diff --git a/Rethinking_2/End_of_chapter_problems/Chapter_4.ipynb b/Rethinking_2/End_of_chapter_problems/Chapter_4.ipynb
index 35ae1f53..a97d3d07 100644
--- a/Rethinking_2/End_of_chapter_problems/Chapter_4.ipynb
+++ b/Rethinking_2/End_of_chapter_problems/Chapter_4.ipynb
@@ -157,7 +157,10 @@
   },
   {
    "cell_type": "markdown",
-   "metadata": {},
+   "metadata": {
+    "jp-MarkdownHeadingCollapsed": true,
+    "tags": []
+   },
    "source": [
     "### 4M2.\n",
     "Translate the model just above into a quap formula.\n",
@@ -188,7 +191,7 @@
     "\n",
     "$\\beta$ ~ $logN(0,1)$ #It seems perfectly sensible that the amount an average student's height changes per year should only be positive. Additionally, I have no idea what the growth rate is so this enocodes ignorance about the rate of change of height per year\n",
     "\n",
-    "$\\sigma$ ~ $Unif(0,50)$"
+    "$\\sigma$ ~ $Unif(0,0.5)$ #This is the standard deviation of height, at a given age and expected height. This cannot conceivably be more than 1 metre. Half is a more sensible upper bound."
    ]
   },
   {
@@ -215,7 +218,9 @@
     "more than 64cm. How does this lead you to revise your priors?\n",
     "\n",
     "#### Answer \n",
-    "This would cause me to use an even larger range of plausable values in my prior for $\\sigma$. Given my assumption for the students ages, increasing the range seems to be not sensible for the data at hand.\n",
+    "A maximum variance ($\\sigma^2$) can be enforced through an upper bound on the prior for $\\sigma$, set to $\\sqrt(64) = 8$ cm, or 0.08 m. Therefore the prior becomes\n",
+    "\n",
+    "$\\sigma$ ~ $Unif(0,0.08)$\n",
     "\n",
     "### 4M7.\n",
     "\n",
@@ -1621,9 +1626,7 @@
   {
    "cell_type": "code",
    "execution_count": 43,
-   "metadata": {
-    "scrolled": false
-   },
+   "metadata": {},
    "outputs": [
     {
      "name": "stderr",
@@ -1714,7 +1717,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.8.5"
+   "version": "3.10.6"
   }
  },
  "nbformat": 4,