First make sure that the galvanometer (aka meter movement) is functional! I test these by connecting a weak camera battery and seeing if there is any movement. 0.5 volts @ very low current, in the microamp range, ought to suffice, but I just use an old LR44 cell.
Fortunately, the gavanometer in my camera tested fine, but the selenium solar cell was dead. I purchased a number of small bare silicon solar cells and cut them to size in this manner:
The cells are extremely fragile, and I needed several tries before I got something usable, and it was not so neat-looking, but it worked! Only maybe 1/4 the size of the original, but this is also fine, because crystalline solar cells are orders of magnitude more efficient than legacy selenium cells. Maximum voltage for either is about the same: 0.5 volts. But silicon produces a lot more amperage (which is no problem at all in this application).
It would have been much easier if I had used some sort of flexible photovoltaic cell, some of which can be cut with an ordinary pair of scissors. But I like old fashioned crystalline cells because they don't degrade much over time.