A series was here. A series were here. A thorough reanalysis of Brown et al. (2005) %u201CDance reveals symmetry especially in young men%u201D shows that all of the major results appear to be based on hidden procedures designed to produce the results later derived. These procedures include the pre-selection of animations of Jamaicans dancing, apparently based on preliminary evaluation in New Jersey, so as to exclude symmetrical individuals who danced poorly and asymmetrical ones who danced well (N = 10 out of 10, P < 0.001). There are also systematic biases in averaging dance evaluations so as to produce significant results where none exist and more highly significant ones than do, in fact, exist. This appears primarily to have been achieved by reducing the variance in within-group dance evaluations thus making between- group comparisons more significant. How this reduction was achieved is obscure to us, as is the source of other biases in the data analysis, but all show the common pattern of making the evidence appear to be more striking than it really is. Using the same fluctuating asymmetry (FA) values used in Brown et al. one set of correlations is confirmed nearly exactly, namely, the sex difference in importance placed on symmetry in dance evaluations. This was a between-evaluator analysis that relied on the same grid of values used in the other analyses. This makes it all the stranger that the two sets of average dance evaluations do not match up. In addition, the significant negative correlation between male fluctuating asymmetry and preference for the animations of relatively symmetrical females also disappears (even though this is also a between-evaluator analysis). #13; #13;After conducting these analyses we were astonished to discover an additional, major source of bias in Brown et al. Values of FA were modified for 65 out of 80 cases of the dancers chosen (1996 and 2002 FAs combined) so as to place good dancers in the symmetrical category and poor dancers in the asymmetrical one. Meanwhile values for individuals not selected as dancers remain (with one exception) unchanged. Since the incorrect values were used in the between-evaluator comparisons of males and females, there is no way now to confirm these findings. The probability that all of these biases could have resulted from chance is well less than 1 in 10,000,000,000. Thus Brown et al.%u2019s results appear to be entirely artificially constructed%u2014that is, fraudulent. An analysis of the full data of (~2) Rutgers University evaluators, an unbiased set of data, reveals at best a weak positive relationship between symmetry and dancing ability, with no sex difference, using 2002 FA values only. #13; #13;Finally we turn to an analysis of some of the factors that may have contributed to the fraud, especially decisions taken by Dr Trivers. We discuss the role that Nature played and we summarize the reactions of the other co-authors of Brown et al. (2005) to the discoveries described in this book. More generally, we mention some of the factors that reduce the chance that fraud will be discovered or (if discovered) revealed. We mention some recent cases and end with a summary of Fisher%u2019s famous reanalysis of Mendel%u2019s genetics work. #13; #13;#13;