Comments on"Analysis of permanent magnets as elasmobranch bycatch reduction devices in hook-and-line and longline trials"

A recent study (Fish. Bull. 109:394–401 (2011)) purportedly tests two hypotheses: 1. that the capture of elasmobranchs would be reduced with hooks containing magnets in comparison with control hooks in hook-and-line and longline studies. 2. that the presence of permanent magnets on hooks would not alter teleost capture because teleosts lack the ampullary organ. Review of this paper shows some inconsistencies in the data supporting the first hypothesis and insufficient data and poor experimental design to adequately test the second hypothesis. Further, since several orders of teleosts are known to possess ampullary organs and demonstrate electroreception, grouping all teleosts in a study design or data analysis of magnetic hook catch rates is not warranted. Adequate tests of the hypothesis that permanent magnets or magnetized hooks do not alter teleost capture requires a more careful study design and much larger sample sizes than O'Connell et al. (Fish. Bull. 109:394–401 (2011)). A recent study (O'Connell et al., 2011) purportedly tests two hypotheses. The first hypothesis is that “the capture of elasmobranchs would be reduced with hooks containing magnets in comparison with control hooks in hook-and-line and longline studies.” Both the study design and the data presented seem adequate to support this hypothesis for the species where support is claimed (Rhizoprionodon terraenovae, Mustelus canis, Carcharhinus limbatus, and Dasyatis americana), but there are some inconsistencies in the data and analyses that are presented. First, Table 3 lists the total number of elasmobranchs caught on all treatments in the hook-and-line study as 147, but then lists 119 caught on the control and 57 caught on the magnet treatment, a total of 176. Summing the elasmobranchs in the n column suggests a total of 300 elasmobranchs on all treatments, which would imply a total of 124 elasmobranchs caught on the procedural control (sham, lead weight treatment). The second and more significant issue with the analyses in support of the first hypothesis is the P value used to determine significance in the study. Both Table 2 and Table 3 state, “Asterisks indicate significant (P<0.005) differences between control and magnetic treatments in chi-square analysis.” However, the lines marked with asterisks in Table 2 have P values of P=0.0396 and P=0.0348 showing that these groups were either not significant, or that significance was determined with a P threshold different from that stated in Table 2. Similarly, P=0.0067 is reported for Mustelus canis in Table 3. While this would be considered significant with some common P thresholds for determining significance, it is not consistent with the stated threshold P<0.005. The most objectionable issue with the paper is that neither the study design nor the quantity of data is adequate for testing the second hypothesis that “that the presence of permanent magnets on hooks would not alter teleost capture because teleosts lack the ampullary organ.” In spite of the fact that neither the discussion nor the conclusion contain any consideration of the second hypothesis or the results in teleost species, the abstract makes the unsupported assertion that “Teleosts, such as red drum (Sciaenops ocellatus), Atlantic croaker (Micropogonias undulatus), oyster toadfish (Opsanus tau), black sea bass (Centropristis striata),