Referee 2's review:

The manuscript has improved. Thank you for considering my previous comments.

I have a few suggestions for minor revisions:

(1) It is still not clear to the reader where exactly you applied the paint in your first set of experiments. Please indicate paint locations in Fig. 2a.
(2) You also do not discuss the need for animals to face each other during the test, if facial features are to be crucial in the recognition process. My previous comment 1c did not address the assessment during the fight, but during your test procedures.
(3) P9, pg1, line 1: It is not true that "tethered fiddler crabs use natural and artificial color patches ...". Detto et al used tethered females to show that resident male crabs respond differently to neighbouring females when their appearance was altered by modifying their natural colouration; they also showed that females prefer to approach tethered males who's major claw carries the species-specific colour, independent of the male's species identity.
(4) P9, pg3, line 4: With reference to Detto et al. it is also not true that "addition of artificial colour patches ...permit crustaceans to identify individuals". The implication of that study was that fiddler crabs identify neighbours by their individually distinct natural carapace colour patterns.
(5) P10, pg3, l1: Please indicate paint patch placements in Fig. 2a.
(6) P19, pg 1, l1ff: I understand the reason why you randomly varied the pen containing the familiar crayfish to control for positional effects in the arena. But the reason you give here, namely "to eliminate the possibility of positional learning" is plainly wrong, because you tested each crayfish only once, as you state on page 17, 2nd paragraph.
(7) P20, pg2ff: Several statements in this paragraph do not make sense (at least to this reader): "Our experiments were designed to relate the test conditions to some of the range of conditions that C destructor operate in the wild." There is both an English language problem with this sentence and one with the content it tries to deliver. "Based on ... the light corresponds...crayfish could view only one of the choice animal's appearance in detail..." Ditto. "Also if crayfish viewed the choice animal from angles other than that taken for the analysis..." This is true, but it is especially important for the "facial width" measurements you have taken that animals face each other during the recognition process.
(8) Ref 28 & 46: Hippolytidae

**********
N.B. These are the comments made by the referee when reviewing an earlier version of this paper. Prior to publication the manuscript has been revised in light of these comments and to address other editorial requirements.