The differences are so small and variation so wide it seem impossible to separate the species on the underside alone.
Some species do have a more distinct underside eg P. philota and sumatrae.
The two species Z. ogygia and ogygiodes are impossible to separate in the field.
The males are separated by Z. ogygia having streaks of brand on the FW, but not in Z. ogygioides.
The females are separated ,as Eliot state mainly by guesswork.
It is unfortunate that a large part of the HW is gone, right where most of the ID marks are.
The remaining black spot is the one at the tornal angle, not the one in space 2.
The saving grace is that a bit of the UpH is seen showing cupreous brown.
Salient features.
1. Male with UpH brownish.
2. UpF postdiscal band straight.
P. phama is stated to have the FW main bands straight so that the intervening pale spacing is also fairly straight and broad.
However this is a very variable feature.
For example in Otsuka's Bornean Lycaenidae (available for download) P. phama have narrow crooked bands and P. hewitsoni have straight broad bands, the reverse of what the key states.
The IDs there was based on the upperside.
Furthermore Saito and Seki in a revision in 2006 considered phama and hewitsoni the same species and phama falls as a junior synonym.
This is followed by Yutaka.
The taxa regia and rajata formerly as ssp. of P. phama are now under P. hewitsoni. https://yutaka.it-n.jp/lyc1/80060020.html
I leave this as P. phama.
Perhaps some one will do a molecular analysis.
Try to get an upperside.
It will give some clue of what it really is. PORITWP.png