C305 in the rev 1 A1200 is a 220nF (0.22uF) SMD ceramic. Don't have one in front of me, but the package size is probably either 1206 or 1210. Its job is AC decoupling (meaning it's a short circuit for noise) of the +2.5V bias supply required for the op-amp U15.
Out of interest, why are you fitting a socket? Handy for testing spare parts, else it generally just introduces an element of unreliability.
A helpful tip for the capacitor list mentioned above is that I'd strongly suggest not using radial capacitors in place of SMD ones. Electrically they'll work fine, but mechanically it's not great as PCB pads can be easily damaged when the capacitor gets knocked. You can easily get good quality modern SMD electrolytics with equally as good temperature/lifetime ratings, plus the final job is tidy and robust.
And the myth about having to use non-polarised capacitors for AC coupling/DC blocking in audio stages is an incredibly annoying old housewife's tale. The value of the capacitor is selected so that the AC cycle time constant will never charge/discharge the capacitor. It remains charged with the DC bias applied to the surrounding stage and the AC signal passing through effectively sees the capacitor as a short circuit or another piece of wire. If the capacitor was somehow magically charging/discharging in each half cycle, then the AC component wouldn't be passed at all and the circuit wouldn't work. For further detail, read one of the many websites explaining RC Time Constants.
The reason the capacitors leak is because the liquid electrolyte becomes unstable over time and with temperature.
