * vjp_fun_mu(x.mu)[0] ** ω)
+                    - x.p_1**ω
+                ).ω,
+                jax.jvp(self.constrain, (x.q_1,), (term_2.vf(t1, x.p_1, args),))[1],
+            )
+            return zero
+
+        cs = jax.eval_shape(self.constrain, q0)
+
+        init_vars = RattleVars(
+            p_1_2=p0,
+            q_1=(q0**ω * 2).ω,
+            p_1=p0,
+            lam=jtu.tree_map(jnp.zeros_like, cs),
+            mu=jtu.tree_map(jnp.zeros_like, cs),
+        )
+
+        sol = self.nonlinear_solver(eq, init_vars, None)
+
+        y1 = (sol.root.p_1, sol.root.q_1)
+        dense_info = dict(y0=y0, y1=y1)
+        return y1, None, dense_info, None, RESULTS.successful
+
+    def func(
+        self,
+        terms: Tuple[AbstractTerm, AbstractTerm],
+        t0: Scalar,
+        y0: Tuple[PyTree, PyTree],
+        args: PyTree,
+    ) -> Tuple[PyTree, PyTree]:
+        term_1, term_2 = terms
+        y0_1, y0_2 = y0
+        f1 = term_1.func(t0, y0_2, args)
+        f2 = term_2.func(t0, y0_1, args)
+        return (f1, f2)
diff --git a/test/test_solver.py b/test/test_solver.py
index 67ed0701..ff51c375 100644
--- a/test/test_solver.py
+++ b/test/test_solver.py
@@ -470,3 +470,41 @@ def vector_field(t, y, args):
         return out.ys[0]
 
     f(1.0)
+
+
+def test_rattle():
+    import numpy as np
+
+    def constrain(q):
+        return jnp.sqrt(jnp.sum(q**2, keepdims=True)) - 1.0
+
+    rat = diffrax.Rattle(
+        nonlinear_solver=diffrax.NewtonNonlinearSolver(rtol=1e-4, atol=1e-6),
+        constrain=constrain,
+    )
+
+    # Potential free movement on a circle
+    def H(p, q):
+        del q
+        return p @ p.T / 2.0
+
+    # V = p^2/2m m=1, v=1
+
+    terms = (
+        diffrax.ODETerm(
+            lambda t, q, args: -jax.grad(H, argnums=1)(jnp.zeros_like(q), q)
+        ),
+        diffrax.ODETerm(
+            lambda t, p, args: jax.grad(H, argnums=0)(p, jnp.zeros_like(p))
+        ),
+    )
+    # p,q
+    y0 = (jnp.asarray([1.0, 0.0]), jnp.asarray([0.0, 1.0]))
+    t1 = 2 * jnp.pi / 4
+    n = 2**12
+    dt = t1 / n
+    saveat = diffrax.SaveAt(t1=True)
+    solution = diffrax.diffeqsolve(terms, rat, 0.0, t1, dt0=dt, y0=y0, saveat=saveat)
+    p1, q1 = solution.ys
+    assert np.allclose(p1, jnp.asarray([0.0, -1.0]), rtol=1e-4, atol=1e-4)
+    assert np.allclose(q1, jnp.asarray([1.0, 0.0]), rtol=1e-4, atol=1e-4)