Gyro's measure angular rotation. Unlike accelerometers a gyro's output is linear to the rate of rotation measured in degrees per second. Each Gyro channel measures rotation about one axis. It is common to have single axis and dual axis gyro's which measure rotation around the x and y axis (i.e. roll and pitch). It is also common to have separate single axis gyros which measure rotation around the y axis (yaw). It is less common to have a 3 axis gyro, although they are becomming more available. Gyro's in general are more expensive than accelerometers.
Gyro reading conversion
You do not normally use the values from a gyro directly. First, they must be converted to a value in degrees per second. Depending on the gyro used, the datasheet will show what voltage output the device will give for a specific rotation
For example, the IDG500 datasheet tells us that 500 degrees per second gives an output of 1V.
Assuming the microcontroller you are using has a 10 bit ADC and is powered by 5V (or ADV ref voltage is 5V) then we can see that 1V will give an ADC reading of 1023/5 = 204, so the value for degree's per second = ADC * 500 / 204
Therefore to use a value for degrees per second in our calculation we use ADC * 500 / 204
This only applies to the example above, substitute your own values for microcontroller voltage and Gyro used
Note that like an accelerometer, a gyro normally outputs a voltage that is half its supply voltage when not rotating. So what i term the ADC value is actually the ADC value minus this baseline voltage.