diff --git a/README.md b/README.md
index 59aceab4d..4cac34cd0 100644
--- a/README.md
+++ b/README.md
@@ -354,7 +354,7 @@ You can read all about the philosophy behind Lazer [here](http://phaser.io/news/
 * Removed a `console.log` from the TilingSprite generator.
 * Sound.position can no longer become negative, meaning calls to AudioContextNode.start with negative position offsets will no longer throw errors (thanks @Weedshaker #2351 #2368)
 * The default state of the internal property `_boundDispatch` in Phaser.Signal is now `false`, which allows for use of boundDispatches (thanks @alvinlao #2346)
-* The Tiled parser only supports un-compressed layer data. Previously it would silently fail, now it detects if layer compression is used and displays a console warning instead (thanks @MannyC #2413)
+* The Tiled parser only supports uncompressed layer data. Previously it would silently fail, now it detects if layer compression is used and displays a console warning instead (thanks @MannyC #2413)
 * The Tiled parser now removes the `encoding` parameter so that a subsequent process doesn't try to decode the data again (thanks @MannyC #2412)
 * Ensure a parent container is a Group before removing from its hash (thanks @rblopes #2397)
 * The Game Object Input Handler now checks to see if the Object was destroyed during the `onInputDown` phase, and bails out early if so (thanks @zeterain #2394)
@@ -368,7 +368,7 @@ You can read all about the philosophy behind Lazer [here](http://phaser.io/news/
 * Video now uses MediaStreamTrack.stop() instead of MediaStream.stop() where possible, as the later is now deprecated in some browsers (thanks @stoneman1 #2371)
 * The Physics Manager will now throw a console warning if you try to enable a physics body using an unknown physics engine type (thanks @jakewilson #2415)
 * The Tileset class will tell you the name of the tileset image throwing the uneven size error (thanks @jakewilson #2415)
-* Emitter.start when used with a false `explode` parameter would cummulatively add particles to the current total. With quantity 10 the first call would emit 10 particles, the next 20, and so on. Calls to start will now reset the quantity each time. This is a behavior change from earlier versions, so if you relied on the old way please account for it in your code (thanks @BdR76 #2187)
+* Emitter.start when used with a false `explode` parameter would cumulatively add particles to the current total. With quantity 10 the first call would emit 10 particles, the next 20, and so on. Calls to start will now reset the quantity each time. This is a behavior change from earlier versions, so if you relied on the old way please account for it in your code (thanks @BdR76 #2187)
 
 ### Bug Fixes
 
@@ -387,12 +387,13 @@ You can read all about the philosophy behind Lazer [here](http://phaser.io/news/
 * Camera.position would return the view rectangles centerX/Y coordinates, instead of view.x/y (which is what Camera.x/y returns), so it has been updated to return view.x/y instead (thanks @kamparR #2120)
 * Passing a BitmapData to a TileSprite as a texture would fail if the BitmapData had not been previously added to the cache. It now uses the new frameData property (thanks @mzamateo @lucap86 #2380)
 * When setting a global volume for the SoundManager it would previously incorrectly calculate the volumes of AudioTag based Sound objects that were not played at volume 1. The new approach uses Sound.updateGlobalVolume which adjusts the Sound volume to be a percentage of the global volume. So if the global volume is 0.5 and the Sound volume is 0.5, the Sound will play with an actual volume of 0.25 (thanks @VitaZheltyakov #2325)
-* Sound.play when using an AudioTag would ignore the muted state of the SoundManager and play regardless. It now checks the SoundManager.mute state on play, and sets the volume accorindingly (thanks @brianbunch #2139)
+* Sound.play when using an AudioTag would ignore the muted state of the SoundManager and play regardless. It now checks the SoundManager.mute state on play, and sets the volume accordingly (thanks @brianbunch #2139)
 * Graphics objects can now have a Physics Body directly attached to them, where-as before it would throw an error due to a lack of anchor property (thanks @NLilley #2400)
 * A Game Object with `fixedToCamera = true` that was then set for Input, and enabled for dragging from its center (`input.enableDrag(true)`) would throw an error upon being dragged (thanks @solusipse #2367)
 * P2.World.updateBoundsCollisionGroup wouldn't use the `boundsCollisionGroup` mask if you passed `true` as the argument, only if it was left undefined.
 * P2.World.updateBoundsCollisionGroup didn't set the `_boundsOwnGroup` private var, meaning the `World.setBounds` method wasn't able to restore previously set collision masks automatically (thanks @jmp909 #2183)
-* P2.World.setBounds has been re-written completely. If the World is resized it no longer removes the P2 body instances and re-creates them. Instead it checks to see which walls are required and then just moves the position of the shapes instead, or updates them, or creates or destroys them as required. This is far more efficnent, especially in a game which see's a lot of world bounds changes (i.e. resizes responsively in browser)
+* P2.World.setBounds has been re-written completely. If the World is resized it no longer removes the P2 body instances and re-creates them. Instead it checks to see which walls are required and then just moves the position of the shapes instead, or updates them, or creates or destroys them as required. This is far more efficient, especially in a game which sees a lot of world bounds changes (i.e. resizes responsively in browser)
+* BitmapText would throw an error if you passed in a number as the text property to the constructor. It worked if you used the text accessor directly because it cast the value to a string, but the constructor missed out this step (thanks @lewispollard #2429)
 
 ### Pixi Updates