Referee 2's review:

Generally, this is an interesting paper which deserves to be published. There are, however two basic problems, which have to be solved (however, it should not be a problem, to address and solve them).

The methodical problem:
1) The authors do not provide any data on age and sex of the involved bats. They explain the case, when animals came back even despite being magnetically treated, by a possibility that they were experienced animals which were "familiar with the release site and simply flew home, ignoring conflicting compass information." However, the problem is that the same explanation can be applied also to those bats which did not find the way home - simply they were not experienced! Bats are long-living animals, and males have larger home-ranges than females. Therefore we can assume that older bats are more experienced than younger ones and males are better at homing than females. Unless sex and age are provided for each group and their effects are analyzed, the results are questionable.

2) The second problem is more an ethical one.
I do not understand why the authors decided to ignore relevant sources - which they must know - They "sell" the used method as their invention and imply that in this case it was used for the first time to solve the given question.
"This is the first direct evidence that magnetite is used to detect the polarity of the magnetic field for compass use by an animal."

This is not true.

Let me cite a relevant paragraph from the recent review by Nemec et al. Naturwiss. 2005, 92: 151-157: "Experiments utilizing disruptive tools specifically address either the magnetite-based or photoreceptor-based magneto-sensory systems. A short, strong magnetic pulse, a treatment designed to alter the magnetisation of magnetite, has a conspicuous effect on the magnetic orientation behaviour of birds (Wiltschko et al. 2002a and citations therein) and mole-rats (Marhold et al. 1997a). Such a procedure might alter or destroy magnetite-based magnetoreceptors (Lohmann and Johnsen 2000)...." By the way, the cited paper Wiltschko et al. 2002a (Wiltschko W, Munro U, Wiltschko R, Kirschvink JL (2002): Magnetite-based magnetoreception in birds: the effect of a biasing field and a pulse on migratory behaviour. J Exp Biol 205:3031-3037) is a paper co-authored also by one of the co-authors of the present manuscript. In spite of this fact and in spite that it reads: "... effect of a .... pulse ..." it is not cited in the present paper. The - already 10 year old - paper by Marhold et al. refers to mole-rats - it means also to mammals, also living in darkness, in this study also treated by magnetic pulse.

Generally, the choice of the quoted literature is rather non-representative and biased. Thus for instance, it is written (lines 167-169): "In birds and in fish they are located in the nasal region [23,35] and innervated by the trigeminal nerve [36,37], which is present in essentially all vertebrate groups." Here it would be appropriate to cite studies by Nemec et al. (Science 2001) and Wegner et al. (JEB 2006) showing that in mole-rats (mammals!) trigeminal nerve is also involved in magnetoreception.

Minor point:
Apart from being vertebrates, mammals and birds are phylogenetically not related at all. Hence it is not surprising that they may have evolved quite different mechanisms of magnetoreception (or retained a different component from an original duplex system). Also the authors should talk of turtles (referring to the study of Lohman) rather than of "reptiles". "Reptilia" is a polyphyletic artificial group, which has been rejected by taxonomists already decades ago. Turtles are the only representative of traditional "Reptilia" which have been studied from the point of view of magnetoreception. They are not related to birds and also not related to mammals. Magnetoreception in other reptiles (like snakes, lizards, crocodiles) has not been studied.

**********
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.